Tests for a future article.
Quantum ESPRESSO Quantum ESPRESSO is an integrated suite of Open-Source computer codes for electronic-structure calculations and materials modeling at the nanoscale. It is based on density-functional theory, plane waves, and pseudopotentials. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Quantum ESPRESSO 6.7 Input: AUSURF112 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 SE +/- 11.50, N = 3 SE +/- 0.33, N = 3 SE +/- 1.52, N = 3 SE +/- 1.64, N = 3 SE +/- 22.87, N = 9 SE +/- 16.58, N = 4 SE +/- 3.48, N = 3 SE +/- 6.34, N = 3 SE +/- 5.50, N = 3 SE +/- 1.29, N = 3 SE +/- 5.19, N = 3 SE +/- 4.40, N = 3 SE +/- 1.18, N = 3 SE +/- 0.83, N = 3 1357.30 1356.26 1208.31 1216.50 1372.73 1329.90 1317.27 1403.87 1386.78 1342.14 1403.52 1456.82 1520.43 1656.88 1. (F9X) gfortran options: -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Quantum ESPRESSO 6.7 Input: AUSURF112 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 Min: 1338.99 / Avg: 1357.3 / Max: 1378.5 Min: 1355.84 / Avg: 1356.26 / Max: 1356.92 Min: 1205.82 / Avg: 1208.31 / Max: 1211.05 Min: 1214.11 / Avg: 1216.5 / Max: 1219.63 Min: 1251.31 / Avg: 1372.73 / Max: 1438.64 Min: 1306.89 / Avg: 1329.9 / Max: 1377.55 Min: 1313.6 / Avg: 1317.27 / Max: 1324.22 Min: 1396.5 / Avg: 1403.87 / Max: 1416.5 Min: 1377.07 / Avg: 1386.78 / Max: 1396.11 Min: 1339.87 / Avg: 1342.14 / Max: 1344.32 Min: 1394.1 / Avg: 1403.52 / Max: 1411.99 Min: 1450.92 / Avg: 1456.82 / Max: 1465.43 Min: 1518.1 / Avg: 1520.43 / Max: 1521.97 Min: 1655.62 / Avg: 1656.88 / Max: 1658.44 1. (F9X) gfortran options: -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
NWChem NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better NWChem 7.0.2 Input: C240 Buckyball EPYC 7F52 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 2K 4K 6K 8K 10K 5684.6 2247.0 2220.7 2716.7 3653.9 3653.6 3678.8 4709.5 6676.4 7056.0 8844.1 1. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lcomex -lm -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2
LeelaChessZero LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: Eigen EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 600 1200 1800 2400 3000 SE +/- 17.43, N = 5 SE +/- 16.37, N = 9 SE +/- 26.03, N = 3 SE +/- 29.49, N = 9 SE +/- 37.70, N = 9 SE +/- 31.31, N = 9 SE +/- 9.84, N = 3 SE +/- 20.90, N = 3 SE +/- 14.82, N = 9 SE +/- 5.90, N = 3 SE +/- 12.27, N = 9 SE +/- 11.49, N = 9 SE +/- 11.79, N = 9 SE +/- 8.88, N = 3 1699 1104 2686 2408 2311 1927 1617 1769 1539 1439 1233 1051 925 681 1. (CXX) g++ options: -flto -pthread
Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better LeelaChessZero 0.26 Backend: Eigen EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 8.39 8.98 15.06 14.75 13.41 12.55 10.30 11.26 11.09 10.56 10.97 11.06 10.05 8.61
Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: Eigen EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 Min: 1639 / Avg: 1699.2 / Max: 1747 Min: 1024 / Avg: 1103.67 / Max: 1176 Min: 2634 / Avg: 2685.67 / Max: 2717 Min: 2310 / Avg: 2408 / Max: 2551 Min: 2112 / Avg: 2311.33 / Max: 2527 Min: 1773 / Avg: 1926.89 / Max: 2050 Min: 1597 / Avg: 1616.67 / Max: 1627 Min: 1729 / Avg: 1769.33 / Max: 1799 Min: 1454 / Avg: 1538.78 / Max: 1593 Min: 1432 / Avg: 1439.33 / Max: 1451 Min: 1194 / Avg: 1232.78 / Max: 1292 Min: 1008 / Avg: 1051.22 / Max: 1107 Min: 870 / Avg: 924.56 / Max: 993 Min: 663 / Avg: 680.67 / Max: 691 1. (CXX) g++ options: -flto -pthread
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: 20k Atoms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.053, N = 3 SE +/- 0.004, N = 3 SE +/- 0.024, N = 3 SE +/- 0.056, N = 3 SE +/- 0.029, N = 3 SE +/- 0.112, N = 3 SE +/- 0.062, N = 3 SE +/- 0.016, N = 3 SE +/- 0.059, N = 3 SE +/- 0.010, N = 3 SE +/- 0.009, N = 3 SE +/- 0.021, N = 3 SE +/- 0.023, N = 3 SE +/- 0.007, N = 3 11.757 6.705 24.818 25.206 22.442 22.046 18.156 17.525 17.614 14.907 10.602 9.889 7.737 5.406 1. (CXX) g++ options: -O3 -pthread -lm
ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: 20k Atoms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0315 0.063 0.0945 0.126 0.1575 0.06 0.06 0.13 0.14 0.11 0.13 0.11 0.10 0.12 0.10 0.09 0.10 0.08 0.07
Result Confidence
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: 20k Atoms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 11.66 / Avg: 11.76 / Max: 11.85 Min: 6.7 / Avg: 6.7 / Max: 6.71 Min: 24.78 / Avg: 24.82 / Max: 24.86 Min: 25.11 / Avg: 25.21 / Max: 25.31 Min: 22.4 / Avg: 22.44 / Max: 22.5 Min: 21.83 / Avg: 22.05 / Max: 22.19 Min: 18.03 / Avg: 18.16 / Max: 18.22 Min: 17.5 / Avg: 17.53 / Max: 17.55 Min: 17.51 / Avg: 17.61 / Max: 17.71 Min: 14.89 / Avg: 14.91 / Max: 14.93 Min: 10.58 / Avg: 10.6 / Max: 10.62 Min: 9.85 / Avg: 9.89 / Max: 9.92 Min: 7.71 / Avg: 7.74 / Max: 7.78 Min: 5.39 / Avg: 5.41 / Max: 5.41 1. (CXX) g++ options: -O3 -pthread -lm
LeelaChessZero LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: BLAS EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 600 1200 1800 2400 3000 SE +/- 31.35, N = 9 SE +/- 22.57, N = 9 SE +/- 14.38, N = 3 SE +/- 41.44, N = 9 SE +/- 26.91, N = 3 SE +/- 27.18, N = 3 SE +/- 20.09, N = 3 SE +/- 22.27, N = 9 SE +/- 23.71, N = 9 SE +/- 20.26, N = 3 SE +/- 13.76, N = 9 SE +/- 3.71, N = 3 SE +/- 7.54, N = 3 SE +/- 4.36, N = 3 1758 1052 2699 2376 2203 1969 1666 1735 1559 1521 1253 1042 946 747 1. (CXX) g++ options: -flto -pthread
Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better LeelaChessZero 0.26 Backend: BLAS EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 8.60 8.98 14.95 14.58 12.86 12.82 10.83 11.33 11.37 11.26 11.29 11.09 10.33 9.44
Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: BLAS EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 Min: 1652 / Avg: 1758 / Max: 1943 Min: 958 / Avg: 1052.44 / Max: 1142 Min: 2683 / Avg: 2699.33 / Max: 2728 Min: 2146 / Avg: 2376 / Max: 2558 Min: 2155 / Avg: 2203.33 / Max: 2248 Min: 1916 / Avg: 1969 / Max: 2006 Min: 1626 / Avg: 1665.67 / Max: 1691 Min: 1624 / Avg: 1734.78 / Max: 1815 Min: 1445 / Avg: 1559.33 / Max: 1649 Min: 1492 / Avg: 1521 / Max: 1560 Min: 1190 / Avg: 1253.33 / Max: 1313 Min: 1035 / Avg: 1042.33 / Max: 1047 Min: 931 / Avg: 945.67 / Max: 956 Min: 739 / Avg: 747 / Max: 754 1. (CXX) g++ options: -flto -pthread
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 100, Lossless Compression EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 SE +/- 0.50, N = 3 SE +/- 0.43, N = 3 SE +/- 2.26, N = 3 SE +/- 0.41, N = 3 SE +/- 0.22, N = 3 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.13, N = 3 SE +/- 0.20, N = 3 SE +/- 0.25, N = 3 SE +/- 1.69, N = 3 SE +/- 0.30, N = 3 SE +/- 0.80, N = 3 SE +/- 0.56, N = 3 475.39 778.66 481.95 474.55 473.76 481.50 467.51 474.59 467.43 467.70 562.07 580.07 680.90 954.80 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 100, Lossless Compression EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 474.62 / Avg: 475.39 / Max: 476.33 Min: 777.8 / Avg: 778.66 / Max: 779.18 Min: 477.74 / Avg: 481.95 / Max: 485.47 Min: 473.79 / Avg: 474.55 / Max: 475.18 Min: 473.32 / Avg: 473.76 / Max: 474.03 Min: 481.32 / Avg: 481.5 / Max: 481.6 Min: 467.15 / Avg: 467.51 / Max: 467.82 Min: 474.43 / Avg: 474.59 / Max: 474.85 Min: 467.12 / Avg: 467.43 / Max: 467.79 Min: 467.23 / Avg: 467.7 / Max: 468.09 Min: 558.73 / Avg: 562.07 / Max: 564.17 Min: 579.69 / Avg: 580.07 / Max: 580.65 Min: 679.66 / Avg: 680.9 / Max: 682.4 Min: 953.7 / Avg: 954.8 / Max: 955.56 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
Caffe This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 200 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90K 180K 270K 360K 450K SE +/- 1699.46, N = 3 SE +/- 95.91, N = 3 SE +/- 730.51, N = 3 SE +/- 1226.45, N = 3 SE +/- 10886.13, N = 9 SE +/- 18239.03, N = 9 SE +/- 147.16, N = 3 SE +/- 422.64, N = 3 SE +/- 440.86, N = 3 SE +/- 338.07, N = 3 SE +/- 60.14, N = 3 SE +/- 456.29, N = 3 SE +/- 226.68, N = 3 SE +/- 533.40, N = 3 382020 340168 347484 345626 429875 399610 411681 441978 427967 391015 387173 388699 378367 369027 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
Result Confidence
OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 200 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80K 160K 240K 320K 400K Min: 379249 / Avg: 382019.67 / Max: 385110 Min: 339980 / Avg: 340168 / Max: 340295 Min: 346044 / Avg: 347483.67 / Max: 348419 Min: 344068 / Avg: 345626.33 / Max: 348046 Min: 364614 / Avg: 429875.11 / Max: 446994 Min: 331340 / Avg: 399610.44 / Max: 446964 Min: 411435 / Avg: 411681.33 / Max: 411944 Min: 441242 / Avg: 441977.67 / Max: 442706 Min: 427345 / Avg: 427966.67 / Max: 428819 Min: 390425 / Avg: 391015.33 / Max: 391596 Min: 387058 / Avg: 387173 / Max: 387261 Min: 387821 / Avg: 388699.33 / Max: 389353 Min: 378044 / Avg: 378367 / Max: 378804 Min: 367965 / Avg: 369027.33 / Max: 369643 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K SE +/- 146.33, N = 3 SE +/- 233.09, N = 3 SE +/- 52.30, N = 3 SE +/- 32.36, N = 3 SE +/- 142.34, N = 3 SE +/- 159.60, N = 7 SE +/- 37.78, N = 3 SE +/- 135.81, N = 9 SE +/- 33.50, N = 3 SE +/- 31.25, N = 3 SE +/- 28.18, N = 3 SE +/- 49.84, N = 3 SE +/- 167.66, N = 3 SE +/- 181.85, N = 3 19037 19036 17793 17681 17609 17090 18047 17308 17664 17644 17600 17559 16820 14713 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Mflops Per Watt
OpenBenchmarking.org Mflops Per Watt, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90 180 270 360 450 250.92 332.96 241.00 246.91 257.05 277.86 350.90 239.28 339.99 344.76 337.27 395.06 387.92 338.94
Result Confidence
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 18745 / Avg: 19037 / Max: 19200 Min: 18580 / Avg: 19036.33 / Max: 19347 Min: 17693 / Avg: 17792.67 / Max: 17870 Min: 17626 / Avg: 17680.67 / Max: 17738 Min: 17324 / Avg: 17608.67 / Max: 17754 Min: 16499 / Avg: 17090.14 / Max: 17664 Min: 18004 / Avg: 18046.67 / Max: 18122 Min: 16389 / Avg: 17308.33 / Max: 17615 Min: 17611 / Avg: 17664 / Max: 17726 Min: 17597 / Avg: 17643.67 / Max: 17703 Min: 17559 / Avg: 17600 / Max: 17654 Min: 17469 / Avg: 17559.33 / Max: 17641 Min: 16499 / Avg: 16820.33 / Max: 17064 Min: 14527 / Avg: 14713.33 / Max: 15077 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Timed LLVM Compilation This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 10.0 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160 320 480 640 800 SE +/- 2.67, N = 3 SE +/- 0.89, N = 3 SE +/- 1.73, N = 3 SE +/- 0.89, N = 3 SE +/- 1.29, N = 3 SE +/- 1.44, N = 3 SE +/- 1.12, N = 3 SE +/- 1.56, N = 3 SE +/- 2.29, N = 3 SE +/- 3.16, N = 9 SE +/- 5.04, N = 4 SE +/- 3.68, N = 3 SE +/- 0.47, N = 3 SE +/- 2.05, N = 3 343.39 581.30 234.21 233.51 243.92 245.40 275.35 280.78 289.36 312.70 404.33 440.00 529.62 763.56
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 10.0 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 130 260 390 520 650 Min: 340.62 / Avg: 343.39 / Max: 348.73 Min: 579.68 / Avg: 581.3 / Max: 582.76 Min: 230.9 / Avg: 234.2 / Max: 236.76 Min: 232.12 / Avg: 233.51 / Max: 235.17 Min: 241.53 / Avg: 243.91 / Max: 245.95 Min: 242.68 / Avg: 245.4 / Max: 247.61 Min: 273.43 / Avg: 275.35 / Max: 277.3 Min: 278.05 / Avg: 280.78 / Max: 283.44 Min: 286.12 / Avg: 289.36 / Max: 293.79 Min: 297.71 / Avg: 312.7 / Max: 321.97 Min: 391.86 / Avg: 404.33 / Max: 414.08 Min: 432.97 / Avg: 440 / Max: 445.42 Min: 528.7 / Avg: 529.62 / Max: 530.26 Min: 759.57 / Avg: 763.56 / Max: 766.35
OpenFOAM OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 8 Input: Motorbike 60M EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 120 240 360 480 600 SE +/- 1.71, N = 3 SE +/- 0.36, N = 3 SE +/- 0.04, N = 3 SE +/- 0.24, N = 3 SE +/- 0.31, N = 3 SE +/- 0.13, N = 3 SE +/- 0.59, N = 3 SE +/- 0.57, N = 3 SE +/- 0.44, N = 3 SE +/- 0.21, N = 3 SE +/- 0.15, N = 3 SE +/- 0.84, N = 3 SE +/- 1.13, N = 3 SE +/- 0.29, N = 3 454.84 362.74 233.34 232.98 233.73 260.78 319.87 236.99 320.20 314.06 313.64 520.45 530.10 573.04 1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -ldecompose -lgenericPatchFields -lmetisDecomp -lscotchDecomp -llagrangian -lregionModels -lOpenFOAM -ldl -lm
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 8 Input: Motorbike 60M EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 Min: 452.49 / Avg: 454.84 / Max: 458.16 Min: 362.09 / Avg: 362.74 / Max: 363.35 Min: 233.29 / Avg: 233.34 / Max: 233.42 Min: 232.55 / Avg: 232.98 / Max: 233.38 Min: 233.28 / Avg: 233.73 / Max: 234.33 Min: 260.58 / Avg: 260.78 / Max: 261.02 Min: 319.1 / Avg: 319.87 / Max: 321.04 Min: 235.98 / Avg: 236.99 / Max: 237.94 Min: 319.73 / Avg: 320.2 / Max: 321.08 Min: 313.84 / Avg: 314.06 / Max: 314.48 Min: 313.39 / Avg: 313.64 / Max: 313.91 Min: 518.85 / Avg: 520.45 / Max: 521.68 Min: 528.05 / Avg: 530.1 / Max: 531.94 Min: 572.49 / Avg: 573.04 / Max: 573.47 1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -ldecompose -lgenericPatchFields -lmetisDecomp -lscotchDecomp -llagrangian -lregionModels -lOpenFOAM -ldl -lm
BlogBench BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400K 800K 1200K 1600K 2000K SE +/- 13461.20, N = 3 SE +/- 14389.61, N = 3 SE +/- 10548.76, N = 3 SE +/- 9874.65, N = 3 SE +/- 14031.09, N = 3 SE +/- 4883.16, N = 3 SE +/- 7106.01, N = 3 SE +/- 13194.50, N = 3 SE +/- 7760.40, N = 3 SE +/- 7805.03, N = 3 SE +/- 7235.14, N = 3 SE +/- 12175.94, N = 3 SE +/- 20232.46, N = 3 SE +/- 16960.13, N = 3 1676516 1772037 1372955 1397475 1583350 1771624 1923368 1636092 1950080 1959615 1898428 2043037 2023180 1944496 1. (CC) gcc options: -O2 -pthread
Final Score Per Watt
OpenBenchmarking.org Final Score Per Watt, More Is Better BlogBench 1.1 Test: Read EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6K 12K 18K 24K 30K 12966.29 19861.21 10196.91 10885.37 12835.03 15592.36 20274.62 14192.86 20067.14 21034.07 22851.81 27647.84 28304.98 29655.38
Result Confidence
OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400K 800K 1200K 1600K 2000K Min: 1650220 / Avg: 1676515.67 / Max: 1694665 Min: 1743950 / Avg: 1772037 / Max: 1791514 Min: 1360224 / Avg: 1372955 / Max: 1393890 Min: 1384215 / Avg: 1397475 / Max: 1416780 Min: 1565457 / Avg: 1583350 / Max: 1611018 Min: 1764285 / Avg: 1771624.33 / Max: 1780874 Min: 1909275 / Avg: 1923368 / Max: 1932004 Min: 1610844 / Avg: 1636092.33 / Max: 1655363 Min: 1935120 / Avg: 1950080.33 / Max: 1961140 Min: 1944836 / Avg: 1959614.67 / Max: 1971357 Min: 1890301 / Avg: 1898428 / Max: 1912860 Min: 2022967 / Avg: 2043037.33 / Max: 2065016 Min: 1992185 / Avg: 2023179.67 / Max: 2061206 Min: 1916141 / Avg: 1944496.33 / Max: 1974796 1. (CC) gcc options: -O2 -pthread
PlaidML This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.3478 2.6956 4.0434 5.3912 6.739 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 5.99 5.96 5.56 5.76 5.72 5.67 5.84 5.71 5.76 5.68 5.67 5.57 5.53 5.17
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better PlaidML FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0248 0.0496 0.0744 0.0992 0.124 0.07 0.09 0.06 0.07 0.07 0.08 0.10 0.07 0.09 0.09 0.09 0.11 0.11 0.10
Result Confidence
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 5.9 / Avg: 5.99 / Max: 6.05 Min: 5.9 / Avg: 5.96 / Max: 6.02 Min: 5.54 / Avg: 5.56 / Max: 5.58 Min: 5.73 / Avg: 5.76 / Max: 5.79 Min: 5.7 / Avg: 5.72 / Max: 5.73 Min: 5.64 / Avg: 5.67 / Max: 5.69 Min: 5.82 / Avg: 5.84 / Max: 5.86 Min: 5.68 / Avg: 5.71 / Max: 5.73 Min: 5.73 / Avg: 5.76 / Max: 5.78 Min: 5.62 / Avg: 5.68 / Max: 5.72 Min: 5.65 / Avg: 5.67 / Max: 5.68 Min: 5.55 / Avg: 5.57 / Max: 5.59 Min: 5.46 / Avg: 5.53 / Max: 5.61 Min: 5.14 / Avg: 5.17 / Max: 5.2
Hierarchical INTegration This test runs the U.S. Department of Energy's Ames Laboratory Hierarchical INTegration (HINT) benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70M 140M 210M 280M 350M SE +/- 33906.80, N = 3 SE +/- 60046.42, N = 3 SE +/- 320354.97, N = 3 SE +/- 136985.64, N = 3 SE +/- 234513.76, N = 3 SE +/- 40543.01, N = 3 SE +/- 12833.10, N = 3 SE +/- 131222.60, N = 3 SE +/- 465752.47, N = 3 SE +/- 206566.86, N = 3 SE +/- 38644.36, N = 3 SE +/- 380619.49, N = 3 SE +/- 333820.85, N = 3 346505619.64 346733990.00 297478250.43 293169027.15 293216264.80 302406827.18 293437386.11 297816529.28 297827555.09 292597759.78 283588722.34 284710986.66 284332324.15 1. (CC) gcc options: -O3 -march=native -lm
QUIPs Per Watt
OpenBenchmarking.org QUIPs Per Watt, More Is Better Hierarchical INTegration 1.0 Test: FLOAT EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.5M 3M 4.5M 6M 7.5M 4710525.85 6400547.89 4189997.27 4208867.42 5029503.05 6161220.61 4210427.87 6028009.87 6095028.20 5937216.36 6710210.44 6892716.30 6962207.10
Result Confidence
OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60M 120M 180M 240M 300M Min: 346439918.8 / Avg: 346505619.64 / Max: 346553015.3 Min: 346619184.22 / Avg: 346733990 / Max: 346821912.43 Min: 296837585.98 / Avg: 297478250.43 / Max: 297805193.98 Min: 292897063.96 / Avg: 293169027.15 / Max: 293333683.01 Min: 292759126.74 / Avg: 293216264.8 / Max: 293535711.28 Min: 302325902.16 / Avg: 302406827.18 / Max: 302451712.79 Min: 293416576.95 / Avg: 293437386.11 / Max: 293460802.19 Min: 297554997.89 / Avg: 297816529.28 / Max: 297966245.2 Min: 297325728.53 / Avg: 297827555.09 / Max: 298758103.32 Min: 292233343.29 / Avg: 292597759.78 / Max: 292948520.88 Min: 283512887.38 / Avg: 283588722.34 / Max: 283639560.88 Min: 284060087.32 / Avg: 284710986.66 / Max: 285378285.89 Min: 283664869.62 / Avg: 284332324.15 / Max: 284679741.15 1. (CC) gcc options: -O3 -march=native -lm
AI Benchmark Alpha AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Score, More Is Better AI Benchmark Alpha 0.1.2 Device Training Score EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 1159 1173 1124 1219 1403 1348 1534 1475 1488 1546 1459 1384 1307 1094
OpenBenchmarking.org Score, More Is Better AI Benchmark Alpha 0.1.2 Device Inference Score EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 1456 1301 1965 2122 2125 2049 2107 2043 2022 1926 1660 1538 1372 1112
Crypto++ Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Keyed Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 140 280 420 560 700 SE +/- 0.14, N = 3 SE +/- 0.29, N = 3 SE +/- 0.06, N = 3 SE +/- 0.63, N = 3 SE +/- 0.31, N = 3 SE +/- 0.13, N = 3 SE +/- 0.77, N = 3 SE +/- 0.27, N = 3 SE +/- 0.31, N = 3 SE +/- 0.56, N = 3 SE +/- 0.13, N = 3 SE +/- 0.21, N = 3 SE +/- 0.07, N = 3 SE +/- 0.59, N = 3 628.21 627.44 539.45 529.62 529.60 530.80 546.72 531.14 539.17 538.09 530.36 513.60 514.85 513.65 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
MiB/second Per Watt
OpenBenchmarking.org MiB/second Per Watt, More Is Better Crypto++ 8.2 Test: Keyed Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 8.43 11.44 7.54 7.58 7.98 9.00 11.01 7.57 10.78 10.84 10.63 12.03 12.28 12.39
Result Confidence
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Keyed Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110 220 330 440 550 Min: 628.01 / Avg: 628.21 / Max: 628.49 Min: 627.02 / Avg: 627.44 / Max: 627.99 Min: 539.37 / Avg: 539.45 / Max: 539.56 Min: 528.75 / Avg: 529.62 / Max: 530.84 Min: 528.99 / Avg: 529.6 / Max: 530.03 Min: 530.54 / Avg: 530.8 / Max: 530.93 Min: 545.3 / Avg: 546.72 / Max: 547.95 Min: 530.69 / Avg: 531.14 / Max: 531.61 Min: 538.79 / Avg: 539.17 / Max: 539.79 Min: 537.38 / Avg: 538.09 / Max: 539.19 Min: 530.22 / Avg: 530.36 / Max: 530.61 Min: 513.18 / Avg: 513.6 / Max: 513.82 Min: 514.7 / Avg: 514.85 / Max: 514.94 Min: 512.6 / Avg: 513.65 / Max: 514.65 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
Blender Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Barbershop - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 SE +/- 0.69, N = 3 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.20, N = 3 SE +/- 0.23, N = 3 SE +/- 0.31, N = 3 SE +/- 0.07, N = 3 SE +/- 0.14, N = 3 SE +/- 0.46, N = 3 SE +/- 0.35, N = 3 SE +/- 0.25, N = 3 SE +/- 0.65, N = 3 SE +/- 1.20, N = 3 SE +/- 0.62, N = 3 356.90 683.81 154.42 154.23 179.18 180.94 223.08 243.58 237.65 291.52 411.56 432.92 554.64 866.85
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Barbershop - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 150 300 450 600 750 Min: 356.09 / Avg: 356.9 / Max: 358.27 Min: 683.61 / Avg: 683.81 / Max: 683.97 Min: 154.37 / Avg: 154.42 / Max: 154.49 Min: 153.83 / Avg: 154.23 / Max: 154.49 Min: 178.74 / Avg: 179.18 / Max: 179.54 Min: 180.32 / Avg: 180.94 / Max: 181.27 Min: 222.95 / Avg: 223.08 / Max: 223.18 Min: 243.32 / Avg: 243.58 / Max: 243.79 Min: 237.18 / Avg: 237.65 / Max: 238.58 Min: 291.13 / Avg: 291.52 / Max: 292.22 Min: 411.08 / Avg: 411.56 / Max: 411.94 Min: 432.21 / Avg: 432.92 / Max: 434.21 Min: 552.98 / Avg: 554.64 / Max: 556.98 Min: 865.62 / Avg: 866.85 / Max: 867.52
JPEG XL The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance. Learn more via the OpenBenchmarking.org test page.
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: super-resolution-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1100 2200 3300 4400 5500 SE +/- 88.35, N = 12 SE +/- 5.63, N = 3 SE +/- 84.35, N = 9 SE +/- 95.52, N = 12 SE +/- 52.31, N = 12 SE +/- 27.98, N = 3 SE +/- 67.46, N = 11 SE +/- 35.49, N = 11 SE +/- 62.74, N = 3 SE +/- 71.97, N = 12 SE +/- 73.62, N = 12 SE +/- 24.57, N = 3 SE +/- 16.48, N = 3 SE +/- 3.69, N = 3 5212 4518 4087 4175 4506 4642 4986 4864 4755 4946 4522 4044 3943 3941 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Inferences Per Minute Per Watt
OpenBenchmarking.org Inferences Per Minute Per Watt, More Is Better ONNX Runtime 1.6 Model: super-resolution-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11 22 33 44 55 28.68 36.02 25.73 25.67 28.36 33.35 40.89 32.78 38.41 39.90 43.67 45.86 44.61 47.91
Result Confidence
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: super-resolution-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 Min: 4665 / Avg: 5211.92 / Max: 5596 Min: 4508 / Avg: 4517.67 / Max: 4527.5 Min: 3791.5 / Avg: 4086.5 / Max: 4435 Min: 3788 / Avg: 4174.71 / Max: 4608.5 Min: 4219.5 / Avg: 4506.42 / Max: 4722 Min: 4586 / Avg: 4641.67 / Max: 4674.5 Min: 4318 / Avg: 4986.05 / Max: 5084 Min: 4539 / Avg: 4864.36 / Max: 4940.5 Min: 4633.5 / Avg: 4754.67 / Max: 4843.5 Min: 4494.5 / Avg: 4946.25 / Max: 5164.5 Min: 4171 / Avg: 4521.92 / Max: 5107 Min: 3996 / Avg: 4043.83 / Max: 4077.5 Min: 3922 / Avg: 3943 / Max: 3975.5 Min: 3934 / Avg: 3941 / Max: 3946.5 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
WireGuard + Linux Networking Stack Stress Test This is a benchmark of the WireGuard secure VPN tunnel and Linux networking stack stress test. The test runs on the local host but does require root permissions to run. The way it works is it creates three namespaces. ns0 has a loopback device. ns1 and ns2 each have wireguard devices. Those two wireguard devices send traffic through the loopback device of ns0. The end result of this is that tests wind up testing encryption and decryption at the same time -- a pretty CPU and scheduler-heavy workflow. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better WireGuard + Linux Networking Stack Stress Test EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70 140 210 280 350 SE +/- 0.12, N = 3 SE +/- 0.79, N = 3 SE +/- 1.60, N = 3 SE +/- 1.82, N = 3 SE +/- 0.75, N = 3 SE +/- 0.63, N = 3 SE +/- 0.79, N = 3 SE +/- 1.08, N = 3 SE +/- 1.52, N = 3 SE +/- 0.22, N = 3 SE +/- 0.25, N = 3 SE +/- 0.55, N = 3 SE +/- 0.07, N = 3 299.33 302.70 307.85 302.42 302.11 291.08 305.06 292.42 297.53 308.49 296.16 305.22 326.23
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better WireGuard + Linux Networking Stack Stress Test EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 299.09 / Avg: 299.33 / Max: 299.49 Min: 301.87 / Avg: 302.7 / Max: 304.28 Min: 305.74 / Avg: 307.85 / Max: 310.98 Min: 299.36 / Avg: 302.42 / Max: 305.67 Min: 301.02 / Avg: 302.11 / Max: 303.55 Min: 290.32 / Avg: 291.08 / Max: 292.34 Min: 303.92 / Avg: 305.06 / Max: 306.58 Min: 291.12 / Avg: 292.42 / Max: 294.56 Min: 295.48 / Avg: 297.53 / Max: 300.5 Min: 308.17 / Avg: 308.49 / Max: 308.9 Min: 295.71 / Avg: 296.16 / Max: 296.59 Min: 304.14 / Avg: 305.22 / Max: 305.91 Min: 326.14 / Avg: 326.23 / Max: 326.37
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 95, Compression Effort 7 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110 220 330 440 550 SE +/- 1.43, N = 3 SE +/- 0.26, N = 3 SE +/- 0.12, N = 3 SE +/- 0.17, N = 3 SE +/- 0.24, N = 3 SE +/- 0.12, N = 3 SE +/- 0.19, N = 3 SE +/- 0.16, N = 3 SE +/- 0.13, N = 3 SE +/- 0.27, N = 3 SE +/- 0.98, N = 3 SE +/- 2.55, N = 3 SE +/- 2.81, N = 3 SE +/- 0.90, N = 3 254.76 416.27 257.01 256.56 254.88 258.37 250.41 260.46 253.10 254.24 301.52 309.98 361.48 507.74 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 95, Compression Effort 7 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90 180 270 360 450 Min: 252.29 / Avg: 254.76 / Max: 257.23 Min: 415.89 / Avg: 416.27 / Max: 416.76 Min: 256.86 / Avg: 257 / Max: 257.24 Min: 256.37 / Avg: 256.56 / Max: 256.9 Min: 254.57 / Avg: 254.88 / Max: 255.36 Min: 258.22 / Avg: 258.37 / Max: 258.61 Min: 250.08 / Avg: 250.41 / Max: 250.74 Min: 260.18 / Avg: 260.46 / Max: 260.72 Min: 252.84 / Avg: 253.1 / Max: 253.26 Min: 253.9 / Avg: 254.24 / Max: 254.77 Min: 300.37 / Avg: 301.52 / Max: 303.47 Min: 305.77 / Avg: 309.98 / Max: 314.59 Min: 356.31 / Avg: 361.48 / Max: 365.98 Min: 506.68 / Avg: 507.74 / Max: 509.53 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
Numpy Benchmark This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better Numpy Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 SE +/- 0.20, N = 3 SE +/- 0.65, N = 3 SE +/- 0.70, N = 3 SE +/- 0.41, N = 3 SE +/- 0.24, N = 3 SE +/- 0.35, N = 3 SE +/- 0.40, N = 3 SE +/- 0.48, N = 3 SE +/- 0.31, N = 3 SE +/- 0.30, N = 3 SE +/- 1.35, N = 3 SE +/- 0.76, N = 3 SE +/- 0.30, N = 3 SE +/- 0.53, N = 3 348.66 345.36 306.50 303.21 303.75 303.45 311.97 303.33 309.10 307.47 301.28 296.96 298.27 270.83
Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better Numpy Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 4.56 6.03 4.18 4.22 4.46 4.97 6.05 4.23 5.96 5.98 5.85 6.72 6.82 6.25
Result Confidence
OpenBenchmarking.org Score, More Is Better Numpy Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 348.4 / Avg: 348.66 / Max: 349.06 Min: 344.12 / Avg: 345.36 / Max: 346.34 Min: 305.69 / Avg: 306.5 / Max: 307.9 Min: 302.77 / Avg: 303.21 / Max: 304.03 Min: 303.29 / Avg: 303.75 / Max: 304.08 Min: 302.78 / Avg: 303.45 / Max: 303.96 Min: 311.24 / Avg: 311.97 / Max: 312.62 Min: 302.7 / Avg: 303.33 / Max: 304.28 Min: 308.57 / Avg: 309.1 / Max: 309.64 Min: 307 / Avg: 307.47 / Max: 308.03 Min: 298.67 / Avg: 301.28 / Max: 303.16 Min: 295.48 / Avg: 296.96 / Max: 298.01 Min: 297.91 / Avg: 298.27 / Max: 298.86 Min: 269.87 / Avg: 270.83 / Max: 271.69
Ngspice Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Ngspice 34 Circuit: C7552 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 SE +/- 2.04, N = 12 SE +/- 1.27, N = 12 SE +/- 1.19, N = 3 SE +/- 1.10, N = 3 SE +/- 1.35, N = 3 SE +/- 1.31, N = 12 SE +/- 1.09, N = 12 SE +/- 1.66, N = 3 SE +/- 1.05, N = 3 SE +/- 1.75, N = 3 SE +/- 2.36, N = 3 122.44 119.44 132.62 136.87 131.53 130.09 132.04 128.92 135.68 137.59 212.89 1. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Ngspice 34 Circuit: C7552 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 113.19 / Avg: 122.44 / Max: 133.84 Min: 112.44 / Avg: 119.44 / Max: 126.77 Min: 130.9 / Avg: 132.62 / Max: 134.89 Min: 134.99 / Avg: 136.87 / Max: 138.8 Min: 128.85 / Avg: 131.53 / Max: 133.13 Min: 122.93 / Avg: 130.09 / Max: 137.37 Min: 125.23 / Avg: 132.04 / Max: 137.09 Min: 126.46 / Avg: 128.92 / Max: 132.08 Min: 133.59 / Avg: 135.68 / Max: 136.92 Min: 134.17 / Avg: 137.59 / Max: 139.97 Min: 209.86 / Avg: 212.89 / Max: 217.53 1. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
Blender Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Pabellon Barcelona - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 140 280 420 560 700 SE +/- 0.98, N = 3 SE +/- 0.51, N = 3 SE +/- 0.15, N = 3 SE +/- 0.39, N = 3 SE +/- 0.14, N = 3 SE +/- 0.92, N = 3 SE +/- 0.07, N = 3 SE +/- 0.33, N = 3 SE +/- 0.14, N = 3 SE +/- 1.69, N = 3 SE +/- 0.50, N = 3 SE +/- 1.19, N = 3 SE +/- 0.64, N = 3 SE +/- 0.72, N = 3 265.13 508.26 119.17 118.26 137.83 139.21 167.99 189.09 182.76 219.75 317.27 338.52 426.59 664.29
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Pabellon Barcelona - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 120 240 360 480 600 Min: 263.71 / Avg: 265.13 / Max: 267.02 Min: 507.47 / Avg: 508.26 / Max: 509.22 Min: 118.86 / Avg: 119.17 / Max: 119.35 Min: 117.87 / Avg: 118.26 / Max: 119.05 Min: 137.55 / Avg: 137.83 / Max: 138.03 Min: 138.06 / Avg: 139.21 / Max: 141.03 Min: 167.9 / Avg: 167.99 / Max: 168.12 Min: 188.42 / Avg: 189.09 / Max: 189.45 Min: 182.48 / Avg: 182.76 / Max: 182.95 Min: 217.98 / Avg: 219.75 / Max: 223.14 Min: 316.57 / Avg: 317.27 / Max: 318.25 Min: 336.65 / Avg: 338.52 / Max: 340.74 Min: 425.64 / Avg: 426.59 / Max: 427.81 Min: 662.88 / Avg: 664.29 / Max: 665.27
Gcrypt Library Libgcrypt is a general purpose cryptographic library developed as part of the GnuPG project. This is a benchmark of libgcrypt's integrated benchmark and is measuring the time to run the benchmark command with a cipher/mac/hash repetition count set for 50 times as simple, high level look at the overall crypto performance of the system under test. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Gcrypt Library 1.9 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 SE +/- 0.70, N = 3 SE +/- 0.81, N = 3 SE +/- 1.00, N = 3 SE +/- 0.57, N = 3 SE +/- 0.52, N = 3 SE +/- 0.48, N = 3 SE +/- 1.03, N = 3 SE +/- 1.43, N = 3 SE +/- 0.38, N = 3 SE +/- 0.34, N = 3 SE +/- 0.37, N = 3 SE +/- 0.64, N = 3 223.75 223.71 261.45 265.19 265.92 264.99 257.57 264.68 262.12 262.91 268.68 274.30 273.20 273.02 1. (CC) gcc options: -O2 -fvisibility=hidden
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Gcrypt Library 1.9 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 223.6 / Avg: 223.75 / Max: 223.97 Min: 223.61 / Avg: 223.71 / Max: 223.89 Min: 260.68 / Avg: 261.45 / Max: 262.85 Min: 263.73 / Avg: 265.19 / Max: 266.52 Min: 264.29 / Avg: 265.92 / Max: 267.74 Min: 264.26 / Avg: 264.98 / Max: 266.12 Min: 256.94 / Avg: 257.57 / Max: 258.61 Min: 263.89 / Avg: 264.68 / Max: 265.54 Min: 260.44 / Avg: 262.12 / Max: 264 Min: 261.48 / Avg: 262.91 / Max: 265.77 Min: 267.92 / Avg: 268.68 / Max: 269.11 Min: 273.81 / Avg: 274.3 / Max: 274.96 Min: 272.5 / Avg: 273.19 / Max: 273.74 Min: 271.81 / Avg: 273.02 / Max: 274.01 1. (CC) gcc options: -O2 -fvisibility=hidden
ECP-CANDLE The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ECP-CANDLE 0.3 Benchmark: P3B2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 785.07 775.67 817.66 789.56 792.74 788.52 739.71 783.25 755.01 749.67 763.80 736.22 748.54 783.24
Crypto++ Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Integer + Elliptic Curve Public Key Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1000 2000 3000 4000 5000 SE +/- 3.14, N = 3 SE +/- 2.49, N = 3 SE +/- 5.35, N = 3 SE +/- 2.07, N = 3 SE +/- 3.12, N = 3 SE +/- 2.54, N = 3 SE +/- 5.67, N = 3 SE +/- 3.97, N = 3 SE +/- 4.11, N = 3 SE +/- 2.96, N = 3 SE +/- 1.09, N = 3 SE +/- 2.72, N = 3 SE +/- 1.24, N = 3 SE +/- 5.48, N = 3 4834.05 4835.02 4152.60 4091.78 4072.76 4085.53 4207.91 4089.12 4139.98 4151.29 4086.08 3949.00 3965.03 3949.44 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
MiB/second Per Watt
OpenBenchmarking.org MiB/second Per Watt, More Is Better Crypto++ 8.2 Test: Integer + Elliptic Curve Public Key Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 64.14 87.24 57.63 58.12 61.08 68.50 83.94 57.18 81.90 82.69 81.20 91.59 93.36 94.07
Result Confidence
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Integer + Elliptic Curve Public Key Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 4828.11 / Avg: 4834.05 / Max: 4838.82 Min: 4830.71 / Avg: 4835.02 / Max: 4839.34 Min: 4142.26 / Avg: 4152.6 / Max: 4160.14 Min: 4087.91 / Avg: 4091.78 / Max: 4095.01 Min: 4068.43 / Avg: 4072.76 / Max: 4078.81 Min: 4080.99 / Avg: 4085.53 / Max: 4089.76 Min: 4199.27 / Avg: 4207.91 / Max: 4218.6 Min: 4081.58 / Avg: 4089.12 / Max: 4095.03 Min: 4132.23 / Avg: 4139.98 / Max: 4146.24 Min: 4145.43 / Avg: 4151.29 / Max: 4154.96 Min: 4084.12 / Avg: 4086.08 / Max: 4087.91 Min: 3943.6 / Avg: 3949 / Max: 3952.3 Min: 3963.31 / Avg: 3965.03 / Max: 3967.43 Min: 3939.83 / Avg: 3949.44 / Max: 3958.81 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
Incompact3D Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Incompact3D 2020-09-17 Input: Cylinder EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 SE +/- 0.68, N = 3 SE +/- 0.12, N = 3 SE +/- 0.50, N = 3 SE +/- 0.75, N = 3 SE +/- 0.73, N = 3 SE +/- 1.22, N = 3 SE +/- 0.39, N = 3 SE +/- 0.57, N = 3 SE +/- 0.52, N = 3 SE +/- 0.20, N = 3 SE +/- 0.25, N = 3 SE +/- 0.47, N = 3 SE +/- 0.37, N = 3 SE +/- 1.04, N = 3 230.94 375.70 224.62 206.95 195.34 206.58 187.60 193.92 189.66 200.21 262.63 270.72 325.84 450.85 1. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Incompact3D 2020-09-17 Input: Cylinder EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 Min: 229.71 / Avg: 230.94 / Max: 232.08 Min: 375.47 / Avg: 375.7 / Max: 375.88 Min: 224.11 / Avg: 224.62 / Max: 225.62 Min: 206.04 / Avg: 206.95 / Max: 208.44 Min: 194.23 / Avg: 195.34 / Max: 196.73 Min: 204.79 / Avg: 206.58 / Max: 208.92 Min: 186.85 / Avg: 187.6 / Max: 188.17 Min: 192.99 / Avg: 193.92 / Max: 194.97 Min: 188.81 / Avg: 189.66 / Max: 190.61 Min: 199.84 / Avg: 200.21 / Max: 200.55 Min: 262.32 / Avg: 262.63 / Max: 263.13 Min: 270.13 / Avg: 270.72 / Max: 271.65 Min: 325.11 / Avg: 325.84 / Max: 326.32 Min: 449.53 / Avg: 450.85 / Max: 452.91 1. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: shufflenet-v2-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 112.30, N = 4 SE +/- 45.76, N = 3 SE +/- 53.19, N = 12 SE +/- 46.88, N = 3 SE +/- 73.86, N = 12 SE +/- 70.17, N = 12 SE +/- 29.60, N = 3 SE +/- 105.59, N = 12 SE +/- 63.96, N = 12 SE +/- 18.32, N = 3 SE +/- 1.83, N = 3 SE +/- 45.05, N = 3 SE +/- 2.92, N = 3 SE +/- 15.37, N = 3 9348 9823 6478 6686 7221 7373 8698 8070 8421 8810 8975 8611 8721 8836 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Inferences Per Minute Per Watt
OpenBenchmarking.org Inferences Per Minute Per Watt, More Is Better ONNX Runtime 1.6 Model: shufflenet-v2-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 82.09 135.13 46.15 50.67 59.42 68.55 101.89 74.72 96.92 108.45 129.62 142.11 154.91 172.02
Result Confidence
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: shufflenet-v2-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 9037.5 / Avg: 9348.13 / Max: 9565.5 Min: 9743 / Avg: 9822.67 / Max: 9901.5 Min: 6037.5 / Avg: 6477.5 / Max: 6804 Min: 6618 / Avg: 6686.17 / Max: 6776 Min: 6810.5 / Avg: 7220.58 / Max: 7710.5 Min: 6932 / Avg: 7373.04 / Max: 7736 Min: 8644 / Avg: 8698 / Max: 8746 Min: 7492 / Avg: 8070 / Max: 8524 Min: 7942 / Avg: 8421.42 / Max: 8675.5 Min: 8780.5 / Avg: 8809.83 / Max: 8843.5 Min: 8973 / Avg: 8974.83 / Max: 8978.5 Min: 8522 / Avg: 8611.17 / Max: 8667 Min: 8717.5 / Avg: 8720.67 / Max: 8726.5 Min: 8814.5 / Avg: 8836.33 / Max: 8866 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Tinymembench This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K SE +/- 39.37, N = 3 SE +/- 28.30, N = 3 SE +/- 217.98, N = 3 SE +/- 52.61, N = 3 SE +/- 217.68, N = 3 SE +/- 39.21, N = 3 SE +/- 27.08, N = 3 SE +/- 13.93, N = 3 SE +/- 46.42, N = 3 SE +/- 20.90, N = 3 SE +/- 18.80, N = 3 SE +/- 37.39, N = 3 SE +/- 21.61, N = 3 15585.7 16357.2 17329.8 15640.1 16494.4 15872.0 14786.9 15097.1 14961.4 14820.3 14921.3 14776.4 14571.7 1. (CC) gcc options: -O2 -lm
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Tinymembench 2018-05-28 Standard Memset EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70 140 210 280 350 209.76 286.11 236.78 215.62 268.71 303.95 202.28 289.31 284.81 279.72 333.88 334.79 332.00
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 15512.6 / Avg: 15585.7 / Max: 15647.6 Min: 16301.7 / Avg: 16357.2 / Max: 16394.6 Min: 16900.6 / Avg: 17329.83 / Max: 17610.5 Min: 15534.9 / Avg: 15640.1 / Max: 15694.5 Min: 16092.5 / Avg: 16494.43 / Max: 16840.3 Min: 15794.6 / Avg: 15872.03 / Max: 15921.5 Min: 14734.2 / Avg: 14786.93 / Max: 14824 Min: 15074.8 / Avg: 15097.07 / Max: 15122.7 Min: 14876.5 / Avg: 14961.37 / Max: 15036.4 Min: 14781.5 / Avg: 14820.27 / Max: 14853.2 Min: 14884.6 / Avg: 14921.27 / Max: 14946.8 Min: 14704.8 / Avg: 14776.4 / Max: 14830.9 Min: 14546.4 / Avg: 14571.7 / Max: 14614.7 1. (CC) gcc options: -O2 -lm
Result
OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 1.52, N = 3 SE +/- 2.03, N = 3 SE +/- 44.65, N = 3 SE +/- 14.82, N = 3 SE +/- 35.12, N = 3 SE +/- 17.46, N = 3 SE +/- 2.22, N = 3 SE +/- 2.05, N = 3 SE +/- 19.03, N = 3 SE +/- 2.29, N = 3 SE +/- 7.90, N = 3 SE +/- 22.30, N = 3 SE +/- 2.95, N = 3 9055.1 9025.4 9314.0 8852.0 9087.1 8907.7 8829.9 8854.2 8902.8 8842.7 8850.6 8825.7 8805.9 1. (CC) gcc options: -O2 -lm
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1600 3200 4800 6400 8000 Min: 9052.3 / Avg: 9055.13 / Max: 9057.5 Min: 9023.3 / Avg: 9025.43 / Max: 9029.5 Min: 9224.8 / Avg: 9313.97 / Max: 9362.9 Min: 8829.6 / Avg: 8851.97 / Max: 8880 Min: 9051.6 / Avg: 9087.07 / Max: 9157.3 Min: 8873.4 / Avg: 8907.73 / Max: 8930.4 Min: 8825.5 / Avg: 8829.87 / Max: 8832.7 Min: 8851 / Avg: 8854.17 / Max: 8858 Min: 8864.9 / Avg: 8902.83 / Max: 8924.5 Min: 8838.4 / Avg: 8842.73 / Max: 8846.2 Min: 8835.2 / Avg: 8850.57 / Max: 8861.4 Min: 8781.6 / Avg: 8825.67 / Max: 8853.7 Min: 8800 / Avg: 8805.9 / Max: 8809 1. (CC) gcc options: -O2 -lm
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: bertsquad-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110 220 330 440 550 SE +/- 6.13, N = 12 SE +/- 0.50, N = 3 SE +/- 7.16, N = 12 SE +/- 6.16, N = 12 SE +/- 3.97, N = 3 SE +/- 0.60, N = 3 SE +/- 1.15, N = 3 SE +/- 0.44, N = 3 SE +/- 4.69, N = 12 SE +/- 4.15, N = 9 SE +/- 5.01, N = 3 SE +/- 0.29, N = 3 SE +/- 1.26, N = 3 SE +/- 0.50, N = 3 372 480 390 395 431 440 475 466 459 500 456 400 403 409 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Inferences Per Minute Per Watt
OpenBenchmarking.org Inferences Per Minute Per Watt, More Is Better ONNX Runtime 1.6 Model: bertsquad-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.179 2.358 3.537 4.716 5.895 2.50 4.10 2.39 2.50 2.84 3.22 4.08 3.27 3.86 4.25 4.50 4.71 4.79 5.24
Result Confidence
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: bertsquad-10 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90 180 270 360 450 Min: 329.5 / Avg: 371.92 / Max: 403.5 Min: 479.5 / Avg: 480 / Max: 481 Min: 335.5 / Avg: 389.54 / Max: 410.5 Min: 343.5 / Avg: 395 / Max: 412.5 Min: 424.5 / Avg: 430.5 / Max: 438 Min: 439.5 / Avg: 440.33 / Max: 441.5 Min: 472.5 / Avg: 474.5 / Max: 476.5 Min: 465.5 / Avg: 466.33 / Max: 467 Min: 425.5 / Avg: 459.13 / Max: 469.5 Min: 468 / Avg: 500.22 / Max: 507 Min: 450.5 / Avg: 456 / Max: 466 Min: 399.5 / Avg: 400 / Max: 400.5 Min: 401.5 / Avg: 403 / Max: 405.5 Min: 408 / Avg: 409 / Max: 409.5 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Blender Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Classroom - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 120 240 360 480 600 SE +/- 0.74, N = 3 SE +/- 0.25, N = 3 SE +/- 0.41, N = 3 SE +/- 0.25, N = 3 SE +/- 0.77, N = 3 SE +/- 0.69, N = 3 SE +/- 0.41, N = 3 SE +/- 1.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.04, N = 3 SE +/- 0.64, N = 3 SE +/- 0.83, N = 3 SE +/- 0.31, N = 3 SE +/- 0.21, N = 3 237.79 454.49 103.55 101.68 121.21 122.03 148.50 168.40 162.29 196.70 287.81 304.34 387.23 558.87
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Classroom - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 Min: 236.43 / Avg: 237.79 / Max: 238.96 Min: 454.13 / Avg: 454.49 / Max: 454.97 Min: 103.02 / Avg: 103.55 / Max: 104.35 Min: 101.18 / Avg: 101.68 / Max: 101.94 Min: 120.06 / Avg: 121.21 / Max: 122.68 Min: 120.66 / Avg: 122.03 / Max: 122.86 Min: 147.97 / Avg: 148.5 / Max: 149.3 Min: 167.27 / Avg: 168.4 / Max: 170.57 Min: 161.98 / Avg: 162.29 / Max: 162.65 Min: 196.63 / Avg: 196.7 / Max: 196.77 Min: 286.64 / Avg: 287.81 / Max: 288.83 Min: 303.05 / Avg: 304.34 / Max: 305.88 Min: 386.68 / Avg: 387.23 / Max: 387.74 Min: 558.51 / Avg: 558.87 / Max: 559.22
Radiance Benchmark This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Radiance Benchmark 5.0 Test: Serial EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160 320 480 640 800 612.89 613.55 713.56 723.32 725.23 727.34 705.16 724.56 721.79 714.24 724.99 746.18 746.72 748.59
Monte Carlo Simulations of Ionised Nebulae Mocassin is the Monte Carlo Simulations of Ionised Nebulae. MOCASSIN is a fully 3D or 2D photoionisation and dust radiative transfer code which employs a Monte Carlo approach to the transfer of radiation through media of arbitrary geometry and density distribution. Learn more via the OpenBenchmarking.org test page.
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.05299, N = 11 SE +/- 0.06375, N = 3 SE +/- 0.03096, N = 3 SE +/- 0.00378, N = 3 SE +/- 0.00479, N = 3 SE +/- 0.00667, N = 3 SE +/- 0.00275, N = 3 SE +/- 0.00249, N = 3 SE +/- 0.00324, N = 3 SE +/- 0.01008, N = 3 SE +/- 0.07483, N = 3 SE +/- 0.00263, N = 3 SE +/- 0.00448, N = 3 SE +/- 0.00542, N = 3 7.08925 12.95600 17.30750 17.38630 17.68500 16.64940 15.27940 17.96770 15.29200 15.56290 15.63290 9.05154 9.09743 8.64661 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
GFLOP/s Per Watt
OpenBenchmarking.org GFLOP/s Per Watt, More Is Better High Performance Conjugate Gradient 3.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0315 0.063 0.0945 0.126 0.1575 0.05 0.11 0.09 0.09 0.10 0.10 0.12 0.10 0.11 0.12 0.14 0.10 0.11 0.11
Result Confidence
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 6.64 / Avg: 7.09 / Max: 7.27 Min: 12.86 / Avg: 12.96 / Max: 13.08 Min: 17.25 / Avg: 17.31 / Max: 17.34 Min: 17.38 / Avg: 17.39 / Max: 17.39 Min: 17.68 / Avg: 17.69 / Max: 17.69 Min: 16.64 / Avg: 16.65 / Max: 16.66 Min: 15.28 / Avg: 15.28 / Max: 15.28 Min: 17.96 / Avg: 17.97 / Max: 17.97 Min: 15.29 / Avg: 15.29 / Max: 15.3 Min: 15.55 / Avg: 15.56 / Max: 15.58 Min: 15.51 / Avg: 15.63 / Max: 15.77 Min: 9.05 / Avg: 9.05 / Max: 9.05 Min: 9.09 / Avg: 9.1 / Max: 9.1 Min: 8.64 / Avg: 8.65 / Max: 8.66 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
Ngspice Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Ngspice 34 Circuit: C2670 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 SE +/- 2.33, N = 9 SE +/- 1.05, N = 3 SE +/- 1.88, N = 4 SE +/- 1.33, N = 3 SE +/- 1.89, N = 3 SE +/- 1.90, N = 3 SE +/- 1.85, N = 3 SE +/- 1.93, N = 5 SE +/- 1.23, N = 3 SE +/- 1.59, N = 3 SE +/- 2.54, N = 3 151.46 158.79 177.62 172.37 170.22 165.66 175.85 173.90 183.22 174.20 211.49 1. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Ngspice 34 Circuit: C2670 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 142 / Avg: 151.46 / Max: 162.3 Min: 157.45 / Avg: 158.79 / Max: 160.86 Min: 172.48 / Avg: 177.62 / Max: 181.39 Min: 170.3 / Avg: 172.37 / Max: 174.84 Min: 168.07 / Avg: 170.22 / Max: 173.98 Min: 162.62 / Avg: 165.66 / Max: 169.15 Min: 172.42 / Avg: 175.85 / Max: 178.76 Min: 168.11 / Avg: 173.89 / Max: 179.12 Min: 181.83 / Avg: 183.22 / Max: 185.67 Min: 171.02 / Avg: 174.2 / Max: 175.94 Min: 206.5 / Avg: 211.49 / Max: 214.78 1. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
ECP-CANDLE The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ECP-CANDLE 0.3 Benchmark: P3B1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 1108.26 668.99 649.31 604.15 590.56 584.61 566.99 586.67 577.65 576.98 574.77 655.45 623.78 715.21
Mobile Neural Network MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: inception-v3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.27, N = 3 SE +/- 0.02, N = 14 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 15 SE +/- 0.05, N = 15 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.70, N = 3 SE +/- 0.09, N = 15 SE +/- 0.09, N = 3 SE +/- 0.09, N = 4 SE +/- 0.12, N = 11 43.36 34.91 38.00 33.58 32.86 35.47 31.87 32.47 32.92 36.88 41.12 43.13 38.62 38.33 MIN: 40.47 / MAX: 57.18 MIN: 33.81 / MAX: 50.1 MIN: 36.26 / MAX: 38.93 MIN: 32.98 / MAX: 34.68 MIN: 32.05 / MAX: 34.89 MIN: 34.49 / MAX: 38.11 MIN: 30.79 / MAX: 34.77 MIN: 32.03 / MAX: 34.7 MIN: 32.02 / MAX: 35.9 MIN: 35.1 / MAX: 40.26 MIN: 39.63 / MAX: 81.35 MIN: 42.07 / MAX: 59.48 MIN: 38.09 / MAX: 52.97 MIN: 37.06 / MAX: 55.19 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: inception-v3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 42.82 / Avg: 43.36 / Max: 43.65 Min: 34.79 / Avg: 34.91 / Max: 35.08 Min: 37.92 / Avg: 38 / Max: 38.08 Min: 33.57 / Avg: 33.58 / Max: 33.59 Min: 32.62 / Avg: 32.86 / Max: 33.02 Min: 34.98 / Avg: 35.47 / Max: 35.68 Min: 31.52 / Avg: 31.87 / Max: 32.13 Min: 32.35 / Avg: 32.47 / Max: 32.57 Min: 32.64 / Avg: 32.92 / Max: 33.11 Min: 35.66 / Avg: 36.88 / Max: 38.08 Min: 40.41 / Avg: 41.12 / Max: 41.65 Min: 43 / Avg: 43.13 / Max: 43.3 Min: 38.45 / Avg: 38.62 / Max: 38.79 Min: 37.94 / Avg: 38.33 / Max: 39.34 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: mobilenet-v1-1.0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.3001 2.6002 3.9003 5.2004 6.5005 SE +/- 0.401, N = 3 SE +/- 0.002, N = 14 SE +/- 0.015, N = 3 SE +/- 0.006, N = 3 SE +/- 0.024, N = 3 SE +/- 0.004, N = 15 SE +/- 0.005, N = 15 SE +/- 0.008, N = 3 SE +/- 0.010, N = 3 SE +/- 0.012, N = 3 SE +/- 0.005, N = 15 SE +/- 0.060, N = 3 SE +/- 0.015, N = 4 SE +/- 0.007, N = 11 5.778 3.625 3.187 3.073 3.056 3.022 2.836 3.003 2.927 3.213 3.856 4.081 5.322 3.917 MIN: 4.65 / MAX: 7.24 MIN: 3.57 / MAX: 17.34 MIN: 3.13 / MAX: 3.28 MIN: 3.03 / MAX: 3.22 MIN: 2.98 / MAX: 3.31 MIN: 2.96 / MAX: 3.4 MIN: 2.77 / MAX: 4.94 MIN: 2.95 / MAX: 3.25 MIN: 2.87 / MAX: 5.27 MIN: 3.15 / MAX: 4.81 MIN: 3.74 / MAX: 20.71 MIN: 3.8 / MAX: 20.13 MIN: 5.22 / MAX: 13.18 MIN: 3.79 / MAX: 19.86 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: mobilenet-v1-1.0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 5 / Avg: 5.78 / Max: 6.34 Min: 3.61 / Avg: 3.62 / Max: 3.64 Min: 3.17 / Avg: 3.19 / Max: 3.22 Min: 3.06 / Avg: 3.07 / Max: 3.09 Min: 3.02 / Avg: 3.06 / Max: 3.1 Min: 3.01 / Avg: 3.02 / Max: 3.05 Min: 2.81 / Avg: 2.84 / Max: 2.88 Min: 2.99 / Avg: 3 / Max: 3.02 Min: 2.92 / Avg: 2.93 / Max: 2.95 Min: 3.19 / Avg: 3.21 / Max: 3.23 Min: 3.82 / Avg: 3.86 / Max: 3.88 Min: 3.98 / Avg: 4.08 / Max: 4.18 Min: 5.3 / Avg: 5.32 / Max: 5.36 Min: 3.86 / Avg: 3.92 / Max: 3.95 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: MobileNetV2_224 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.391, N = 3 SE +/- 0.017, N = 14 SE +/- 0.016, N = 3 SE +/- 0.022, N = 3 SE +/- 0.016, N = 3 SE +/- 0.008, N = 15 SE +/- 0.009, N = 15 SE +/- 0.022, N = 3 SE +/- 0.028, N = 3 SE +/- 0.026, N = 3 SE +/- 0.046, N = 15 SE +/- 0.033, N = 3 SE +/- 0.155, N = 4 SE +/- 0.030, N = 11 8.171 4.578 5.067 4.920 4.825 4.852 4.713 4.821 4.820 5.270 5.772 5.855 5.930 4.944 MIN: 7.28 / MAX: 12.17 MIN: 4.43 / MAX: 10.46 MIN: 4.95 / MAX: 5.21 MIN: 4.78 / MAX: 5.1 MIN: 4.71 / MAX: 5.16 MIN: 4.69 / MAX: 5.21 MIN: 4.51 / MAX: 6.85 MIN: 4.72 / MAX: 5.31 MIN: 4.67 / MAX: 6.92 MIN: 5.12 / MAX: 6.52 MIN: 5.46 / MAX: 20.47 MIN: 5.58 / MAX: 23.16 MIN: 5.53 / MAX: 20.6 MIN: 4.74 / MAX: 7.04 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: MobileNetV2_224 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 7.72 / Avg: 8.17 / Max: 8.95 Min: 4.51 / Avg: 4.58 / Max: 4.67 Min: 5.05 / Avg: 5.07 / Max: 5.1 Min: 4.88 / Avg: 4.92 / Max: 4.96 Min: 4.79 / Avg: 4.83 / Max: 4.84 Min: 4.78 / Avg: 4.85 / Max: 4.89 Min: 4.6 / Avg: 4.71 / Max: 4.75 Min: 4.8 / Avg: 4.82 / Max: 4.87 Min: 4.78 / Avg: 4.82 / Max: 4.87 Min: 5.23 / Avg: 5.27 / Max: 5.32 Min: 5.57 / Avg: 5.77 / Max: 6.19 Min: 5.81 / Avg: 5.86 / Max: 5.92 Min: 5.66 / Avg: 5.93 / Max: 6.37 Min: 4.81 / Avg: 4.94 / Max: 5.12 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: resnet-v2-50 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.19, N = 3 SE +/- 0.01, N = 14 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 15 SE +/- 0.05, N = 15 SE +/- 0.05, N = 3 SE +/- 0.12, N = 3 SE +/- 0.97, N = 3 SE +/- 0.09, N = 15 SE +/- 0.12, N = 3 SE +/- 0.11, N = 4 SE +/- 0.12, N = 11 33.50 27.45 26.92 25.22 24.78 25.44 24.30 25.27 25.06 29.02 31.54 33.59 38.10 31.44 MIN: 32.85 / MAX: 46.41 MIN: 27.01 / MAX: 58.07 MIN: 26.49 / MAX: 27.67 MIN: 24.73 / MAX: 27.24 MIN: 24.46 / MAX: 25.37 MIN: 24.73 / MAX: 28.58 MIN: 23.72 / MAX: 27.28 MIN: 24.96 / MAX: 28.42 MIN: 24.51 / MAX: 27.82 MIN: 27.23 / MAX: 33.12 MIN: 30.05 / MAX: 48.95 MIN: 32.54 / MAX: 48.77 MIN: 36.85 / MAX: 55.75 MIN: 30.44 / MAX: 49.82 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: resnet-v2-50 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 33.28 / Avg: 33.5 / Max: 33.89 Min: 27.35 / Avg: 27.44 / Max: 27.53 Min: 26.83 / Avg: 26.92 / Max: 27.06 Min: 25.03 / Avg: 25.22 / Max: 25.35 Min: 24.73 / Avg: 24.78 / Max: 24.82 Min: 25.04 / Avg: 25.44 / Max: 25.64 Min: 23.9 / Avg: 24.3 / Max: 24.57 Min: 25.2 / Avg: 25.27 / Max: 25.36 Min: 24.81 / Avg: 25.06 / Max: 25.18 Min: 27.93 / Avg: 29.02 / Max: 30.96 Min: 30.84 / Avg: 31.54 / Max: 32 Min: 33.39 / Avg: 33.59 / Max: 33.8 Min: 37.9 / Avg: 38.1 / Max: 38.42 Min: 30.9 / Avg: 31.44 / Max: 32.5 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: SqueezeNetV1.0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.064, N = 3 SE +/- 0.062, N = 14 SE +/- 0.014, N = 3 SE +/- 0.018, N = 3 SE +/- 0.003, N = 3 SE +/- 0.271, N = 15 SE +/- 0.252, N = 15 SE +/- 0.016, N = 3 SE +/- 0.048, N = 3 SE +/- 0.114, N = 3 SE +/- 0.061, N = 15 SE +/- 0.027, N = 3 SE +/- 0.115, N = 4 SE +/- 0.071, N = 11 10.209 9.357 8.375 7.542 7.479 8.142 7.625 7.338 7.493 8.978 9.504 9.610 9.768 9.650 MIN: 10.06 / MAX: 11.23 MIN: 8.98 / MAX: 23.95 MIN: 8.25 / MAX: 8.55 MIN: 7.38 / MAX: 8.01 MIN: 7.34 / MAX: 8.04 MIN: 7.43 / MAX: 13.26 MIN: 7.08 / MAX: 14.04 MIN: 7.23 / MAX: 7.87 MIN: 7.29 / MAX: 9.78 MIN: 8.7 / MAX: 10.42 MIN: 9.01 / MAX: 25.56 MIN: 9.28 / MAX: 25.71 MIN: 9.3 / MAX: 12.44 MIN: 9.16 / MAX: 25.96 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: SqueezeNetV1.0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 10.14 / Avg: 10.21 / Max: 10.34 Min: 9.16 / Avg: 9.36 / Max: 10 Min: 8.35 / Avg: 8.37 / Max: 8.4 Min: 7.51 / Avg: 7.54 / Max: 7.56 Min: 7.47 / Avg: 7.48 / Max: 7.48 Min: 7.63 / Avg: 8.14 / Max: 11.51 Min: 7.17 / Avg: 7.63 / Max: 10.86 Min: 7.32 / Avg: 7.34 / Max: 7.37 Min: 7.44 / Avg: 7.49 / Max: 7.59 Min: 8.84 / Avg: 8.98 / Max: 9.2 Min: 9.26 / Avg: 9.5 / Max: 10.27 Min: 9.58 / Avg: 9.61 / Max: 9.66 Min: 9.49 / Avg: 9.77 / Max: 10.05 Min: 9.3 / Avg: 9.65 / Max: 10.14 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Darmstadt Automotive Parallel Heterogeneous Suite DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6K 12K 18K 24K 30K SE +/- 205.94, N = 13 SE +/- 296.05, N = 4 SE +/- 189.69, N = 6 SE +/- 203.19, N = 6 SE +/- 208.12, N = 5 SE +/- 149.83, N = 11 SE +/- 300.22, N = 3 SE +/- 233.41, N = 5 SE +/- 303.68, N = 3 SE +/- 228.79, N = 5 SE +/- 278.67, N = 4 SE +/- 162.08, N = 13 SE +/- 251.88, N = 3 SE +/- 130.62, N = 3 24205.54 27135.63 19062.92 19915.00 20036.19 20634.20 21442.00 20967.48 21272.04 21653.91 22339.85 22574.80 23807.67 23230.19 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Test Cases Per Minute Per Watt
OpenBenchmarking.org Test Cases Per Minute Per Watt, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110 220 330 440 550 283.96 446.35 211.71 231.24 242.20 278.09 365.15 261.72 356.46 372.78 405.87 475.56 527.48 531.57
Result Confidence
OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K Min: 22569.97 / Avg: 24205.54 / Max: 25313.47 Min: 26266.6 / Avg: 27135.63 / Max: 27598.01 Min: 18118.9 / Avg: 19062.92 / Max: 19320.87 Min: 18959.86 / Avg: 19915 / Max: 20292.89 Min: 19216.66 / Avg: 20036.19 / Max: 20355.61 Min: 19629.33 / Avg: 20634.2 / Max: 21282.63 Min: 20857.09 / Avg: 21442 / Max: 21851.95 Min: 20100.44 / Avg: 20967.48 / Max: 21472.74 Min: 20665.78 / Avg: 21272.04 / Max: 21606.83 Min: 20776.27 / Avg: 21653.91 / Max: 22118.67 Min: 21516.1 / Avg: 22339.85 / Max: 22741.4 Min: 20715.3 / Avg: 22574.8 / Max: 23030.24 Min: 23305.32 / Avg: 23807.67 / Max: 24091.45 Min: 22989.39 / Avg: 23230.19 / Max: 23438.32 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6K 12K 18K 24K 30K SE +/- 239.23, N = 3 SE +/- 204.43, N = 3 SE +/- 294.57, N = 15 SE +/- 287.13, N = 3 SE +/- 82.48, N = 3 SE +/- 82.21, N = 3 SE +/- 218.98, N = 3 SE +/- 124.32, N = 3 SE +/- 275.94, N = 15 SE +/- 250.51, N = 6 SE +/- 238.46, N = 3 SE +/- 144.50, N = 3 SE +/- 254.46, N = 15 SE +/- 93.54, N = 3 26636 30037 26143 26867 26289 26303 26724 26305 26186 26769 26045 25601 24876 18378 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Mflops Per Watt
OpenBenchmarking.org Mflops Per Watt, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 130 260 390 520 650 365.69 541.24 361.64 383.02 393.21 437.62 536.35 372.69 515.11 530.99 516.81 588.80 580.44 440.07
Result Confidence
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K Min: 26158 / Avg: 26636.33 / Max: 26885 Min: 29628 / Avg: 30036.67 / Max: 30252 Min: 23624 / Avg: 26142.87 / Max: 27261 Min: 26432 / Avg: 26866.67 / Max: 27409 Min: 26194 / Avg: 26288.67 / Max: 26453 Min: 26216 / Avg: 26302.67 / Max: 26467 Min: 26452 / Avg: 26723.67 / Max: 27157 Min: 26075 / Avg: 26304.67 / Max: 26502 Min: 23734 / Avg: 26185.6 / Max: 27586 Min: 26241 / Avg: 26768.83 / Max: 27891 Min: 25635 / Avg: 26045.33 / Max: 26461 Min: 25375 / Avg: 25601 / Max: 25870 Min: 23020 / Avg: 24876 / Max: 25801 Min: 18209 / Avg: 18378 / Max: 18532 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: regnety_400m EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.20, N = 3 SE +/- 0.88, N = 9 SE +/- 1.58, N = 9 SE +/- 1.44, N = 12 SE +/- 2.91, N = 3 SE +/- 1.20, N = 3 SE +/- 0.60, N = 11 SE +/- 0.89, N = 3 SE +/- 0.47, N = 3 SE +/- 0.51, N = 3 SE +/- 0.34, N = 3 SE +/- 0.04, N = 3 SE +/- 0.30, N = 3 46.30 32.16 120.47 115.48 93.95 91.90 61.29 66.90 60.77 51.09 41.35 40.27 34.24 31.16 MIN: 45.45 / MAX: 121.75 MIN: 31.34 / MAX: 32.97 MIN: 113.86 / MAX: 575.1 MIN: 104.38 / MAX: 272.84 MIN: 86.48 / MAX: 1086.99 MIN: 86.84 / MAX: 102.42 MIN: 59.46 / MAX: 72.41 MIN: 63.42 / MAX: 212.32 MIN: 58.26 / MAX: 140.56 MIN: 49.4 / MAX: 53.35 MIN: 40.02 / MAX: 44.02 MIN: 38.44 / MAX: 167.34 MIN: 33.82 / MAX: 47.56 MIN: 30.33 / MAX: 33.9 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: regnety_400m EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 46.2 / Avg: 46.3 / Max: 46.36 Min: 31.76 / Avg: 32.16 / Max: 32.37 Min: 117.06 / Avg: 120.47 / Max: 124.46 Min: 105.72 / Avg: 115.48 / Max: 120.7 Min: 87.38 / Avg: 93.95 / Max: 100.86 Min: 88.94 / Avg: 91.9 / Max: 97.72 Min: 60 / Avg: 61.29 / Max: 63.69 Min: 64.25 / Avg: 66.9 / Max: 69.82 Min: 59.06 / Avg: 60.77 / Max: 62.06 Min: 50.14 / Avg: 51.09 / Max: 51.59 Min: 40.41 / Avg: 41.35 / Max: 42.16 Min: 39.89 / Avg: 40.27 / Max: 40.94 Min: 34.18 / Avg: 34.24 / Max: 34.32 Min: 30.61 / Avg: 31.16 / Max: 31.62 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: squeezenet_ssd EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.20, N = 9 SE +/- 0.15, N = 9 SE +/- 0.09, N = 12 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 11 SE +/- 0.14, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.01, N = 3 25.20 25.76 33.77 32.09 28.68 28.94 24.52 25.94 23.73 23.02 22.45 22.85 22.62 27.97 MIN: 24.28 / MAX: 26.66 MIN: 24.63 / MAX: 28.33 MIN: 32.17 / MAX: 177.38 MIN: 30.51 / MAX: 174.53 MIN: 27.54 / MAX: 42.27 MIN: 28.1 / MAX: 41.11 MIN: 24.07 / MAX: 27.01 MIN: 25.15 / MAX: 38.37 MIN: 22.99 / MAX: 25.94 MIN: 22.47 / MAX: 24.28 MIN: 21.7 / MAX: 24.03 MIN: 21.95 / MAX: 99.38 MIN: 22.05 / MAX: 23.63 MIN: 27.46 / MAX: 42.1 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: squeezenet_ssd EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 24.91 / Avg: 25.2 / Max: 25.41 Min: 25.69 / Avg: 25.76 / Max: 25.86 Min: 33.05 / Avg: 33.77 / Max: 34.89 Min: 31.56 / Avg: 32.09 / Max: 32.7 Min: 28.28 / Avg: 28.68 / Max: 29.56 Min: 28.89 / Avg: 28.94 / Max: 28.98 Min: 24.41 / Avg: 24.52 / Max: 24.65 Min: 25.63 / Avg: 25.94 / Max: 26.21 Min: 23.45 / Avg: 23.73 / Max: 23.93 Min: 22.92 / Avg: 23.02 / Max: 23.09 Min: 22.32 / Avg: 22.45 / Max: 22.58 Min: 22.7 / Avg: 22.85 / Max: 22.98 Min: 22.44 / Avg: 22.62 / Max: 22.74 Min: 27.95 / Avg: 27.97 / Max: 27.99 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: yolov4-tiny EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.28, N = 3 SE +/- 0.29, N = 3 SE +/- 0.22, N = 9 SE +/- 0.12, N = 9 SE +/- 0.06, N = 12 SE +/- 0.10, N = 3 SE +/- 0.04, N = 3 SE +/- 0.17, N = 11 SE +/- 0.33, N = 3 SE +/- 0.02, N = 3 SE +/- 0.17, N = 3 SE +/- 0.45, N = 3 SE +/- 0.61, N = 3 SE +/- 0.50, N = 3 35.38 27.61 35.66 34.15 30.63 31.61 28.27 29.66 29.11 27.78 27.87 29.78 29.23 29.67 MIN: 34.22 / MAX: 37.33 MIN: 26.76 / MAX: 28.5 MIN: 34 / MAX: 102.9 MIN: 32.54 / MAX: 132.52 MIN: 29.81 / MAX: 126.09 MIN: 30.9 / MAX: 36.86 MIN: 27.87 / MAX: 31.14 MIN: 28.65 / MAX: 167.51 MIN: 28.33 / MAX: 32.38 MIN: 27.43 / MAX: 29.51 MIN: 27.31 / MAX: 41.48 MIN: 28.36 / MAX: 108.17 MIN: 28.35 / MAX: 32.28 MIN: 28.72 / MAX: 42.97 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: yolov4-tiny EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 34.83 / Avg: 35.38 / Max: 35.68 Min: 27.05 / Avg: 27.61 / Max: 28 Min: 34.86 / Avg: 35.66 / Max: 36.9 Min: 33.31 / Avg: 34.15 / Max: 34.59 Min: 30.33 / Avg: 30.63 / Max: 30.98 Min: 31.41 / Avg: 31.61 / Max: 31.75 Min: 28.2 / Avg: 28.27 / Max: 28.34 Min: 29.28 / Avg: 29.66 / Max: 31.16 Min: 28.7 / Avg: 29.11 / Max: 29.76 Min: 27.74 / Avg: 27.78 / Max: 27.82 Min: 27.62 / Avg: 27.87 / Max: 28.19 Min: 28.88 / Avg: 29.78 / Max: 30.25 Min: 28.56 / Avg: 29.23 / Max: 30.45 Min: 29 / Avg: 29.67 / Max: 30.66 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet50 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.33, N = 3 SE +/- 0.09, N = 3 SE +/- 0.25, N = 9 SE +/- 0.12, N = 9 SE +/- 0.19, N = 12 SE +/- 0.25, N = 3 SE +/- 0.15, N = 3 SE +/- 0.04, N = 11 SE +/- 0.26, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.33, N = 3 SE +/- 0.22, N = 3 35.47 23.75 34.51 32.73 27.83 28.57 23.55 24.62 24.32 22.79 21.88 25.17 26.83 28.19 MIN: 34.12 / MAX: 114.69 MIN: 23.29 / MAX: 24.37 MIN: 31.68 / MAX: 186.94 MIN: 31.74 / MAX: 44.34 MIN: 26.6 / MAX: 234.67 MIN: 27.64 / MAX: 33.13 MIN: 23.01 / MAX: 26.62 MIN: 23.94 / MAX: 27.83 MIN: 23.62 / MAX: 27.07 MIN: 22.48 / MAX: 33.96 MIN: 21.64 / MAX: 23.37 MIN: 24.53 / MAX: 50.53 MIN: 26.22 / MAX: 41.74 MIN: 27.81 / MAX: 29.82 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet50 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 34.82 / Avg: 35.47 / Max: 35.92 Min: 23.57 / Avg: 23.75 / Max: 23.86 Min: 33.19 / Avg: 34.51 / Max: 35.72 Min: 32.29 / Avg: 32.73 / Max: 33.15 Min: 27.17 / Avg: 27.83 / Max: 29.25 Min: 28.08 / Avg: 28.57 / Max: 28.9 Min: 23.25 / Avg: 23.55 / Max: 23.73 Min: 24.38 / Avg: 24.62 / Max: 24.86 Min: 23.81 / Avg: 24.32 / Max: 24.65 Min: 22.65 / Avg: 22.79 / Max: 22.86 Min: 21.81 / Avg: 21.88 / Max: 21.92 Min: 25.05 / Avg: 25.17 / Max: 25.35 Min: 26.36 / Avg: 26.83 / Max: 27.48 Min: 27.94 / Avg: 28.19 / Max: 28.64 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: alexnet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.14, N = 3 SE +/- 0.00, N = 3 SE +/- 0.09, N = 9 SE +/- 0.08, N = 9 SE +/- 0.07, N = 12 SE +/- 0.32, N = 3 SE +/- 0.16, N = 3 SE +/- 0.06, N = 11 SE +/- 0.16, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 2 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 9.87 7.60 9.04 8.81 8.56 9.49 8.28 7.81 8.55 7.80 7.44 10.16 9.86 9.86 MIN: 9.45 / MAX: 10.5 MIN: 7.51 / MAX: 8.13 MIN: 8.61 / MAX: 17.95 MIN: 8.44 / MAX: 32.96 MIN: 8.22 / MAX: 24.11 MIN: 8.54 / MAX: 13.63 MIN: 8 / MAX: 10.85 MIN: 7.55 / MAX: 10.47 MIN: 8.18 / MAX: 10.72 MIN: 7.66 / MAX: 9.41 MIN: 7.34 / MAX: 8.53 MIN: 9.88 / MAX: 21.62 MIN: 9.77 / MAX: 10.66 MIN: 9.77 / MAX: 25.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: alexnet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 9.7 / Avg: 9.87 / Max: 10.15 Min: 7.6 / Avg: 7.6 / Max: 7.61 Min: 8.83 / Avg: 9.04 / Max: 9.5 Min: 8.55 / Avg: 8.81 / Max: 9.42 Min: 8.41 / Avg: 8.56 / Max: 9.07 Min: 8.86 / Avg: 9.49 / Max: 9.82 Min: 8.12 / Avg: 8.28 / Max: 8.6 Min: 7.65 / Avg: 7.81 / Max: 8.24 Min: 8.38 / Avg: 8.55 / Max: 8.86 Min: 7.78 / Avg: 7.8 / Max: 7.82 Min: 7.43 / Avg: 7.44 / Max: 7.45 Min: 10.05 / Avg: 10.16 / Max: 10.31 Min: 9.82 / Avg: 9.86 / Max: 9.89 Min: 9.84 / Avg: 9.86 / Max: 9.9 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: vgg16 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 SE +/- 0.15, N = 3 SE +/- 0.18, N = 3 SE +/- 0.24, N = 9 SE +/- 0.27, N = 9 SE +/- 0.19, N = 12 SE +/- 0.32, N = 3 SE +/- 0.07, N = 3 SE +/- 0.15, N = 11 SE +/- 0.12, N = 3 SE +/- 0.01, N = 3 SE +/- 0.12, N = 3 SE +/- 0.27, N = 3 SE +/- 0.13, N = 3 SE +/- 0.05, N = 3 59.81 33.81 39.03 37.35 34.90 34.20 31.48 33.27 32.04 31.46 30.84 39.42 40.01 40.67 MIN: 57.95 / MAX: 122.99 MIN: 33.39 / MAX: 80.53 MIN: 36.68 / MAX: 159.21 MIN: 35.01 / MAX: 111.55 MIN: 33.25 / MAX: 146.34 MIN: 33.04 / MAX: 37.52 MIN: 31.08 / MAX: 45.9 MIN: 31.92 / MAX: 108.33 MIN: 31.32 / MAX: 109.6 MIN: 31.24 / MAX: 33.08 MIN: 30.45 / MAX: 32.11 MIN: 38.3 / MAX: 143.19 MIN: 39.54 / MAX: 41.05 MIN: 40.35 / MAX: 43.91 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: vgg16 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 Min: 59.51 / Avg: 59.81 / Max: 60.01 Min: 33.62 / Avg: 33.81 / Max: 34.17 Min: 38.1 / Avg: 39.03 / Max: 40.32 Min: 35.99 / Avg: 37.35 / Max: 38.19 Min: 34.19 / Avg: 34.9 / Max: 36.02 Min: 33.59 / Avg: 34.2 / Max: 34.65 Min: 31.34 / Avg: 31.48 / Max: 31.56 Min: 32.69 / Avg: 33.27 / Max: 34.04 Min: 31.88 / Avg: 32.04 / Max: 32.28 Min: 31.45 / Avg: 31.46 / Max: 31.48 Min: 30.69 / Avg: 30.84 / Max: 31.08 Min: 38.92 / Avg: 39.42 / Max: 39.85 Min: 39.75 / Avg: 40.01 / Max: 40.15 Min: 40.61 / Avg: 40.67 / Max: 40.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mobilenet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.72, N = 9 SE +/- 0.82, N = 9 SE +/- 0.64, N = 12 SE +/- 0.37, N = 3 SE +/- 0.03, N = 3 SE +/- 0.18, N = 11 SE +/- 0.15, N = 3 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.16, N = 3 SE +/- 0.10, N = 3 24.14 20.54 36.67 35.78 28.92 26.50 20.43 23.59 21.14 19.52 19.70 20.46 20.06 22.33 MIN: 23.55 / MAX: 25.73 MIN: 20.19 / MAX: 21.27 MIN: 30.26 / MAX: 274.51 MIN: 28.31 / MAX: 170.87 MIN: 24.32 / MAX: 41.88 MIN: 24.55 / MAX: 33.31 MIN: 19.91 / MAX: 33.27 MIN: 22.06 / MAX: 161.6 MIN: 20.56 / MAX: 35 MIN: 19.11 / MAX: 21.65 MIN: 19.2 / MAX: 20.45 MIN: 19.72 / MAX: 77.23 MIN: 19.5 / MAX: 23.62 MIN: 21.92 / MAX: 23.79 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mobilenet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 24 / Avg: 24.14 / Max: 24.32 Min: 20.4 / Avg: 20.54 / Max: 20.73 Min: 32.61 / Avg: 36.67 / Max: 39.06 Min: 31.86 / Avg: 35.78 / Max: 39.06 Min: 26.74 / Avg: 28.92 / Max: 33.72 Min: 25.8 / Avg: 26.5 / Max: 27.04 Min: 20.38 / Avg: 20.43 / Max: 20.48 Min: 22.94 / Avg: 23.59 / Max: 24.92 Min: 20.98 / Avg: 21.14 / Max: 21.44 Min: 19.39 / Avg: 19.52 / Max: 19.6 Min: 19.56 / Avg: 19.7 / Max: 19.85 Min: 20.36 / Avg: 20.46 / Max: 20.66 Min: 19.75 / Avg: 20.06 / Max: 20.29 Min: 22.23 / Avg: 22.33 / Max: 22.53 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Caffe This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: AlexNet - Acceleration: CPU - Iterations: 200 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40K 80K 120K 160K 200K SE +/- 1333.26, N = 3 SE +/- 101.93, N = 3 SE +/- 182.06, N = 3 SE +/- 146.69, N = 3 SE +/- 114.49, N = 3 SE +/- 1549.72, N = 7 SE +/- 120.93, N = 3 SE +/- 106.86, N = 3 SE +/- 187.42, N = 3 SE +/- 92.46, N = 3 SE +/- 37.21, N = 3 SE +/- 76.96, N = 3 SE +/- 272.29, N = 3 SE +/- 139.37, N = 3 151327 140558 128965 127699 173812 171885 162544 172508 168728 154693 152539 155763 154871 151559 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
Result Confidence
OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: AlexNet - Acceleration: CPU - Iterations: 200 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K Min: 149779 / Avg: 151326.67 / Max: 153981 Min: 140379 / Avg: 140558 / Max: 140732 Min: 128622 / Avg: 128965.33 / Max: 129242 Min: 127449 / Avg: 127699.33 / Max: 127957 Min: 173597 / Avg: 173811.67 / Max: 173988 Min: 162598 / Avg: 171885 / Max: 173740 Min: 162302 / Avg: 162543.67 / Max: 162673 Min: 172364 / Avg: 172508.33 / Max: 172717 Min: 168512 / Avg: 168727.67 / Max: 169101 Min: 154545 / Avg: 154693 / Max: 154863 Min: 152490 / Avg: 152539 / Max: 152612 Min: 155611 / Avg: 155763 / Max: 155860 Min: 154328 / Avg: 154871.33 / Max: 155175 Min: 151334 / Avg: 151559 / Max: 151814 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
asmFish This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30M 60M 90M 120M 150M SE +/- 491333.28, N = 3 SE +/- 83002.90, N = 3 SE +/- 688813.01, N = 3 SE +/- 1202983.30, N = 3 SE +/- 373542.05, N = 3 SE +/- 394553.31, N = 3 SE +/- 939330.26, N = 3 SE +/- 375595.74, N = 3 SE +/- 395352.82, N = 3 SE +/- 230890.74, N = 3 SE +/- 215295.19, N = 3 SE +/- 91373.33, N = 3 SE +/- 57670.67, N = 3 46342665 24675753 122849461 126290077 105304578 82716262 74422699 78084684 62581775 42112219 41583844 31791813 21027795
Nodes/second Per Watt
OpenBenchmarking.org Nodes/second Per Watt, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 140K 280K 420K 560K 700K 231626.04 228242.13 630429.66 665647.19 586425.64 473583.96 437974.87 513107.47 390336.76 359777.90 394905.45 333068.50 282725.36
Result Confidence
OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20M 40M 60M 80M 100M Min: 45648904 / Avg: 46342664.67 / Max: 47292245 Min: 24510321 / Avg: 24675752.67 / Max: 24770415 Min: 122057435 / Avg: 122849461.33 / Max: 124221646 Min: 124900585 / Avg: 126290076.67 / Max: 128685849 Min: 104841596 / Avg: 105304577.67 / Max: 106043845 Min: 82311763 / Avg: 82716261.67 / Max: 83505284 Min: 72577275 / Avg: 74422699 / Max: 75650094 Min: 77335639 / Avg: 78084684 / Max: 78508351 Min: 61945670 / Avg: 62581775.33 / Max: 63306570 Min: 41683109 / Avg: 42112219 / Max: 42474524 Min: 41298645 / Avg: 41583844.33 / Max: 42005822 Min: 31610049 / Avg: 31791813.33 / Max: 31899083 Min: 20912491 / Avg: 21027794.67 / Max: 21087999
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 75, Compression Effort 7 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.46, N = 3 SE +/- 0.90, N = 3 SE +/- 0.47, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 SE +/- 1.62, N = 3 SE +/- 1.08, N = 3 SE +/- 0.65, N = 3 SE +/- 0.87, N = 3 140.61 227.13 139.36 139.56 138.08 140.18 135.97 142.20 137.44 138.04 164.81 173.72 199.54 277.67 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 75, Compression Effort 7 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 140.07 / Avg: 140.61 / Max: 141.52 Min: 225.67 / Avg: 227.13 / Max: 228.77 Min: 138.86 / Avg: 139.36 / Max: 140.29 Min: 139.49 / Avg: 139.56 / Max: 139.6 Min: 138.03 / Avg: 138.08 / Max: 138.13 Min: 140 / Avg: 140.18 / Max: 140.43 Min: 135.9 / Avg: 135.97 / Max: 136.04 Min: 142.07 / Avg: 142.2 / Max: 142.38 Min: 137.39 / Avg: 137.44 / Max: 137.53 Min: 137.91 / Avg: 138.04 / Max: 138.21 Min: 161.8 / Avg: 164.81 / Max: 167.37 Min: 172.62 / Avg: 173.72 / Max: 175.87 Min: 198.32 / Avg: 199.54 / Max: 200.54 Min: 275.95 / Avg: 277.67 / Max: 278.78 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet18 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.13, N = 9 SE +/- 0.08, N = 9 SE +/- 0.50, N = 12 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 11 SE +/- 0.22, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.23, N = 3 SE +/- 0.16, N = 3 SE +/- 0.01, N = 3 15.83 11.88 16.80 16.19 15.35 14.95 13.01 13.09 13.17 12.53 12.02 14.52 14.50 14.14 MIN: 15.56 / MAX: 17.09 MIN: 11.62 / MAX: 12.46 MIN: 15.34 / MAX: 148.17 MIN: 14.96 / MAX: 24.1 MIN: 13.83 / MAX: 1062.41 MIN: 14.44 / MAX: 17.51 MIN: 12.71 / MAX: 16.73 MIN: 12.54 / MAX: 15.91 MIN: 12.72 / MAX: 15.99 MIN: 12.29 / MAX: 14.54 MIN: 11.87 / MAX: 12.25 MIN: 14.04 / MAX: 90.85 MIN: 14.03 / MAX: 30.23 MIN: 14.02 / MAX: 14.84 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet18 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 15.79 / Avg: 15.83 / Max: 15.87 Min: 11.76 / Avg: 11.88 / Max: 11.97 Min: 16.19 / Avg: 16.8 / Max: 17.25 Min: 15.8 / Avg: 16.19 / Max: 16.66 Min: 14.54 / Avg: 15.35 / Max: 20.77 Min: 14.77 / Avg: 14.95 / Max: 15.05 Min: 12.88 / Avg: 13.01 / Max: 13.19 Min: 12.94 / Avg: 13.09 / Max: 13.29 Min: 12.85 / Avg: 13.17 / Max: 13.59 Min: 12.49 / Avg: 12.53 / Max: 12.57 Min: 11.98 / Avg: 12.02 / Max: 12.05 Min: 14.26 / Avg: 14.52 / Max: 14.98 Min: 14.19 / Avg: 14.5 / Max: 14.66 Min: 14.12 / Avg: 14.14 / Max: 14.15 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: googlenet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.27, N = 3 SE +/- 0.10, N = 3 SE +/- 1.15, N = 9 SE +/- 1.29, N = 9 SE +/- 0.86, N = 12 SE +/- 0.14, N = 3 SE +/- 0.12, N = 3 SE +/- 0.29, N = 11 SE +/- 0.33, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.48, N = 3 SE +/- 0.28, N = 3 SE +/- 0.03, N = 3 21.84 16.32 36.76 34.00 27.68 23.97 19.94 21.93 20.35 17.64 16.83 18.84 17.58 17.87 MIN: 20.88 / MAX: 25.6 MIN: 15.69 / MAX: 79.67 MIN: 28.46 / MAX: 174.17 MIN: 27.04 / MAX: 97.01 MIN: 23.41 / MAX: 138.56 MIN: 23.45 / MAX: 28.72 MIN: 19.43 / MAX: 22.28 MIN: 20.32 / MAX: 95.63 MIN: 19.39 / MAX: 62.1 MIN: 17.29 / MAX: 30.94 MIN: 16.42 / MAX: 20.42 MIN: 17.66 / MAX: 132.29 MIN: 17.05 / MAX: 18.64 MIN: 17.59 / MAX: 19.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: googlenet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 21.32 / Avg: 21.84 / Max: 22.24 Min: 16.12 / Avg: 16.32 / Max: 16.47 Min: 32.1 / Avg: 36.76 / Max: 42.47 Min: 29.81 / Avg: 34 / Max: 38.69 Min: 23.94 / Avg: 27.68 / Max: 33.93 Min: 23.7 / Avg: 23.97 / Max: 24.12 Min: 19.7 / Avg: 19.94 / Max: 20.06 Min: 20.81 / Avg: 21.93 / Max: 23.79 Min: 19.9 / Avg: 20.35 / Max: 20.99 Min: 17.62 / Avg: 17.64 / Max: 17.68 Min: 16.81 / Avg: 16.83 / Max: 16.84 Min: 18.11 / Avg: 18.84 / Max: 19.74 Min: 17.28 / Avg: 17.58 / Max: 18.13 Min: 17.81 / Avg: 17.87 / Max: 17.93 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: blazeface EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.25, N = 9 SE +/- 0.37, N = 9 SE +/- 0.26, N = 12 SE +/- 0.67, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 11 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 3.82 3.30 9.32 8.68 6.80 6.57 4.61 4.95 4.60 4.21 3.72 3.74 3.54 3.49 MIN: 3.72 / MAX: 4.1 MIN: 3.2 / MAX: 3.51 MIN: 7.36 / MAX: 18.28 MIN: 6.7 / MAX: 15.99 MIN: 5.67 / MAX: 11.73 MIN: 5.73 / MAX: 18.41 MIN: 4.45 / MAX: 4.81 MIN: 4.63 / MAX: 6.53 MIN: 4.38 / MAX: 6.31 MIN: 4.06 / MAX: 5.91 MIN: 3.58 / MAX: 5.16 MIN: 3.58 / MAX: 8.88 MIN: 3.43 / MAX: 3.74 MIN: 3.4 / MAX: 3.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: blazeface EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 3.79 / Avg: 3.82 / Max: 3.83 Min: 3.26 / Avg: 3.3 / Max: 3.36 Min: 8.39 / Avg: 9.32 / Max: 10.48 Min: 6.85 / Avg: 8.68 / Max: 9.99 Min: 5.79 / Avg: 6.8 / Max: 8.57 Min: 5.85 / Avg: 6.57 / Max: 7.9 Min: 4.55 / Avg: 4.61 / Max: 4.67 Min: 4.75 / Avg: 4.95 / Max: 5.37 Min: 4.49 / Avg: 4.6 / Max: 4.66 Min: 4.19 / Avg: 4.21 / Max: 4.23 Min: 3.68 / Avg: 3.72 / Max: 3.75 Min: 3.67 / Avg: 3.74 / Max: 3.82 Min: 3.5 / Avg: 3.54 / Max: 3.58 Min: 3.46 / Avg: 3.49 / Max: 3.51 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: efficientnet-b0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 SE +/- 0.85, N = 9 SE +/- 1.16, N = 9 SE +/- 0.75, N = 12 SE +/- 2.24, N = 3 SE +/- 0.11, N = 3 SE +/- 0.26, N = 11 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 12.47 10.55 26.76 26.52 21.71 19.66 12.94 14.63 12.92 11.69 10.80 10.40 9.84 10.87 MIN: 12.02 / MAX: 14.07 MIN: 10.24 / MAX: 69.96 MIN: 19.65 / MAX: 171.03 MIN: 18.34 / MAX: 159.39 MIN: 16.06 / MAX: 33.9 MIN: 15.98 / MAX: 32.16 MIN: 12.46 / MAX: 15.4 MIN: 13.29 / MAX: 20.58 MIN: 12.36 / MAX: 16.63 MIN: 11.36 / MAX: 14.13 MIN: 10.45 / MAX: 26 MIN: 9.83 / MAX: 20.38 MIN: 9.58 / MAX: 10.75 MIN: 10.68 / MAX: 11.57 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: efficientnet-b0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 12.29 / Avg: 12.47 / Max: 12.8 Min: 10.38 / Avg: 10.55 / Max: 10.8 Min: 23.41 / Avg: 26.76 / Max: 31.76 Min: 20 / Avg: 26.52 / Max: 29.87 Min: 18.48 / Avg: 21.71 / Max: 26.84 Min: 17.26 / Avg: 19.66 / Max: 24.13 Min: 12.72 / Avg: 12.94 / Max: 13.1 Min: 13.74 / Avg: 14.63 / Max: 16.32 Min: 12.84 / Avg: 12.92 / Max: 13.06 Min: 11.58 / Avg: 11.69 / Max: 11.76 Min: 10.65 / Avg: 10.8 / Max: 10.94 Min: 10.24 / Avg: 10.4 / Max: 10.57 Min: 9.77 / Avg: 9.84 / Max: 9.88 Min: 10.82 / Avg: 10.87 / Max: 10.92 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mnasnet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.31, N = 3 SE +/- 0.12, N = 3 SE +/- 0.77, N = 9 SE +/- 1.14, N = 9 SE +/- 0.94, N = 12 SE +/- 2.35, N = 3 SE +/- 0.37, N = 3 SE +/- 0.23, N = 11 SE +/- 0.05, N = 3 SE +/- 0.04, N = 2 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 8.72 6.42 22.34 22.05 16.71 15.49 9.28 10.38 9.07 8.16 7.32 6.95 6.49 6.64 MIN: 8.12 / MAX: 10.61 MIN: 6.09 / MAX: 7.09 MIN: 14.54 / MAX: 162.38 MIN: 13.54 / MAX: 37.68 MIN: 11.37 / MAX: 30.58 MIN: 11.65 / MAX: 28.64 MIN: 8.66 / MAX: 11.37 MIN: 9.3 / MAX: 16.29 MIN: 8.75 / MAX: 11.1 MIN: 7.75 / MAX: 9.91 MIN: 7.05 / MAX: 8.22 MIN: 6.71 / MAX: 14.84 MIN: 6.37 / MAX: 6.92 MIN: 6.54 / MAX: 7.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mnasnet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 8.32 / Avg: 8.72 / Max: 9.32 Min: 6.18 / Avg: 6.42 / Max: 6.58 Min: 19.2 / Avg: 22.34 / Max: 26.98 Min: 14.87 / Avg: 22.05 / Max: 25.24 Min: 12.43 / Avg: 16.71 / Max: 22.78 Min: 12.98 / Avg: 15.49 / Max: 20.18 Min: 8.87 / Avg: 9.28 / Max: 10.01 Min: 9.65 / Avg: 10.38 / Max: 12.34 Min: 8.98 / Avg: 9.07 / Max: 9.14 Min: 8.12 / Avg: 8.16 / Max: 8.19 Min: 7.22 / Avg: 7.32 / Max: 7.41 Min: 6.9 / Avg: 6.95 / Max: 7.02 Min: 6.47 / Avg: 6.49 / Max: 6.52 Min: 6.59 / Avg: 6.64 / Max: 6.7 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: shufflenet-v2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.36, N = 3 SE +/- 0.04, N = 3 SE +/- 0.35, N = 9 SE +/- 0.61, N = 9 SE +/- 0.42, N = 12 SE +/- 0.60, N = 3 SE +/- 0.07, N = 3 SE +/- 0.07, N = 11 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 9.99 9.60 16.71 17.29 14.06 12.82 9.97 10.68 9.94 9.52 8.91 8.87 9.61 10.20 MIN: 9.5 / MAX: 11.56 MIN: 9.43 / MAX: 10.04 MIN: 14.39 / MAX: 153.63 MIN: 13.68 / MAX: 128.19 MIN: 11.71 / MAX: 20.88 MIN: 12.07 / MAX: 17.58 MIN: 9.68 / MAX: 12.72 MIN: 10.17 / MAX: 154.31 MIN: 9.64 / MAX: 11.99 MIN: 9.28 / MAX: 10.99 MIN: 8.74 / MAX: 12.61 MIN: 8.52 / MAX: 34.05 MIN: 9.51 / MAX: 13.18 MIN: 10.07 / MAX: 10.8 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: shufflenet-v2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 9.61 / Avg: 9.99 / Max: 10.71 Min: 9.56 / Avg: 9.6 / Max: 9.68 Min: 14.5 / Avg: 16.71 / Max: 17.77 Min: 13.79 / Avg: 17.29 / Max: 19.27 Min: 12.14 / Avg: 14.06 / Max: 16.52 Min: 12.17 / Avg: 12.82 / Max: 14.01 Min: 9.83 / Avg: 9.97 / Max: 10.05 Min: 10.49 / Avg: 10.68 / Max: 11.36 Min: 9.85 / Avg: 9.94 / Max: 9.98 Min: 9.47 / Avg: 9.52 / Max: 9.55 Min: 8.84 / Avg: 8.91 / Max: 8.99 Min: 8.8 / Avg: 8.87 / Max: 8.94 Min: 9.56 / Avg: 9.61 / Max: 9.65 Min: 10.18 / Avg: 10.2 / Max: 10.21 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v3-v3 - Model: mobilenet-v3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.20, N = 3 SE +/- 0.02, N = 3 SE +/- 0.75, N = 9 SE +/- 1.03, N = 9 SE +/- 0.60, N = 12 SE +/- 1.69, N = 3 SE +/- 0.19, N = 3 SE +/- 0.18, N = 11 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 8.74 6.58 20.70 20.29 15.94 14.21 9.20 10.51 9.11 8.13 7.31 6.94 6.54 6.72 MIN: 8.27 / MAX: 12.35 MIN: 6.13 / MAX: 7.08 MIN: 14.71 / MAX: 286.24 MIN: 13.64 / MAX: 155.26 MIN: 11.89 / MAX: 108.3 MIN: 11.65 / MAX: 24.12 MIN: 8.74 / MAX: 13.15 MIN: 9.5 / MAX: 16.4 MIN: 8.69 / MAX: 12.94 MIN: 7.77 / MAX: 9.9 MIN: 7.02 / MAX: 19.94 MIN: 6.68 / MAX: 16.72 MIN: 6.36 / MAX: 10.78 MIN: 6.58 / MAX: 8.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v3-v3 - Model: mobilenet-v3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 8.53 / Avg: 8.74 / Max: 9.13 Min: 6.55 / Avg: 6.58 / Max: 6.62 Min: 17.42 / Avg: 20.7 / Max: 24.59 Min: 14.9 / Avg: 20.29 / Max: 23.39 Min: 13.25 / Avg: 15.94 / Max: 19.97 Min: 12.47 / Avg: 14.21 / Max: 17.58 Min: 8.99 / Avg: 9.2 / Max: 9.58 Min: 9.94 / Avg: 10.51 / Max: 11.53 Min: 8.97 / Avg: 9.11 / Max: 9.18 Min: 8.05 / Avg: 8.13 / Max: 8.19 Min: 7.21 / Avg: 7.31 / Max: 7.46 Min: 6.9 / Avg: 6.94 / Max: 6.96 Min: 6.53 / Avg: 6.54 / Max: 6.55 Min: 6.68 / Avg: 6.72 / Max: 6.76 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v2-v2 - Model: mobilenet-v2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.27, N = 3 SE +/- 0.03, N = 3 SE +/- 0.94, N = 9 SE +/- 1.26, N = 9 SE +/- 1.02, N = 12 SE +/- 2.17, N = 3 SE +/- 0.17, N = 3 SE +/- 0.22, N = 11 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 10.11 7.13 22.98 23.01 17.78 16.05 10.15 11.55 10.07 8.41 7.85 7.88 7.32 7.61 MIN: 9.12 / MAX: 13.81 MIN: 6.91 / MAX: 7.7 MIN: 15.37 / MAX: 193.31 MIN: 14.46 / MAX: 130.77 MIN: 12.27 / MAX: 35.04 MIN: 12.41 / MAX: 30.85 MIN: 9.54 / MAX: 14.14 MIN: 10.24 / MAX: 25.89 MIN: 9.55 / MAX: 13.68 MIN: 7.96 / MAX: 10.39 MIN: 7.44 / MAX: 11.79 MIN: 7.53 / MAX: 17.78 MIN: 7.06 / MAX: 11.87 MIN: 7.45 / MAX: 8.18 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v2-v2 - Model: mobilenet-v2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 9.76 / Avg: 10.11 / Max: 10.65 Min: 7.08 / Avg: 7.13 / Max: 7.18 Min: 19.59 / Avg: 22.98 / Max: 29.47 Min: 16.68 / Avg: 23.01 / Max: 27.84 Min: 12.86 / Avg: 17.78 / Max: 24.89 Min: 13.72 / Avg: 16.05 / Max: 20.38 Min: 9.86 / Avg: 10.15 / Max: 10.44 Min: 10.7 / Avg: 11.55 / Max: 13.04 Min: 10 / Avg: 10.07 / Max: 10.13 Min: 8.33 / Avg: 8.41 / Max: 8.48 Min: 7.8 / Avg: 7.85 / Max: 7.92 Min: 7.86 / Avg: 7.88 / Max: 7.9 Min: 7.3 / Avg: 7.32 / Max: 7.34 Min: 7.59 / Avg: 7.61 / Max: 7.64 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Appleseed Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Emily EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110 220 330 440 550 226.80 381.37 154.44 153.46 153.98 154.73 161.64 172.91 170.48 194.53 260.98 275.81 333.25 487.37
YafaRay YafaRay is an open-source physically based montecarlo ray-tracing engine. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better YafaRay 3.4.1 Total Time For Sample Scene EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 1.73, N = 3 SE +/- 0.41, N = 3 SE +/- 0.72, N = 15 SE +/- 0.74, N = 3 SE +/- 0.60, N = 15 SE +/- 0.43, N = 14 SE +/- 0.43, N = 3 SE +/- 0.21, N = 3 SE +/- 0.38, N = 3 SE +/- 1.05, N = 3 SE +/- 1.37, N = 3 SE +/- 1.06, N = 3 SE +/- 0.80, N = 3 SE +/- 0.97, N = 3 127.45 172.35 67.25 66.48 67.66 64.88 64.92 85.56 70.70 80.50 110.94 102.53 131.03 186.18 1. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better YafaRay 3.4.1 Total Time For Sample Scene EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 124 / Avg: 127.45 / Max: 129.28 Min: 171.56 / Avg: 172.35 / Max: 172.91 Min: 61.8 / Avg: 67.25 / Max: 71.05 Min: 65.27 / Avg: 66.48 / Max: 67.84 Min: 64.7 / Avg: 67.66 / Max: 71.97 Min: 61.7 / Avg: 64.88 / Max: 67.09 Min: 64.15 / Avg: 64.92 / Max: 65.62 Min: 85.16 / Avg: 85.56 / Max: 85.86 Min: 69.95 / Avg: 70.7 / Max: 71.18 Min: 79.1 / Avg: 80.5 / Max: 82.54 Min: 108.85 / Avg: 110.94 / Max: 113.51 Min: 100.59 / Avg: 102.53 / Max: 104.24 Min: 129.53 / Avg: 131.03 / Max: 132.24 Min: 184.7 / Avg: 186.18 / Max: 188.01 1. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: fcn-resnet101-11 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.60, N = 3 SE +/- 0.67, N = 3 SE +/- 0.44, N = 3 SE +/- 0.33, N = 3 SE +/- 0.17, N = 3 SE +/- 1.04, N = 3 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 SE +/- 0.53, N = 12 SE +/- 0.29, N = 3 SE +/- 0.67, N = 3 SE +/- 0.29, N = 3 SE +/- 0.17, N = 3 SE +/- 0.00, N = 3 69 59 74 79 80 79 69 79 64 72 69 77 70 59 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Inferences Per Minute Per Watt
OpenBenchmarking.org Inferences Per Minute Per Watt, More Is Better ONNX Runtime 1.6 Model: fcn-resnet101-11 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1845 0.369 0.5535 0.738 0.9225 0.39 0.46 0.57 0.55 0.58 0.67 0.60 0.55 0.58 0.61 0.66 0.82 0.76 0.68
Result Confidence
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: fcn-resnet101-11 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 15 30 45 60 75 Min: 68 / Avg: 69.17 / Max: 70 Min: 58 / Avg: 59.33 / Max: 60 Min: 73.5 / Avg: 74.33 / Max: 75 Min: 78.5 / Avg: 79.17 / Max: 79.5 Min: 80 / Avg: 80.17 / Max: 80.5 Min: 77 / Avg: 79 / Max: 80.5 Min: 68 / Avg: 69 / Max: 70 Min: 78.5 / Avg: 79.17 / Max: 80.5 Min: 61 / Avg: 64 / Max: 67.5 Min: 71 / Avg: 71.5 / Max: 72 Min: 68 / Avg: 69.33 / Max: 70 Min: 76.5 / Avg: 77 / Max: 77.5 Min: 69.5 / Avg: 69.83 / Max: 70 Min: 58.5 / Avg: 58.5 / Max: 58.5 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Timed HMMer Search This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.1 Pfam Database Search EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 SE +/- 0.18, N = 3 SE +/- 0.16, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 110.43 104.26 191.23 191.15 166.11 165.55 139.82 145.01 141.65 133.76 128.93 132.39 129.44 126.21 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.1 Pfam Database Search EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 110.31 / Avg: 110.43 / Max: 110.57 Min: 104.21 / Avg: 104.26 / Max: 104.31 Min: 190.86 / Avg: 191.23 / Max: 191.42 Min: 190.84 / Avg: 191.15 / Max: 191.38 Min: 165.99 / Avg: 166.11 / Max: 166.19 Min: 165.45 / Avg: 165.55 / Max: 165.66 Min: 139.74 / Avg: 139.82 / Max: 139.89 Min: 144.87 / Avg: 145.01 / Max: 145.14 Min: 141.6 / Avg: 141.65 / Max: 141.72 Min: 133.6 / Avg: 133.76 / Max: 133.9 Min: 128.87 / Avg: 128.93 / Max: 129.01 Min: 132.35 / Avg: 132.39 / Max: 132.48 Min: 129.26 / Avg: 129.44 / Max: 129.62 Min: 126.1 / Avg: 126.21 / Max: 126.28 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.1228 2.2456 3.3684 4.4912 5.614 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.48 0.95 4.92 4.99 4.21 4.17 3.23 2.97 3.03 2.45 1.65 1.59 1.22 0.76 MIN: 1.47 / MAX: 1.49 MIN: 4.88 / MAX: 4.95 MIN: 4.95 / MAX: 5.03 MIN: 4.1 / MAX: 4.24 MIN: 4.12 / MAX: 4.18 MIN: 3.19 / MAX: 3.24 MIN: 2.93 / MAX: 2.99 MIN: 2.99 / MAX: 3.04 MIN: 2.43 / MAX: 2.46 MIN: 1.63 MIN: 1.58 / MAX: 1.6 MIN: 1.21 MAX: 0.77
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.009 0.018 0.027 0.036 0.045 0.01 0.01 0.03 0.04 0.03 0.03 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.01
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.48 / Avg: 1.48 / Max: 1.48 Min: 0.95 / Avg: 0.95 / Max: 0.95 Min: 4.9 / Avg: 4.92 / Max: 4.93 Min: 4.98 / Avg: 4.99 / Max: 5 Min: 4.2 / Avg: 4.21 / Max: 4.22 Min: 4.17 / Avg: 4.17 / Max: 4.17 Min: 3.23 / Avg: 3.23 / Max: 3.23 Min: 2.97 / Avg: 2.97 / Max: 2.97 Min: 3.03 / Avg: 3.03 / Max: 3.03 Min: 2.44 / Avg: 2.45 / Max: 2.45 Min: 1.65 / Avg: 1.65 / Max: 1.65 Min: 1.59 / Avg: 1.59 / Max: 1.59 Min: 1.22 / Avg: 1.22 / Max: 1.22 Min: 0.76 / Avg: 0.76 / Max: 0.76
GPAW GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better GPAW 20.1 Input: Carbon Nanotube EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.87, N = 3 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.26, N = 3 SE +/- 0.11, N = 3 SE +/- 0.27, N = 3 SE +/- 0.02, N = 3 SE +/- 0.20, N = 3 SE +/- 0.46, N = 3 SE +/- 0.19, N = 3 SE +/- 0.09, N = 3 SE +/- 0.37, N = 3 SE +/- 0.67, N = 3 165.41 200.41 81.06 78.39 79.98 86.74 103.30 93.24 105.67 116.41 139.15 177.91 205.89 271.16 1. (CC) gcc options: -pthread -shared -fwrapv -O2 -lxc -lblas -lmpi
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better GPAW 20.1 Input: Carbon Nanotube EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 163.66 / Avg: 165.41 / Max: 166.3 Min: 200.15 / Avg: 200.41 / Max: 200.81 Min: 80.78 / Avg: 81.06 / Max: 81.29 Min: 78.29 / Avg: 78.39 / Max: 78.44 Min: 79.56 / Avg: 79.98 / Max: 80.47 Min: 86.54 / Avg: 86.74 / Max: 86.91 Min: 102.87 / Avg: 103.3 / Max: 103.79 Min: 93.21 / Avg: 93.24 / Max: 93.28 Min: 105.35 / Avg: 105.67 / Max: 106.04 Min: 115.76 / Avg: 116.41 / Max: 117.3 Min: 138.85 / Avg: 139.15 / Max: 139.52 Min: 177.8 / Avg: 177.9 / Max: 178.09 Min: 205.17 / Avg: 205.89 / Max: 206.37 Min: 270.41 / Avg: 271.16 / Max: 272.49 1. (CC) gcc options: -pthread -shared -fwrapv -O2 -lxc -lblas -lmpi
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Leukocyte EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.52, N = 3 SE +/- 0.11, N = 3 SE +/- 0.61, N = 14 SE +/- 0.47, N = 3 SE +/- 0.47, N = 15 SE +/- 0.82, N = 15 SE +/- 0.28, N = 3 SE +/- 0.42, N = 3 SE +/- 0.26, N = 3 SE +/- 0.25, N = 3 SE +/- 1.14, N = 3 SE +/- 0.76, N = 10 SE +/- 0.53, N = 3 SE +/- 0.41, N = 3 90.52 130.98 48.72 47.37 46.53 47.43 57.68 59.44 57.98 59.65 98.19 99.72 107.97 148.20 1. (CXX) g++ options: -O2 -lOpenCL
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Leukocyte EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 89.7 / Avg: 90.52 / Max: 91.49 Min: 130.76 / Avg: 130.98 / Max: 131.14 Min: 46.16 / Avg: 48.72 / Max: 52.95 Min: 46.63 / Avg: 47.37 / Max: 48.23 Min: 44.25 / Avg: 46.53 / Max: 49.64 Min: 44.96 / Avg: 47.42 / Max: 54.58 Min: 57.12 / Avg: 57.68 / Max: 57.99 Min: 58.61 / Avg: 59.44 / Max: 59.95 Min: 57.58 / Avg: 57.98 / Max: 58.48 Min: 59.34 / Avg: 59.65 / Max: 60.14 Min: 96.7 / Avg: 98.19 / Max: 100.44 Min: 96.18 / Avg: 99.72 / Max: 103.91 Min: 106.91 / Avg: 107.97 / Max: 108.52 Min: 147.55 / Avg: 148.2 / Max: 148.96 1. (CXX) g++ options: -O2 -lOpenCL
Minion Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.12, N = 3 SE +/- 0.39, N = 3 SE +/- 0.49, N = 3 SE +/- 0.28, N = 3 SE +/- 0.18, N = 3 SE +/- 0.31, N = 3 SE +/- 0.55, N = 3 SE +/- 0.37, N = 3 SE +/- 0.46, N = 3 SE +/- 0.44, N = 3 SE +/- 0.37, N = 3 SE +/- 0.80, N = 3 SE +/- 0.38, N = 3 SE +/- 0.50, N = 3 114.91 114.95 134.93 135.90 135.09 136.07 132.23 135.33 133.23 133.73 136.52 140.13 140.43 140.53 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 114.67 / Avg: 114.91 / Max: 115.06 Min: 114.45 / Avg: 114.95 / Max: 115.72 Min: 133.97 / Avg: 134.93 / Max: 135.63 Min: 135.39 / Avg: 135.9 / Max: 136.37 Min: 134.86 / Avg: 135.09 / Max: 135.44 Min: 135.55 / Avg: 136.07 / Max: 136.63 Min: 131.17 / Avg: 132.23 / Max: 133 Min: 134.6 / Avg: 135.33 / Max: 135.75 Min: 132.34 / Avg: 133.23 / Max: 133.85 Min: 133.03 / Avg: 133.73 / Max: 134.55 Min: 135.79 / Avg: 136.52 / Max: 136.95 Min: 138.59 / Avg: 140.13 / Max: 141.26 Min: 139.94 / Avg: 140.43 / Max: 141.19 Min: 139.78 / Avg: 140.53 / Max: 141.47 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Appleseed Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Material Tester EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 192.28 247.66 188.84 187.07 185.01 182.38 173.16 186.27 173.53 173.05 191.51 177.85 210.16 278.32
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.19, N = 3 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.12, N = 3 SE +/- 0.11, N = 3 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 SE +/- 0.20, N = 3 SE +/- 0.51, N = 3 SE +/- 0.05, N = 3 125.53 243.84 54.71 54.65 63.47 64.21 79.11 87.56 85.01 103.00 147.38 157.17 200.35 296.69 1. (CXX) g++ options: -O2 -lOpenCL
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 125.35 / Avg: 125.53 / Max: 125.64 Min: 243.81 / Avg: 243.84 / Max: 243.89 Min: 54.68 / Avg: 54.71 / Max: 54.75 Min: 54.27 / Avg: 54.65 / Max: 54.89 Min: 63.31 / Avg: 63.47 / Max: 63.56 Min: 64.09 / Avg: 64.21 / Max: 64.3 Min: 79.03 / Avg: 79.11 / Max: 79.27 Min: 87.32 / Avg: 87.56 / Max: 87.75 Min: 84.8 / Avg: 85 / Max: 85.18 Min: 102.75 / Avg: 103 / Max: 103.2 Min: 147.07 / Avg: 147.38 / Max: 147.56 Min: 156.97 / Avg: 157.17 / Max: 157.56 Min: 199.81 / Avg: 200.35 / Max: 201.37 Min: 296.63 / Avg: 296.69 / Max: 296.79 1. (CXX) g++ options: -O2 -lOpenCL
PlaidML This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG19 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.26, N = 3 SE +/- 0.12, N = 3 SE +/- 0.02, N = 3 SE +/- 0.14, N = 3 SE +/- 0.04, N = 3 SE +/- 0.21, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 20.00 12.06 26.49 31.84 29.37 27.50 27.67 25.35 26.51 23.34 17.39 16.61 13.67 9.30
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG19 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.054 0.108 0.162 0.216 0.27 0.11 0.09 0.21 0.24 0.20 0.22 0.21 0.19 0.22 0.19 0.16 0.18 0.15 0.11
Result Confidence
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG19 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 19.79 / Avg: 20 / Max: 20.19 Min: 11.98 / Avg: 12.06 / Max: 12.18 Min: 26.42 / Avg: 26.49 / Max: 26.59 Min: 31.48 / Avg: 31.84 / Max: 32.35 Min: 29.23 / Avg: 29.37 / Max: 29.61 Min: 27.47 / Avg: 27.5 / Max: 27.53 Min: 27.53 / Avg: 27.67 / Max: 27.96 Min: 25.31 / Avg: 25.35 / Max: 25.42 Min: 26.11 / Avg: 26.51 / Max: 26.82 Min: 23.11 / Avg: 23.34 / Max: 23.52 Min: 17.26 / Avg: 17.39 / Max: 17.46 Min: 16.52 / Avg: 16.61 / Max: 16.72 Min: 13.44 / Avg: 13.67 / Max: 13.83 Min: 9.23 / Avg: 9.3 / Max: 9.41
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: yolov4 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 1.00, N = 3 SE +/- 1.30, N = 3 SE +/- 0.93, N = 3 SE +/- 1.01, N = 3 SE +/- 0.50, N = 3 SE +/- 0.67, N = 3 SE +/- 0.17, N = 3 SE +/- 0.88, N = 3 SE +/- 0.17, N = 3 SE +/- 0.29, N = 3 SE +/- 0.33, N = 3 SE +/- 0.29, N = 3 SE +/- 0.60, N = 3 248 262 240 243 263 264 285 271 277 280 277 267 259 243 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
Inferences Per Minute Per Watt
OpenBenchmarking.org Inferences Per Minute Per Watt, More Is Better ONNX Runtime 1.6 Model: yolov4 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.7695 1.539 2.3085 3.078 3.8475 1.71 2.55 1.55 1.58 1.75 2.01 2.57 1.98 2.49 2.60 2.95 3.28 3.26 3.42
Result Confidence
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.6 Model: yolov4 - Device: OpenMP CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 260 / Avg: 262 / Max: 263 Min: 238 / Avg: 240.33 / Max: 242.5 Min: 242 / Avg: 243.17 / Max: 245 Min: 261 / Avg: 262.83 / Max: 264.5 Min: 263 / Avg: 263.5 / Max: 264.5 Min: 284.5 / Avg: 285.17 / Max: 286.5 Min: 270.5 / Avg: 270.67 / Max: 271 Min: 275.5 / Avg: 276.83 / Max: 278.5 Min: 279.5 / Avg: 279.67 / Max: 280 Min: 276.5 / Avg: 277 / Max: 277.5 Min: 266.5 / Avg: 267.17 / Max: 267.5 Min: 258.5 / Avg: 259 / Max: 259.5 Min: 242.5 / Avg: 243.33 / Max: 244.5 1. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Exhaustive EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.12, N = 3 109.17 214.44 45.80 45.84 54.34 67.95 75.27 73.38 89.07 129.26 137.78 176.04 262.14 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Exhaustive EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 108.86 / Avg: 109.17 / Max: 109.35 Min: 214.41 / Avg: 214.44 / Max: 214.47 Min: 45.79 / Avg: 45.8 / Max: 45.82 Min: 45.83 / Avg: 45.84 / Max: 45.85 Min: 54.32 / Avg: 54.34 / Max: 54.35 Min: 67.93 / Avg: 67.95 / Max: 67.96 Min: 75.26 / Avg: 75.27 / Max: 75.27 Min: 73.37 / Avg: 73.38 / Max: 73.4 Min: 89.06 / Avg: 89.07 / Max: 89.08 Min: 129.21 / Avg: 129.26 / Max: 129.36 Min: 137.74 / Avg: 137.78 / Max: 137.82 Min: 176.03 / Avg: 176.04 / Max: 176.05 Min: 262.01 / Avg: 262.14 / Max: 262.39 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: CPU Cache EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 SE +/- 1.23, N = 12 SE +/- 0.72, N = 15 SE +/- 0.81, N = 15 SE +/- 0.36, N = 3 SE +/- 0.96, N = 12 SE +/- 0.67, N = 3 SE +/- 1.06, N = 15 SE +/- 1.28, N = 12 SE +/- 0.69, N = 12 SE +/- 0.68, N = 15 SE +/- 0.37, N = 15 SE +/- 0.77, N = 12 SE +/- 0.49, N = 13 46.39 24.85 43.70 44.36 49.78 49.19 51.39 45.23 41.11 32.72 30.12 25.17 32.53 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: CPU Cache EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1553 0.3106 0.4659 0.6212 0.7765 0.47 0.38 0.50 0.53 0.63 0.69 0.58 0.66 0.62 0.54 0.55 0.50 0.63
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: CPU Cache EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 Min: 38.66 / Avg: 46.39 / Max: 52.29 Min: 20.06 / Avg: 24.85 / Max: 30.63 Min: 38.22 / Avg: 43.7 / Max: 48.67 Min: 43.65 / Avg: 44.36 / Max: 44.75 Min: 45.03 / Avg: 49.78 / Max: 56.32 Min: 48.23 / Avg: 49.19 / Max: 50.47 Min: 45.08 / Avg: 51.39 / Max: 61.54 Min: 35.87 / Avg: 45.23 / Max: 51.94 Min: 35.35 / Avg: 41.11 / Max: 44.78 Min: 28.49 / Avg: 32.72 / Max: 38.56 Min: 28.09 / Avg: 30.12 / Max: 33.69 Min: 20.46 / Avg: 25.17 / Max: 29.16 Min: 29.83 / Avg: 32.53 / Max: 36.16 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Chaos Group V-RAY This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org vsamples, More Is Better Chaos Group V-RAY 5 Mode: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 EPYC 7232P 10K 20K 30K 40K 50K SE +/- 165.84, N = 3 SE +/- 117.08, N = 4 SE +/- 469.86, N = 3 SE +/- 392.90, N = 8 SE +/- 284.86, N = 11 SE +/- 305.20, N = 6 SE +/- 144.16, N = 3 SE +/- 298.29, N = 3 SE +/- 120.35, N = 3 SE +/- 140.41, N = 4 SE +/- 50.21, N = 3 19521 10044 45292 44658 37843 30632 28401 28448 15246 12198 7864
vsamples Per Watt
OpenBenchmarking.org vsamples Per Watt, More Is Better Chaos Group V-RAY 5 Mode: CPU EPYC 7F52 EPYC 7F32 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 60 120 180 240 300 94.82 84.74 272.08 244.21 202.52 193.49 212.71 168.70 135.50
Result Confidence
OpenBenchmarking.org vsamples, More Is Better Chaos Group V-RAY 5 Mode: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 EPYC 7232P 8K 16K 24K 32K 40K Min: 19190 / Avg: 19521 / Max: 19705 Min: 9719 / Avg: 10044.25 / Max: 10235 Min: 44695 / Avg: 45292 / Max: 46219 Min: 43124 / Avg: 44658.38 / Max: 46027 Min: 36221 / Avg: 37842.73 / Max: 38879 Min: 29652 / Avg: 30632 / Max: 31469 Min: 28141 / Avg: 28400.67 / Max: 28639 Min: 28095 / Avg: 28448 / Max: 29041 Min: 15016 / Avg: 15246.33 / Max: 15422 Min: 11788 / Avg: 12198.25 / Max: 12425 Min: 7769 / Avg: 7863.67 / Max: 7940
Blender Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Fishy Cat - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.20, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.14, N = 3 SE +/- 0.22, N = 3 SE +/- 0.12, N = 3 SE +/- 0.30, N = 3 SE +/- 0.04, N = 3 SE +/- 0.46, N = 3 SE +/- 0.08, N = 3 SE +/- 0.24, N = 3 SE +/- 0.81, N = 3 SE +/- 0.21, N = 3 108.40 207.12 55.40 55.82 61.93 62.72 72.27 78.94 78.01 91.81 129.81 139.46 177.65 267.24
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Fishy Cat - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 108.03 / Avg: 108.4 / Max: 108.72 Min: 207.06 / Avg: 207.12 / Max: 207.19 Min: 55.2 / Avg: 55.4 / Max: 55.58 Min: 55.77 / Avg: 55.82 / Max: 55.85 Min: 61.67 / Avg: 61.93 / Max: 62.15 Min: 62.45 / Avg: 62.72 / Max: 63.15 Min: 72.03 / Avg: 72.27 / Max: 72.43 Min: 78.44 / Avg: 78.94 / Max: 79.47 Min: 77.95 / Avg: 78.01 / Max: 78.08 Min: 90.99 / Avg: 91.81 / Max: 92.58 Min: 129.68 / Avg: 129.81 / Max: 129.95 Min: 139.16 / Avg: 139.46 / Max: 139.93 Min: 176.74 / Avg: 177.65 / Max: 179.26 Min: 267.03 / Avg: 267.24 / Max: 267.66
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP HotSpot3D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.58, N = 3 SE +/- 0.07, N = 3 SE +/- 0.46, N = 3 SE +/- 0.11, N = 3 SE +/- 0.98, N = 3 SE +/- 0.05, N = 3 SE +/- 0.18, N = 3 SE +/- 0.36, N = 3 SE +/- 0.50, N = 3 SE +/- 0.65, N = 3 SE +/- 0.56, N = 3 SE +/- 0.58, N = 3 SE +/- 0.54, N = 3 SE +/- 1.23, N = 5 92.08 93.08 106.58 107.88 107.70 108.25 103.80 107.48 105.24 105.28 106.77 108.86 109.60 110.50 1. (CXX) g++ options: -O2 -lOpenCL
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP HotSpot3D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 91.23 / Avg: 92.08 / Max: 93.18 Min: 92.93 / Avg: 93.08 / Max: 93.17 Min: 105.66 / Avg: 106.58 / Max: 107.16 Min: 107.69 / Avg: 107.88 / Max: 108.06 Min: 105.86 / Avg: 107.7 / Max: 109.21 Min: 108.18 / Avg: 108.25 / Max: 108.34 Min: 103.43 / Avg: 103.8 / Max: 104 Min: 106.77 / Avg: 107.48 / Max: 107.96 Min: 104.53 / Avg: 105.24 / Max: 106.2 Min: 104.57 / Avg: 105.28 / Max: 106.58 Min: 106.14 / Avg: 106.77 / Max: 107.88 Min: 108.23 / Avg: 108.86 / Max: 110.01 Min: 108.57 / Avg: 109.6 / Max: 110.38 Min: 108.22 / Avg: 110.5 / Max: 115.08 1. (CXX) g++ options: -O2 -lOpenCL
LZ4 Compression This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 33.96, N = 3 SE +/- 12.37, N = 15 SE +/- 25.22, N = 14 SE +/- 42.91, N = 3 SE +/- 28.28, N = 5 SE +/- 71.90, N = 3 SE +/- 51.28, N = 3 SE +/- 31.02, N = 3 SE +/- 36.49, N = 3 SE +/- 35.59, N = 3 SE +/- 51.98, N = 3 SE +/- 37.27, N = 3 SE +/- 28.32, N = 5 10068.5 10561.9 10219.5 10249.3 10156.5 10292.5 10188.6 10184.5 10256.3 10132.1 10110.9 10070.2 10057.8 1. (CC) gcc options: -O3
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 134.54 183.94 139.63 144.64 166.46 201.45 142.36 197.68 201.08 197.87 228.69 231.30 234.32
Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 10031 / Avg: 10068.5 / Max: 10136.3 Min: 10508.8 / Avg: 10561.89 / Max: 10609.7 Min: 10124.1 / Avg: 10219.52 / Max: 10411.5 Min: 10163.9 / Avg: 10249.27 / Max: 10299.5 Min: 10088.8 / Avg: 10156.48 / Max: 10226.8 Min: 10179.5 / Avg: 10292.47 / Max: 10426 Min: 10089.5 / Avg: 10188.6 / Max: 10261 Min: 10134.8 / Avg: 10184.5 / Max: 10241.5 Min: 10188.2 / Avg: 10256.27 / Max: 10313.1 Min: 10093.6 / Avg: 10132.1 / Max: 10203.2 Min: 10010.1 / Avg: 10110.87 / Max: 10183.4 Min: 10010.8 / Avg: 10070.17 / Max: 10138.9 Min: 10025.8 / Avg: 10057.76 / Max: 10170.9 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 SE +/- 0.42, N = 3 SE +/- 0.36, N = 15 SE +/- 0.30, N = 14 SE +/- 0.42, N = 3 SE +/- 0.48, N = 5 SE +/- 0.28, N = 3 SE +/- 0.02, N = 3 SE +/- 0.39, N = 3 SE +/- 0.34, N = 3 SE +/- 0.42, N = 3 SE +/- 0.13, N = 3 SE +/- 0.61, N = 3 SE +/- 0.44, N = 5 52.02 51.95 44.54 44.80 44.49 45.13 43.69 44.68 45.09 43.98 42.25 42.94 43.17 1. (CC) gcc options: -O3
Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 Min: 51.53 / Avg: 52.02 / Max: 52.86 Min: 49.68 / Avg: 51.95 / Max: 54.03 Min: 42.74 / Avg: 44.54 / Max: 46.27 Min: 44.34 / Avg: 44.8 / Max: 45.63 Min: 43.45 / Avg: 44.49 / Max: 45.67 Min: 44.77 / Avg: 45.13 / Max: 45.69 Min: 43.65 / Avg: 43.69 / Max: 43.71 Min: 44.24 / Avg: 44.68 / Max: 45.45 Min: 44.42 / Avg: 45.09 / Max: 45.46 Min: 43.51 / Avg: 43.98 / Max: 44.82 Min: 42.11 / Avg: 42.25 / Max: 42.52 Min: 42.28 / Avg: 42.94 / Max: 44.16 Min: 42.2 / Avg: 43.17 / Max: 44.21 1. (CC) gcc options: -O3
C-Blosc A simple, compressed, fast and persistent data store library for C. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better C-Blosc 2.0 Beta 5 Compressor: blosclz EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 27.57, N = 3 SE +/- 15.51, N = 3 SE +/- 20.77, N = 3 SE +/- 18.47, N = 3 SE +/- 9.12, N = 3 SE +/- 22.60, N = 3 SE +/- 0.44, N = 3 SE +/- 8.53, N = 3 SE +/- 6.00, N = 3 SE +/- 4.79, N = 3 SE +/- 5.79, N = 3 SE +/- 21.69, N = 3 SE +/- 6.54, N = 3 SE +/- 13.17, N = 3 10910.1 11206.8 8372.7 8328.2 9301.0 9155.0 9878.0 10253.3 9808.0 10178.3 10476.8 9400.7 9442.9 9386.8 1. (CXX) g++ options: -rdynamic
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better C-Blosc 2.0 Beta 5 Compressor: blosclz EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 109.47 165.32 77.37 83.21 93.88 104.37 138.76 106.55 135.67 145.89 160.12 165.18 173.90 182.90
Result Confidence
OpenBenchmarking.org MB/s, More Is Better C-Blosc 2.0 Beta 5 Compressor: blosclz EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 10869.2 / Avg: 10910.13 / Max: 10962.6 Min: 11190.3 / Avg: 11206.8 / Max: 11237.8 Min: 8331.9 / Avg: 8372.7 / Max: 8399.9 Min: 8294.6 / Avg: 8328.2 / Max: 8358.3 Min: 9282.8 / Avg: 9301 / Max: 9311.2 Min: 9132 / Avg: 9155 / Max: 9200.2 Min: 9877.3 / Avg: 9878 / Max: 9878.8 Min: 10242.8 / Avg: 10253.3 / Max: 10270.2 Min: 9796 / Avg: 9808 / Max: 9814.1 Min: 10170 / Avg: 10178.27 / Max: 10186.6 Min: 10468.4 / Avg: 10476.8 / Max: 10487.9 Min: 9370.8 / Avg: 9400.73 / Max: 9442.9 Min: 9434.1 / Avg: 9442.93 / Max: 9455.7 Min: 9360.9 / Avg: 9386.83 / Max: 9403.8 1. (CXX) g++ options: -rdynamic
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K SE +/- 7660.92, N = 6 SE +/- 3955.87, N = 3 SE +/- 1138.19, N = 3 SE +/- 1119.16, N = 3 SE +/- 1276.30, N = 3 SE +/- 6981.82, N = 3 SE +/- 7271.02, N = 3 SE +/- 4243.12, N = 3 SE +/- 6414.70, N = 3 SE +/- 9904.69, N = 4 SE +/- 6451.00, N = 3 SE +/- 3857.18, N = 3 SE +/- 198.51, N = 3 SE +/- 6504.38, N = 3 756676 624502 451003 447191 527541 536242 885722 825374 881986 838224 756268 775796 705855 611302 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K 4243.87 6995.00 2844.47 2634.77 3278.22 3850.07 5622.62 4999.98 6049.54 5925.99 7541.24 8813.61 9067.03 10036.15
Result Confidence
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 150K 300K 450K 600K 750K Min: 736622 / Avg: 756676 / Max: 785060 Min: 617818 / Avg: 624501.67 / Max: 631510 Min: 449363 / Avg: 451002.67 / Max: 453190 Min: 445589 / Avg: 447191.33 / Max: 449346 Min: 525500 / Avg: 527540.67 / Max: 529889 Min: 528910 / Avg: 536242.33 / Max: 550200 Min: 875249 / Avg: 885722.33 / Max: 899696 Min: 820254 / Avg: 825374 / Max: 833795 Min: 872797 / Avg: 881986 / Max: 894334 Min: 809652 / Avg: 838224 / Max: 852941 Min: 749121 / Avg: 756268 / Max: 769144 Min: 771919 / Avg: 775795.67 / Max: 783510 Min: 705589 / Avg: 705854.67 / Max: 706243 Min: 598299 / Avg: 611302 / Max: 618139 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Apache Cassandra This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 3.11.4 Test: Writes EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50K 100K 150K 200K 250K SE +/- 1803.14, N = 3 SE +/- 224.51, N = 3 SE +/- 1230.73, N = 3 SE +/- 2776.27, N = 3 SE +/- 1690.31, N = 3 SE +/- 1126.54, N = 3 SE +/- 2207.46, N = 15 SE +/- 2221.55, N = 3 SE +/- 2790.65, N = 3 SE +/- 2693.44, N = 3 SE +/- 1793.27, N = 3 SE +/- 936.51, N = 3 SE +/- 388.59, N = 3 SE +/- 244.34, N = 3 144123 58118 227564 219380 215576 230871 236524 211361 233730 200705 135260 136442 93155 51734
Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Apache Cassandra 3.11.4 Test: Writes EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 945.02 654.20 1563.40 1537.85 1507.25 1765.84 2007.98 1502.63 1934.28 1745.18 1456.94 1607.35 1222.04 837.71
Result Confidence
OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 3.11.4 Test: Writes EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40K 80K 120K 160K 200K Min: 141196 / Avg: 144123.33 / Max: 147411 Min: 57842 / Avg: 58118.33 / Max: 58563 Min: 225103 / Avg: 227564.33 / Max: 228816 Min: 214454 / Avg: 219380 / Max: 224062 Min: 212562 / Avg: 215576 / Max: 218409 Min: 229505 / Avg: 230871.33 / Max: 233106 Min: 227162 / Avg: 236524.33 / Max: 255532 Min: 208632 / Avg: 211361 / Max: 215762 Min: 228668 / Avg: 233730 / Max: 238297 Min: 197409 / Avg: 200705 / Max: 206043 Min: 131775 / Avg: 135259.67 / Max: 137737 Min: 134572 / Avg: 136442 / Max: 137469 Min: 92380 / Avg: 93154.67 / Max: 93596 Min: 51248 / Avg: 51734.33 / Max: 52019
Cpuminer-Opt Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Garlicoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 3.02, N = 3 SE +/- 8.24, N = 3 SE +/- 215.48, N = 14 SE +/- 72.85, N = 13 SE +/- 113.98, N = 14 SE +/- 49.30, N = 3 SE +/- 157.39, N = 15 SE +/- 135.96, N = 15 SE +/- 179.55, N = 12 SE +/- 8.85, N = 3 SE +/- 6.92, N = 3 SE +/- 46.80, N = 15 SE +/- 22.88, N = 15 SE +/- 24.63, N = 15 3522.44 1769.46 9581.06 9507.95 7961.64 7811.99 6242.21 5725.79 6104.27 4490.86 2965.29 2991.73 2192.82 1473.31 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Garlicoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 33.55 27.22 101.85 105.24 88.03 97.57 89.06 63.17 86.48 62.91 45.58 52.46 41.66 31.40
Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Garlicoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1700 3400 5100 6800 8500 Min: 3516.78 / Avg: 3522.44 / Max: 3527.08 Min: 1758.86 / Avg: 1769.46 / Max: 1785.69 Min: 9331.1 / Avg: 9581.06 / Max: 12370 Min: 9387.02 / Avg: 9507.95 / Max: 10330 Min: 7778.26 / Avg: 7961.64 / Max: 9440.93 Min: 7760.8 / Avg: 7811.99 / Max: 7910.57 Min: 5942.72 / Avg: 6242.21 / Max: 7690.85 Min: 5501.12 / Avg: 5725.79 / Max: 7108.86 Min: 5581 / Avg: 6104.27 / Max: 7538.77 Min: 4477.91 / Avg: 4490.86 / Max: 4507.79 Min: 2956.9 / Avg: 2965.29 / Max: 2979.02 Min: 2871.77 / Avg: 2991.73 / Max: 3335.77 Min: 2113.61 / Avg: 2192.82 / Max: 2419.18 Min: 1428.33 / Avg: 1473.31 / Max: 1776.69 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
PlaidML This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.28, N = 3 SE +/- 0.46, N = 4 SE +/- 0.45, N = 3 SE +/- 0.16, N = 3 SE +/- 0.37, N = 3 SE +/- 0.11, N = 3 SE +/- 0.19, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.22, N = 3 SE +/- 0.15, N = 3 24.29 14.67 31.89 37.38 35.25 32.68 33.33 30.36 32.05 27.81 21.24 20.34 16.79 11.41
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.063 0.126 0.189 0.252 0.315 0.13 0.11 0.26 0.28 0.25 0.27 0.27 0.23 0.28 0.23 0.20 0.23 0.19 0.14
Result Confidence
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 24.18 / Avg: 24.29 / Max: 24.46 Min: 14.6 / Avg: 14.67 / Max: 14.82 Min: 31.34 / Avg: 31.89 / Max: 32.28 Min: 36.33 / Avg: 37.38 / Max: 38.57 Min: 34.41 / Avg: 35.25 / Max: 35.93 Min: 32.38 / Avg: 32.68 / Max: 32.94 Min: 32.65 / Avg: 33.33 / Max: 33.94 Min: 30.21 / Avg: 30.36 / Max: 30.58 Min: 31.69 / Avg: 32.05 / Max: 32.3 Min: 27.57 / Avg: 27.81 / Max: 27.96 Min: 21.13 / Avg: 21.24 / Max: 21.44 Min: 20.24 / Avg: 20.34 / Max: 20.43 Min: 16.5 / Avg: 16.79 / Max: 17.22 Min: 11.13 / Avg: 11.41 / Max: 11.63
Cpuminer-Opt Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Deepcoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10K 20K 30K 40K 50K SE +/- 333.56, N = 15 SE +/- 73.09, N = 6 SE +/- 2134.76, N = 15 SE +/- 1295.28, N = 15 SE +/- 1521.20, N = 15 SE +/- 964.94, N = 15 SE +/- 991.42, N = 15 SE +/- 38.44, N = 3 SE +/- 1900.37, N = 15 SE +/- 94.04, N = 3 SE +/- 14.53, N = 3 SE +/- 407.34, N = 15 SE +/- 78.72, N = 3 SE +/- 0.23, N = 3 15338.00 7644.10 46113.00 45051.00 34363.00 33589.00 26493.00 23013.00 26546.00 19363.00 12697.00 12793.00 9361.06 6207.89 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Deepcoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 120 240 360 480 600 175.98 128.88 562.57 565.67 432.56 474.77 438.65 286.60 441.39 318.51 224.80 255.88 201.02 140.99
Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Deepcoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8K 16K 24K 32K 40K Min: 14690 / Avg: 15338 / Max: 19990 Min: 7535.18 / Avg: 7644.1 / Max: 7997.88 Min: 42730 / Avg: 46112.67 / Max: 75810 Min: 42590 / Avg: 45050.67 / Max: 63100 Min: 32480 / Avg: 34363.33 / Max: 55610 Min: 32160 / Avg: 33588.67 / Max: 46960 Min: 25310 / Avg: 26492.67 / Max: 40340 Min: 22940 / Avg: 23013.33 / Max: 23070 Min: 23580 / Avg: 26546 / Max: 44860 Min: 19250 / Avg: 19363.33 / Max: 19550 Min: 12670 / Avg: 12696.67 / Max: 12720 Min: 12330 / Avg: 12793.33 / Max: 18480 Min: 9269.8 / Avg: 9361.06 / Max: 9517.79 Min: 6207.47 / Avg: 6207.89 / Max: 6208.24 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
dav1d Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p 10-bit EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 SE +/- 0.31, N = 3 SE +/- 0.56, N = 3 SE +/- 0.28, N = 3 SE +/- 0.12, N = 3 SE +/- 0.16, N = 3 SE +/- 0.13, N = 3 SE +/- 0.08, N = 3 SE +/- 0.20, N = 3 SE +/- 0.09, N = 3 SE +/- 0.38, N = 3 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 129.77 109.55 185.31 190.19 179.72 178.70 152.78 146.29 152.15 135.69 116.68 114.49 107.25 94.36 MIN: 84.35 / MAX: 268.95 MIN: 70.77 / MAX: 249.84 MIN: 125.96 / MAX: 294.64 MIN: 126.93 / MAX: 307.7 MIN: 121.84 / MAX: 276.81 MIN: 121.07 / MAX: 277.42 MIN: 98.52 / MAX: 278.09 MIN: 94.75 / MAX: 256.37 MIN: 97.46 / MAX: 272.72 MIN: 85.88 / MAX: 272.31 MIN: 74.3 / MAX: 261.93 MIN: 72.02 / MAX: 259.87 MIN: 67.93 / MAX: 243.83 MIN: 60.14 / MAX: 225.56 1. (CC) gcc options: -pthread
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p 10-bit EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.5828 1.1656 1.7484 2.3312 2.914 1.47 1.62 2.22 2.37 2.34 2.59 2.58 1.84 2.53 2.26 2.00 2.29 2.15 1.95
Result Confidence
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p 10-bit EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 129.43 / Avg: 129.77 / Max: 129.97 Min: 109.41 / Avg: 109.55 / Max: 109.8 Min: 184.71 / Avg: 185.31 / Max: 185.73 Min: 189.1 / Avg: 190.19 / Max: 190.97 Min: 179.44 / Avg: 179.72 / Max: 180.27 Min: 178.47 / Avg: 178.7 / Max: 178.84 Min: 152.47 / Avg: 152.78 / Max: 153.01 Min: 146.03 / Avg: 146.29 / Max: 146.46 Min: 152.02 / Avg: 152.15 / Max: 152.28 Min: 135.36 / Avg: 135.69 / Max: 136.06 Min: 116.54 / Avg: 116.68 / Max: 116.86 Min: 113.88 / Avg: 114.49 / Max: 115.2 Min: 106.92 / Avg: 107.25 / Max: 107.46 Min: 94.18 / Avg: 94.36 / Max: 94.61 1. (CC) gcc options: -pthread
Numenta Anomaly Benchmark Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Earthgecko Skyline EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.59, N = 3 SE +/- 0.92, N = 7 SE +/- 0.05, N = 3 SE +/- 0.45, N = 3 SE +/- 0.18, N = 3 SE +/- 1.10, N = 3 SE +/- 0.80, N = 3 SE +/- 0.65, N = 3 SE +/- 0.60, N = 3 SE +/- 0.58, N = 3 SE +/- 0.45, N = 3 SE +/- 1.07, N = 3 SE +/- 0.64, N = 3 SE +/- 1.44, N = 3 75.23 98.52 84.83 84.79 85.03 86.06 83.83 86.70 85.73 85.76 88.32 96.85 101.59 141.25
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Earthgecko Skyline EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 74.45 / Avg: 75.23 / Max: 76.39 Min: 95.01 / Avg: 98.52 / Max: 100.63 Min: 84.74 / Avg: 84.83 / Max: 84.9 Min: 84.12 / Avg: 84.79 / Max: 85.64 Min: 84.8 / Avg: 85.03 / Max: 85.39 Min: 83.86 / Avg: 86.06 / Max: 87.25 Min: 82.24 / Avg: 83.83 / Max: 84.72 Min: 85.78 / Avg: 86.7 / Max: 87.96 Min: 85.04 / Avg: 85.73 / Max: 86.93 Min: 84.6 / Avg: 85.76 / Max: 86.43 Min: 87.53 / Avg: 88.32 / Max: 89.11 Min: 94.95 / Avg: 96.85 / Max: 98.65 Min: 100.3 / Avg: 101.59 / Max: 102.33 Min: 138.42 / Avg: 141.25 / Max: 143.09
TensorFlow Lite This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: NASNet Mobile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50K 100K 150K 200K 250K SE +/- 244.85, N = 3 SE +/- 54.60, N = 3 SE +/- 718.53, N = 15 SE +/- 833.99, N = 6 SE +/- 34.51, N = 3 SE +/- 275.99, N = 3 SE +/- 115.93, N = 3 SE +/- 115.78, N = 3 SE +/- 797.17, N = 15 SE +/- 414.30, N = 3 SE +/- 652.17, N = 3 SE +/- 183.23, N = 3 SE +/- 219.52, N = 3 SE +/- 42.10, N = 3 126743.0 177928.0 95873.7 88093.7 86140.6 92153.2 79313.6 84263.5 85462.8 105046.0 144445.0 151235.0 177540.0 212253.0
Result Confidence
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: NASNet Mobile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40K 80K 120K 160K 200K Min: 126253 / Avg: 126742.67 / Max: 126993 Min: 177860 / Avg: 177928 / Max: 178036 Min: 91993.6 / Avg: 95873.69 / Max: 100262 Min: 85520.1 / Avg: 88093.72 / Max: 91636.7 Min: 86071.8 / Avg: 86140.57 / Max: 86180.1 Min: 91707.3 / Avg: 92153.17 / Max: 92657.9 Min: 79189.4 / Avg: 79313.63 / Max: 79545.3 Min: 84034.6 / Avg: 84263.47 / Max: 84408.4 Min: 82675.3 / Avg: 85462.75 / Max: 91289.3 Min: 104457 / Avg: 105045.67 / Max: 105845 Min: 143270 / Avg: 144444.67 / Max: 145523 Min: 150900 / Avg: 151235.33 / Max: 151531 Min: 177299 / Avg: 177539.67 / Max: 177978 Min: 212188 / Avg: 212253.33 / Max: 212332
FinanceBench FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Bonds OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 395.61, N = 3 SE +/- 185.79, N = 3 SE +/- 232.73, N = 3 SE +/- 966.88, N = 3 SE +/- 75.52, N = 3 SE +/- 26.41, N = 3 SE +/- 287.97, N = 3 SE +/- 294.03, N = 3 SE +/- 397.99, N = 3 SE +/- 461.99, N = 3 SE +/- 402.31, N = 3 SE +/- 203.38, N = 3 SE +/- 200.17, N = 3 76410.11 76027.98 88553.78 90672.45 90310.90 86946.61 90157.28 88487.57 88634.31 89863.26 92896.62 92387.25 92426.49 1. (CXX) g++ options: -O3 -march=native -fopenmp
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Bonds OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16K 32K 48K 64K 80K Min: 75619.2 / Avg: 76410.11 / Max: 76824.91 Min: 75656.88 / Avg: 76027.98 / Max: 76230.05 Min: 88306.55 / Avg: 88553.78 / Max: 89018.95 Min: 89694.92 / Avg: 90672.45 / Max: 92606.17 Min: 90159.87 / Avg: 90310.9 / Max: 90387.48 Min: 86896.07 / Avg: 86946.61 / Max: 86985.19 Min: 89581.38 / Avg: 90157.28 / Max: 90451.57 Min: 88176.03 / Avg: 88487.57 / Max: 89075.27 Min: 88093.03 / Avg: 88634.31 / Max: 89410.36 Min: 89400.2 / Avg: 89863.26 / Max: 90787.24 Min: 92266.52 / Avg: 92896.62 / Max: 93645.02 Min: 92154.74 / Avg: 92387.25 / Max: 92792.54 Min: 92207.57 / Avg: 92426.49 / Max: 92826.23 1. (CXX) g++ options: -O3 -march=native -fopenmp
JPEG XL The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance. Learn more via the OpenBenchmarking.org test page.
Timed Eigen Compilation This test times how long it takes to build all Eigen examples. The Eigen examples are compiled serially. Eigen is a C++ template library for linear algebra. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Eigen Compilation 3.3.9 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 83.02 83.50 96.09 97.55 97.99 97.22 94.97 97.37 95.81 94.84 97.20 99.92 99.84 103.98
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Eigen Compilation 3.3.9 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 82.99 / Avg: 83.02 / Max: 83.06 Min: 83.48 / Avg: 83.5 / Max: 83.55 Min: 96 / Avg: 96.09 / Max: 96.18 Min: 97.52 / Avg: 97.55 / Max: 97.6 Min: 97.94 / Avg: 97.99 / Max: 98.06 Min: 97.2 / Avg: 97.22 / Max: 97.24 Min: 94.92 / Avg: 94.97 / Max: 95.02 Min: 97.37 / Avg: 97.37 / Max: 97.38 Min: 95.78 / Avg: 95.81 / Max: 95.83 Min: 94.73 / Avg: 94.83 / Max: 94.91 Min: 97.15 / Avg: 97.2 / Max: 97.27 Min: 99.9 / Avg: 99.92 / Max: 99.96 Min: 99.78 / Avg: 99.84 / Max: 99.91 Min: 103.94 / Avg: 103.98 / Max: 104.04
Chaos Group V-RAY This is a test of Chaos Group's V-RAY benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Ksamples, More Is Better Chaos Group V-RAY 4.10.07 Mode: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13K 26K 39K 52K 65K SE +/- 71.92, N = 3 SE +/- 31.21, N = 3 SE +/- 487.40, N = 3 SE +/- 467.21, N = 3 SE +/- 179.29, N = 3 SE +/- 328.54, N = 3 SE +/- 391.02, N = 6 SE +/- 397.54, N = 5 SE +/- 264.62, N = 15 SE +/- 87.62, N = 3 SE +/- 114.11, N = 3 SE +/- 122.98, N = 3 SE +/- 80.88, N = 3 SE +/- 44.44, N = 3 27132 14204 62754 62295 54445 53570 38904 36223 37085 33184 23486 22022 17550 11136
Ksamples Per Watt
OpenBenchmarking.org Ksamples Per Watt, More Is Better Chaos Group V-RAY 4.10.07 Mode: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 127.66 124.47 353.18 364.02 306.53 328.86 257.36 243.18 275.92 235.77 202.24 233.63 185.79 152.41
Result Confidence
OpenBenchmarking.org Ksamples, More Is Better Chaos Group V-RAY 4.10.07 Mode: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11K 22K 33K 44K 55K Min: 27010 / Avg: 27132 / Max: 27259 Min: 14151 / Avg: 14203.67 / Max: 14259 Min: 62183 / Avg: 62754.33 / Max: 63724 Min: 61378 / Avg: 62294.67 / Max: 62910 Min: 54235 / Avg: 54445.33 / Max: 54802 Min: 53078 / Avg: 53569.67 / Max: 54193 Min: 37112 / Avg: 38903.83 / Max: 39603 Min: 35373 / Avg: 36223.2 / Max: 37595 Min: 35376 / Avg: 37084.67 / Max: 38925 Min: 33018 / Avg: 33183.67 / Max: 33316 Min: 23305 / Avg: 23486.33 / Max: 23697 Min: 21875 / Avg: 22021.67 / Max: 22266 Min: 17397 / Avg: 17550 / Max: 17672 Min: 11048 / Avg: 11136.33 / Max: 11189
Timed MrBayes Analysis This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.24, N = 3 SE +/- 0.18, N = 3 SE +/- 0.25, N = 3 SE +/- 1.02, N = 3 SE +/- 0.17, N = 3 SE +/- 1.17, N = 4 SE +/- 0.07, N = 3 SE +/- 0.34, N = 3 SE +/- 0.21, N = 3 SE +/- 0.29, N = 3 SE +/- 0.21, N = 3 SE +/- 0.16, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 72.43 72.89 115.27 116.15 102.45 104.52 86.72 96.81 94.26 86.00 85.93 89.20 87.60 88.67 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 72.17 / Avg: 72.43 / Max: 72.9 Min: 72.56 / Avg: 72.89 / Max: 73.16 Min: 114.96 / Avg: 115.27 / Max: 115.77 Min: 115.11 / Avg: 116.15 / Max: 118.18 Min: 102.2 / Avg: 102.45 / Max: 102.77 Min: 103.18 / Avg: 104.52 / Max: 108.01 Min: 86.61 / Avg: 86.72 / Max: 86.86 Min: 96.44 / Avg: 96.81 / Max: 97.49 Min: 93.87 / Avg: 94.26 / Max: 94.59 Min: 85.64 / Avg: 86 / Max: 86.58 Min: 85.7 / Avg: 85.93 / Max: 86.35 Min: 88.88 / Avg: 89.2 / Max: 89.39 Min: 87.51 / Avg: 87.6 / Max: 87.67 Min: 88.6 / Avg: 88.67 / Max: 88.81 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 9.06, N = 3 SE +/- 9.36, N = 3 SE +/- 8.46, N = 3 SE +/- 6.05, N = 3 SE +/- 5.63, N = 3 SE +/- 1.11, N = 3 SE +/- 4.32, N = 3 SE +/- 1.11, N = 3 SE +/- 3.28, N = 3 SE +/- 1.25, N = 3 SE +/- 7.84, N = 3 SE +/- 0.03, N = 3 1410.59 705.21 3989.79 4019.09 3322.99 3292.84 2389.76 2333.54 2380.27 1813.29 1190.48 1156.26 860.01 579.18 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 8.30 7.56 23.49 24.18 18.05 20.64 17.20 14.32 16.63 14.01 12.24 13.47 11.07 9.24
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700 1400 2100 2800 3500 Min: 1410.49 / Avg: 1410.59 / Max: 1410.66 Min: 705.16 / Avg: 705.21 / Max: 705.27 Min: 3972.06 / Avg: 3989.79 / Max: 4001.86 Min: 4001.46 / Avg: 4019.09 / Max: 4033.33 Min: 3306.1 / Avg: 3322.99 / Max: 3332.1 Min: 3282.03 / Avg: 3292.84 / Max: 3302.95 Min: 2378.57 / Avg: 2389.76 / Max: 2396.35 Min: 2331.32 / Avg: 2333.54 / Max: 2334.74 Min: 2371.63 / Avg: 2380.27 / Max: 2384.67 Min: 1811.43 / Avg: 1813.29 / Max: 1815.27 Min: 1183.95 / Avg: 1190.48 / Max: 1194.25 Min: 1153.76 / Avg: 1156.26 / Max: 1157.7 Min: 844.33 / Avg: 860.01 / Max: 868.26 Min: 579.13 / Avg: 579.18 / Max: 579.23 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 3.26, N = 3 SE +/- 5.81, N = 3 SE +/- 18.50, N = 3 SE +/- 4.98, N = 3 SE +/- 10.51, N = 3 SE +/- 2.76, N = 3 SE +/- 27.37, N = 3 SE +/- 16.71, N = 3 SE +/- 18.97, N = 3 SE +/- 19.18, N = 3 SE +/- 16.74, N = 3 SE +/- 6.09, N = 3 SE +/- 7.74, N = 3 8872.56 9215.24 9048.74 9073.57 9022.45 9117.75 8963.64 9078.88 9012.11 8995.85 8945.24 8970.77 8983.76 1. (CC) gcc options: -O3 -march=native
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 108.64 149.02 111.88 117.41 134.77 160.00 112.76 158.03 159.14 154.41 179.48 183.37 185.90
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1600 3200 4800 6400 8000 Min: 8866.67 / Avg: 8872.56 / Max: 8877.91 Min: 9203.97 / Avg: 9215.24 / Max: 9223.37 Min: 9023.71 / Avg: 9048.74 / Max: 9084.85 Min: 9065.01 / Avg: 9073.56 / Max: 9082.25 Min: 9010.61 / Avg: 9022.45 / Max: 9043.42 Min: 9112.63 / Avg: 9117.75 / Max: 9122.1 Min: 8910.42 / Avg: 8963.64 / Max: 9001.37 Min: 9045.61 / Avg: 9078.88 / Max: 9098.27 Min: 8974.68 / Avg: 9012.11 / Max: 9036.23 Min: 8971.17 / Avg: 8995.85 / Max: 9033.63 Min: 8912.98 / Avg: 8945.24 / Max: 8969.13 Min: 8960.57 / Avg: 8970.77 / Max: 8981.63 Min: 8971.61 / Avg: 8983.76 / Max: 8998.15 1. (CC) gcc options: -O3 -march=native
Apache CouchDB This is a bulk insertion benchmark of Apache CouchDB. CouchDB is a document-oriented NoSQL database implemented in Erlang. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Apache CouchDB 3.1.1 Bulk Size: 100 - Inserts: 1000 - Rounds: 24 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.45, N = 3 SE +/- 0.48, N = 3 SE +/- 0.55, N = 3 SE +/- 0.34, N = 3 SE +/- 0.03, N = 3 SE +/- 0.35, N = 3 SE +/- 0.45, N = 3 SE +/- 0.64, N = 3 SE +/- 1.14, N = 3 SE +/- 0.52, N = 3 SE +/- 0.51, N = 3 SE +/- 0.89, N = 3 SE +/- 1.02, N = 4 SE +/- 0.41, N = 3 83.28 93.79 97.95 97.30 91.51 92.59 86.57 91.32 86.23 86.68 89.43 87.94 92.35 107.01 1. (CXX) g++ options: -std=c++14 -lmozjs-68 -lm -lerl_interface -lei -fPIC -MMD
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Apache CouchDB 3.1.1 Bulk Size: 100 - Inserts: 1000 - Rounds: 24 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 82.43 / Avg: 83.28 / Max: 83.97 Min: 92.83 / Avg: 93.79 / Max: 94.31 Min: 96.92 / Avg: 97.95 / Max: 98.78 Min: 96.62 / Avg: 97.29 / Max: 97.72 Min: 91.47 / Avg: 91.51 / Max: 91.55 Min: 91.9 / Avg: 92.59 / Max: 93.05 Min: 86.01 / Avg: 86.57 / Max: 87.47 Min: 90.24 / Avg: 91.31 / Max: 92.46 Min: 84.36 / Avg: 86.23 / Max: 88.29 Min: 85.65 / Avg: 86.68 / Max: 87.25 Min: 88.41 / Avg: 89.43 / Max: 89.96 Min: 86.34 / Avg: 87.94 / Max: 89.41 Min: 90.47 / Avg: 92.35 / Max: 94.9 Min: 106.25 / Avg: 107.01 / Max: 107.66 1. (CXX) g++ options: -std=c++14 -lmozjs-68 -lm -lerl_interface -lei -fPIC -MMD
Montage Astronomical Image Mosaic Engine Montage is an open-source astronomical image mosaic engine. This BSD-licensed astronomy software is developed by the California Institute of Technology, Pasadena. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Montage Astronomical Image Mosaic Engine 6.0 Mosaic of M17, K band, 1.5 deg x 1.5 deg EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.14, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.19, N = 3 SE +/- 0.28, N = 3 SE +/- 0.31, N = 3 SE +/- 0.08, N = 3 SE +/- 0.17, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.12, N = 3 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 80.94 80.85 94.76 95.56 95.39 95.53 92.77 95.48 93.86 93.92 95.45 98.24 98.31 98.41 1. (CC) gcc options: -std=gnu99 -lcfitsio -lm -O2
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Montage Astronomical Image Mosaic Engine 6.0 Mosaic of M17, K band, 1.5 deg x 1.5 deg EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 80.67 / Avg: 80.94 / Max: 81.13 Min: 80.76 / Avg: 80.85 / Max: 80.91 Min: 94.65 / Avg: 94.76 / Max: 94.97 Min: 95.43 / Avg: 95.56 / Max: 95.8 Min: 95.12 / Avg: 95.39 / Max: 95.75 Min: 95.11 / Avg: 95.53 / Max: 96.06 Min: 92.29 / Avg: 92.77 / Max: 93.36 Min: 95.35 / Avg: 95.48 / Max: 95.62 Min: 93.53 / Avg: 93.86 / Max: 94.1 Min: 93.8 / Avg: 93.92 / Max: 94.11 Min: 95.27 / Avg: 95.45 / Max: 95.58 Min: 98.08 / Avg: 98.24 / Max: 98.48 Min: 98.25 / Avg: 98.31 / Max: 98.39 Min: 98.12 / Avg: 98.41 / Max: 98.67 1. (CC) gcc options: -std=gnu99 -lcfitsio -lm -O2
Timed GDB GNU Debugger Compilation This test times how long it takes to build the GNU Debugger (GDB) in a default configuration. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed GDB GNU Debugger Compilation 9.1 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.11, N = 3 82.92 94.01 86.37 87.02 87.21 86.95 84.83 88.42 85.96 86.97 92.26 94.48 99.22 114.83
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed GDB GNU Debugger Compilation 9.1 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 82.8 / Avg: 82.92 / Max: 83.07 Min: 93.84 / Avg: 94.01 / Max: 94.13 Min: 86.23 / Avg: 86.37 / Max: 86.48 Min: 86.98 / Avg: 87.02 / Max: 87.09 Min: 87.2 / Avg: 87.21 / Max: 87.22 Min: 86.95 / Avg: 86.95 / Max: 86.96 Min: 84.74 / Avg: 84.83 / Max: 84.97 Min: 88.21 / Avg: 88.42 / Max: 88.6 Min: 85.93 / Avg: 85.96 / Max: 86.02 Min: 86.94 / Avg: 86.97 / Max: 87.03 Min: 92.07 / Avg: 92.26 / Max: 92.37 Min: 94.41 / Avg: 94.48 / Max: 94.59 Min: 99.11 / Avg: 99.22 / Max: 99.3 Min: 114.63 / Avg: 114.83 / Max: 115
InfluxDB This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K SE +/- 771.81, N = 3 SE +/- 2229.96, N = 3 SE +/- 1936.89, N = 3 SE +/- 2041.67, N = 3 SE +/- 472.72, N = 3 SE +/- 2507.53, N = 3 SE +/- 577.19, N = 3 SE +/- 2350.83, N = 3 SE +/- 1998.87, N = 3 SE +/- 607.09, N = 3 SE +/- 842.53, N = 3 SE +/- 1258.02, N = 3 SE +/- 1165.03, N = 3 SE +/- 1045.78, N = 3 1247037.6 1155871.8 1212918.2 1209698.0 1215056.1 1208480.8 1162690.9 1197778.9 1262530.4 1248284.1 1188387.2 1172021.1 1143360.5 1041996.8
val/sec Per Watt
OpenBenchmarking.org val/sec Per Watt, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K 11716.29 15038.78 12097.16 12804.70 12867.29 14523.73 16668.21 12972.75 17421.85 17298.69 17671.85 19191.04 19247.10 19005.43
Result Confidence
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K Min: 1245534.1 / Avg: 1247037.63 / Max: 1248092.1 Min: 1152601.1 / Avg: 1155871.8 / Max: 1160133 Min: 1209097.9 / Avg: 1212918.2 / Max: 1215383.9 Min: 1206399.7 / Avg: 1209698.03 / Max: 1213431.9 Min: 1214294.3 / Avg: 1215056.1 / Max: 1215921.9 Min: 1203466.2 / Avg: 1208480.8 / Max: 1211047.5 Min: 1161606.1 / Avg: 1162690.93 / Max: 1163575.1 Min: 1194820.8 / Avg: 1197778.87 / Max: 1202422.8 Min: 1259246.9 / Avg: 1262530.43 / Max: 1266147.1 Min: 1247624.2 / Avg: 1248284.1 / Max: 1249496.7 Min: 1187003 / Avg: 1188387.17 / Max: 1189911.5 Min: 1169540.4 / Avg: 1172021.13 / Max: 1173625.2 Min: 1141134.6 / Avg: 1143360.47 / Max: 1145070.1 Min: 1040104.9 / Avg: 1041996.77 / Max: 1043715.1
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 2.34 1.27 5.92 6.00 5.15 5.10 4.04 3.70 3.75 3.07 2.15 1.98 1.55 0.99 MIN: 2.29 / MAX: 2.38 MIN: 1.26 / MAX: 1.28 MIN: 5.85 / MAX: 5.99 MIN: 5.92 / MAX: 6.06 MIN: 5.08 / MAX: 5.21 MIN: 5.05 / MAX: 5.15 MIN: 3.98 / MAX: 4.07 MIN: 3.64 / MAX: 3.73 MIN: 3.72 / MAX: 3.77 MIN: 3.02 / MAX: 3.11 MIN: 2.13 / MAX: 2.16 MIN: 1.96 / MAX: 1.99 MIN: 1.54 / MAX: 1.57 MAX: 1
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.009 0.018 0.027 0.036 0.045 0.01 0.01 0.03 0.04 0.03 0.03 0.02 0.02 0.03 0.02 0.02 0.02 0.02 0.01
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 2.33 / Avg: 2.34 / Max: 2.35 Min: 1.27 / Avg: 1.27 / Max: 1.27 Min: 5.92 / Avg: 5.92 / Max: 5.92 Min: 5.99 / Avg: 6 / Max: 6.02 Min: 5.13 / Avg: 5.15 / Max: 5.15 Min: 5.1 / Avg: 5.1 / Max: 5.1 Min: 4.03 / Avg: 4.04 / Max: 4.05 Min: 3.7 / Avg: 3.7 / Max: 3.7 Min: 3.75 / Avg: 3.75 / Max: 3.75 Min: 3.06 / Avg: 3.07 / Max: 3.08 Min: 2.15 / Avg: 2.15 / Max: 2.15 Min: 1.98 / Avg: 1.98 / Max: 1.98 Min: 1.55 / Avg: 1.55 / Max: 1.55 Min: 0.99 / Avg: 0.99 / Max: 0.99
Timed Godot Game Engine Compilation This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 3.2.3 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 0.32, N = 3 SE +/- 0.21, N = 3 SE +/- 0.31, N = 3 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.17, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.17, N = 3 SE +/- 0.12, N = 3 SE +/- 0.07, N = 3 SE +/- 0.26, N = 3 SE +/- 0.14, N = 3 SE +/- 0.30, N = 3 84.21 136.62 62.56 62.27 61.81 62.48 66.57 69.62 69.42 76.16 96.77 103.74 122.41 172.13
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 3.2.3 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 83.58 / Avg: 84.21 / Max: 84.54 Min: 136.37 / Avg: 136.62 / Max: 137.03 Min: 62.22 / Avg: 62.56 / Max: 63.19 Min: 61.99 / Avg: 62.27 / Max: 62.58 Min: 61.69 / Avg: 61.81 / Max: 61.9 Min: 62.24 / Avg: 62.48 / Max: 62.81 Min: 66.45 / Avg: 66.57 / Max: 66.76 Min: 69.45 / Avg: 69.62 / Max: 69.79 Min: 69.16 / Avg: 69.42 / Max: 69.74 Min: 75.93 / Avg: 76.16 / Max: 76.36 Min: 96.64 / Avg: 96.77 / Max: 96.88 Min: 103.26 / Avg: 103.74 / Max: 104.13 Min: 122.19 / Avg: 122.41 / Max: 122.67 Min: 171.71 / Avg: 172.13 / Max: 172.7
Blender Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: BMW27 - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 0.27, N = 3 SE +/- 0.25, N = 3 SE +/- 0.12, N = 3 SE +/- 0.27, N = 3 SE +/- 0.19, N = 3 SE +/- 0.42, N = 3 SE +/- 0.06, N = 3 SE +/- 0.21, N = 3 SE +/- 0.37, N = 3 SE +/- 0.08, N = 3 SE +/- 0.95, N = 3 SE +/- 0.34, N = 3 SE +/- 0.34, N = 3 SE +/- 1.01, N = 3 83.45 156.31 40.07 40.33 45.78 46.59 55.32 61.25 60.75 69.84 102.29 108.34 136.20 198.45
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: BMW27 - Compute: CPU-Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 83.1 / Avg: 83.45 / Max: 83.98 Min: 155.93 / Avg: 156.31 / Max: 156.78 Min: 39.84 / Avg: 40.07 / Max: 40.23 Min: 39.99 / Avg: 40.33 / Max: 40.86 Min: 45.39 / Avg: 45.78 / Max: 45.99 Min: 45.95 / Avg: 46.59 / Max: 47.39 Min: 55.22 / Avg: 55.32 / Max: 55.43 Min: 60.93 / Avg: 61.25 / Max: 61.64 Min: 60 / Avg: 60.75 / Max: 61.17 Min: 69.76 / Avg: 69.84 / Max: 69.99 Min: 100.63 / Avg: 102.29 / Max: 103.93 Min: 107.74 / Avg: 108.34 / Max: 108.92 Min: 135.6 / Avg: 136.2 / Max: 136.76 Min: 196.87 / Avg: 198.45 / Max: 200.33
GROMACS The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.3 Water Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.0217 2.0434 3.0651 4.0868 5.1085 SE +/- 0.005, N = 3 SE +/- 0.003, N = 3 SE +/- 0.003, N = 3 SE +/- 0.008, N = 3 SE +/- 0.013, N = 3 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.008, N = 3 SE +/- 0.011, N = 3 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.003, N = 3 1.995 1.345 4.373 4.541 4.117 3.863 3.323 3.267 3.128 2.741 2.014 1.677 1.409 0.985 1. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
Ns Per Day Per Watt
OpenBenchmarking.org Ns Per Day Per Watt, More Is Better GROMACS 2020.3 Water Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0068 0.0136 0.0204 0.0272 0.034 0.01 0.01 0.02 0.03 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.01
Result Confidence
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.3 Water Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.99 / Avg: 2 / Max: 2.01 Min: 1.34 / Avg: 1.35 / Max: 1.35 Min: 4.37 / Avg: 4.37 / Max: 4.38 Min: 4.53 / Avg: 4.54 / Max: 4.55 Min: 4.09 / Avg: 4.12 / Max: 4.13 Min: 3.86 / Avg: 3.86 / Max: 3.87 Min: 3.32 / Avg: 3.32 / Max: 3.33 Min: 3.25 / Avg: 3.27 / Max: 3.28 Min: 3.11 / Avg: 3.13 / Max: 3.14 Min: 2.74 / Avg: 2.74 / Max: 2.74 Min: 2.01 / Avg: 2.01 / Max: 2.02 Min: 1.68 / Avg: 1.68 / Max: 1.68 Min: 1.41 / Avg: 1.41 / Max: 1.41 Min: 0.98 / Avg: 0.99 / Max: 0.99 1. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
TensorFlow Lite This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception ResNet V2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700K 1400K 2100K 2800K 3500K SE +/- 1646.01, N = 3 SE +/- 314.32, N = 3 SE +/- 6234.46, N = 7 SE +/- 4181.60, N = 3 SE +/- 837.49, N = 3 SE +/- 1818.31, N = 3 SE +/- 149.07, N = 3 SE +/- 386.91, N = 3 SE +/- 466.83, N = 3 SE +/- 1259.14, N = 3 SE +/- 366.38, N = 3 SE +/- 1027.95, N = 3 SE +/- 518.11, N = 3 SE +/- 326.24, N = 3 1346153 2592970 702607 655279 830730 893517 840604 921865 897884 1206790 1567070 1639167 2242950 3136920
Result Confidence
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception ResNet V2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500K 1000K 1500K 2000K 2500K Min: 1343550 / Avg: 1346153.33 / Max: 1349200 Min: 2592570 / Avg: 2592970 / Max: 2593590 Min: 690630 / Avg: 702607 / Max: 737249 Min: 649807 / Avg: 655278.67 / Max: 663492 Min: 829366 / Avg: 830730.33 / Max: 832254 Min: 889933 / Avg: 893516.67 / Max: 895844 Min: 840370 / Avg: 840604 / Max: 840881 Min: 921416 / Avg: 921864.67 / Max: 922635 Min: 897006 / Avg: 897884 / Max: 898598 Min: 1205130 / Avg: 1206790 / Max: 1209260 Min: 1566650 / Avg: 1567070 / Max: 1567800 Min: 1637410 / Avg: 1639166.67 / Max: 1640970 Min: 2241960 / Avg: 2242950 / Max: 2243710 Min: 3136350 / Avg: 3136920 / Max: 3137480
Build2 This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.13 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.30, N = 3 SE +/- 0.15, N = 3 SE +/- 0.22, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.16, N = 3 SE +/- 0.32, N = 3 SE +/- 0.16, N = 3 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 SE +/- 0.14, N = 3 SE +/- 0.07, N = 3 SE +/- 0.23, N = 3 SE +/- 0.29, N = 3 75.85 113.26 67.97 68.29 69.43 69.86 71.46 74.48 74.20 76.65 87.66 94.35 105.71 145.44
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.13 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 75.26 / Avg: 75.85 / Max: 76.19 Min: 113.02 / Avg: 113.26 / Max: 113.53 Min: 67.65 / Avg: 67.97 / Max: 68.39 Min: 68.11 / Avg: 68.29 / Max: 68.39 Min: 69.34 / Avg: 69.43 / Max: 69.57 Min: 69.58 / Avg: 69.86 / Max: 70.15 Min: 70.83 / Avg: 71.46 / Max: 71.85 Min: 74.18 / Avg: 74.48 / Max: 74.74 Min: 74.09 / Avg: 74.2 / Max: 74.34 Min: 76.5 / Avg: 76.65 / Max: 76.84 Min: 87.37 / Avg: 87.66 / Max: 87.82 Min: 94.22 / Avg: 94.35 / Max: 94.43 Min: 105.37 / Avg: 105.71 / Max: 106.15 Min: 145.08 / Avg: 145.44 / Max: 146.02
InfluxDB This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K SE +/- 2686.24, N = 3 SE +/- 1176.92, N = 3 SE +/- 628.66, N = 3 SE +/- 762.66, N = 3 SE +/- 1733.41, N = 3 SE +/- 686.09, N = 3 SE +/- 295.02, N = 3 SE +/- 759.90, N = 3 SE +/- 702.94, N = 3 SE +/- 2008.07, N = 3 SE +/- 1121.87, N = 3 SE +/- 1061.91, N = 3 SE +/- 1242.45, N = 3 SE +/- 1442.60, N = 3 1438050.4 1279851.8 1308517.2 1294664.5 1295588.8 1288738.6 1206497.8 1281490.5 1303437.6 1290897.8 1249264.6 1211354.3 1185857.8 1113861.3
val/sec Per Watt
OpenBenchmarking.org val/sec Per Watt, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K 13202.43 16288.24 12919.70 13662.86 13661.39 15448.09 17199.25 13818.71 17835.71 17743.64 18467.75 19723.37 19817.67 19958.05
Result Confidence
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K Min: 1433192.9 / Avg: 1438050.37 / Max: 1442466.9 Min: 1277678.1 / Avg: 1279851.77 / Max: 1281720.8 Min: 1307504.2 / Avg: 1308517.23 / Max: 1309668.7 Min: 1293775.5 / Avg: 1294664.5 / Max: 1296182.4 Min: 1293258.5 / Avg: 1295588.83 / Max: 1298976.9 Min: 1287797.7 / Avg: 1288738.57 / Max: 1290074 Min: 1205942.3 / Avg: 1206497.83 / Max: 1206947.8 Min: 1279979.8 / Avg: 1281490.47 / Max: 1282389.9 Min: 1302037.7 / Avg: 1303437.6 / Max: 1304249.7 Min: 1287374.1 / Avg: 1290897.83 / Max: 1294328.4 Min: 1247865.2 / Avg: 1249264.6 / Max: 1251483.2 Min: 1209909.6 / Avg: 1211354.27 / Max: 1213424.8 Min: 1183868.7 / Avg: 1185857.83 / Max: 1188142.2 Min: 1110976.3 / Avg: 1113861.27 / Max: 1115334.9
TensorFlow Lite This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception V4 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700K 1400K 2100K 2800K 3500K SE +/- 192.21, N = 3 SE +/- 227.03, N = 3 SE +/- 6069.08, N = 3 SE +/- 3136.75, N = 3 SE +/- 1781.71, N = 3 SE +/- 907.34, N = 3 SE +/- 425.74, N = 3 SE +/- 143.80, N = 3 SE +/- 487.90, N = 3 SE +/- 1233.46, N = 3 SE +/- 272.21, N = 3 SE +/- 565.25, N = 3 SE +/- 436.36, N = 3 SE +/- 1129.29, N = 3 1499183 2872227 764269 700833 938691 978828 943229 1034367 1006347 1366410 1742890 1820153 2493747 3472940
Result Confidence
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception V4 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 600K 1200K 1800K 2400K 3000K Min: 1498800 / Avg: 1499183.33 / Max: 1499400 Min: 2871920 / Avg: 2872226.67 / Max: 2872670 Min: 752152 / Avg: 764269 / Max: 770948 Min: 695393 / Avg: 700833.33 / Max: 706259 Min: 935608 / Avg: 938691 / Max: 941780 Min: 977917 / Avg: 978828.33 / Max: 980643 Min: 942378 / Avg: 943229 / Max: 943679 Min: 1034080 / Avg: 1034366.67 / Max: 1034530 Min: 1005560 / Avg: 1006346.67 / Max: 1007240 Min: 1365020 / Avg: 1366410 / Max: 1368870 Min: 1742350 / Avg: 1742890 / Max: 1743220 Min: 1819400 / Avg: 1820153.33 / Max: 1821260 Min: 2492910 / Avg: 2493746.67 / Max: 2494380 Min: 3471470 / Avg: 3472940 / Max: 3475160
GnuPG This test times how long it takes to encrypt a sample file using GnuPG. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better GnuPG 2.2.27 2.7GB Sample File Encryption EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.29, N = 3 SE +/- 0.36, N = 3 SE +/- 0.76, N = 3 SE +/- 0.88, N = 3 SE +/- 0.80, N = 3 SE +/- 0.71, N = 3 SE +/- 0.30, N = 3 SE +/- 0.48, N = 3 SE +/- 0.75, N = 3 SE +/- 0.84, N = 3 SE +/- 0.95, N = 3 SE +/- 0.25, N = 3 SE +/- 0.94, N = 3 SE +/- 0.53, N = 3 73.85 73.75 85.24 85.73 86.36 86.37 83.90 85.79 85.02 84.43 85.82 89.68 88.44 88.44 1. (CC) gcc options: -O2
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better GnuPG 2.2.27 2.7GB Sample File Encryption EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 73.37 / Avg: 73.85 / Max: 74.37 Min: 73.34 / Avg: 73.75 / Max: 74.47 Min: 83.83 / Avg: 85.24 / Max: 86.45 Min: 84.82 / Avg: 85.73 / Max: 87.48 Min: 84.87 / Avg: 86.36 / Max: 87.6 Min: 85.07 / Avg: 86.37 / Max: 87.5 Min: 83.3 / Avg: 83.9 / Max: 84.23 Min: 85.06 / Avg: 85.79 / Max: 86.69 Min: 83.6 / Avg: 85.02 / Max: 86.12 Min: 83.56 / Avg: 84.43 / Max: 86.11 Min: 84.82 / Avg: 85.82 / Max: 87.72 Min: 89.26 / Avg: 89.68 / Max: 90.12 Min: 87.5 / Avg: 88.44 / Max: 90.31 Min: 87.54 / Avg: 88.44 / Max: 89.39 1. (CC) gcc options: -O2
Appleseed Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Disney Material EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 118.48 225.09 67.74 67.33 70.49 73.33 81.83 89.01 87.51 105.69 149.34 151.86 195.08 265.54
Kripke Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50M 100M 150M 200M 250M SE +/- 574642.20, N = 15 SE +/- 1036502.10, N = 15 SE +/- 2966178.44, N = 15 SE +/- 2208966.30, N = 15 SE +/- 5572341.62, N = 12 SE +/- 1883204.11, N = 15 SE +/- 3811837.38, N = 15 SE +/- 5452949.69, N = 15 SE +/- 2175939.87, N = 3 SE +/- 481377.88, N = 4 SE +/- 3144064.63, N = 15 SE +/- 927929.45, N = 9 SE +/- 1366026.89, N = 3 SE +/- 1075776.09, N = 4 71413394 129536920 199683960 215771227 230851783 211310307 187397007 216525133 176798867 209811150 162898387 112489722 121256433 100898758 1. (CXX) g++ options: -O3 -fopenmp
Throughput FoM Per Watt
OpenBenchmarking.org Throughput FoM Per Watt, More Is Better Kripke 1.2.4 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400K 800K 1200K 1600K 2000K 514628.71 1163515.81 1423158.29 1604877.74 1663875.88 1673028.98 1635500.08 1673190.41 1575405.35 1932212.23 1638766.48 1357552.35 1508621.96 1349588.10
Result Confidence
OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40M 80M 120M 160M 200M Min: 68795100 / Avg: 71413394 / Max: 74273840 Min: 124322100 / Avg: 129536920 / Max: 133041300 Min: 179253900 / Avg: 199683960 / Max: 214308000 Min: 199388700 / Avg: 215771226.67 / Max: 228027900 Min: 196106600 / Avg: 230851783.33 / Max: 255760700 Min: 200163900 / Avg: 211310306.67 / Max: 221141000 Min: 160767700 / Avg: 187397006.67 / Max: 215708300 Min: 170848300 / Avg: 216525133.33 / Max: 232847700 Min: 172473100 / Avg: 176798866.67 / Max: 179374000 Min: 208902500 / Avg: 209811150 / Max: 211002600 Min: 148262800 / Avg: 162898386.67 / Max: 178935200 Min: 107412100 / Avg: 112489722.22 / Max: 117685700 Min: 119380500 / Avg: 121256433.33 / Max: 123914500 Min: 97675130 / Avg: 100898757.5 / Max: 102097200 1. (CXX) g++ options: -O3 -fopenmp
GROMACS The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2021 Input: water_GMX50_bare EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 1.0199 2.0398 3.0597 4.0796 5.0995 SE +/- 0.006, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.005, N = 3 SE +/- 0.004, N = 3 SE +/- 0.007, N = 3 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 2.313 1.348 4.371 4.533 3.317 3.256 3.140 1.675 1.407 1. (CXX) g++ options: -O3 -pthread
Ns Per Day Per Watt
OpenBenchmarking.org Ns Per Day Per Watt, More Is Better GROMACS 2021 Input: water_GMX50_bare EPYC 7F52 EPYC 7F32 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 0.0068 0.0136 0.0204 0.0272 0.034 0.01 0.01 0.03 0.02 0.02 0.02 0.02
Result Confidence
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2021 Input: water_GMX50_bare EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7282 EPYC 7272 2 4 6 8 10 Min: 2.31 / Avg: 2.31 / Max: 2.33 Min: 1.35 / Avg: 1.35 / Max: 1.35 Min: 4.37 / Avg: 4.37 / Max: 4.37 Min: 4.52 / Avg: 4.53 / Max: 4.54 Min: 3.31 / Avg: 3.32 / Max: 3.33 Min: 3.24 / Avg: 3.26 / Max: 3.27 Min: 3.14 / Avg: 3.14 / Max: 3.14 Min: 1.67 / Avg: 1.68 / Max: 1.68 Min: 1.41 / Avg: 1.41 / Max: 1.41 1. (CXX) g++ options: -O3 -pthread
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.24, N = 3 SE +/- 0.15, N = 3 SE +/- 0.77, N = 3 SE +/- 0.63, N = 3 SE +/- 0.83, N = 3 SE +/- 0.29, N = 3 SE +/- 0.29, N = 3 SE +/- 0.28, N = 3 SE +/- 0.22, N = 3 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.21, N = 3 SE +/- 0.24, N = 3 SE +/- 0.88, N = 3 49.01 34.04 110.47 109.30 95.40 93.48 75.44 128.29 76.19 65.42 95.77 62.29 51.65 75.43 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 48.58 / Avg: 49.01 / Max: 49.42 Min: 33.82 / Avg: 34.04 / Max: 34.32 Min: 109.25 / Avg: 110.47 / Max: 111.89 Min: 108.54 / Avg: 109.3 / Max: 110.55 Min: 93.75 / Avg: 95.4 / Max: 96.31 Min: 93.04 / Avg: 93.48 / Max: 94.02 Min: 75.01 / Avg: 75.44 / Max: 76 Min: 127.98 / Avg: 128.29 / Max: 128.85 Min: 75.76 / Avg: 76.19 / Max: 76.51 Min: 65.38 / Avg: 65.42 / Max: 65.48 Min: 95.59 / Avg: 95.77 / Max: 95.96 Min: 61.91 / Avg: 62.29 / Max: 62.65 Min: 51.36 / Avg: 51.65 / Max: 52.13 Min: 73.67 / Avg: 75.43 / Max: 76.37 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 2.05, N = 3 SE +/- 0.86, N = 3 SE +/- 5.75, N = 3 SE +/- 8.27, N = 3 SE +/- 0.57, N = 3 SE +/- 1.32, N = 3 SE +/- 1.42, N = 3 SE +/- 2.63, N = 3 SE +/- 5.60, N = 3 SE +/- 0.54, N = 3 SE +/- 2.99, N = 3 SE +/- 1.43, N = 3 SE +/- 1.59, N = 3 SE +/- 5.10, N = 3 2013.75 3345.10 2306.33 2221.33 1213.58 1350.55 2718.68 2214.79 2738.42 1672.24 2412.79 2836.68 3056.05 4345.12 MIN: 1995.38 MIN: 3322.56 MIN: 2274.33 MIN: 2189.19 MIN: 1190.99 MIN: 1331.05 MIN: 2705.41 MIN: 2191.84 MIN: 2722.76 MIN: 1658.48 MIN: 2397.68 MIN: 2798.53 MIN: 3041.35 MIN: 4298.14 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 2010.58 / Avg: 2013.75 / Max: 2017.58 Min: 3343.83 / Avg: 3345.1 / Max: 3346.73 Min: 2294.83 / Avg: 2306.33 / Max: 2312.48 Min: 2206.05 / Avg: 2221.33 / Max: 2234.44 Min: 1212.83 / Avg: 1213.58 / Max: 1214.69 Min: 1347.94 / Avg: 1350.55 / Max: 1352.2 Min: 2716.43 / Avg: 2718.68 / Max: 2721.31 Min: 2209.9 / Avg: 2214.79 / Max: 2218.93 Min: 2731.04 / Avg: 2738.42 / Max: 2749.41 Min: 1671.4 / Avg: 1672.24 / Max: 1673.25 Min: 2408.58 / Avg: 2412.79 / Max: 2418.57 Min: 2834.78 / Avg: 2836.68 / Max: 2839.49 Min: 3053.73 / Avg: 3056.05 / Max: 3059.09 Min: 4339.98 / Avg: 4345.12 / Max: 4355.32 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 0.63, N = 3 SE +/- 2.55, N = 3 SE +/- 9.75, N = 3 SE +/- 0.34, N = 3 SE +/- 3.66, N = 3 SE +/- 1.28, N = 3 SE +/- 2.51, N = 3 SE +/- 5.96, N = 3 SE +/- 4.79, N = 3 SE +/- 0.94, N = 3 SE +/- 3.71, N = 3 SE +/- 1.77, N = 3 SE +/- 0.81, N = 3 SE +/- 4.35, N = 3 1997.39 3348.06 2314.36 2203.16 1211.54 1350.13 2716.54 2221.37 2740.52 1673.31 2407.93 2835.35 3059.19 4348.88 MIN: 1989.1 MIN: 3331.21 MIN: 2276.78 MIN: 2183.71 MIN: 1187.15 MIN: 1333.75 MIN: 2705.25 MIN: 2198.48 MIN: 2722.51 MIN: 1660.31 MIN: 2392.45 MIN: 2800.15 MIN: 3051.11 MIN: 4311.94 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 1996.27 / Avg: 1997.39 / Max: 1998.44 Min: 3342.99 / Avg: 3348.06 / Max: 3351.09 Min: 2296.67 / Avg: 2314.36 / Max: 2330.29 Min: 2202.51 / Avg: 2203.16 / Max: 2203.65 Min: 1207.18 / Avg: 1211.54 / Max: 1218.81 Min: 1347.63 / Avg: 1350.13 / Max: 1351.85 Min: 2713.68 / Avg: 2716.54 / Max: 2721.54 Min: 2213.13 / Avg: 2221.37 / Max: 2232.96 Min: 2731.74 / Avg: 2740.52 / Max: 2748.21 Min: 1671.47 / Avg: 1673.31 / Max: 1674.53 Min: 2401.58 / Avg: 2407.93 / Max: 2414.43 Min: 2831.82 / Avg: 2835.35 / Max: 2837.39 Min: 3057.83 / Avg: 3059.19 / Max: 3060.62 Min: 4341.28 / Avg: 4348.88 / Max: 4356.33 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 6.65, N = 3 SE +/- 0.89, N = 3 SE +/- 3.86, N = 3 SE +/- 11.35, N = 3 SE +/- 2.03, N = 3 SE +/- 1.92, N = 3 SE +/- 2.43, N = 3 SE +/- 5.23, N = 3 SE +/- 2.18, N = 3 SE +/- 1.08, N = 3 SE +/- 0.38, N = 3 SE +/- 2.25, N = 3 SE +/- 3.17, N = 3 SE +/- 1.97, N = 3 2013.24 3345.67 2296.53 2230.18 1210.04 1352.62 2721.87 2230.35 2743.71 1674.66 2410.23 2834.09 3057.08 4347.66 MIN: 1992.69 MIN: 3327 MIN: 2270.01 MIN: 2194.78 MIN: 1190.59 MIN: 1333.53 MIN: 2708.11 MIN: 2212.66 MIN: 2727.14 MIN: 1660.31 MIN: 2395.31 MIN: 2799.34 MIN: 3044.36 MIN: 4308.62 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 2003.33 / Avg: 2013.24 / Max: 2025.88 Min: 3344.19 / Avg: 3345.67 / Max: 3347.26 Min: 2288.81 / Avg: 2296.53 / Max: 2300.61 Min: 2212.51 / Avg: 2230.18 / Max: 2251.36 Min: 1206.71 / Avg: 1210.04 / Max: 1213.71 Min: 1350.2 / Avg: 1352.62 / Max: 1356.42 Min: 2718.51 / Avg: 2721.87 / Max: 2726.6 Min: 2221.2 / Avg: 2230.35 / Max: 2239.31 Min: 2741.09 / Avg: 2743.71 / Max: 2748.05 Min: 1672.87 / Avg: 1674.66 / Max: 1676.6 Min: 2409.5 / Avg: 2410.23 / Max: 2410.77 Min: 2829.71 / Avg: 2834.09 / Max: 2837.17 Min: 3051.79 / Avg: 3057.08 / Max: 3062.75 Min: 4343.96 / Avg: 4347.66 / Max: 4350.69 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Read While Writing EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2M 4M 6M 8M 10M SE +/- 8186.58, N = 3 SE +/- 10805.74, N = 13 SE +/- 47749.78, N = 3 SE +/- 27617.12, N = 3 SE +/- 69440.76, N = 6 SE +/- 44027.23, N = 3 SE +/- 55384.00, N = 3 SE +/- 32730.71, N = 3 SE +/- 24292.51, N = 3 SE +/- 43537.48, N = 3 SE +/- 25144.15, N = 3 SE +/- 26792.49, N = 3 SE +/- 24217.23, N = 3 SE +/- 1710.84, N = 3 3012387 1603673 8232474 8175383 7131600 7183579 5564433 5214302 5486066 4397695 3086237 3023239 2357928 1470621 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Facebook RocksDB 6.3.6 Test: Read While Writing EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9K 18K 27K 36K 45K 14841.44 14725.01 43072.97 44350.10 37151.16 40538.44 32038.20 31284.41 36657.22 28233.69 24738.50 29403.11 23416.29 18803.69
Result Confidence
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Read While Writing EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.4M 2.8M 4.2M 5.6M 7M Min: 3002918 / Avg: 3012386.67 / Max: 3028689 Min: 1550246 / Avg: 1603672.85 / Max: 1683898 Min: 8181136 / Avg: 8232473.67 / Max: 8327881 Min: 8142180 / Avg: 8175382.67 / Max: 8230211 Min: 7048399 / Avg: 7131599.5 / Max: 7477000 Min: 7101279 / Avg: 7183579 / Max: 7251844 Min: 5463565 / Avg: 5564432.67 / Max: 5654508 Min: 5150993 / Avg: 5214301.67 / Max: 5260375 Min: 5439082 / Avg: 5486065.67 / Max: 5520271 Min: 4337144 / Avg: 4397695 / Max: 4482162 Min: 3036500 / Avg: 3086236.67 / Max: 3117538 Min: 2989438 / Avg: 3023238.67 / Max: 3076148 Min: 2309505 / Avg: 2357928 / Max: 2383052 Min: 1467509 / Avg: 1470620.67 / Max: 1473409 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Minion Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.11, N = 3 SE +/- 0.23, N = 3 SE +/- 0.14, N = 3 SE +/- 0.07, N = 3 SE +/- 0.32, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 0.11, N = 3 SE +/- 0.33, N = 3 SE +/- 0.19, N = 3 SE +/- 0.13, N = 3 67.74 67.80 79.39 80.12 79.72 80.15 77.80 79.85 79.00 79.03 80.33 82.15 81.97 82.84 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 Min: 67.63 / Avg: 67.74 / Max: 67.95 Min: 67.36 / Avg: 67.8 / Max: 68.07 Min: 79.2 / Avg: 79.39 / Max: 79.66 Min: 80.05 / Avg: 80.12 / Max: 80.26 Min: 79.1 / Avg: 79.72 / Max: 80.13 Min: 80.03 / Avg: 80.15 / Max: 80.32 Min: 77.68 / Avg: 77.8 / Max: 77.93 Min: 79.51 / Avg: 79.85 / Max: 80.03 Min: 78.98 / Avg: 79 / Max: 79.03 Min: 78.84 / Avg: 79.03 / Max: 79.18 Min: 80.18 / Avg: 80.33 / Max: 80.55 Min: 81.53 / Avg: 82.15 / Max: 82.68 Min: 81.59 / Avg: 81.97 / Max: 82.16 Min: 82.61 / Avg: 82.84 / Max: 83.06 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
JPEG XL Decoding The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. Learn more via the OpenBenchmarking.org test page.
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.2723 0.5446 0.8169 1.0892 1.3615 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 15 0.81 0.79 1.21 1.06 1.02 0.99 0.77 0.95 0.84 0.92 0.84 0.78 0.89 0.98
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.8 / Avg: 0.81 / Max: 0.82 Min: 0.78 / Avg: 0.79 / Max: 0.81 Min: 1.21 / Avg: 1.21 / Max: 1.21 Min: 1.06 / Avg: 1.06 / Max: 1.07 Min: 1.01 / Avg: 1.02 / Max: 1.02 Min: 0.99 / Avg: 0.99 / Max: 0.99 Min: 0.77 / Avg: 0.77 / Max: 0.78 Min: 0.95 / Avg: 0.95 / Max: 0.95 Min: 0.83 / Avg: 0.84 / Max: 0.84 Min: 0.91 / Avg: 0.92 / Max: 0.93 Min: 0.83 / Avg: 0.84 / Max: 0.84 Min: 0.77 / Avg: 0.78 / Max: 0.79 Min: 0.87 / Avg: 0.89 / Max: 0.91 Min: 0.91 / Avg: 0.98 / Max: 1.03
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6K 12K 18K 24K 30K SE +/- 43.76, N = 3 SE +/- 52.69, N = 3 SE +/- 28.57, N = 3 SE +/- 46.32, N = 3 SE +/- 31.70, N = 3 SE +/- 55.79, N = 3 SE +/- 85.72, N = 3 SE +/- 22.86, N = 3 SE +/- 42.04, N = 3 SE +/- 95.61, N = 3 SE +/- 21.57, N = 3 SE +/- 50.79, N = 3 SE +/- 89.67, N = 3 SE +/- 41.71, N = 15 9663.43 4953.65 25067.04 28284.87 22986.05 23441.91 19922.01 16446.88 18612.54 12800.59 9316.80 10017.98 6689.95 4012.40
Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K Min: 9575.98 / Avg: 9663.43 / Max: 9710.01 Min: 4855.38 / Avg: 4953.65 / Max: 5035.74 Min: 25025.67 / Avg: 25067.04 / Max: 25121.85 Min: 28192.24 / Avg: 28284.87 / Max: 28331.27 Min: 22927.38 / Avg: 22986.05 / Max: 23036.21 Min: 23347.44 / Avg: 23441.91 / Max: 23540.57 Min: 19751.03 / Avg: 19922.01 / Max: 20018.47 Min: 16403.4 / Avg: 16446.88 / Max: 16480.83 Min: 18567.53 / Avg: 18612.54 / Max: 18696.54 Min: 12615.39 / Avg: 12800.59 / Max: 12934.43 Min: 9293.16 / Avg: 9316.8 / Max: 9359.87 Min: 9917.01 / Avg: 10017.98 / Max: 10078.03 Min: 6524.74 / Avg: 6689.95 / Max: 6832.97 Min: 3827.5 / Avg: 4012.4 / Max: 4339.41
Cpuminer-Opt Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Skeincoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 140K 280K 420K 560K 700K SE +/- 335.77, N = 3 SE +/- 654.80, N = 15 SE +/- 12582.41, N = 12 SE +/- 5351.86, N = 13 SE +/- 9107.04, N = 12 SE +/- 6158.44, N = 12 SE +/- 2411.69, N = 3 SE +/- 2127.92, N = 3 SE +/- 4313.15, N = 3 SE +/- 3944.26, N = 15 SE +/- 981.65, N = 3 SE +/- 1328.09, N = 3 SE +/- 728.58, N = 3 SE +/- 611.43, N = 6 182403 79365 605852 631038 470502 468227 324420 317250 316350 210539 151030 150287 105140 63648 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Skeincoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K 2197.63 1374.22 8011.78 8559.50 6520.94 7328.52 5932.59 4208.20 5770.73 3879.07 2891.08 3270.93 2410.51 1483.87
Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: Skeincoin EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110K 220K 330K 440K 550K Min: 181900 / Avg: 182403.33 / Max: 183040 Min: 75330 / Avg: 79365.33 / Max: 84270 Min: 470320 / Avg: 605851.67 / Max: 630960 Min: 570830 / Avg: 631038.46 / Max: 649140 Min: 373060 / Avg: 470501.67 / Max: 494480 Min: 404270 / Avg: 468226.67 / Max: 483970 Min: 319750 / Avg: 324420 / Max: 327800 Min: 313810 / Avg: 317250 / Max: 321140 Min: 310740 / Avg: 316350 / Max: 324830 Min: 187570 / Avg: 210538.67 / Max: 242310 Min: 149150 / Avg: 151030 / Max: 152460 Min: 148430 / Avg: 150286.67 / Max: 152860 Min: 103990 / Avg: 105140 / Max: 106490 Min: 61790 / Avg: 63648.33 / Max: 66140 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K SE +/- 6124.02, N = 3 SE +/- 6471.76, N = 5 SE +/- 1303.40, N = 3 SE +/- 227.93, N = 3 SE +/- 3532.96, N = 3 SE +/- 4097.81, N = 3 SE +/- 4338.82, N = 3 SE +/- 2994.50, N = 3 SE +/- 6323.38, N = 3 SE +/- 10119.73, N = 3 SE +/- 5524.24, N = 9 SE +/- 9036.88, N = 4 SE +/- 6225.31, N = 6 SE +/- 6988.75, N = 3 699077 581407 451547 442864 518446 521910 860225 806814 846887 805066 692044 735381 656055 545816 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K 3901.69 6147.22 3013.23 2783.45 3303.48 3855.32 5587.32 4922.99 5906.76 5637.77 6716.57 8078.17 8044.80 8529.54
Result Confidence
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 150K 300K 450K 600K 750K Min: 687578 / Avg: 699077 / Max: 708479 Min: 569871 / Avg: 581407 / Max: 602220 Min: 449913 / Avg: 451547 / Max: 454123 Min: 442493 / Avg: 442864.33 / Max: 443279 Min: 513881 / Avg: 518445.67 / Max: 525399 Min: 516355 / Avg: 521910 / Max: 529906 Min: 851801 / Avg: 860225.33 / Max: 866240 Min: 801674 / Avg: 806813.67 / Max: 812046 Min: 835350 / Avg: 846887 / Max: 857142 Min: 794424 / Avg: 805065.67 / Max: 825296 Min: 669061 / Avg: 692043.67 / Max: 723569 Min: 722999 / Avg: 735381.25 / Max: 761712 Min: 638259 / Avg: 656055.33 / Max: 681884 Min: 533331 / Avg: 545816.33 / Max: 557501 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
SQLite Speedtest This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.19, N = 3 SE +/- 0.27, N = 3 SE +/- 0.36, N = 3 SE +/- 0.54, N = 3 SE +/- 0.05, N = 3 SE +/- 0.18, N = 3 SE +/- 0.06, N = 3 SE +/- 0.25, N = 3 66.52 67.30 77.61 77.94 78.23 78.26 76.16 78.07 77.19 75.91 77.74 80.05 80.19 81.51 1. (CC) gcc options: -O2 -ldl -lz -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 Min: 66.35 / Avg: 66.52 / Max: 66.7 Min: 67.2 / Avg: 67.3 / Max: 67.4 Min: 77.55 / Avg: 77.61 / Max: 77.67 Min: 77.74 / Avg: 77.94 / Max: 78.1 Min: 78.09 / Avg: 78.23 / Max: 78.45 Min: 78.14 / Avg: 78.26 / Max: 78.37 Min: 75.89 / Avg: 76.16 / Max: 76.52 Min: 77.69 / Avg: 78.06 / Max: 78.59 Min: 76.8 / Avg: 77.18 / Max: 77.9 Min: 74.83 / Avg: 75.9 / Max: 76.53 Min: 77.69 / Avg: 77.74 / Max: 77.85 Min: 79.76 / Avg: 80.05 / Max: 80.38 Min: 80.11 / Avg: 80.19 / Max: 80.3 Min: 81.25 / Avg: 81.51 / Max: 82.01 1. (CC) gcc options: -O2 -ldl -lz -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 SE +/- 3.97, N = 3 SE +/- 0.95, N = 3 SE +/- 2.81, N = 3 SE +/- 1.88, N = 3 SE +/- 3.28, N = 3 SE +/- 2.09, N = 3 SE +/- 1.35, N = 3 SE +/- 1.32, N = 3 SE +/- 2.58, N = 3 SE +/- 1.57, N = 3 SE +/- 0.54, N = 3 SE +/- 5.33, N = 3 SE +/- 4.61, N = 3 SE +/- 3.61, N = 3 998.67 1742.79 878.20 793.23 734.40 784.82 886.88 812.99 919.61 974.12 1227.56 1886.30 2032.88 2397.22 MIN: 987.09 MIN: 1728.82 MIN: 851.72 MIN: 777.03 MIN: 716.82 MIN: 771.05 MIN: 865.5 MIN: 799.51 MIN: 906.32 MIN: 962.55 MIN: 1218.44 MIN: 1851.55 MIN: 2016.92 MIN: 2370.41 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 Min: 991.29 / Avg: 998.67 / Max: 1004.91 Min: 1741.68 / Avg: 1742.79 / Max: 1744.69 Min: 873.24 / Avg: 878.2 / Max: 882.97 Min: 791.29 / Avg: 793.23 / Max: 796.99 Min: 728.04 / Avg: 734.4 / Max: 738.94 Min: 780.67 / Avg: 784.82 / Max: 787.4 Min: 884.18 / Avg: 886.88 / Max: 888.33 Min: 810.37 / Avg: 812.99 / Max: 814.64 Min: 916.98 / Avg: 919.61 / Max: 924.77 Min: 971.07 / Avg: 974.12 / Max: 976.32 Min: 1226.72 / Avg: 1227.56 / Max: 1228.57 Min: 1876.6 / Avg: 1886.3 / Max: 1894.99 Min: 2023.66 / Avg: 2032.88 / Max: 2037.71 Min: 2392.08 / Avg: 2397.22 / Max: 2404.17 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 SE +/- 1.61, N = 3 SE +/- 0.51, N = 3 SE +/- 3.54, N = 3 SE +/- 1.17, N = 3 SE +/- 1.28, N = 3 SE +/- 2.70, N = 3 SE +/- 4.31, N = 3 SE +/- 1.51, N = 3 SE +/- 5.72, N = 3 SE +/- 0.91, N = 3 SE +/- 0.19, N = 3 SE +/- 3.12, N = 3 SE +/- 2.69, N = 3 SE +/- 2.26, N = 3 997.85 1745.54 874.44 793.20 732.81 787.43 890.48 813.12 902.26 972.93 1229.69 1882.64 2022.45 2397.56 MIN: 987.15 MIN: 1733.89 MIN: 847.07 MIN: 776.66 MIN: 712.9 MIN: 769.99 MIN: 867.03 MIN: 799.88 MIN: 882.8 MIN: 960.02 MIN: 1221.64 MIN: 1853.21 MIN: 2012.67 MIN: 2367.09 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 Min: 995.25 / Avg: 997.85 / Max: 1000.79 Min: 1744.59 / Avg: 1745.54 / Max: 1746.31 Min: 868.68 / Avg: 874.44 / Max: 880.89 Min: 790.85 / Avg: 793.2 / Max: 794.46 Min: 731.06 / Avg: 732.81 / Max: 735.31 Min: 782.55 / Avg: 787.43 / Max: 791.88 Min: 882.05 / Avg: 890.48 / Max: 896.25 Min: 810.1 / Avg: 813.12 / Max: 814.7 Min: 890.83 / Avg: 902.26 / Max: 908.5 Min: 971.59 / Avg: 972.93 / Max: 974.66 Min: 1229.46 / Avg: 1229.69 / Max: 1230.06 Min: 1876.41 / Avg: 1882.64 / Max: 1885.98 Min: 2018.73 / Avg: 2022.45 / Max: 2027.68 Min: 2394.02 / Avg: 2397.56 / Max: 2401.75 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 SE +/- 0.99, N = 3 SE +/- 2.16, N = 3 SE +/- 3.99, N = 3 SE +/- 0.96, N = 3 SE +/- 2.51, N = 3 SE +/- 1.66, N = 3 SE +/- 1.82, N = 3 SE +/- 2.11, N = 3 SE +/- 5.94, N = 3 SE +/- 0.65, N = 3 SE +/- 1.03, N = 3 SE +/- 4.32, N = 3 SE +/- 1.28, N = 3 SE +/- 4.09, N = 3 999.14 1744.75 875.71 792.55 735.82 782.65 888.54 814.03 917.61 972.77 1228.50 1884.70 2029.28 2398.98 MIN: 991.16 MIN: 1731.94 MIN: 846.67 MIN: 773.96 MIN: 715.63 MIN: 769.36 MIN: 872.96 MIN: 801.22 MIN: 892.04 MIN: 962.66 MIN: 1219.05 MIN: 1853.9 MIN: 2020.38 MIN: 2367.55 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 Min: 997.98 / Avg: 999.14 / Max: 1001.1 Min: 1742.54 / Avg: 1744.75 / Max: 1749.07 Min: 869.12 / Avg: 875.71 / Max: 882.9 Min: 791.43 / Avg: 792.55 / Max: 794.46 Min: 731.42 / Avg: 735.82 / Max: 740.12 Min: 779.51 / Avg: 782.65 / Max: 785.17 Min: 885.37 / Avg: 888.54 / Max: 891.69 Min: 810.97 / Avg: 814.03 / Max: 818.07 Min: 907.1 / Avg: 917.61 / Max: 927.67 Min: 971.9 / Avg: 972.77 / Max: 974.05 Min: 1226.65 / Avg: 1228.5 / Max: 1230.2 Min: 1878.26 / Avg: 1884.7 / Max: 1892.91 Min: 2026.92 / Avg: 2029.28 / Max: 2031.33 Min: 2391.8 / Avg: 2398.98 / Max: 2405.96 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Perl Benchmarks Perl benchmark suite that can be used to compare the relative speed of different versions of perl. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0352 0.0704 0.1056 0.1408 0.176 SE +/- 0.00013172, N = 3 SE +/- 0.00007037, N = 3 SE +/- 0.00022828, N = 3 SE +/- 0.00009609, N = 3 SE +/- 0.00113052, N = 3 SE +/- 0.00076232, N = 3 SE +/- 0.00010921, N = 3 SE +/- 0.00097851, N = 3 SE +/- 0.00050264, N = 3 SE +/- 0.00052181, N = 3 SE +/- 0.00104595, N = 3 SE +/- 0.00047942, N = 3 SE +/- 0.00070595, N = 3 SE +/- 0.00101629, N = 3 0.12924285 0.12819545 0.14954339 0.14976795 0.14868253 0.15124976 0.14850812 0.14988293 0.14859273 0.14573597 0.14978489 0.15341705 0.15457890 0.15653208
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1 2 3 4 5 Min: 0.13 / Avg: 0.13 / Max: 0.13 Min: 0.13 / Avg: 0.13 / Max: 0.13 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.14 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.15 / Avg: 0.15 / Max: 0.16 Min: 0.16 / Avg: 0.16 / Max: 0.16
Coremark This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400K 800K 1200K 1600K 2000K SE +/- 1470.27, N = 3 SE +/- 818.34, N = 3 SE +/- 1046.01, N = 3 SE +/- 473.83, N = 3 SE +/- 10873.39, N = 3 SE +/- 1150.50, N = 3 SE +/- 707.15, N = 3 SE +/- 609.31, N = 3 SE +/- 1542.09, N = 3 SE +/- 1815.81, N = 3 SE +/- 538.09, N = 3 SE +/- 270.65, N = 3 SE +/- 102.44, N = 3 716393.88 356570.41 1845888.80 1867023.46 1530338.79 1203386.72 1123555.30 1147733.95 902070.15 603381.33 586659.10 440879.64 293387.40 1. (CC) gcc options: -O2 -lrt" -lrt
Iterations/Sec Per Watt
OpenBenchmarking.org Iterations/Sec Per Watt, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K 3863.56 3489.81 11585.51 11942.55 10306.40 8324.83 7324.67 8528.23 6564.90 5810.65 6395.91 5295.21 4345.65
Result Confidence
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K Min: 713469.52 / Avg: 716393.88 / Max: 718122.78 Min: 355244.72 / Avg: 356570.41 / Max: 358064.49 Min: 1844712.66 / Avg: 1845888.8 / Max: 1847975.17 Min: 1866161.25 / Avg: 1867023.46 / Max: 1867795.13 Min: 1517351.75 / Avg: 1530338.79 / Max: 1551938.41 Min: 1201201.2 / Avg: 1203386.72 / Max: 1205102.86 Min: 1122462.4 / Avg: 1123555.3 / Max: 1124879.16 Min: 1146516.78 / Avg: 1147733.95 / Max: 1148394.04 Min: 899666.14 / Avg: 902070.15 / Max: 904945.39 Min: 599995.31 / Avg: 603381.33 / Max: 606211.3 Min: 585752.04 / Avg: 586659.1 / Max: 587614.19 Min: 440356.87 / Avg: 440879.64 / Max: 441262.65 Min: 293258.49 / Avg: 293387.4 / Max: 293589.77 1. (CC) gcc options: -O2 -lrt" -lrt
Himeno Benchmark The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 38.21, N = 3 SE +/- 33.69, N = 3 SE +/- 36.48, N = 3 SE +/- 47.38, N = 4 SE +/- 44.49, N = 4 SE +/- 31.29, N = 9 SE +/- 9.42, N = 3 SE +/- 42.46, N = 3 SE +/- 43.10, N = 4 SE +/- 37.84, N = 3 SE +/- 46.66, N = 4 SE +/- 17.87, N = 3 SE +/- 37.21, N = 3 SE +/- 25.83, N = 3 4367.12 4336.41 3762.49 3837.56 3816.08 3793.17 3988.94 3830.20 3853.50 3765.85 3792.81 3756.96 3634.17 3589.93 1. (CC) gcc options: -O3 -mavx2
MFLOPS Per Watt
OpenBenchmarking.org MFLOPS Per Watt, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 57.55 75.95 51.62 53.85 56.78 62.42 77.43 53.96 75.05 73.50 73.24 84.36 82.98 83.09
Result Confidence
OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 4297.85 / Avg: 4367.12 / Max: 4429.7 Min: 4277.41 / Avg: 4336.41 / Max: 4394.1 Min: 3718.35 / Avg: 3762.49 / Max: 3834.86 Min: 3701.86 / Avg: 3837.56 / Max: 3910.17 Min: 3692 / Avg: 3816.08 / Max: 3902.55 Min: 3632 / Avg: 3793.17 / Max: 3925.79 Min: 3976.02 / Avg: 3988.94 / Max: 4007.28 Min: 3768.21 / Avg: 3830.2 / Max: 3911.46 Min: 3731.7 / Avg: 3853.5 / Max: 3929.87 Min: 3719.42 / Avg: 3765.85 / Max: 3840.83 Min: 3699.39 / Avg: 3792.81 / Max: 3917.69 Min: 3721.38 / Avg: 3756.96 / Max: 3777.75 Min: 3590.61 / Avg: 3634.17 / Max: 3708.22 Min: 3541.75 / Avg: 3589.93 / Max: 3630.15 1. (CC) gcc options: -O3 -mavx2
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: python_startup EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 7.54 7.47 8.71 8.80 8.84 8.83 8.57 8.79 8.62 8.72 8.82 9.13 9.06 9.49
Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: python_startup EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 7.54 / Avg: 7.54 / Max: 7.55 Min: 7.47 / Avg: 7.47 / Max: 7.48 Min: 8.71 / Avg: 8.71 / Max: 8.71 Min: 8.8 / Avg: 8.8 / Max: 8.8 Min: 8.83 / Avg: 8.84 / Max: 8.85 Min: 8.81 / Avg: 8.83 / Max: 8.84 Min: 8.57 / Avg: 8.57 / Max: 8.57 Min: 8.78 / Avg: 8.79 / Max: 8.79 Min: 8.61 / Avg: 8.62 / Max: 8.62 Min: 8.72 / Avg: 8.72 / Max: 8.73 Min: 8.81 / Avg: 8.82 / Max: 8.82 Min: 9.12 / Avg: 9.13 / Max: 9.13 Min: 9.05 / Avg: 9.06 / Max: 9.06 Min: 9.49 / Avg: 9.49 / Max: 9.49
Cpuminer-Opt Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: x25x EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 SE +/- 1.25, N = 3 SE +/- 1.31, N = 3 SE +/- 29.63, N = 14 SE +/- 32.30, N = 15 SE +/- 1.46, N = 3 SE +/- 11.98, N = 15 SE +/- 2.23, N = 3 SE +/- 0.86, N = 3 SE +/- 37.08, N = 15 SE +/- 39.35, N = 15 SE +/- 0.74, N = 3 SE +/- 1.04, N = 3 SE +/- 0.69, N = 3 SE +/- 1.33, N = 3 524.31 262.90 1360.06 1398.77 1147.36 1139.38 891.02 807.02 883.87 862.81 447.01 429.51 322.83 216.58 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
kH/s Per Watt
OpenBenchmarking.org kH/s Per Watt, More Is Better Cpuminer-Opt 3.15.5 Algorithm: x25x EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 2.99 2.72 9.94 9.95 7.80 8.88 7.25 5.88 7.53 7.07 4.46 4.95 3.88 3.24
Result Confidence
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.15.5 Algorithm: x25x EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 521.85 / Avg: 524.31 / Max: 525.94 Min: 260.27 / Avg: 262.9 / Max: 264.3 Min: 1313.18 / Avg: 1360.06 / Max: 1735.96 Min: 1361.4 / Avg: 1398.77 / Max: 1850.16 Min: 1144.44 / Avg: 1147.36 / Max: 1148.92 Min: 1114.29 / Avg: 1139.38 / Max: 1274.78 Min: 887.85 / Avg: 891.02 / Max: 895.31 Min: 806.02 / Avg: 807.02 / Max: 808.73 Min: 822.42 / Avg: 883.87 / Max: 1393.5 Min: 675.58 / Avg: 862.81 / Max: 1074.46 Min: 445.61 / Avg: 447.01 / Max: 448.15 Min: 427.95 / Avg: 429.51 / Max: 431.48 Min: 321.57 / Avg: 322.83 / Max: 323.96 Min: 214.04 / Avg: 216.58 / Max: 218.55 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
LZ4 Compression This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 89.78, N = 3 SE +/- 34.53, N = 3 SE +/- 18.64, N = 7 SE +/- 44.06, N = 3 SE +/- 54.81, N = 3 SE +/- 29.33, N = 3 SE +/- 43.54, N = 3 SE +/- 14.46, N = 3 SE +/- 36.01, N = 4 SE +/- 93.14, N = 3 SE +/- 66.67, N = 3 SE +/- 29.05, N = 3 SE +/- 44.49, N = 3 10099.6 10567.2 10162.0 10170.7 10159.7 10222.4 10118.0 10232.8 10209.3 10172.7 10112.2 10101.0 10063.0 1. (CC) gcc options: -O3
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 134.23 188.33 139.07 143.59 166.50 199.63 141.26 198.84 199.84 198.83 227.31 232.15 233.97
Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 9990.9 / Avg: 10099.57 / Max: 10277.7 Min: 10498.8 / Avg: 10567.23 / Max: 10609.5 Min: 10103.5 / Avg: 10162.01 / Max: 10237.8 Min: 10082.6 / Avg: 10170.7 / Max: 10215.9 Min: 10076.5 / Avg: 10159.67 / Max: 10263.1 Min: 10167.9 / Avg: 10222.43 / Max: 10268.4 Min: 10073.2 / Avg: 10118.03 / Max: 10205.1 Min: 10214 / Avg: 10232.77 / Max: 10261.2 Min: 10125 / Avg: 10209.25 / Max: 10270.7 Min: 10073 / Avg: 10172.67 / Max: 10358.8 Min: 9978.9 / Avg: 10112.23 / Max: 10180.1 Min: 10050 / Avg: 10101 / Max: 10150.6 Min: 10013.3 / Avg: 10063.03 / Max: 10151.8 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 SE +/- 0.01, N = 3 SE +/- 0.35, N = 3 SE +/- 0.42, N = 7 SE +/- 0.31, N = 3 SE +/- 0.65, N = 3 SE +/- 0.04, N = 3 SE +/- 0.37, N = 3 SE +/- 0.29, N = 3 SE +/- 0.55, N = 4 SE +/- 0.53, N = 3 SE +/- 0.15, N = 3 SE +/- 0.15, N = 3 SE +/- 0.43, N = 3 52.82 52.79 45.69 45.19 45.38 45.90 44.69 45.77 45.98 44.53 44.38 43.23 44.93 1. (CC) gcc options: -O3
Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11 22 33 44 55 Min: 52.79 / Avg: 52.82 / Max: 52.84 Min: 52.36 / Avg: 52.79 / Max: 53.48 Min: 44.72 / Avg: 45.69 / Max: 47.44 Min: 44.56 / Avg: 45.19 / Max: 45.52 Min: 44.32 / Avg: 45.38 / Max: 46.55 Min: 45.84 / Avg: 45.9 / Max: 45.97 Min: 44.12 / Avg: 44.69 / Max: 45.37 Min: 45.44 / Avg: 45.77 / Max: 46.36 Min: 45.35 / Avg: 45.98 / Max: 47.62 Min: 44 / Avg: 44.53 / Max: 45.58 Min: 44.15 / Avg: 44.38 / Max: 44.66 Min: 42.93 / Avg: 43.23 / Max: 43.4 Min: 44.07 / Avg: 44.93 / Max: 45.44 1. (CC) gcc options: -O3
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.0553 2.1106 3.1659 4.2212 5.2765 SE +/- 0.024, N = 15 SE +/- 0.011, N = 3 SE +/- 0.016, N = 3 SE +/- 0.006, N = 3 SE +/- 0.016, N = 3 SE +/- 0.022, N = 3 SE +/- 0.024, N = 3 SE +/- 0.008, N = 3 SE +/- 0.015, N = 15 SE +/- 0.017, N = 3 SE +/- 0.029, N = 12 SE +/- 0.021, N = 15 SE +/- 0.009, N = 3 SE +/- 0.004, N = 3 2.697 4.145 1.798 1.643 1.701 1.759 1.915 2.103 1.935 2.101 2.606 2.596 3.137 4.690 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 2.5 / Avg: 2.7 / Max: 2.86 Min: 4.13 / Avg: 4.14 / Max: 4.16 Min: 1.77 / Avg: 1.8 / Max: 1.82 Min: 1.63 / Avg: 1.64 / Max: 1.65 Min: 1.68 / Avg: 1.7 / Max: 1.73 Min: 1.73 / Avg: 1.76 / Max: 1.8 Min: 1.87 / Avg: 1.91 / Max: 1.95 Min: 2.09 / Avg: 2.1 / Max: 2.11 Min: 1.84 / Avg: 1.94 / Max: 2.02 Min: 2.07 / Avg: 2.1 / Max: 2.12 Min: 2.42 / Avg: 2.61 / Max: 2.71 Min: 2.49 / Avg: 2.6 / Max: 2.73 Min: 3.13 / Avg: 3.14 / Max: 3.16 Min: 4.68 / Avg: 4.69 / Max: 4.7 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Write EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13K 26K 39K 52K 65K SE +/- 337.73, N = 15 SE +/- 63.04, N = 3 SE +/- 486.09, N = 3 SE +/- 229.52, N = 3 SE +/- 539.78, N = 3 SE +/- 709.54, N = 3 SE +/- 676.58, N = 3 SE +/- 174.06, N = 3 SE +/- 406.89, N = 15 SE +/- 387.48, N = 3 SE +/- 447.90, N = 12 SE +/- 309.38, N = 15 SE +/- 91.21, N = 3 SE +/- 18.98, N = 3 37152 24137 55699 60949 58889 56945 52325 47615 51794 47662 38458 38586 31896 21334 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Write EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11K 22K 33K 44K 55K Min: 35001.56 / Avg: 37152.01 / Max: 39986.56 Min: 24025.48 / Avg: 24137.33 / Max: 24243.66 Min: 54882.69 / Avg: 55699.33 / Max: 56564.47 Min: 60636.54 / Avg: 60949.34 / Max: 61396.7 Min: 57853.09 / Avg: 58888.92 / Max: 59670.2 Min: 55594.88 / Avg: 56944.6 / Max: 57999 Min: 51433.97 / Avg: 52325.08 / Max: 53652.51 Min: 47424.2 / Avg: 47614.55 / Max: 47962.14 Min: 49652.94 / Avg: 51793.61 / Max: 54520.15 Min: 47265.69 / Avg: 47661.51 / Max: 48436.42 Min: 36956.38 / Avg: 38457.92 / Max: 41415.74 Min: 36728.54 / Avg: 38586.13 / Max: 40215.78 Min: 31714.82 / Avg: 31895.88 / Max: 32005.63 Min: 21301.63 / Avg: 21333.63 / Max: 21367.31 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Stockfish This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 12 Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20M 40M 60M 80M 100M SE +/- 416432.88, N = 3 SE +/- 79798.13, N = 3 SE +/- 597478.91, N = 3 SE +/- 1212155.43, N = 4 SE +/- 82765.80, N = 3 SE +/- 88880.09, N = 3 SE +/- 708563.50, N = 4 SE +/- 436034.06, N = 15 SE +/- 304616.08, N = 3 SE +/- 332801.18, N = 6 SE +/- 236966.05, N = 10 SE +/- 71263.56, N = 3 SE +/- 150250.14, N = 3 39043410 19191199 100908453 98847469 82397132 64152929 58384026 60864616 48804732 32973749 31246137 24314443 16034994 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better Stockfish 12 Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 120K 240K 360K 480K 600K 179617.03 162559.29 552076.25 560733.59 491891.78 391203.16 369416.83 419124.02 325661.01 271259.26 311552.67 246303.77 205655.17
Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 12 Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20M 40M 60M 80M 100M Min: 38530677 / Avg: 39043410.33 / Max: 39868176 Min: 19064701 / Avg: 19191199.33 / Max: 19338721 Min: 99825755 / Avg: 100908453 / Max: 101887714 Min: 95322228 / Avg: 98847468.5 / Max: 100858323 Min: 82260216 / Avg: 82397132 / Max: 82546157 Min: 63992140 / Avg: 64152928.67 / Max: 64298968 Min: 56735799 / Avg: 58384025.75 / Max: 60090226 Min: 57227030 / Avg: 60864616.13 / Max: 63501145 Min: 48420286 / Avg: 48804732.33 / Max: 49406252 Min: 32141990 / Avg: 32973748.83 / Max: 34537511 Min: 30342075 / Avg: 31246136.8 / Max: 32525396 Min: 24195092 / Avg: 24314443.33 / Max: 24441587 Min: 15735361 / Avg: 16034994 / Max: 16204568 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
Radiance Benchmark This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Radiance Benchmark 5.0 Test: SMP Parallel EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 184.49 184.74 214.77 219.26 218.42 217.90 214.26 217.33 215.16 217.55 218.14 224.52 223.38 226.64
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1100 2200 3300 4400 5500 SE +/- 1.31, N = 3 SE +/- 4.62, N = 3 SE +/- 6.77, N = 3 SE +/- 6.15, N = 3 SE +/- 4.94, N = 3 SE +/- 5.33, N = 3 SE +/- 2.78, N = 3 SE +/- 10.25, N = 3 SE +/- 1.53, N = 3 SE +/- 14.72, N = 3 SE +/- 5.57, N = 3 SE +/- 2.79, N = 3 SE +/- 9.03, N = 3 SE +/- 3.19, N = 3 2643.39 2399.19 5170.99 4731.72 4132.13 4197.60 3276.52 3757.35 3578.09 3310.91 3145.89 3344.39 3264.49 3071.91
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 Min: 2641.61 / Avg: 2643.39 / Max: 2645.95 Min: 2393.6 / Avg: 2399.19 / Max: 2408.36 Min: 5162.09 / Avg: 5170.99 / Max: 5184.28 Min: 4723.88 / Avg: 4731.72 / Max: 4743.85 Min: 4122.78 / Avg: 4132.13 / Max: 4139.56 Min: 4187.64 / Avg: 4197.6 / Max: 4205.86 Min: 3273.17 / Avg: 3276.52 / Max: 3282.04 Min: 3741.86 / Avg: 3757.35 / Max: 3776.73 Min: 3576.1 / Avg: 3578.09 / Max: 3581.09 Min: 3286.85 / Avg: 3310.91 / Max: 3337.63 Min: 3138.42 / Avg: 3145.89 / Max: 3156.77 Min: 3340.81 / Avg: 3344.39 / Max: 3349.89 Min: 3253.32 / Avg: 3264.49 / Max: 3282.36 Min: 3065.98 / Avg: 3071.91 / Max: 3076.9
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 2.99 1.66 6.02 6.62 5.67 5.53 4.74 4.20 4.37 3.53 2.51 2.37 1.81 1.29
Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 2.97 / Avg: 2.99 / Max: 3.01 Min: 1.65 / Avg: 1.66 / Max: 1.67 Min: 6.01 / Avg: 6.02 / Max: 6.03 Min: 6.6 / Avg: 6.62 / Max: 6.64 Min: 5.66 / Avg: 5.67 / Max: 5.68 Min: 5.51 / Avg: 5.53 / Max: 5.56 Min: 4.72 / Avg: 4.74 / Max: 4.75 Min: 4.19 / Avg: 4.2 / Max: 4.21 Min: 4.33 / Avg: 4.37 / Max: 4.4 Min: 3.52 / Avg: 3.53 / Max: 3.56 Min: 2.5 / Avg: 2.51 / Max: 2.52 Min: 2.34 / Avg: 2.37 / Max: 2.38 Min: 1.8 / Avg: 1.81 / Max: 1.82 Min: 1.29 / Avg: 1.29 / Max: 1.3
KeyDB A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.0.16 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90K 180K 270K 360K 450K SE +/- 428.23, N = 3 SE +/- 183.84, N = 3 SE +/- 4507.61, N = 3 SE +/- 4382.81, N = 3 SE +/- 4280.31, N = 3 SE +/- 5135.26, N = 3 SE +/- 3270.77, N = 3 SE +/- 2791.26, N = 3 SE +/- 430.27, N = 3 SE +/- 3460.73, N = 3 SE +/- 639.41, N = 3 SE +/- 1079.81, N = 3 SE +/- 1365.59, N = 3 SE +/- 533.37, N = 3 433091.73 424600.07 371448.21 376241.11 386824.54 394639.23 422944.46 404313.15 413354.59 415583.96 405905.78 420324.96 418080.47 399110.88 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Ops/sec Per Watt
OpenBenchmarking.org Ops/sec Per Watt, More Is Better KeyDB 6.0.16 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1600 3200 4800 6400 8000 4143.95 5640.60 3833.65 4095.78 4228.11 4838.07 6102.85 4517.76 5856.34 5903.16 6206.07 7158.88 7232.18 7415.88
Result Confidence
OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.0.16 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80K 160K 240K 320K 400K Min: 432304.57 / Avg: 433091.73 / Max: 433777.59 Min: 424252.48 / Avg: 424600.07 / Max: 424877.7 Min: 362545.96 / Avg: 371448.21 / Max: 377131.41 Min: 368183.94 / Avg: 376241.11 / Max: 383259.45 Min: 379984.9 / Avg: 386824.54 / Max: 394702.8 Min: 385396.59 / Avg: 394639.23 / Max: 403139.08 Min: 416505.35 / Avg: 422944.46 / Max: 427162.66 Min: 399536.5 / Avg: 404313.15 / Max: 409203.69 Min: 412498.94 / Avg: 413354.59 / Max: 413861.79 Min: 412054.17 / Avg: 415583.96 / Max: 422504.97 Min: 404993.42 / Avg: 405905.78 / Max: 407138 Min: 419061.55 / Avg: 420324.96 / Max: 422473.5 Min: 416043.78 / Avg: 418080.47 / Max: 420674.69 Min: 398044.81 / Avg: 399110.88 / Max: 399676.81 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1100 2200 3300 4400 5500 SE +/- 4.58, N = 3 SE +/- 1.90, N = 3 SE +/- 6.37, N = 3 SE +/- 4.63, N = 3 SE +/- 14.85, N = 3 SE +/- 8.65, N = 3 SE +/- 5.09, N = 3 SE +/- 3.83, N = 3 SE +/- 1.62, N = 3 SE +/- 4.18, N = 3 SE +/- 5.33, N = 3 SE +/- 1.17, N = 3 SE +/- 8.60, N = 3 SE +/- 10.38, N = 3 2649.35 2409.03 5153.99 4732.73 4138.69 4203.81 3271.26 3754.85 3582.56 3315.63 3155.85 3342.13 3265.93 3090.58
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 Min: 2640.2 / Avg: 2649.35 / Max: 2654.02 Min: 2405.41 / Avg: 2409.03 / Max: 2411.84 Min: 5142.04 / Avg: 5153.99 / Max: 5163.81 Min: 4726.94 / Avg: 4732.73 / Max: 4741.89 Min: 4119.13 / Avg: 4138.69 / Max: 4167.82 Min: 4186.54 / Avg: 4203.81 / Max: 4213.33 Min: 3264.96 / Avg: 3271.26 / Max: 3281.33 Min: 3748.56 / Avg: 3754.85 / Max: 3761.77 Min: 3579.4 / Avg: 3582.56 / Max: 3584.73 Min: 3309.61 / Avg: 3315.63 / Max: 3323.66 Min: 3145.96 / Avg: 3155.85 / Max: 3164.23 Min: 3339.82 / Avg: 3342.13 / Max: 3343.55 Min: 3248.92 / Avg: 3265.93 / Max: 3276.67 Min: 3072.59 / Avg: 3090.58 / Max: 3108.55
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 2.98 1.64 6.05 6.60 5.67 5.53 4.70 4.19 4.36 3.51 2.51 2.38 1.81 1.28
Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Person Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 2.97 / Avg: 2.98 / Max: 2.99 Min: 1.64 / Avg: 1.64 / Max: 1.65 Min: 6.03 / Avg: 6.05 / Max: 6.06 Min: 6.58 / Avg: 6.6 / Max: 6.63 Min: 5.66 / Avg: 5.67 / Max: 5.68 Min: 5.52 / Avg: 5.53 / Max: 5.54 Min: 4.69 / Avg: 4.7 / Max: 4.72 Min: 4.18 / Avg: 4.19 / Max: 4.2 Min: 4.32 / Avg: 4.36 / Max: 4.4 Min: 3.5 / Avg: 3.51 / Max: 3.53 Min: 2.5 / Avg: 2.51 / Max: 2.52 Min: 2.37 / Avg: 2.38 / Max: 2.39 Min: 1.8 / Avg: 1.81 / Max: 1.82 Min: 1.27 / Avg: 1.28 / Max: 1.29
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 3.08, N = 3 SE +/- 19.31, N = 3 SE +/- 12.73, N = 3 SE +/- 0.31, N = 3 SE +/- 2.50, N = 3 SE +/- 2.44, N = 3 SE +/- 3.65, N = 3 SE +/- 1.33, N = 3 SE +/- 1.04, N = 3 SE +/- 12.54, N = 3 SE +/- 1.52, N = 3 SE +/- 7.78, N = 3 SE +/- 9.26, N = 3 SE +/- 14.48, N = 3 1976.70 1836.51 4029.98 3562.63 3059.05 3325.99 2579.79 2815.80 2794.25 2507.58 2414.74 2527.86 2434.17 2298.63
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700 1400 2100 2800 3500 Min: 1971.22 / Avg: 1976.7 / Max: 1981.88 Min: 1797.95 / Avg: 1836.51 / Max: 1857.52 Min: 4015.57 / Avg: 4029.98 / Max: 4055.36 Min: 3562.19 / Avg: 3562.63 / Max: 3563.22 Min: 3056.48 / Avg: 3059.05 / Max: 3064.04 Min: 3322.37 / Avg: 3325.99 / Max: 3330.62 Min: 2575.39 / Avg: 2579.79 / Max: 2587.03 Min: 2813.23 / Avg: 2815.8 / Max: 2817.7 Min: 2792.4 / Avg: 2794.25 / Max: 2795.99 Min: 2482.49 / Avg: 2507.58 / Max: 2520.14 Min: 2412.09 / Avg: 2414.74 / Max: 2417.36 Min: 2515.93 / Avg: 2527.86 / Max: 2542.47 Min: 2418.87 / Avg: 2434.17 / Max: 2450.86 Min: 2271.79 / Avg: 2298.63 / Max: 2321.46
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 4.04 2.16 7.88 8.97 7.81 7.20 6.17 5.68 5.73 4.76 3.31 3.16 2.47 1.75
Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP16 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 4.02 / Avg: 4.04 / Max: 4.07 Min: 2.14 / Avg: 2.16 / Max: 2.2 Min: 7.84 / Avg: 7.88 / Max: 7.9 Min: 8.97 / Avg: 8.97 / Max: 8.97 Min: 7.8 / Avg: 7.81 / Max: 7.83 Min: 7.19 / Avg: 7.2 / Max: 7.22 Min: 6.13 / Avg: 6.17 / Max: 6.19 Min: 5.67 / Avg: 5.68 / Max: 5.68 Min: 5.73 / Avg: 5.73 / Max: 5.73 Min: 4.74 / Avg: 4.76 / Max: 4.8 Min: 3.3 / Avg: 3.31 / Max: 3.32 Min: 3.14 / Avg: 3.16 / Max: 3.18 Min: 2.45 / Avg: 2.47 / Max: 2.48 Min: 1.72 / Avg: 1.75 / Max: 1.76
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 1.86, N = 3 SE +/- 14.55, N = 3 SE +/- 8.92, N = 3 SE +/- 0.45, N = 3 SE +/- 1.92, N = 3 SE +/- 3.41, N = 3 SE +/- 2.29, N = 3 SE +/- 2.24, N = 3 SE +/- 0.81, N = 3 SE +/- 3.50, N = 3 SE +/- 2.67, N = 3 SE +/- 13.37, N = 3 SE +/- 34.12, N = 3 SE +/- 17.95, N = 3 1980.56 1844.31 4030.53 3558.09 3069.38 3327.50 2572.36 2814.15 2791.54 2522.61 2413.45 2446.02 2406.22 2314.66
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700 1400 2100 2800 3500 Min: 1978.38 / Avg: 1980.56 / Max: 1984.26 Min: 1815.75 / Avg: 1844.31 / Max: 1863.4 Min: 4020.18 / Avg: 4030.53 / Max: 4048.29 Min: 3557.24 / Avg: 3558.09 / Max: 3558.76 Min: 3065.61 / Avg: 3069.38 / Max: 3071.93 Min: 3322.21 / Avg: 3327.5 / Max: 3333.86 Min: 2567.84 / Avg: 2572.36 / Max: 2575.2 Min: 2809.91 / Avg: 2814.15 / Max: 2817.55 Min: 2789.93 / Avg: 2791.54 / Max: 2792.44 Min: 2516.2 / Avg: 2522.61 / Max: 2528.27 Min: 2410.34 / Avg: 2413.45 / Max: 2418.77 Min: 2426.14 / Avg: 2446.02 / Max: 2471.46 Min: 2355.5 / Avg: 2406.22 / Max: 2471.13 Min: 2283.58 / Avg: 2314.66 / Max: 2345.75
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 4.02 2.16 7.88 8.98 7.79 7.22 6.13 5.68 5.72 4.73 3.32 3.23 2.47 1.74
Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Face Detection 0106 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 4.02 / Avg: 4.02 / Max: 4.03 Min: 2.14 / Avg: 2.16 / Max: 2.18 Min: 7.84 / Avg: 7.88 / Max: 7.9 Min: 8.97 / Avg: 8.98 / Max: 9.01 Min: 7.78 / Avg: 7.79 / Max: 7.81 Min: 7.19 / Avg: 7.22 / Max: 7.24 Min: 6.03 / Avg: 6.13 / Max: 6.19 Min: 5.67 / Avg: 5.68 / Max: 5.69 Min: 5.7 / Avg: 5.72 / Max: 5.74 Min: 4.68 / Avg: 4.73 / Max: 4.79 Min: 3.31 / Avg: 3.32 / Max: 3.32 Min: 3.2 / Avg: 3.23 / Max: 3.25 Min: 2.44 / Avg: 2.47 / Max: 2.5 Min: 1.71 / Avg: 1.74 / Max: 1.76
Crypto++ Crypto++ is a C++ class library of cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70 140 210 280 350 SE +/- 0.17, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.18, N = 3 SE +/- 0.93, N = 3 SE +/- 0.10, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.25, N = 3 SE +/- 0.17, N = 3 SE +/- 0.14, N = 3 342.93 343.61 298.73 293.81 292.60 293.88 302.93 294.27 298.71 298.49 294.44 285.25 286.18 286.19 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
MiB/second Per Watt
OpenBenchmarking.org MiB/second Per Watt, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 4.76 6.50 4.29 4.31 4.52 5.14 6.26 4.26 6.13 6.18 5.98 6.83 7.01 7.09
Result Confidence
OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 342.6 / Avg: 342.93 / Max: 343.11 Min: 343.56 / Avg: 343.61 / Max: 343.68 Min: 298.7 / Avg: 298.73 / Max: 298.75 Min: 293.61 / Avg: 293.81 / Max: 294.16 Min: 290.74 / Avg: 292.6 / Max: 293.53 Min: 293.7 / Avg: 293.88 / Max: 294.04 Min: 302.75 / Avg: 302.93 / Max: 303.14 Min: 294.13 / Avg: 294.27 / Max: 294.49 Min: 298.57 / Avg: 298.71 / Max: 298.93 Min: 298.33 / Avg: 298.49 / Max: 298.7 Min: 294.21 / Avg: 294.44 / Max: 294.59 Min: 284.78 / Avg: 285.25 / Max: 285.61 Min: 286.01 / Avg: 286.18 / Max: 286.52 Min: 285.95 / Avg: 286.19 / Max: 286.43 1. (CXX) g++ options: -g2 -O3 -fPIC -pthread -pipe
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Read EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50M 100M 150M 200M 250M SE +/- 33050.19, N = 3 SE +/- 173513.15, N = 3 SE +/- 2628540.04, N = 3 SE +/- 1573991.80, N = 3 SE +/- 1627088.40, N = 3 SE +/- 1503518.94, N = 3 SE +/- 979987.31, N = 3 SE +/- 577476.24, N = 3 SE +/- 298644.62, N = 3 SE +/- 701112.41, N = 3 SE +/- 731623.27, N = 6 SE +/- 860580.20, N = 3 SE +/- 532212.35, N = 3 SE +/- 468827.18, N = 3 93883130 48331735 218469824 214717760 185712982 185095008 147488280 133038550 139338056 114391434 76972090 74816560 56928967 39206156 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Facebook RocksDB 6.3.6 Test: Random Read EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K 405107.75 383902.17 1144599.80 1165531.87 966385.49 1055932.79 861998.88 820332.86 931173.20 752967.14 623360.93 744401.17 561814.90 482400.32
Result Confidence
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Read EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40M 80M 120M 160M 200M Min: 93839853 / Avg: 93883129.67 / Max: 93948038 Min: 48063107 / Avg: 48331734.67 / Max: 48656311 Min: 213921113 / Avg: 218469824.33 / Max: 223026632 Min: 211571907 / Avg: 214717760.33 / Max: 216390965 Min: 182486843 / Avg: 185712982.33 / Max: 187695198 Min: 182409655 / Avg: 185095007.67 / Max: 187609592 Min: 146389489 / Avg: 147488280.33 / Max: 149443243 Min: 131941123 / Avg: 133038549.67 / Max: 133898996 Min: 138995330 / Avg: 139338056 / Max: 139933058 Min: 112997116 / Avg: 114391434.33 / Max: 115217369 Min: 75222240 / Avg: 76972090.17 / Max: 79656987 Min: 73275140 / Avg: 74816560.33 / Max: 76250454 Min: 55956147 / Avg: 56928967.33 / Max: 57789497 Min: 38308495 / Avg: 39206155.67 / Max: 39889615 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.2745 0.549 0.8235 1.098 1.3725 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 0.79 0.78 1.22 1.07 1.02 0.99 0.77 0.95 0.84 0.88 0.84 0.79 0.87 0.96
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.78 / Avg: 0.79 / Max: 0.79 Min: 0.78 / Avg: 0.78 / Max: 0.79 Min: 1.21 / Avg: 1.22 / Max: 1.22 Min: 1.06 / Avg: 1.07 / Max: 1.07 Min: 1.01 / Avg: 1.02 / Max: 1.02 Min: 0.98 / Avg: 0.99 / Max: 0.99 Min: 0.77 / Avg: 0.77 / Max: 0.78 Min: 0.95 / Avg: 0.95 / Max: 0.95 Min: 0.83 / Avg: 0.84 / Max: 0.84 Min: 0.87 / Avg: 0.88 / Max: 0.91 Min: 0.83 / Avg: 0.84 / Max: 0.84 Min: 0.77 / Avg: 0.79 / Max: 0.81 Min: 0.87 / Avg: 0.87 / Max: 0.87 Min: 0.94 / Avg: 0.96 / Max: 0.99
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6K 12K 18K 24K 30K SE +/- 47.83, N = 3 SE +/- 6.26, N = 3 SE +/- 48.27, N = 3 SE +/- 30.27, N = 3 SE +/- 4.60, N = 3 SE +/- 49.13, N = 3 SE +/- 75.75, N = 3 SE +/- 24.60, N = 3 SE +/- 60.57, N = 3 SE +/- 157.24, N = 3 SE +/- 6.65, N = 3 SE +/- 103.67, N = 4 SE +/- 21.76, N = 3 SE +/- 46.05, N = 4 9866.08 4984.35 25012.98 28316.50 22996.93 23559.17 19958.41 16443.73 18514.60 13220.95 9314.82 9884.62 6803.09 4115.44
Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2021.1 Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K Min: 9816.8 / Avg: 9866.08 / Max: 9961.72 Min: 4971.91 / Avg: 4984.35 / Max: 4991.8 Min: 24963.18 / Avg: 25012.98 / Max: 25109.5 Min: 28279.61 / Avg: 28316.5 / Max: 28376.51 Min: 22990.44 / Avg: 22996.93 / Max: 23005.82 Min: 23467.34 / Avg: 23559.17 / Max: 23635.37 Min: 19820.95 / Avg: 19958.41 / Max: 20082.32 Min: 16395.26 / Avg: 16443.73 / Max: 16475.33 Min: 18416.86 / Avg: 18514.6 / Max: 18625.46 Min: 12906.62 / Avg: 13220.95 / Max: 13386.45 Min: 9307.85 / Avg: 9314.82 / Max: 9328.12 Min: 9618.02 / Avg: 9884.62 / Max: 10112.13 Min: 6767.48 / Avg: 6803.09 / Max: 6842.56 Min: 4003.8 / Avg: 4115.44 / Max: 4226.83
IndigoBench This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Bedroom EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.006, N = 3 SE +/- 0.013, N = 3 SE +/- 0.020, N = 3 SE +/- 0.007, N = 3 SE +/- 0.002, N = 3 SE +/- 0.008, N = 3 SE +/- 0.006, N = 3 SE +/- 0.009, N = 3 SE +/- 0.006, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 3.461 1.808 8.811 8.898 7.596 7.555 5.950 5.415 5.477 4.494 3.129 2.932 2.289 1.456
M samples/s Per Watt
OpenBenchmarking.org M samples/s Per Watt, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Bedroom EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0113 0.0226 0.0339 0.0452 0.0565 0.02 0.02 0.05 0.05 0.04 0.04 0.03 0.03 0.04 0.03 0.03 0.03 0.02 0.02
Result Confidence
OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Bedroom EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 3.46 / Avg: 3.46 / Max: 3.47 Min: 1.8 / Avg: 1.81 / Max: 1.81 Min: 8.8 / Avg: 8.81 / Max: 8.82 Min: 8.88 / Avg: 8.9 / Max: 8.93 Min: 7.57 / Avg: 7.6 / Max: 7.64 Min: 7.54 / Avg: 7.56 / Max: 7.57 Min: 5.95 / Avg: 5.95 / Max: 5.96 Min: 5.4 / Avg: 5.42 / Max: 5.43 Min: 5.47 / Avg: 5.48 / Max: 5.49 Min: 4.48 / Avg: 4.49 / Max: 4.51 Min: 3.12 / Avg: 3.13 / Max: 3.14 Min: 2.93 / Avg: 2.93 / Max: 2.93 Min: 2.29 / Avg: 2.29 / Max: 2.29 Min: 1.45 / Avg: 1.46 / Max: 1.46
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.4 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.43, N = 3 SE +/- 0.46, N = 3 SE +/- 0.26, N = 7 SE +/- 0.28, N = 6 SE +/- 0.30, N = 6 SE +/- 0.30, N = 6 SE +/- 0.41, N = 4 SE +/- 0.42, N = 4 SE +/- 0.42, N = 4 SE +/- 0.47, N = 4 SE +/- 0.65, N = 4 SE +/- 0.75, N = 3 SE +/- 0.59, N = 3 SE +/- 0.42, N = 3 46.91 79.66 28.13 27.98 30.08 30.21 34.55 36.85 36.39 41.50 53.40 57.91 69.61 101.09
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.4 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 46.21 / Avg: 46.91 / Max: 47.7 Min: 79.06 / Avg: 79.66 / Max: 80.57 Min: 27.79 / Avg: 28.13 / Max: 29.66 Min: 27.59 / Avg: 27.98 / Max: 29.37 Min: 29.67 / Avg: 30.08 / Max: 31.58 Min: 29.87 / Avg: 30.21 / Max: 31.71 Min: 34.07 / Avg: 34.55 / Max: 35.78 Min: 36.38 / Avg: 36.85 / Max: 38.1 Min: 35.93 / Avg: 36.39 / Max: 37.65 Min: 40.71 / Avg: 41.5 / Max: 42.85 Min: 52.39 / Avg: 53.4 / Max: 55.3 Min: 56.64 / Avg: 57.91 / Max: 59.25 Min: 68.52 / Avg: 69.61 / Max: 70.53 Min: 100.61 / Avg: 101.09 / Max: 101.92
IndigoBench This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Supercar EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.013, N = 3 SE +/- 0.002, N = 3 SE +/- 0.043, N = 3 SE +/- 0.063, N = 3 SE +/- 0.035, N = 3 SE +/- 0.038, N = 3 SE +/- 0.008, N = 3 SE +/- 0.016, N = 3 SE +/- 0.031, N = 3 SE +/- 0.014, N = 3 SE +/- 0.006, N = 3 SE +/- 0.005, N = 3 SE +/- 0.010, N = 3 SE +/- 0.002, N = 3 7.684 3.883 19.095 19.062 16.230 15.984 12.738 11.531 11.854 9.630 6.575 6.214 4.855 3.117
M samples/s Per Watt
OpenBenchmarking.org M samples/s Per Watt, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Supercar EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0225 0.045 0.0675 0.09 0.1125 0.03 0.03 0.10 0.10 0.09 0.09 0.08 0.07 0.08 0.06 0.05 0.06 0.05 0.04
Result Confidence
OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Supercar EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 7.67 / Avg: 7.68 / Max: 7.71 Min: 3.88 / Avg: 3.88 / Max: 3.89 Min: 19.02 / Avg: 19.1 / Max: 19.16 Min: 18.94 / Avg: 19.06 / Max: 19.14 Min: 16.18 / Avg: 16.23 / Max: 16.3 Min: 15.93 / Avg: 15.98 / Max: 16.06 Min: 12.72 / Avg: 12.74 / Max: 12.75 Min: 11.52 / Avg: 11.53 / Max: 11.56 Min: 11.8 / Avg: 11.85 / Max: 11.9 Min: 9.6 / Avg: 9.63 / Max: 9.65 Min: 6.57 / Avg: 6.58 / Max: 6.59 Min: 6.21 / Avg: 6.21 / Max: 6.22 Min: 4.84 / Avg: 4.85 / Max: 4.87 Min: 3.11 / Avg: 3.12 / Max: 3.12
LuxCoreRender LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on the CPU as opposed to the OpenCL version. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.3 Scene: DLSC EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.27 1.68 7.04 7.21 6.19 6.13 5.03 4.54 4.65 3.83 2.68 2.53 1.98 1.38 MIN: 3.18 / MAX: 3.41 MIN: 1.61 / MAX: 1.72 MIN: 6.88 / MAX: 7.39 MIN: 7.03 / MAX: 7.62 MIN: 6.04 / MAX: 6.75 MIN: 5.97 / MAX: 6.57 MIN: 4.8 / MAX: 5.35 MIN: 4.38 / MAX: 4.8 MIN: 4.49 / MAX: 5.07 MIN: 3.74 / MAX: 4.01 MIN: 2.58 / MAX: 2.77 MIN: 2.46 / MAX: 2.62 MIN: 1.91 / MAX: 2.03 MIN: 1.34 / MAX: 1.42
M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.3 Scene: DLSC EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.009 0.018 0.027 0.036 0.045 0.01 0.01 0.04 0.04 0.03 0.04 0.03 0.03 0.03 0.03 0.02 0.03 0.02 0.02
Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.3 Scene: DLSC EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 3.26 / Avg: 3.27 / Max: 3.29 Min: 1.66 / Avg: 1.68 / Max: 1.69 Min: 6.98 / Avg: 7.04 / Max: 7.07 Min: 7.13 / Avg: 7.21 / Max: 7.27 Min: 6.14 / Avg: 6.19 / Max: 6.26 Min: 6.07 / Avg: 6.13 / Max: 6.22 Min: 4.96 / Avg: 5.03 / Max: 5.09 Min: 4.47 / Avg: 4.54 / Max: 4.64 Min: 4.58 / Avg: 4.65 / Max: 4.76 Min: 3.81 / Avg: 3.83 / Max: 3.84 Min: 2.66 / Avg: 2.68 / Max: 2.71 Min: 2.52 / Avg: 2.53 / Max: 2.55 Min: 1.97 / Avg: 1.98 / Max: 1.99 Min: 1.37 / Avg: 1.38 / Max: 1.39
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.3 Scene: Rainbow Colors and Prism EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 3.57 1.84 7.82 7.92 6.81 6.75 5.54 4.92 5.04 4.17 2.91 2.74 2.14 1.50 MIN: 3.48 / MAX: 3.63 MIN: 1.81 / MAX: 1.92 MIN: 7.53 / MAX: 7.98 MIN: 7.72 / MAX: 8.01 MIN: 6.61 / MAX: 6.89 MIN: 6.64 / MAX: 6.79 MIN: 5.5 / MAX: 5.57 MIN: 4.81 / MAX: 5.05 MIN: 4.94 / MAX: 5.12 MIN: 4.11 / MAX: 4.26 MIN: 2.88 / MAX: 2.96 MIN: 2.69 / MAX: 2.78 MIN: 2.08 / MAX: 2.17 MIN: 1.48 / MAX: 1.54
M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.3 Scene: Rainbow Colors and Prism EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.009 0.018 0.027 0.036 0.045 0.02 0.01 0.04 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.02 0.03 0.02 0.02
Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.3 Scene: Rainbow Colors and Prism EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 3.53 / Avg: 3.57 / Max: 3.6 Min: 1.83 / Avg: 1.84 / Max: 1.87 Min: 7.71 / Avg: 7.82 / Max: 7.94 Min: 7.83 / Avg: 7.92 / Max: 7.98 Min: 6.71 / Avg: 6.81 / Max: 6.88 Min: 6.73 / Avg: 6.75 / Max: 6.78 Min: 5.52 / Avg: 5.54 / Max: 5.55 Min: 4.81 / Avg: 4.92 / Max: 5.01 Min: 4.94 / Avg: 5.04 / Max: 5.12 Min: 4.11 / Avg: 4.17 / Max: 4.22 Min: 2.89 / Avg: 2.91 / Max: 2.95 Min: 2.73 / Avg: 2.74 / Max: 2.76 Min: 2.12 / Avg: 2.14 / Max: 2.15 Min: 1.5 / Avg: 1.5 / Max: 1.51
Kvazaar This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 10.66 5.73 14.94 16.44 15.94 15.50 14.94 13.65 13.97 11.22 9.15 8.61 6.49 4.72 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0225 0.045 0.0675 0.09 0.1125 0.05 0.05 0.09 0.10 0.09 0.10 0.09 0.09 0.10 0.08 0.08 0.09 0.07 0.06
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 10.64 / Avg: 10.66 / Max: 10.69 Min: 5.71 / Avg: 5.73 / Max: 5.74 Min: 14.93 / Avg: 14.94 / Max: 14.97 Min: 16.42 / Avg: 16.44 / Max: 16.48 Min: 15.92 / Avg: 15.94 / Max: 15.97 Min: 15.49 / Avg: 15.5 / Max: 15.51 Min: 14.93 / Avg: 14.94 / Max: 14.95 Min: 13.61 / Avg: 13.65 / Max: 13.67 Min: 13.95 / Avg: 13.97 / Max: 13.98 Min: 11.18 / Avg: 11.22 / Max: 11.24 Min: 9.13 / Avg: 9.15 / Max: 9.17 Min: 8.58 / Avg: 8.61 / Max: 8.63 Min: 6.48 / Avg: 6.49 / Max: 6.49 Min: 4.72 / Avg: 4.72 / Max: 4.73 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
TensorFlow Lite This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: SqueezeNet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50K 100K 150K 200K 250K SE +/- 132.22, N = 3 SE +/- 44.74, N = 3 SE +/- 39.50, N = 3 SE +/- 174.48, N = 3 SE +/- 111.10, N = 3 SE +/- 74.34, N = 3 SE +/- 77.46, N = 3 SE +/- 39.58, N = 3 SE +/- 126.18, N = 3 SE +/- 6.99, N = 3 SE +/- 25.85, N = 3 SE +/- 38.26, N = 3 SE +/- 82.39, N = 3 SE +/- 202.36, N = 3 107391.0 201258.0 61679.6 56195.5 65104.4 68430.4 70622.9 77264.7 75080.0 93404.1 123947.0 129293.0 173807.0 242480.0
Result Confidence
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: SqueezeNet EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40K 80K 120K 160K 200K Min: 107127 / Avg: 107390.67 / Max: 107540 Min: 201180 / Avg: 201257.67 / Max: 201335 Min: 61602.1 / Avg: 61679.57 / Max: 61731.7 Min: 55903.9 / Avg: 56195.5 / Max: 56507.3 Min: 64904.3 / Avg: 65104.43 / Max: 65288.1 Min: 68340.5 / Avg: 68430.4 / Max: 68577.9 Min: 70510.4 / Avg: 70622.9 / Max: 70771.4 Min: 77193.3 / Avg: 77264.7 / Max: 77330 Min: 74862.2 / Avg: 75080 / Max: 75299.3 Min: 93391.7 / Avg: 93404.1 / Max: 93415.9 Min: 123895 / Avg: 123946.67 / Max: 123974 Min: 129222 / Avg: 129293.33 / Max: 129353 Min: 173701 / Avg: 173806.67 / Max: 173969 Min: 242075 / Avg: 242479.67 / Max: 242688
Result
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Float EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K SE +/- 13.80, N = 3 SE +/- 599.00, N = 3 SE +/- 280.28, N = 3 SE +/- 187.50, N = 3 SE +/- 322.42, N = 3 SE +/- 194.69, N = 3 SE +/- 86.41, N = 3 SE +/- 28.71, N = 3 SE +/- 76.82, N = 3 SE +/- 193.63, N = 3 SE +/- 21.22, N = 3 SE +/- 63.38, N = 3 SE +/- 107.22, N = 3 SE +/- 22.85, N = 3 68637.6 133236.0 35037.4 32085.5 45436.6 48507.4 43977.9 48084.7 47083.4 61419.9 80320.5 84723.5 117060.0 160713.0
Result Confidence
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Float EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K Min: 68610.2 / Avg: 68637.6 / Max: 68654.2 Min: 132636 / Avg: 133236 / Max: 134434 Min: 34603.7 / Avg: 35037.37 / Max: 35561.8 Min: 31772.4 / Avg: 32085.53 / Max: 32420.8 Min: 44937.4 / Avg: 45436.6 / Max: 46039.7 Min: 48205.1 / Avg: 48507.43 / Max: 48871.1 Min: 43815.4 / Avg: 43977.87 / Max: 44110.1 Min: 48044.8 / Avg: 48084.67 / Max: 48140.4 Min: 46946.3 / Avg: 47083.37 / Max: 47212 Min: 61061.3 / Avg: 61419.9 / Max: 61725.8 Min: 80284 / Avg: 80320.5 / Max: 80357.5 Min: 84636.8 / Avg: 84723.47 / Max: 84846.9 Min: 116941 / Avg: 117060 / Max: 117274 Min: 160685 / Avg: 160712.67 / Max: 160758
Result
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Quant EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40K 80K 120K 160K 200K SE +/- 61.58, N = 3 SE +/- 13.05, N = 3 SE +/- 118.11, N = 3 SE +/- 61.64, N = 3 SE +/- 392.20, N = 3 SE +/- 518.49, N = 3 SE +/- 72.07, N = 3 SE +/- 69.37, N = 3 SE +/- 44.27, N = 3 SE +/- 40.77, N = 3 SE +/- 15.21, N = 3 SE +/- 47.75, N = 3 SE +/- 28.48, N = 3 SE +/- 20.79, N = 3 70154.5 136023.0 36468.8 33024.0 45818.8 50401.8 45118.1 49078.5 47911.7 62904.1 82490.4 86277.6 120119.0 165426.0
Result Confidence
OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Quant EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K Min: 70069.3 / Avg: 70154.47 / Max: 70274.1 Min: 135998 / Avg: 136023 / Max: 136042 Min: 36258.5 / Avg: 36468.83 / Max: 36667.1 Min: 32924.1 / Avg: 33024 / Max: 33136.5 Min: 45327.1 / Avg: 45818.77 / Max: 46593.9 Min: 49377.6 / Avg: 50401.83 / Max: 51054.3 Min: 45013.3 / Avg: 45118.1 / Max: 45256.2 Min: 48987 / Avg: 49078.53 / Max: 49214.6 Min: 47856.7 / Avg: 47911.7 / Max: 47999.3 Min: 62851.8 / Avg: 62904.07 / Max: 62984.4 Min: 82465 / Avg: 82490.43 / Max: 82517.6 Min: 86198 / Avg: 86277.57 / Max: 86363.1 Min: 120062 / Avg: 120118.67 / Max: 120152 Min: 165394 / Avg: 165426 / Max: 165465
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Sync EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70K 140K 210K 280K 350K SE +/- 436.15, N = 3 SE +/- 175.05, N = 3 SE +/- 166.32, N = 3 SE +/- 735.95, N = 3 SE +/- 768.33, N = 3 SE +/- 749.32, N = 3 SE +/- 344.84, N = 3 SE +/- 622.13, N = 3 SE +/- 219.22, N = 3 SE +/- 607.97, N = 3 SE +/- 203.74, N = 3 SE +/- 223.01, N = 3 SE +/- 204.78, N = 3 172585 100843 287115 333412 331791 331300 258759 253689 259353 216427 163248 163535 131583 94371 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Sync EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700 1400 2100 2800 3500 1704.39 1654.28 2752.63 2353.30 2741.84 3130.89 3314.42 2535.97 3252.08 3035.68 2693.49 3041.80 2715.04 2127.71
Result Confidence
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Sync EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60K 120K 180K 240K 300K Min: 171773 / Avg: 172585 / Max: 173267 Min: 100631 / Avg: 100842.67 / Max: 101190 Min: 286867 / Avg: 287115 / Max: 287431 Min: 332027 / Avg: 333412 / Max: 334536 Min: 330767 / Avg: 331790.67 / Max: 333295 Min: 329808 / Avg: 331300 / Max: 332168 Min: 258263 / Avg: 258759 / Max: 259422 Min: 252478 / Avg: 253689 / Max: 254542 Min: 259071 / Avg: 259353.33 / Max: 259785 Min: 215402 / Avg: 216427 / Max: 217506 Min: 163163 / Avg: 163535 / Max: 163865 Min: 131216 / Avg: 131583 / Max: 131986 Min: 93963 / Avg: 94371 / Max: 94606 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: MD5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900K 1800K 2700K 3600K 4500K SE +/- 1763.83, N = 3 SE +/- 525.27, N = 3 SE +/- 3282.95, N = 3 SE +/- 881.92, N = 3 SE +/- 881.92, N = 3 SE +/- 2905.93, N = 3 SE +/- 1527.53, N = 3 SE +/- 2185.81, N = 3 SE +/- 1452.97, N = 3 SE +/- 2603.42, N = 3 SE +/- 2000.00, N = 3 SE +/- 881.92, N = 3 SE +/- 798.79, N = 3 1731667 874700 4183000 4195333 3557667 3528667 2798667 2534000 2596333 2127333 1460333 1354000 1072333 717175 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 1.9.0-jumbo-1 Test: MD5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K 7625.61 7334.16 22232.88 22927.63 18839.03 20367.12 16435.49 15619.65 17709.04 13762.68 11851.02 13466.23 10769.05 9459.27
Result Confidence
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: MD5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700K 1400K 2100K 2800K 3500K Min: 1729000 / Avg: 1731666.67 / Max: 1735000 Min: 873984 / Avg: 874700.33 / Max: 875724 Min: 4189000 / Avg: 4195333.33 / Max: 4200000 Min: 3556000 / Avg: 3557666.67 / Max: 3559000 Min: 3527000 / Avg: 3528666.67 / Max: 3530000 Min: 2794000 / Avg: 2798666.67 / Max: 2804000 Min: 2531000 / Avg: 2534000 / Max: 2536000 Min: 2592000 / Avg: 2596333.33 / Max: 2599000 Min: 2125000 / Avg: 2127333.33 / Max: 2130000 Min: 1456000 / Avg: 1460333.33 / Max: 1465000 Min: 1352000 / Avg: 1354000 / Max: 1358000 Min: 1071000 / Avg: 1072333.33 / Max: 1074000 Min: 716134 / Avg: 717175 / Max: 718745 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: LargeRandom EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0855 0.171 0.2565 0.342 0.4275 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.38 0.38 0.33 0.32 0.32 0.32 0.33 0.32 0.33 0.33 0.33 0.32 0.32 0.31 1. (CXX) g++ options: -O3 -pthread
GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 0.7.1 Throughput Test: LargeRandom EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0023 0.0046 0.0069 0.0092 0.0115 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: LargeRandom EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1 2 3 4 5 Min: 0.38 / Avg: 0.38 / Max: 0.38 Min: 0.38 / Avg: 0.38 / Max: 0.39 Min: 0.33 / Avg: 0.33 / Max: 0.33 Min: 0.32 / Avg: 0.32 / Max: 0.33 Min: 0.32 / Avg: 0.32 / Max: 0.33 Min: 0.32 / Avg: 0.32 / Max: 0.33 Min: 0.33 / Avg: 0.33 / Max: 0.34 Min: 0.32 / Avg: 0.32 / Max: 0.32 Min: 0.33 / Avg: 0.33 / Max: 0.33 Min: 0.33 / Avg: 0.33 / Max: 0.33 Min: 0.32 / Avg: 0.33 / Max: 0.33 Min: 0.31 / Avg: 0.32 / Max: 0.32 Min: 0.31 / Avg: 0.32 / Max: 0.32 Min: 0.31 / Avg: 0.31 / Max: 0.31 1. (CXX) g++ options: -O3 -pthread
rav1e Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.2619 0.5238 0.7857 1.0476 1.3095 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.003, N = 3 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 1.156 1.164 1.005 0.993 0.989 0.992 1.024 0.999 1.006 1.009 0.997 0.969 0.972 0.965
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better rav1e 0.4 Speed: 5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0045 0.009 0.0135 0.018 0.0225 0.01 0.02 0.01 0.01 0.01 0.02 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.15 / Avg: 1.16 / Max: 1.16 Min: 1.16 / Avg: 1.16 / Max: 1.17 Min: 1 / Avg: 1.01 / Max: 1.01 Min: 0.99 / Avg: 0.99 / Max: 0.99 Min: 0.99 / Avg: 0.99 / Max: 0.99 Min: 0.99 / Avg: 0.99 / Max: 1 Min: 1.02 / Avg: 1.02 / Max: 1.03 Min: 1 / Avg: 1 / Max: 1 Min: 1 / Avg: 1.01 / Max: 1.01 Min: 1.01 / Avg: 1.01 / Max: 1.01 Min: 0.99 / Avg: 1 / Max: 1 Min: 0.97 / Avg: 0.97 / Max: 0.97 Min: 0.97 / Avg: 0.97 / Max: 0.97 Min: 0.96 / Avg: 0.96 / Max: 0.97
FinanceBench FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Repo OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11K 22K 33K 44K 55K SE +/- 174.69, N = 3 SE +/- 107.65, N = 3 SE +/- 45.54, N = 3 SE +/- 71.36, N = 3 SE +/- 241.95, N = 3 SE +/- 84.26, N = 3 SE +/- 133.22, N = 3 SE +/- 71.59, N = 3 SE +/- 17.38, N = 3 SE +/- 383.79, N = 3 SE +/- 109.50, N = 3 SE +/- 108.67, N = 3 SE +/- 275.73, N = 3 43527.28 43548.63 51015.87 51437.79 51630.50 50076.27 51428.89 50609.04 50474.70 51629.10 52977.60 52936.98 53237.68 1. (CXX) g++ options: -O3 -march=native -fopenmp
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better FinanceBench 2016-07-25 Benchmark: Repo OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9K 18K 27K 36K 45K Min: 43324.05 / Avg: 43527.28 / Max: 43875.02 Min: 43343.38 / Avg: 43548.63 / Max: 43707.56 Min: 50939.99 / Avg: 51015.87 / Max: 51097.43 Min: 51363.7 / Avg: 51437.79 / Max: 51580.48 Min: 51347.45 / Avg: 51630.5 / Max: 52111.92 Min: 49985.34 / Avg: 50076.27 / Max: 50244.61 Min: 51271.63 / Avg: 51428.89 / Max: 51693.79 Min: 50528.84 / Avg: 50609.04 / Max: 50751.85 Min: 50450.23 / Avg: 50474.7 / Max: 50508.32 Min: 51241.3 / Avg: 51629.1 / Max: 52396.66 Min: 52852.14 / Avg: 52977.6 / Max: 53195.79 Min: 52784.59 / Avg: 52936.98 / Max: 53147.39 Min: 52860.58 / Avg: 53237.68 / Max: 53774.7 1. (CXX) g++ options: -O3 -march=native -fopenmp
RawTherapee RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better RawTherapee Total Benchmark Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 SE +/- 0.12, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.59, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 53.15 62.24 56.03 55.04 54.49 52.13 54.77 52.74 53.93 58.51 61.54 62.76 72.47 1. RawTherapee, version 5.8, command line.
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better RawTherapee Total Benchmark Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 14 28 42 56 70 Min: 52.9 / Avg: 53.15 / Max: 53.28 Min: 62.13 / Avg: 62.24 / Max: 62.4 Min: 55.95 / Avg: 56.03 / Max: 56.09 Min: 55.03 / Avg: 55.04 / Max: 55.06 Min: 54.34 / Avg: 54.49 / Max: 54.59 Min: 52.04 / Avg: 52.12 / Max: 52.19 Min: 54.71 / Avg: 54.77 / Max: 54.89 Min: 52.69 / Avg: 52.74 / Max: 52.79 Min: 53.88 / Avg: 53.93 / Max: 54 Min: 58.49 / Avg: 58.51 / Max: 58.52 Min: 60.92 / Avg: 61.54 / Max: 62.73 Min: 62.66 / Avg: 62.76 / Max: 62.82 Min: 72.44 / Avg: 72.47 / Max: 72.51 1. RawTherapee, version 5.8, command line.
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.6132 1.2264 1.8396 2.4528 3.066 SE +/- 0.00173, N = 3 SE +/- 0.00063, N = 3 SE +/- 0.00053, N = 3 SE +/- 0.00014, N = 3 SE +/- 0.00034, N = 3 SE +/- 0.00037, N = 3 SE +/- 0.00071, N = 3 SE +/- 0.00074, N = 3 SE +/- 0.00041, N = 3 SE +/- 0.00125, N = 3 SE +/- 0.00048, N = 3 SE +/- 0.00125, N = 3 SE +/- 0.00045, N = 3 SE +/- 0.00080, N = 3 1.14375 2.22949 0.49264 0.48908 0.57048 0.57484 0.71549 0.79079 0.77439 0.93705 1.35058 1.45145 1.83895 2.72553
Result Confidence
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.14 / Avg: 1.14 / Max: 1.15 Min: 2.23 / Avg: 2.23 / Max: 2.23 Min: 0.49 / Avg: 0.49 / Max: 0.49 Min: 0.49 / Avg: 0.49 / Max: 0.49 Min: 0.57 / Avg: 0.57 / Max: 0.57 Min: 0.57 / Avg: 0.57 / Max: 0.58 Min: 0.71 / Avg: 0.72 / Max: 0.72 Min: 0.79 / Avg: 0.79 / Max: 0.79 Min: 0.77 / Avg: 0.77 / Max: 0.78 Min: 0.94 / Avg: 0.94 / Max: 0.94 Min: 1.35 / Avg: 1.35 / Max: 1.35 Min: 1.45 / Avg: 1.45 / Max: 1.45 Min: 1.84 / Avg: 1.84 / Max: 1.84 Min: 2.72 / Avg: 2.73 / Max: 2.73
DeepSpeech Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better DeepSpeech 0.6 Acceleration: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.45, N = 3 SE +/- 0.14, N = 3 SE +/- 0.28, N = 3 SE +/- 0.10, N = 3 SE +/- 0.58, N = 3 SE +/- 0.13, N = 3 SE +/- 0.14, N = 3 SE +/- 0.09, N = 3 SE +/- 0.52, N = 3 SE +/- 0.20, N = 3 SE +/- 0.22, N = 3 SE +/- 0.07, N = 3 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 71.67 69.51 70.11 70.67 70.17 70.72 72.13 71.56 72.21 72.68 73.17 74.99 73.79 72.99
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better DeepSpeech 0.6 Acceleration: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 14 28 42 56 70 Min: 70.79 / Avg: 71.67 / Max: 72.23 Min: 69.32 / Avg: 69.51 / Max: 69.77 Min: 69.57 / Avg: 70.11 / Max: 70.51 Min: 70.52 / Avg: 70.67 / Max: 70.87 Min: 69.04 / Avg: 70.17 / Max: 70.95 Min: 70.48 / Avg: 70.72 / Max: 70.91 Min: 71.89 / Avg: 72.13 / Max: 72.38 Min: 71.46 / Avg: 71.56 / Max: 71.73 Min: 71.26 / Avg: 72.21 / Max: 73.03 Min: 72.4 / Avg: 72.68 / Max: 73.08 Min: 72.74 / Avg: 73.17 / Max: 73.46 Min: 74.85 / Avg: 74.99 / Max: 75.07 Min: 73.4 / Avg: 73.79 / Max: 74.01 Min: 72.82 / Avg: 72.99 / Max: 73.3
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K SE +/- 1.29, N = 3 SE +/- 4.94, N = 3 SE +/- 67.71, N = 3 SE +/- 2.47, N = 3 SE +/- 23.45, N = 3 SE +/- 24.53, N = 3 SE +/- 36.28, N = 3 SE +/- 83.23, N = 3 SE +/- 93.82, N = 3 SE +/- 60.36, N = 3 SE +/- 87.56, N = 3 SE +/- 85.10, N = 3 SE +/- 33.58, N = 3 14958.19 15666.70 15599.90 15616.90 15459.16 15641.09 15503.05 15621.64 15510.92 15480.29 15482.74 15482.71 15523.14 1. (CC) gcc options: -O3 -march=native
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70 140 210 280 350 182.44 248.71 192.02 201.36 230.35 272.57 194.18 269.18 268.62 266.51 306.50 311.47 315.08
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 14955.9 / Avg: 14958.19 / Max: 14960.35 Min: 15657.32 / Avg: 15666.7 / Max: 15674.1 Min: 15488.47 / Avg: 15599.9 / Max: 15722.26 Min: 15613.31 / Avg: 15616.9 / Max: 15621.62 Min: 15413.52 / Avg: 15459.16 / Max: 15491.31 Min: 15608.52 / Avg: 15641.09 / Max: 15689.15 Min: 15458.28 / Avg: 15503.05 / Max: 15574.89 Min: 15473.05 / Avg: 15621.64 / Max: 15760.91 Min: 15344.4 / Avg: 15510.92 / Max: 15669.08 Min: 15391.63 / Avg: 15480.29 / Max: 15595.56 Min: 15323.81 / Avg: 15482.73 / Max: 15625.9 Min: 15312.58 / Avg: 15482.71 / Max: 15571.59 Min: 15473.92 / Avg: 15523.14 / Max: 15587.31 1. (CC) gcc options: -O3 -march=native
Basis Universal Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: ETC1S EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 14 28 42 56 70 SE +/- 0.03, N = 3 SE +/- 0.23, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.13, N = 3 49.34 53.65 54.10 54.82 54.98 53.93 55.59 54.57 55.14 57.18 58.64 60.10 64.83 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: ETC1S EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 Min: 49.28 / Avg: 49.34 / Max: 49.38 Min: 53.23 / Avg: 53.65 / Max: 54.01 Min: 54.02 / Avg: 54.1 / Max: 54.23 Min: 54.77 / Avg: 54.82 / Max: 54.86 Min: 54.94 / Avg: 54.98 / Max: 55.01 Min: 53.83 / Avg: 53.93 / Max: 54.01 Min: 55.5 / Avg: 55.59 / Max: 55.67 Min: 54.42 / Avg: 54.57 / Max: 54.65 Min: 55.12 / Avg: 55.14 / Max: 55.16 Min: 57.13 / Avg: 57.18 / Max: 57.25 Min: 58.62 / Avg: 58.64 / Max: 58.66 Min: 60.03 / Avg: 60.1 / Max: 60.17 Min: 64.68 / Avg: 64.83 / Max: 65.09 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: PartialTweets EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1395 0.279 0.4185 0.558 0.6975 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 0.61 0.62 0.52 0.52 0.52 0.52 0.53 0.51 0.52 0.53 0.52 0.50 0.50 0.50 1. (CXX) g++ options: -O3 -pthread
GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 0.7.1 Throughput Test: PartialTweets EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0023 0.0046 0.0069 0.0092 0.0115 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: PartialTweets EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.61 / Avg: 0.61 / Max: 0.62 Min: 0.62 / Avg: 0.62 / Max: 0.62 Min: 0.52 / Avg: 0.52 / Max: 0.53 Min: 0.51 / Avg: 0.52 / Max: 0.52 Min: 0.51 / Avg: 0.52 / Max: 0.52 Min: 0.51 / Avg: 0.52 / Max: 0.52 Min: 0.52 / Avg: 0.53 / Max: 0.54 Min: 0.51 / Avg: 0.51 / Max: 0.52 Min: 0.52 / Avg: 0.52 / Max: 0.53 Min: 0.52 / Avg: 0.53 / Max: 0.53 Min: 0.52 / Avg: 0.52 / Max: 0.52 Min: 0.49 / Avg: 0.5 / Max: 0.51 Min: 0.49 / Avg: 0.5 / Max: 0.51 Min: 0.49 / Avg: 0.5 / Max: 0.51 1. (CXX) g++ options: -O3 -pthread
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: DistinctUserID EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1418 0.2836 0.4254 0.5672 0.709 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.62 0.63 0.54 0.53 0.52 0.52 0.54 0.52 0.53 0.53 0.53 0.51 0.51 0.52 1. (CXX) g++ options: -O3 -pthread
GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 0.7.1 Throughput Test: DistinctUserID EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0023 0.0046 0.0069 0.0092 0.0115 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: DistinctUserID EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.61 / Avg: 0.62 / Max: 0.63 Min: 0.62 / Avg: 0.63 / Max: 0.63 Min: 0.53 / Avg: 0.54 / Max: 0.54 Min: 0.52 / Avg: 0.53 / Max: 0.53 Min: 0.52 / Avg: 0.52 / Max: 0.53 Min: 0.52 / Avg: 0.52 / Max: 0.53 Min: 0.54 / Avg: 0.54 / Max: 0.55 Min: 0.52 / Avg: 0.52 / Max: 0.53 Min: 0.53 / Avg: 0.53 / Max: 0.54 Min: 0.53 / Avg: 0.53 / Max: 0.54 Min: 0.53 / Avg: 0.53 / Max: 0.53 Min: 0.51 / Avg: 0.51 / Max: 0.52 Min: 0.51 / Avg: 0.51 / Max: 0.52 Min: 0.51 / Avg: 0.52 / Max: 0.52 1. (CXX) g++ options: -O3 -pthread
Hugin Hugin is an open-source, cross-platform panorama photo stitcher software package. This test profile times how long it takes to run the assistant and panorama photo stitching on a set of images. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Hugin Panorama Photo Assistant + Stitching Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 SE +/- 0.24, N = 3 SE +/- 0.33, N = 3 SE +/- 0.32, N = 3 SE +/- 0.23, N = 3 SE +/- 0.28, N = 3 SE +/- 0.30, N = 3 SE +/- 0.36, N = 3 SE +/- 0.42, N = 3 SE +/- 0.41, N = 3 SE +/- 0.45, N = 3 SE +/- 0.50, N = 3 SE +/- 0.07, N = 3 SE +/- 0.80, N = 3 50.59 52.99 55.68 55.95 55.42 52.99 55.57 53.72 53.36 55.77 56.10 57.24 59.31
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Hugin Panorama Photo Assistant + Stitching Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 Min: 50.13 / Avg: 50.59 / Max: 50.91 Min: 52.38 / Avg: 52.99 / Max: 53.51 Min: 55.04 / Avg: 55.68 / Max: 56.12 Min: 55.51 / Avg: 55.95 / Max: 56.27 Min: 54.86 / Avg: 55.42 / Max: 55.71 Min: 52.43 / Avg: 52.99 / Max: 53.44 Min: 55.11 / Avg: 55.57 / Max: 56.28 Min: 52.98 / Avg: 53.72 / Max: 54.43 Min: 52.68 / Avg: 53.36 / Max: 54.09 Min: 54.93 / Avg: 55.77 / Max: 56.46 Min: 55.21 / Avg: 56.1 / Max: 56.93 Min: 57.12 / Avg: 57.24 / Max: 57.34 Min: 57.78 / Avg: 59.31 / Max: 60.47
Redis Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SADD EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K SE +/- 8827.27, N = 3 SE +/- 11422.05, N = 10 SE +/- 13671.93, N = 15 SE +/- 12423.96, N = 3 SE +/- 16607.14, N = 15 SE +/- 17355.22, N = 15 SE +/- 17960.88, N = 3 SE +/- 3475.48, N = 3 SE +/- 16125.60, N = 15 SE +/- 20897.40, N = 15 SE +/- 8695.76, N = 3 SE +/- 16958.48, N = 15 SE +/- 13873.03, N = 3 SE +/- 8933.33, N = 3 1514902.30 1510395.93 1347163.14 1328786.50 1360852.66 1349388.24 1388206.16 1319402.58 1387616.13 1393300.08 1345235.04 1325778.66 1285763.37 1302498.71 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Requests Per Second Per Watt
OpenBenchmarking.org Requests Per Second Per Watt, More Is Better Redis 6.0.9 Test: SADD EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8K 16K 24K 32K 40K 27216.48 35697.64 25016.28 24141.05 25475.97 29367.10 34766.37 23687.13 34715.91 36325.60 34516.59 36856.97 37338.23 39084.39
Result Confidence
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SADD EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K Min: 1500849.5 / Avg: 1514902.33 / Max: 1531183.5 Min: 1461584.38 / Avg: 1510395.93 / Max: 1591637.12 Min: 1270656.12 / Avg: 1347163.14 / Max: 1487873.75 Min: 1314595.5 / Avg: 1328786.5 / Max: 1353546.25 Min: 1297029.25 / Avg: 1360852.66 / Max: 1498136.88 Min: 1255650.38 / Avg: 1349388.24 / Max: 1542049.62 Min: 1357040.62 / Avg: 1388206.16 / Max: 1419258.62 Min: 1313197.62 / Avg: 1319402.58 / Max: 1325218.12 Min: 1293828.38 / Avg: 1387616.13 / Max: 1488321.5 Min: 1309088.62 / Avg: 1393300.08 / Max: 1560067.38 Min: 1327844.88 / Avg: 1345235.04 / Max: 1354117.75 Min: 1238091.38 / Avg: 1325778.66 / Max: 1481061.62 Min: 1260088.75 / Avg: 1285763.37 / Max: 1307710.75 Min: 1285520.5 / Avg: 1302498.71 / Max: 1315806.38 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenFOAM OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 8 Input: Motorbike 30M EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.18, N = 3 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.17, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.30, N = 3 39.89 60.50 23.35 22.83 23.43 25.17 29.62 26.53 30.62 33.81 40.87 56.57 67.23 113.16 1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -ldecompose -lgenericPatchFields -lmetisDecomp -lscotchDecomp -llagrangian -lregionModels -lOpenFOAM -ldl -lm
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 8 Input: Motorbike 30M EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 39.79 / Avg: 39.89 / Max: 39.94 Min: 60.32 / Avg: 60.5 / Max: 60.7 Min: 23.06 / Avg: 23.35 / Max: 23.68 Min: 22.63 / Avg: 22.83 / Max: 22.97 Min: 23.3 / Avg: 23.43 / Max: 23.53 Min: 25.06 / Avg: 25.17 / Max: 25.35 Min: 29.5 / Avg: 29.62 / Max: 29.8 Min: 26.35 / Avg: 26.53 / Max: 26.65 Min: 30.52 / Avg: 30.62 / Max: 30.71 Min: 33.73 / Avg: 33.81 / Max: 33.97 Min: 40.67 / Avg: 40.87 / Max: 41.21 Min: 56.49 / Avg: 56.57 / Max: 56.65 Min: 67.08 / Avg: 67.23 / Max: 67.35 Min: 112.77 / Avg: 113.16 / Max: 113.75 1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -ldecompose -lgenericPatchFields -lmetisDecomp -lscotchDecomp -llagrangian -lregionModels -lOpenFOAM -ldl -lm
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: Kostya EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1193 0.2386 0.3579 0.4772 0.5965 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.53 0.53 0.46 0.45 0.45 0.45 0.46 0.45 0.46 0.46 0.45 0.44 0.44 0.44 1. (CXX) g++ options: -O3 -pthread
GB/s Per Watt
OpenBenchmarking.org GB/s Per Watt, More Is Better simdjson 0.7.1 Throughput Test: Kostya EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0023 0.0046 0.0069 0.0092 0.0115 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: Kostya EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.53 / Avg: 0.53 / Max: 0.53 Min: 0.53 / Avg: 0.53 / Max: 0.53 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.44 / Avg: 0.44 / Max: 0.44 Min: 0.44 / Avg: 0.44 / Max: 0.44 Min: 0.43 / Avg: 0.44 / Max: 0.44 1. (CXX) g++ options: -O3 -pthread
Minion Minion is an open-source constraint solver that is designed to be very scalable. This test profile uses Minion's integrated benchmarking problems to solve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Graceful EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 44.84 44.92 52.35 53.07 53.20 53.29 51.62 53.18 52.34 52.30 52.93 54.76 54.63 55.08 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Graceful EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11 22 33 44 55 Min: 44.76 / Avg: 44.84 / Max: 44.96 Min: 44.81 / Avg: 44.92 / Max: 45.04 Min: 52.3 / Avg: 52.35 / Max: 52.43 Min: 52.97 / Avg: 53.07 / Max: 53.19 Min: 53.13 / Avg: 53.2 / Max: 53.31 Min: 53.08 / Avg: 53.29 / Max: 53.44 Min: 51.51 / Avg: 51.62 / Max: 51.69 Min: 53.07 / Avg: 53.18 / Max: 53.26 Min: 52.19 / Avg: 52.34 / Max: 52.57 Min: 52.29 / Avg: 52.3 / Max: 52.3 Min: 52.9 / Avg: 52.93 / Max: 52.97 Min: 54.61 / Avg: 54.76 / Max: 54.89 Min: 54.4 / Avg: 54.63 / Max: 54.76 Min: 54.97 / Avg: 55.08 / Max: 55.19 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Perl Benchmarks Perl benchmark suite that can be used to compare the relative speed of different versions of perl. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Interpreter EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0002 0.0004 0.0006 0.0008 0.001 SE +/- 0.00000483, N = 3 SE +/- 0.00001015, N = 3 SE +/- 0.00000621, N = 3 SE +/- 0.00000365, N = 3 SE +/- 0.00000264, N = 3 SE +/- 0.00000921, N = 3 SE +/- 0.00000879, N = 3 SE +/- 0.00000252, N = 3 SE +/- 0.00000724, N = 3 SE +/- 0.00000962, N = 3 SE +/- 0.00000413, N = 3 SE +/- 0.00000497, N = 3 SE +/- 0.00001257, N = 3 SE +/- 0.00000737, N = 3 0.00082570 0.00083563 0.00097084 0.00097455 0.00096579 0.00096153 0.00095485 0.00096753 0.00096082 0.00095586 0.00095478 0.00099210 0.00098074 0.00100608
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Interpreter EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1 2 3 4 5 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0 Min: 0 / Avg: 0 / Max: 0
Timed PHP Compilation This test times how long it takes to build PHP 7. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.4.2 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.21, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 46.54 59.56 42.56 43.06 43.82 43.84 44.55 46.10 45.44 47.41 52.75 55.22 60.20 75.33
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.4.2 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 14 28 42 56 70 Min: 46.12 / Avg: 46.54 / Max: 46.75 Min: 59.55 / Avg: 59.56 / Max: 59.58 Min: 42.5 / Avg: 42.56 / Max: 42.67 Min: 43 / Avg: 43.06 / Max: 43.15 Min: 43.8 / Avg: 43.82 / Max: 43.84 Min: 43.81 / Avg: 43.84 / Max: 43.91 Min: 44.52 / Avg: 44.55 / Max: 44.58 Min: 46.08 / Avg: 46.1 / Max: 46.13 Min: 45.42 / Avg: 45.44 / Max: 45.46 Min: 47.38 / Avg: 47.41 / Max: 47.44 Min: 52.62 / Avg: 52.75 / Max: 52.9 Min: 55.1 / Avg: 55.22 / Max: 55.29 Min: 60.16 / Avg: 60.2 / Max: 60.27 Min: 75.27 / Avg: 75.33 / Max: 75.44
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4.45 2.39 11.07 11.24 9.58 9.52 7.58 6.90 7.01 5.76 4.02 3.72 2.93 1.89 MIN: 4.31 / MAX: 4.52 MIN: 2.37 / MAX: 2.42 MIN: 10.75 / MAX: 11.24 MIN: 10.99 / MAX: 11.36 MIN: 9.17 / MAX: 9.71 MIN: 9.35 / MAX: 9.62 MIN: 7.46 / MAX: 7.63 MIN: 6.8 / MAX: 6.99 MIN: 6.94 / MAX: 7.09 MIN: 5.68 / MAX: 5.81 MIN: 3.98 / MAX: 4.05 MIN: 3.69 / MAX: 3.76 MIN: 2.9 / MAX: 2.97 MIN: 1.84 / MAX: 1.91
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0158 0.0316 0.0474 0.0632 0.079 0.02 0.02 0.07 0.07 0.06 0.06 0.05 0.05 0.05 0.04 0.03 0.04 0.03 0.02
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 4.44 / Avg: 4.45 / Max: 4.46 Min: 2.39 / Avg: 2.39 / Max: 2.39 Min: 10.99 / Avg: 11.07 / Max: 11.11 Min: 11.24 / Avg: 11.24 / Max: 11.24 Min: 9.52 / Avg: 9.58 / Max: 9.62 Min: 9.52 / Avg: 9.52 / Max: 9.52 Min: 7.58 / Avg: 7.58 / Max: 7.58 Min: 6.9 / Avg: 6.9 / Max: 6.9 Min: 6.99 / Avg: 7.01 / Max: 7.04 Min: 5.75 / Avg: 5.76 / Max: 5.78 Min: 4.02 / Avg: 4.02 / Max: 4.02 Min: 3.72 / Avg: 3.72 / Max: 3.72 Min: 2.93 / Avg: 2.93 / Max: 2.93 Min: 1.89 / Avg: 1.89 / Max: 1.89
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.029, N = 3 SE +/- 0.020, N = 3 SE +/- 0.034, N = 3 SE +/- 0.063, N = 3 SE +/- 0.031, N = 3 SE +/- 0.069, N = 3 SE +/- 0.034, N = 13 SE +/- 0.063, N = 5 SE +/- 0.056, N = 3 SE +/- 0.075, N = 4 SE +/- 0.082, N = 6 SE +/- 0.011, N = 3 SE +/- 0.040, N = 3 SE +/- 0.004, N = 3 8.876 12.176 4.804 4.617 4.972 4.872 5.369 5.648 5.481 6.067 8.258 7.791 9.939 14.123 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 8.83 / Avg: 8.88 / Max: 8.93 Min: 12.14 / Avg: 12.18 / Max: 12.2 Min: 4.76 / Avg: 4.8 / Max: 4.87 Min: 4.51 / Avg: 4.62 / Max: 4.73 Min: 4.91 / Avg: 4.97 / Max: 5.01 Min: 4.76 / Avg: 4.87 / Max: 5 Min: 4.97 / Avg: 5.37 / Max: 5.44 Min: 5.51 / Avg: 5.65 / Max: 5.86 Min: 5.37 / Avg: 5.48 / Max: 5.55 Min: 5.94 / Avg: 6.07 / Max: 6.22 Min: 7.91 / Avg: 8.26 / Max: 8.44 Min: 7.77 / Avg: 7.79 / Max: 7.81 Min: 9.87 / Avg: 9.94 / Max: 10.01 Min: 14.12 / Avg: 14.12 / Max: 14.13 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12K 24K 36K 48K 60K SE +/- 92.29, N = 3 SE +/- 32.87, N = 3 SE +/- 365.68, N = 3 SE +/- 734.82, N = 3 SE +/- 308.59, N = 3 SE +/- 725.42, N = 3 SE +/- 320.52, N = 13 SE +/- 489.37, N = 5 SE +/- 468.05, N = 3 SE +/- 510.25, N = 4 SE +/- 306.40, N = 6 SE +/- 44.96, N = 3 SE +/- 98.97, N = 3 SE +/- 5.29, N = 3 28193 20541 52181 54318 50381 51444 46655 44343 45671 41273 30313 32115 25170 17709 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9K 18K 27K 36K 45K Min: 28026.73 / Avg: 28193.29 / Max: 28345.44 Min: 20498.68 / Avg: 20541.36 / Max: 20606 Min: 51458.66 / Avg: 52181.25 / Max: 52640.41 Min: 53010.6 / Avg: 54318.48 / Max: 55552.89 Min: 49982.78 / Avg: 50381.39 / Max: 50988.76 Min: 50134.59 / Avg: 51444.21 / Max: 52639.74 Min: 46038.91 / Avg: 46655.09 / Max: 50394.39 Min: 42733.56 / Avg: 44342.61 / Max: 45416.75 Min: 45084.17 / Avg: 45670.73 / Max: 46595.8 Min: 40219.85 / Avg: 41272.74 / Max: 42149.61 Min: 29631.74 / Avg: 30313.35 / Max: 31648.97 Min: 32033.65 / Avg: 32115.45 / Max: 32188.7 Min: 25002.83 / Avg: 25170.22 / Max: 25345.42 Min: 17702.56 / Avg: 17709.27 / Max: 17719.71 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Redis Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SET EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K SE +/- 8091.69, N = 3 SE +/- 13442.74, N = 3 SE +/- 22602.36, N = 12 SE +/- 18435.82, N = 15 SE +/- 16083.41, N = 14 SE +/- 14106.76, N = 3 SE +/- 15275.38, N = 3 SE +/- 15826.22, N = 15 SE +/- 14615.10, N = 4 SE +/- 4313.29, N = 3 SE +/- 12613.83, N = 5 SE +/- 15930.17, N = 3 SE +/- 17229.52, N = 15 SE +/- 11537.36, N = 4 1336546.21 1315900.46 1228831.79 1200527.31 1183717.45 1170719.46 1189270.87 1186674.32 1180459.91 1147451.38 1173570.00 1143594.63 1156241.74 1126925.50 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Requests Per Second Per Watt
OpenBenchmarking.org Requests Per Second Per Watt, More Is Better Redis 6.0.9 Test: SET EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7K 14K 21K 28K 35K 23705.79 30367.37 22530.98 21301.37 21774.45 24980.53 29411.07 21182.66 29043.85 29418.44 29192.78 31275.55 33520.26 33238.20
Result Confidence
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SET EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K Min: 1320510.25 / Avg: 1336546.21 / Max: 1346451.62 Min: 1289503 / Avg: 1315900.46 / Max: 1333515.38 Min: 1142730.12 / Avg: 1228831.79 / Max: 1378956.5 Min: 1107542.38 / Avg: 1200527.31 / Max: 1349007.38 Min: 1125387.38 / Avg: 1183717.45 / Max: 1363145 Min: 1151024 / Avg: 1170719.46 / Max: 1198062 Min: 1161717.5 / Avg: 1189270.87 / Max: 1214476.5 Min: 1117600.38 / Avg: 1186674.32 / Max: 1295341 Min: 1147582.38 / Avg: 1180459.91 / Max: 1209351.5 Min: 1143020.75 / Avg: 1147451.38 / Max: 1156076.88 Min: 1141950.88 / Avg: 1173570 / Max: 1208467 Min: 1112121 / Avg: 1143594.63 / Max: 1163617.38 Min: 1088613.12 / Avg: 1156241.74 / Max: 1310666.25 Min: 1105957.12 / Avg: 1126925.5 / Max: 1159285.88 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
eSpeak-NG Speech Engine This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. This test profile is now tracking the eSpeak-NG version of eSpeak. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.08, N = 4 SE +/- 0.07, N = 4 SE +/- 0.12, N = 4 SE +/- 0.09, N = 4 SE +/- 0.06, N = 4 SE +/- 0.11, N = 4 SE +/- 0.04, N = 4 SE +/- 0.07, N = 4 SE +/- 0.08, N = 4 SE +/- 0.07, N = 4 SE +/- 0.12, N = 4 SE +/- 0.10, N = 4 SE +/- 0.08, N = 4 SE +/- 0.11, N = 4 30.75 30.84 35.93 36.20 36.28 36.40 35.27 36.24 35.73 35.69 36.36 37.55 37.40 37.40 1. (CC) gcc options: -O2 -std=c99
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 30.52 / Avg: 30.75 / Max: 30.86 Min: 30.64 / Avg: 30.84 / Max: 30.96 Min: 35.63 / Avg: 35.93 / Max: 36.21 Min: 35.97 / Avg: 36.2 / Max: 36.42 Min: 36.17 / Avg: 36.28 / Max: 36.45 Min: 36.11 / Avg: 36.4 / Max: 36.59 Min: 35.16 / Avg: 35.27 / Max: 35.37 Min: 36.12 / Avg: 36.24 / Max: 36.4 Min: 35.5 / Avg: 35.73 / Max: 35.85 Min: 35.55 / Avg: 35.69 / Max: 35.87 Min: 36.06 / Avg: 36.36 / Max: 36.65 Min: 37.36 / Avg: 37.55 / Max: 37.83 Min: 37.26 / Avg: 37.4 / Max: 37.61 Min: 37.13 / Avg: 37.4 / Max: 37.65 1. (CC) gcc options: -O2 -std=c99
JPEG XL Decoding The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. Learn more via the OpenBenchmarking.org test page.
Tachyon This is a test of the threaded Tachyon, a parallel ray-tracing system, measuring the time to ray-trace a sample scene. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.99b6 Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.14, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.27, N = 3 SE +/- 0.04, N = 3 SE +/- 0.26, N = 3 46.59 92.16 19.12 19.08 22.56 22.81 28.59 31.57 30.90 37.73 54.84 58.47 75.04 112.39 1. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.99b6 Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 46.49 / Avg: 46.59 / Max: 46.65 Min: 91.95 / Avg: 92.16 / Max: 92.42 Min: 19.11 / Avg: 19.12 / Max: 19.14 Min: 19.06 / Avg: 19.08 / Max: 19.12 Min: 22.52 / Avg: 22.56 / Max: 22.63 Min: 22.76 / Avg: 22.81 / Max: 22.84 Min: 28.5 / Avg: 28.59 / Max: 28.69 Min: 31.53 / Avg: 31.57 / Max: 31.59 Min: 30.86 / Avg: 30.9 / Max: 30.92 Min: 37.62 / Avg: 37.73 / Max: 37.87 Min: 54.8 / Avg: 54.84 / Max: 54.9 Min: 58.2 / Avg: 58.47 / Max: 59 Min: 74.96 / Avg: 75.04 / Max: 75.08 Min: 112.07 / Avg: 112.39 / Max: 112.9 1. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
Redis Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: GET EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400K 800K 1200K 1600K 2000K SE +/- 17849.12, N = 5 SE +/- 19118.80, N = 4 SE +/- 18256.32, N = 3 SE +/- 13883.40, N = 3 SE +/- 11135.64, N = 15 SE +/- 8071.25, N = 3 SE +/- 13573.21, N = 15 SE +/- 21214.43, N = 3 SE +/- 18437.40, N = 3 SE +/- 13170.58, N = 15 SE +/- 13583.32, N = 6 SE +/- 14367.89, N = 15 SE +/- 12289.67, N = 3 SE +/- 18495.33, N = 15 1635794.57 1601070.50 1482738.92 1417246.67 1437534.23 1450496.71 1508082.41 1481804.00 1419600.13 1466557.03 1438157.73 1414840.41 1372468.88 1421564.35 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Requests Per Second Per Watt
OpenBenchmarking.org Requests Per Second Per Watt, More Is Better Redis 6.0.9 Test: GET EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9K 18K 27K 36K 45K 29783.14 37914.84 27989.20 25840.00 27083.65 31741.99 38176.82 27229.69 35995.93 39042.93 36649.44 39449.14 40587.49 42939.75
Result Confidence
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: GET EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K Min: 1586048 / Avg: 1635794.57 / Max: 1671133.62 Min: 1561051.5 / Avg: 1601070.5 / Max: 1650709.75 Min: 1456664.25 / Avg: 1482738.92 / Max: 1517911.38 Min: 1397042.75 / Avg: 1417246.67 / Max: 1443844 Min: 1363529.88 / Avg: 1437534.23 / Max: 1507163.75 Min: 1439056 / Avg: 1450496.71 / Max: 1466079.5 Min: 1433897.38 / Avg: 1508082.41 / Max: 1604899.38 Min: 1456461.38 / Avg: 1481804 / Max: 1523945.12 Min: 1387942.88 / Avg: 1419600.13 / Max: 1451804.88 Min: 1388893.38 / Avg: 1466557.03 / Max: 1567157.5 Min: 1387351.75 / Avg: 1438157.73 / Max: 1487252.88 Min: 1349900.62 / Avg: 1414840.41 / Max: 1535877.12 Min: 1350621.38 / Avg: 1372468.88 / Max: 1393145.75 Min: 1340662.25 / Avg: 1421564.35 / Max: 1628409.75 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
7-Zip Compression This is a test of 7-Zip using p7zip with its integrated benchmark feature or upstream 7-Zip for the Windows x64 build. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60K 120K 180K 240K 300K SE +/- 558.99, N = 3 SE +/- 288.04, N = 3 SE +/- 226.28, N = 3 SE +/- 419.38, N = 3 SE +/- 453.51, N = 3 SE +/- 190.42, N = 3 SE +/- 894.75, N = 3 SE +/- 249.88, N = 3 SE +/- 315.33, N = 3 SE +/- 192.03, N = 3 SE +/- 313.91, N = 3 SE +/- 108.98, N = 3 SE +/- 84.48, N = 3 109069 56955 264908 270140 229747 176148 171242 171362 138575 95137 90588 69655 45941 1. (CXX) g++ options: -pipe -lpthread
MIPS Per Watt
OpenBenchmarking.org MIPS Per Watt, More Is Better 7-Zip Compression 16.02 Compress Speed Test EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 654.08 601.78 1772.21 1860.21 1667.35 1318.89 1207.72 1397.09 1091.49 969.68 1034.12 872.02 710.08
Result Confidence
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50K 100K 150K 200K 250K Min: 108136 / Avg: 109069.33 / Max: 110069 Min: 56532 / Avg: 56954.67 / Max: 57505 Min: 264563 / Avg: 264907.67 / Max: 265334 Min: 269353 / Avg: 270139.67 / Max: 270785 Min: 229189 / Avg: 229746.67 / Max: 230645 Min: 175784 / Avg: 176148 / Max: 176427 Min: 169527 / Avg: 171242 / Max: 172542 Min: 170978 / Avg: 171362 / Max: 171831 Min: 138164 / Avg: 138575.33 / Max: 139195 Min: 94939 / Avg: 95137 / Max: 95521 Min: 90180 / Avg: 90587.67 / Max: 91205 Min: 69500 / Avg: 69654.67 / Max: 69865 Min: 45801 / Avg: 45941.33 / Max: 46093 1. (CXX) g++ options: -pipe -lpthread
Redis Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: LPUSH EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K SE +/- 8000.25, N = 3 SE +/- 6374.74, N = 3 SE +/- 7054.62, N = 13 SE +/- 5890.12, N = 3 SE +/- 4227.00, N = 3 SE +/- 7701.66, N = 3 SE +/- 11205.82, N = 5 SE +/- 10288.34, N = 15 SE +/- 9569.43, N = 3 SE +/- 11072.90, N = 15 SE +/- 8436.86, N = 3 SE +/- 3774.33, N = 3 SE +/- 15424.51, N = 15 SE +/- 6585.32, N = 3 1174427.00 1174520.46 1031876.33 1039389.17 1018402.69 1013049.13 1067438.77 1059976.32 1045486.81 1045821.59 1030785.79 997652.46 1037523.56 991962.08 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Requests Per Second Per Watt
OpenBenchmarking.org Requests Per Second Per Watt, More Is Better Redis 6.0.9 Test: LPUSH EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6K 12K 18K 24K 30K 20551.40 26954.89 18361.54 18391.09 18554.91 21454.67 26166.33 18680.49 25519.76 26554.24 25453.57 27079.37 29596.02 28993.77
Result Confidence
OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: LPUSH EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K Min: 1165101.25 / Avg: 1174427 / Max: 1190349.75 Min: 1167284.62 / Avg: 1174520.46 / Max: 1187229.25 Min: 1004120.12 / Avg: 1031876.33 / Max: 1093254.62 Min: 1027854.88 / Avg: 1039389.17 / Max: 1047230.12 Min: 1013376.56 / Avg: 1018402.69 / Max: 1026802.69 Min: 1004120.12 / Avg: 1013049.13 / Max: 1028383.38 Min: 1044932.12 / Avg: 1067438.77 / Max: 1106467.75 Min: 1006359.25 / Avg: 1059976.32 / Max: 1149564.75 Min: 1029134.12 / Avg: 1045486.81 / Max: 1062275 Min: 987478.25 / Avg: 1045821.59 / Max: 1105957.12 Min: 1014202.06 / Avg: 1030785.79 / Max: 1041775.19 Min: 990504 / Avg: 997652.46 / Max: 1003327 Min: 976187.38 / Avg: 1037523.56 / Max: 1152915 Min: 978965.06 / Avg: 991962.08 / Max: 1000306.5 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
rav1e Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 6 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.3499 0.6998 1.0497 1.3996 1.7495 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 SE +/- 0.000, N = 3 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.002, N = 3 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 1.542 1.555 1.342 1.333 1.320 1.331 1.368 1.336 1.338 1.351 1.329 1.293 1.296 1.289
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better rav1e 0.4 Speed: 6 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0068 0.0136 0.0204 0.0272 0.034 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.03 0.03 0.03
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 6 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.54 / Avg: 1.54 / Max: 1.54 Min: 1.55 / Avg: 1.55 / Max: 1.56 Min: 1.34 / Avg: 1.34 / Max: 1.34 Min: 1.33 / Avg: 1.33 / Max: 1.34 Min: 1.32 / Avg: 1.32 / Max: 1.32 Min: 1.33 / Avg: 1.33 / Max: 1.33 Min: 1.36 / Avg: 1.37 / Max: 1.37 Min: 1.34 / Avg: 1.34 / Max: 1.34 Min: 1.33 / Avg: 1.34 / Max: 1.34 Min: 1.35 / Avg: 1.35 / Max: 1.35 Min: 1.33 / Avg: 1.33 / Max: 1.33 Min: 1.29 / Avg: 1.29 / Max: 1.3 Min: 1.29 / Avg: 1.3 / Max: 1.3 Min: 1.29 / Avg: 1.29 / Max: 1.29
ACES DGEMM This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.028662, N = 3 SE +/- 0.039059, N = 3 SE +/- 0.163720, N = 4 SE +/- 0.099835, N = 4 SE +/- 0.030087, N = 3 SE +/- 0.083534, N = 3 SE +/- 0.091101, N = 3 SE +/- 0.072162, N = 3 SE +/- 0.066319, N = 3 SE +/- 0.025438, N = 3 SE +/- 0.017716, N = 3 SE +/- 0.037637, N = 3 SE +/- 0.008716, N = 3 SE +/- 0.010754, N = 3 4.897653 3.011595 15.614812 16.881168 13.942465 12.735031 8.863336 9.828871 9.326909 7.208235 4.963715 4.452419 3.464003 1.583704 1. (CC) gcc options: -O3 -march=native -fopenmp
GFLOP/s Per Watt
OpenBenchmarking.org GFLOP/s Per Watt, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0248 0.0496 0.0744 0.0992 0.124 0.03 0.03 0.11 0.11 0.08 0.10 0.07 0.06 0.07 0.06 0.05 0.05 0.05 0.02
Result Confidence
OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 4.85 / Avg: 4.9 / Max: 4.95 Min: 2.95 / Avg: 3.01 / Max: 3.08 Min: 15.39 / Avg: 15.61 / Max: 16.1 Min: 16.59 / Avg: 16.88 / Max: 17.02 Min: 13.91 / Avg: 13.94 / Max: 14 Min: 12.62 / Avg: 12.74 / Max: 12.9 Min: 8.68 / Avg: 8.86 / Max: 8.96 Min: 9.75 / Avg: 9.83 / Max: 9.97 Min: 9.25 / Avg: 9.33 / Max: 9.46 Min: 7.16 / Avg: 7.21 / Max: 7.24 Min: 4.93 / Avg: 4.96 / Max: 4.99 Min: 4.39 / Avg: 4.45 / Max: 4.52 Min: 3.45 / Avg: 3.46 / Max: 3.48 Min: 1.57 / Avg: 1.58 / Max: 1.61 1. (CC) gcc options: -O3 -march=native -fopenmp
Zstd Compression This test measures the time needed to compress a sample file (an Ubuntu ISO) using Zstd compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.79, N = 3 SE +/- 0.03, N = 3 SE +/- 1.22, N = 3 SE +/- 0.55, N = 3 SE +/- 0.12, N = 3 SE +/- 0.15, N = 3 SE +/- 0.15, N = 3 SE +/- 0.00, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 76.1 47.0 147.5 149.6 130.1 125.7 114.4 122.7 114.3 96.9 73.8 66.4 55.2 37.3 1. (CC) gcc options: -O3 -pthread -lz -llzma
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.4.5 Compression Level: 19 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.3218 0.6436 0.9654 1.2872 1.609 0.69 0.63 1.31 1.38 1.24 1.33 1.43 1.17 1.38 1.20 1.00 1.01 0.89 0.66
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 74.9 / Avg: 76.13 / Max: 77.6 Min: 47 / Avg: 47.03 / Max: 47.1 Min: 145.1 / Avg: 147.53 / Max: 148.8 Min: 148.7 / Avg: 149.6 / Max: 150.6 Min: 129.9 / Avg: 130.1 / Max: 130.3 Min: 125.4 / Avg: 125.67 / Max: 125.9 Min: 114.2 / Avg: 114.4 / Max: 114.7 Min: 122.7 / Avg: 122.7 / Max: 122.7 Min: 114.2 / Avg: 114.33 / Max: 114.5 Min: 96.8 / Avg: 96.93 / Max: 97.1 Min: 73.8 / Avg: 73.83 / Max: 73.9 Min: 66.2 / Avg: 66.4 / Max: 66.6 Min: 55.1 / Avg: 55.2 / Max: 55.3 Min: 37.2 / Avg: 37.3 / Max: 37.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 36.44 36.49 42.29 43.14 42.96 42.86 41.66 42.90 42.31 42.30 42.88 44.37 44.27 46.30 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 36.33 / Avg: 36.44 / Max: 36.56 Min: 36.45 / Avg: 36.49 / Max: 36.52 Min: 42.26 / Avg: 42.29 / Max: 42.3 Min: 43 / Avg: 43.13 / Max: 43.38 Min: 42.78 / Avg: 42.96 / Max: 43.08 Min: 42.71 / Avg: 42.86 / Max: 43.03 Min: 41.62 / Avg: 41.66 / Max: 41.7 Min: 42.83 / Avg: 42.9 / Max: 42.94 Min: 42.23 / Avg: 42.31 / Max: 42.35 Min: 42.03 / Avg: 42.3 / Max: 42.64 Min: 42.75 / Avg: 42.88 / Max: 42.95 Min: 44.26 / Avg: 44.37 / Max: 44.53 Min: 44.18 / Avg: 44.27 / Max: 44.39 Min: 46.14 / Avg: 46.3 / Max: 46.42 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.51, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.18, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 34.81 34.84 41.15 41.26 41.17 41.17 39.92 41.19 40.53 40.59 40.98 42.44 42.48 42.44 1. (CC) gcc options: -lm -O3
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 34.79 / Avg: 34.81 / Max: 34.85 Min: 34.8 / Avg: 34.83 / Max: 34.87 Min: 40.64 / Avg: 41.15 / Max: 42.18 Min: 41.23 / Avg: 41.26 / Max: 41.3 Min: 41.14 / Avg: 41.17 / Max: 41.18 Min: 41.14 / Avg: 41.16 / Max: 41.18 Min: 39.91 / Avg: 39.92 / Max: 39.94 Min: 41.18 / Avg: 41.19 / Max: 41.2 Min: 40.51 / Avg: 40.53 / Max: 40.58 Min: 40.58 / Avg: 40.59 / Max: 40.6 Min: 40.63 / Avg: 40.98 / Max: 41.17 Min: 42.4 / Avg: 42.43 / Max: 42.45 Min: 42.42 / Avg: 42.48 / Max: 42.58 Min: 42.43 / Avg: 42.44 / Max: 42.45 1. (CC) gcc options: -lm -O3
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 SE +/- 0.17, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 22.56 12.50 55.56 58.82 50.00 50.00 40.00 35.71 37.04 30.30 20.83 19.61 15.63 10.31 MIN: 21.28 / MAX: 24.39 MIN: 12.2 / MAX: 13.33 MIN: 50 / MAX: 62.5 MIN: 52.63 / MAX: 62.5 MIN: 47.62 / MAX: 52.63 MIN: 47.62 / MAX: 52.63 MIN: 37.04 / MAX: 41.67 MIN: 33.33 / MAX: 38.46 MIN: 34.48 / MAX: 40 MIN: 29.41 / MAX: 32.26 MIN: 20.41 / MAX: 22.73 MIN: 19.23 / MAX: 21.28 MIN: 15.15 / MAX: 16.67 MIN: 10.1 / MAX: 10.99
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.162 0.324 0.486 0.648 0.81 0.21 0.16 0.67 0.72 0.62 0.69 0.59 0.43 0.57 0.44 0.31 0.34 0.26 0.19
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 Min: 22.22 / Avg: 22.56 / Max: 22.73 Min: 12.5 / Avg: 12.5 / Max: 12.5 Min: 55.56 / Avg: 55.56 / Max: 55.56 Min: 58.82 / Avg: 58.82 / Max: 58.82 Min: 35.71 / Avg: 35.71 / Max: 35.71 Min: 37.04 / Avg: 37.04 / Max: 37.04 Min: 30.3 / Avg: 30.3 / Max: 30.3 Min: 20.83 / Avg: 20.83 / Max: 20.83 Min: 19.61 / Avg: 19.61 / Max: 19.61 Min: 15.63 / Avg: 15.63 / Max: 15.63 Min: 10.31 / Avg: 10.31 / Max: 10.31
QuantLib QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.21 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 SE +/- 10.66, N = 3 SE +/- 13.33, N = 3 SE +/- 11.47, N = 3 SE +/- 10.12, N = 3 SE +/- 13.45, N = 3 SE +/- 11.83, N = 3 SE +/- 10.26, N = 3 SE +/- 9.97, N = 3 SE +/- 12.22, N = 3 SE +/- 10.13, N = 3 SE +/- 8.78, N = 3 SE +/- 12.91, N = 3 SE +/- 10.95, N = 3 2300.7 2285.2 1970.1 1955.3 1945.3 2016.4 1922.3 1984.5 1977.3 1952.5 1887.0 1893.2 1892.0 1. (CXX) g++ options: -O3 -march=native -rdynamic
MFLOPS Per Watt
OpenBenchmarking.org MFLOPS Per Watt, More Is Better QuantLib 1.21 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 31.99 42.47 28.04 28.49 33.75 41.04 28.32 40.27 40.44 39.68 44.47 45.42 46.01
Result Confidence
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.21 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 400 800 1200 1600 2000 Min: 2279.8 / Avg: 2300.73 / Max: 2314.7 Min: 2258.5 / Avg: 2285.17 / Max: 2298.8 Min: 1947.7 / Avg: 1970.13 / Max: 1985.5 Min: 1935.1 / Avg: 1955.3 / Max: 1966.5 Min: 1918.7 / Avg: 1945.33 / Max: 1961.9 Min: 1993.3 / Avg: 2016.43 / Max: 2032.3 Min: 1902.5 / Avg: 1922.27 / Max: 1936.9 Min: 1964.7 / Avg: 1984.5 / Max: 1996.4 Min: 1953.5 / Avg: 1977.33 / Max: 1993.9 Min: 1932.3 / Avg: 1952.53 / Max: 1963.6 Min: 1869.4 / Avg: 1886.97 / Max: 1895.9 Min: 1867.5 / Avg: 1893.2 / Max: 1908.2 Min: 1870.4 / Avg: 1892 / Max: 1905.9 1. (CXX) g++ options: -O3 -march=native -rdynamic
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K SE +/- 0.00, N = 3 SE +/- 66.13, N = 4 SE +/- 197.85, N = 2 SE +/- 111.87, N = 3 SE +/- 0.00, N = 3 SE +/- 136.29, N = 3 SE +/- 233.09, N = 3 SE +/- 101.60, N = 3 SE +/- 82.20, N = 3 SE +/- 0.00, N = 3 SE +/- 69.42, N = 3 SE +/- 54.85, N = 9 SE +/- 50.92, N = 3 16399.70 8265.99 20382.20 20991.70 23040.70 17300.80 12699.50 20585.30 12698.20 13892.50 12204.40 7358.19 7130.89 3879.10 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Mpix/sec Per Watt
OpenBenchmarking.org Mpix/sec Per Watt, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 117.06 93.74 116.25 123.10 139.29 117.11 105.62 137.34 104.15 123.63 132.05 92.77 97.15 62.30
Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K Min: 16399.7 / Avg: 16399.7 / Max: 16399.7 Min: 8199.86 / Avg: 8265.99 / Max: 8464.38 Min: 20184.3 / Avg: 20382.15 / Max: 20580 Min: 22817 / Avg: 23040.73 / Max: 23152.6 Min: 17300.8 / Avg: 17300.8 / Max: 17300.8 Min: 12495 / Avg: 12699.47 / Max: 12957.8 Min: 20184.3 / Avg: 20585.33 / Max: 20991.7 Min: 12495 / Avg: 12698.2 / Max: 12799.8 Min: 13810.3 / Avg: 13892.5 / Max: 14056.9 Min: 12204.4 / Avg: 12204.4 / Max: 12204.4 Min: 7288.77 / Avg: 7358.19 / Max: 7497.02 Min: 6786.09 / Avg: 7130.89 / Max: 7288.77 Min: 3802.84 / Avg: 3879.1 / Max: 3975.69 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Result
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Degridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5K 10K 15K 20K 25K SE +/- 118.03, N = 3 SE +/- 74.25, N = 4 SE +/- 112.07, N = 3 SE +/- 121.22, N = 3 SE +/- 197.00, N = 3 SE +/- 61.33, N = 3 SE +/- 162.55, N = 3 SE +/- 215.40, N = 3 SE +/- 137.10, N = 3 SE +/- 63.07, N = 3 SE +/- 68.60, N = 3 SE +/- 27.68, N = 3 SE +/- 48.79, N = 9 SE +/- 28.76, N = 3 13692.30 6907.54 20185.50 20993.10 21472.30 16990.10 11976.90 18311.70 12067.30 12173.70 10427.20 6587.57 6100.18 3350.22 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Degridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K Min: 13456.2 / Avg: 13692.27 / Max: 13810.3 Min: 6728.09 / Avg: 6907.54 / Max: 7091.77 Min: 19992 / Avg: 20185.5 / Max: 20380.2 Min: 20783.8 / Avg: 20993.07 / Max: 21203.7 Min: 21275.3 / Avg: 21472.3 / Max: 21866.3 Min: 16928.8 / Avg: 16990.13 / Max: 17112.8 Min: 11662 / Avg: 11976.87 / Max: 12204.4 Min: 18096.3 / Avg: 18311.7 / Max: 18742.5 Min: 11793.1 / Avg: 12067.3 / Max: 12204.4 Min: 12110.6 / Avg: 12173.67 / Max: 12299.8 Min: 10290 / Avg: 10427.2 / Max: 10495.8 Min: 6559.89 / Avg: 6587.57 / Max: 6642.93 Min: 5788.14 / Avg: 6100.18 / Max: 6247.52 Min: 3321.46 / Avg: 3350.22 / Max: 3407.74 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
m-queens A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 4 SE +/- 0.03, N = 4 SE +/- 0.06, N = 4 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 35.92 71.58 13.84 13.75 16.61 21.37 23.05 22.63 28.19 42.38 43.74 58.22 87.19 1. (CXX) g++ options: -fopenmp -O2 -march=native
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 35.9 / Avg: 35.92 / Max: 35.93 Min: 71.55 / Avg: 71.58 / Max: 71.6 Min: 13.77 / Avg: 13.84 / Max: 13.89 Min: 13.72 / Avg: 13.75 / Max: 13.84 Min: 16.48 / Avg: 16.61 / Max: 16.78 Min: 21.31 / Avg: 21.37 / Max: 21.46 Min: 23.01 / Avg: 23.05 / Max: 23.09 Min: 22.57 / Avg: 22.63 / Max: 22.73 Min: 28.07 / Avg: 28.19 / Max: 28.33 Min: 42.35 / Avg: 42.38 / Max: 42.41 Min: 43.69 / Avg: 43.74 / Max: 43.81 Min: 58.19 / Avg: 58.22 / Max: 58.26 Min: 87.19 / Avg: 87.19 / Max: 87.2 1. (CXX) g++ options: -fopenmp -O2 -march=native
PHPBench PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 130K 260K 390K 520K 650K SE +/- 1476.72, N = 3 SE +/- 7086.33, N = 3 SE +/- 1921.47, N = 3 SE +/- 1174.39, N = 3 SE +/- 215.70, N = 3 SE +/- 142.64, N = 3 SE +/- 741.11, N = 3 SE +/- 4131.15, N = 3 SE +/- 948.15, N = 3 SE +/- 4035.67, N = 3 SE +/- 957.34, N = 3 SE +/- 973.54, N = 3 SE +/- 2157.92, N = 3 SE +/- 569.07, N = 3 614039 621672 529609 524751 521572 523575 539060 532957 531740 542521 524189 511228 513356 511906
Score Per Watt
OpenBenchmarking.org Score Per Watt, More Is Better PHPBench 0.8.1 PHP Benchmark Suite EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K 8583.59 11604.83 7643.25 7697.43 8074.56 9050.74 11143.99 7829.14 10842.61 11170.88 10676.05 12087.81 12418.02 12522.00
Result Confidence
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 110K 220K 330K 440K 550K Min: 611253 / Avg: 614039 / Max: 616281 Min: 613060 / Avg: 621672 / Max: 635726 Min: 525962 / Avg: 529608.67 / Max: 532482 Min: 523197 / Avg: 524750.67 / Max: 527053 Min: 521324 / Avg: 521572.33 / Max: 522002 Min: 523393 / Avg: 523574.67 / Max: 523856 Min: 538010 / Avg: 539060 / Max: 540491 Min: 524902 / Avg: 532956.67 / Max: 538578 Min: 530354 / Avg: 531740.33 / Max: 533554 Min: 534521 / Avg: 542520.67 / Max: 547450 Min: 522517 / Avg: 524188.67 / Max: 525833 Min: 510027 / Avg: 511228.33 / Max: 513156 Min: 510924 / Avg: 513356.33 / Max: 517660 Min: 511300 / Avg: 511905.67 / Max: 513043
Timed FFmpeg Compilation This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 4.2.2 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 15 30 45 60 75 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 33.98 54.15 24.03 23.98 25.21 25.50 27.86 29.23 29.12 32.07 39.74 42.86 49.92 68.50
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 4.2.2 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 Min: 33.92 / Avg: 33.98 / Max: 34.07 Min: 54.13 / Avg: 54.15 / Max: 54.16 Min: 23.95 / Avg: 24.03 / Max: 24.14 Min: 23.93 / Avg: 23.98 / Max: 24.01 Min: 25.18 / Avg: 25.21 / Max: 25.25 Min: 25.47 / Avg: 25.5 / Max: 25.56 Min: 27.84 / Avg: 27.86 / Max: 27.89 Min: 29.13 / Avg: 29.23 / Max: 29.37 Min: 29.07 / Avg: 29.12 / Max: 29.19 Min: 32.01 / Avg: 32.07 / Max: 32.1 Min: 39.64 / Avg: 39.74 / Max: 39.84 Min: 42.78 / Avg: 42.86 / Max: 42.98 Min: 49.75 / Avg: 49.92 / Max: 50.11 Min: 68.42 / Avg: 68.5 / Max: 68.63
LibRaw LibRaw is a RAW image decoder for digital camera photos. This test profile runs LibRaw's post-processing benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.17, N = 3 SE +/- 0.23, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 38.01 38.59 34.50 34.01 33.64 34.38 33.06 34.11 34.03 33.71 33.08 33.12 32.48 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
Mpix/sec Per Watt
OpenBenchmarking.org Mpix/sec Per Watt, More Is Better LibRaw 0.20 Post-Processing Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1395 0.279 0.4185 0.558 0.6975 0.40 0.53 0.40 0.41 0.47 0.55 0.41 0.55 0.55 0.55 0.61 0.62 0.62
Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 37.96 / Avg: 38.01 / Max: 38.06 Min: 38.57 / Avg: 38.59 / Max: 38.62 Min: 34.32 / Avg: 34.5 / Max: 34.63 Min: 33.71 / Avg: 34.01 / Max: 34.31 Min: 33.2 / Avg: 33.64 / Max: 34 Min: 34.31 / Avg: 34.38 / Max: 34.48 Min: 32.93 / Avg: 33.06 / Max: 33.13 Min: 33.96 / Avg: 34.11 / Max: 34.27 Min: 33.89 / Avg: 34.03 / Max: 34.12 Min: 33.67 / Avg: 33.71 / Max: 33.74 Min: 32.96 / Avg: 33.08 / Max: 33.27 Min: 33.04 / Avg: 33.12 / Max: 33.21 Min: 32.44 / Avg: 32.48 / Max: 32.51 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
Basis Universal Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 33.80 60.40 18.87 18.66 20.80 23.97 25.88 25.28 29.16 39.75 41.24 51.77 73.43 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 14 28 42 56 70 Min: 33.79 / Avg: 33.8 / Max: 33.81 Min: 60.39 / Avg: 60.4 / Max: 60.41 Min: 18.86 / Avg: 18.87 / Max: 18.88 Min: 18.65 / Avg: 18.66 / Max: 18.66 Min: 20.77 / Avg: 20.8 / Max: 20.82 Min: 23.96 / Avg: 23.97 / Max: 23.98 Min: 25.87 / Avg: 25.88 / Max: 25.91 Min: 25.27 / Avg: 25.28 / Max: 25.31 Min: 29.15 / Avg: 29.16 / Max: 29.17 Min: 39.75 / Avg: 39.75 / Max: 39.76 Min: 41.2 / Avg: 41.24 / Max: 41.3 Min: 51.75 / Avg: 51.77 / Max: 51.78 Min: 73.43 / Avg: 73.43 / Max: 73.44 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Etcpak Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: ETC2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.00, N = 3 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 159.60 159.79 137.28 135.24 135.14 139.25 135.01 137.34 137.09 135.18 130.88 131.15 131.18 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
Mpx/s Per Watt
OpenBenchmarking.org Mpx/s Per Watt, More Is Better Etcpak 0.7 Configuration: ETC2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.7358 1.4716 2.2074 2.9432 3.679 2.29 3.07 2.00 2.00 2.39 2.91 2.02 2.86 2.87 2.79 3.16 3.24 3.27
Result Confidence
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: ETC2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 159.56 / Avg: 159.6 / Max: 159.65 Min: 159.77 / Avg: 159.79 / Max: 159.8 Min: 137.26 / Avg: 137.28 / Max: 137.3 Min: 135.2 / Avg: 135.23 / Max: 135.25 Min: 135.11 / Avg: 135.14 / Max: 135.2 Min: 139.03 / Avg: 139.25 / Max: 139.39 Min: 134.91 / Avg: 135.01 / Max: 135.17 Min: 137.34 / Avg: 137.34 / Max: 137.35 Min: 136.98 / Avg: 137.09 / Max: 137.3 Min: 135.12 / Avg: 135.18 / Max: 135.24 Min: 130.77 / Avg: 130.88 / Max: 131.05 Min: 131.11 / Avg: 131.15 / Max: 131.19 Min: 131.16 / Avg: 131.18 / Max: 131.22 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.3314 0.6628 0.9942 1.3256 1.657 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 SE +/- 0.002, N = 3 SE +/- 0.002, N = 3 SE +/- 0.005, N = 3 0.630 1.173 0.279 0.267 0.323 0.316 0.382 0.426 0.407 0.480 0.679 0.729 0.908 1.473 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.63 / Avg: 0.63 / Max: 0.63 Min: 1.17 / Avg: 1.17 / Max: 1.18 Min: 0.28 / Avg: 0.28 / Max: 0.28 Min: 0.26 / Avg: 0.27 / Max: 0.27 Min: 0.32 / Avg: 0.32 / Max: 0.32 Min: 0.32 / Avg: 0.32 / Max: 0.32 Min: 0.38 / Avg: 0.38 / Max: 0.38 Min: 0.43 / Avg: 0.43 / Max: 0.43 Min: 0.41 / Avg: 0.41 / Max: 0.41 Min: 0.48 / Avg: 0.48 / Max: 0.48 Min: 0.68 / Avg: 0.68 / Max: 0.68 Min: 0.73 / Avg: 0.73 / Max: 0.73 Min: 0.9 / Avg: 0.91 / Max: 0.91 Min: 1.47 / Avg: 1.47 / Max: 1.48 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K SE +/- 248.26, N = 3 SE +/- 681.51, N = 3 SE +/- 5067.03, N = 3 SE +/- 6923.68, N = 3 SE +/- 1360.28, N = 3 SE +/- 3710.21, N = 3 SE +/- 1178.79, N = 3 SE +/- 853.89, N = 3 SE +/- 1108.32, N = 3 SE +/- 1318.12, N = 3 SE +/- 891.46, N = 3 SE +/- 804.68, N = 3 SE +/- 696.35, N = 3 SE +/- 518.37, N = 3 397038 213345 899992 939750 775687 791456 655774 587548 614183 521653 368449 343229 275677 169832 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160K 320K 480K 640K 800K Min: 396660.01 / Avg: 397038.18 / Max: 397505.91 Min: 212208.72 / Avg: 213344.58 / Max: 214564.99 Min: 889916.01 / Avg: 899992.3 / Max: 905966.24 Min: 926183.25 / Avg: 939750.48 / Max: 948934.04 Min: 773949.79 / Avg: 775687.15 / Max: 778368.93 Min: 784040.65 / Avg: 791455.71 / Max: 795407.24 Min: 654021.41 / Avg: 655774.08 / Max: 658015.96 Min: 585986.61 / Avg: 587547.73 / Max: 588927.93 Min: 613074.34 / Avg: 614182.67 / Max: 616399.31 Min: 519362.06 / Avg: 521653.42 / Max: 523928.08 Min: 367104.95 / Avg: 368449.41 / Max: 370135.74 Min: 341698.14 / Avg: 343228.63 / Max: 344424.83 Min: 274390.74 / Avg: 275676.66 / Max: 276782.78 Min: 168803.67 / Avg: 169832.3 / Max: 170458.78 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1222 0.2444 0.3666 0.4888 0.611 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.263 0.464 0.112 0.109 0.120 0.120 0.153 0.173 0.164 0.201 0.280 0.290 0.359 0.543 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latency EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.26 / Avg: 0.26 / Max: 0.26 Min: 0.46 / Avg: 0.46 / Max: 0.47 Min: 0.11 / Avg: 0.11 / Max: 0.11 Min: 0.11 / Avg: 0.11 / Max: 0.11 Min: 0.12 / Avg: 0.12 / Max: 0.12 Min: 0.12 / Avg: 0.12 / Max: 0.12 Min: 0.15 / Avg: 0.15 / Max: 0.15 Min: 0.17 / Avg: 0.17 / Max: 0.17 Min: 0.16 / Avg: 0.16 / Max: 0.16 Min: 0.2 / Avg: 0.2 / Max: 0.2 Min: 0.28 / Avg: 0.28 / Max: 0.28 Min: 0.29 / Avg: 0.29 / Max: 0.29 Min: 0.36 / Avg: 0.36 / Max: 0.36 Min: 0.54 / Avg: 0.54 / Max: 0.54 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K SE +/- 501.41, N = 3 SE +/- 196.73, N = 3 SE +/- 3714.01, N = 3 SE +/- 2726.39, N = 3 SE +/- 492.11, N = 3 SE +/- 2696.01, N = 3 SE +/- 684.57, N = 3 SE +/- 1146.88, N = 3 SE +/- 1605.99, N = 3 SE +/- 1145.86, N = 3 SE +/- 406.00, N = 3 SE +/- 721.66, N = 3 SE +/- 527.65, N = 3 SE +/- 258.28, N = 3 381184 215617 897453 915523 835457 832509 654914 578340 610667 499138 357781 344598 278510 184251 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Result Confidence
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 100 - Mode: Read Only EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160K 320K 480K 640K 800K Min: 380600.92 / Avg: 381183.76 / Max: 382181.9 Min: 215250.14 / Avg: 215616.98 / Max: 215923.6 Min: 891671.21 / Avg: 897453.24 / Max: 904382.54 Min: 911128.88 / Avg: 915522.69 / Max: 920516.11 Min: 834632.97 / Avg: 835456.65 / Max: 836335.05 Min: 827135.25 / Avg: 832508.89 / Max: 835580.98 Min: 654192.39 / Avg: 654914.4 / Max: 656282.85 Min: 576053.18 / Avg: 578339.82 / Max: 579639.5 Min: 608958.8 / Avg: 610666.61 / Max: 613876.39 Min: 497224.75 / Avg: 499138.43 / Max: 501187.22 Min: 357317.88 / Avg: 357781.31 / Max: 358590.46 Min: 343314.32 / Avg: 344597.63 / Max: 345811.31 Min: 277469.8 / Avg: 278509.56 / Max: 279185.65 Min: 183809.92 / Avg: 184251.3 / Max: 184704.38 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Numenta Anomaly Benchmark Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Bayesian Changepoint EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.20, N = 3 SE +/- 0.22, N = 3 SE +/- 0.26, N = 3 SE +/- 0.35, N = 3 SE +/- 0.24, N = 3 SE +/- 0.14, N = 3 SE +/- 0.20, N = 3 SE +/- 0.11, N = 3 SE +/- 0.27, N = 3 SE +/- 0.24, N = 3 SE +/- 0.10, N = 3 SE +/- 0.48, N = 3 SE +/- 0.13, N = 3 SE +/- 0.11, N = 3 29.45 35.67 33.27 33.41 33.60 33.96 32.87 33.43 32.65 34.31 34.84 36.35 38.07 45.21
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Bayesian Changepoint EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 29.13 / Avg: 29.45 / Max: 29.83 Min: 35.42 / Avg: 35.67 / Max: 36.1 Min: 32.84 / Avg: 33.27 / Max: 33.73 Min: 32.9 / Avg: 33.41 / Max: 34.08 Min: 33.36 / Avg: 33.6 / Max: 34.09 Min: 33.82 / Avg: 33.96 / Max: 34.23 Min: 32.54 / Avg: 32.87 / Max: 33.22 Min: 33.26 / Avg: 33.43 / Max: 33.62 Min: 32.15 / Avg: 32.64 / Max: 33.08 Min: 33.83 / Avg: 34.31 / Max: 34.61 Min: 34.65 / Avg: 34.84 / Max: 35.01 Min: 35.71 / Avg: 36.35 / Max: 37.28 Min: 37.91 / Avg: 38.07 / Max: 38.32 Min: 45 / Avg: 45.21 / Max: 45.37
miniFE MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org CG Mflops, More Is Better miniFE 2.2 Problem Size: Small EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K SE +/- 66.01, N = 14 SE +/- 8.04, N = 3 SE +/- 34.52, N = 3 SE +/- 5.01, N = 3 SE +/- 29.81, N = 3 SE +/- 68.31, N = 3 SE +/- 5.12, N = 3 SE +/- 8.80, N = 3 SE +/- 22.39, N = 3 SE +/- 5.98, N = 3 SE +/- 24.99, N = 3 SE +/- 1.40, N = 3 SE +/- 16.20, N = 3 SE +/- 10.40, N = 3 6787.06 17351.20 19155.40 19256.70 19436.30 18424.10 16649.70 19645.40 16649.50 16743.60 17082.40 10003.60 10066.60 10156.60 1. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi
CG Mflops Per Watt
OpenBenchmarking.org CG Mflops Per Watt, More Is Better miniFE 2.2 Problem Size: Small EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 52.66 167.09 139.02 142.84 140.60 149.38 157.73 152.28 152.21 160.96 181.39 128.37 132.41 147.34
Result Confidence
OpenBenchmarking.org CG Mflops, More Is Better miniFE 2.2 Problem Size: Small EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 5976.74 / Avg: 6787.06 / Max: 6965.65 Min: 17335.9 / Avg: 17351.17 / Max: 17363.2 Min: 19088.2 / Avg: 19155.37 / Max: 19202.8 Min: 19250.6 / Avg: 19256.67 / Max: 19266.6 Min: 19380.5 / Avg: 19436.3 / Max: 19482.4 Min: 18336.1 / Avg: 18424.1 / Max: 18558.6 Min: 16643.8 / Avg: 16649.7 / Max: 16659.9 Min: 19632.2 / Avg: 19645.43 / Max: 19662.1 Min: 16604.7 / Avg: 16649.47 / Max: 16672.4 Min: 16735 / Avg: 16743.6 / Max: 16755.1 Min: 17033.7 / Avg: 17082.43 / Max: 17116.4 Min: 10000.9 / Avg: 10003.63 / Max: 10005.5 Min: 10034.4 / Avg: 10066.63 / Max: 10085.6 Min: 10136.2 / Avg: 10156.6 / Max: 10170.3 1. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Socket Activity EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K SE +/- 17.97, N = 3 SE +/- 13.57, N = 3 SE +/- 57.10, N = 3 SE +/- 64.33, N = 3 SE +/- 24.65, N = 3 SE +/- 32.93, N = 3 SE +/- 125.44, N = 9 SE +/- 30.09, N = 3 SE +/- 16.45, N = 3 SE +/- 6.53, N = 3 SE +/- 16.90, N = 3 SE +/- 8.24, N = 3 SE +/- 24.34, N = 3 8415.19 5638.53 19577.03 19990.75 18192.32 17266.21 15568.24 16396.57 13570.83 9499.33 9134.72 7332.82 4965.96 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: Socket Activity EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 42.43 51.33 109.94 115.18 110.23 104.78 98.20 114.89 92.04 79.41 93.40 75.36 67.53
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Socket Activity EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 8394.92 / Avg: 8415.19 / Max: 8451.03 Min: 5624.62 / Avg: 5638.53 / Max: 5665.66 Min: 19482.9 / Avg: 19577.03 / Max: 19680.11 Min: 19911.11 / Avg: 19990.75 / Max: 20118.07 Min: 18145.03 / Avg: 18192.32 / Max: 18228.03 Min: 17206.64 / Avg: 17266.21 / Max: 17320.33 Min: 14571.02 / Avg: 15568.24 / Max: 15743.07 Min: 16363.28 / Avg: 16396.57 / Max: 16456.63 Min: 13539.78 / Avg: 13570.83 / Max: 13595.76 Min: 9488.78 / Avg: 9499.33 / Max: 9511.28 Min: 9104 / Avg: 9134.72 / Max: 9162.29 Min: 7316.76 / Avg: 7332.82 / Max: 7344.03 Min: 4920.56 / Avg: 4965.96 / Max: 5003.89 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.26, N = 5 SE +/- 0.03, N = 3 SE +/- 0.26, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.14, N = 3 35.42 27.25 23.32 23.82 21.94 26.47 27.64 22.94 27.60 29.02 24.92 46.07 42.95 45.76 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 35.36 / Avg: 35.42 / Max: 35.47 Min: 27.2 / Avg: 27.25 / Max: 27.28 Min: 23.22 / Avg: 23.32 / Max: 23.46 Min: 23.2 / Avg: 23.82 / Max: 24.72 Min: 21.88 / Avg: 21.94 / Max: 21.99 Min: 26.02 / Avg: 26.47 / Max: 26.92 Min: 27.53 / Avg: 27.64 / Max: 27.7 Min: 22.73 / Avg: 22.94 / Max: 23.1 Min: 27.55 / Avg: 27.6 / Max: 27.63 Min: 28.89 / Avg: 29.02 / Max: 29.09 Min: 24.77 / Avg: 24.92 / Max: 25.05 Min: 46.05 / Avg: 46.07 / Max: 46.1 Min: 42.88 / Avg: 42.95 / Max: 43.08 Min: 45.56 / Avg: 45.76 / Max: 46.03 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pathlib EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 17.5 17.2 20.2 20.5 20.5 20.5 19.9 20.4 19.9 20.2 20.4 21.3 21.0 21.1
Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pathlib EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 17.5 / Avg: 17.5 / Max: 17.5 Min: 17.2 / Avg: 17.2 / Max: 17.2 Min: 20.2 / Avg: 20.2 / Max: 20.2 Min: 20.5 / Avg: 20.53 / Max: 20.6 Min: 20.5 / Avg: 20.5 / Max: 20.5 Min: 20.5 / Avg: 20.5 / Max: 20.5 Min: 19.8 / Avg: 19.87 / Max: 19.9 Min: 20.4 / Avg: 20.43 / Max: 20.5 Min: 19.9 / Avg: 19.9 / Max: 19.9 Min: 20.2 / Avg: 20.23 / Max: 20.3 Min: 20.4 / Avg: 20.43 / Max: 20.5 Min: 21.2 / Avg: 21.27 / Max: 21.3 Min: 20.9 / Avg: 20.97 / Max: 21 Min: 21.1 / Avg: 21.1 / Max: 21.1
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 4 SE +/- 0.02, N = 4 SE +/- 0.01, N = 4 SE +/- 0.02, N = 4 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 32.58 64.97 12.85 12.82 15.04 15.21 19.55 21.43 20.91 25.68 38.49 39.66 52.81 79.17 1. (CC) gcc options: -lm -lpthread -O3
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 15 30 45 60 75 Min: 32.55 / Avg: 32.58 / Max: 32.61 Min: 64.93 / Avg: 64.97 / Max: 64.99 Min: 12.83 / Avg: 12.85 / Max: 12.87 Min: 12.79 / Avg: 12.81 / Max: 12.87 Min: 15.01 / Avg: 15.04 / Max: 15.06 Min: 15.17 / Avg: 15.21 / Max: 15.24 Min: 19.49 / Avg: 19.55 / Max: 19.6 Min: 21.4 / Avg: 21.43 / Max: 21.45 Min: 20.89 / Avg: 20.91 / Max: 20.94 Min: 25.67 / Avg: 25.68 / Max: 25.7 Min: 38.48 / Avg: 38.49 / Max: 38.51 Min: 39.61 / Avg: 39.66 / Max: 39.7 Min: 52.8 / Avg: 52.81 / Max: 52.82 Min: 79.13 / Avg: 79.17 / Max: 79.22 1. (CC) gcc options: -lm -lpthread -O3
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6.83 3.65 16.67 16.95 14.49 14.29 11.36 10.42 10.60 8.77 6.07 5.65 4.39 2.82 MIN: 6.67 / MAX: 6.94 MIN: 3.58 / MAX: 3.69 MIN: 16.39 / MAX: 16.95 MIN: 16.39 / MAX: 17.24 MIN: 14.08 / MAX: 14.71 MIN: 13.89 / MAX: 14.49 MIN: 11.11 / MAX: 11.49 MIN: 10.2 / MAX: 10.53 MIN: 10.42 / MAX: 10.75 MIN: 8.62 / MAX: 8.85 MIN: 5.99 / MAX: 6.13 MIN: 5.59 / MAX: 5.75 MIN: 4.33 / MAX: 4.44 MIN: 2.79 / MAX: 2.86
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.027 0.054 0.081 0.108 0.135 0.03 0.03 0.11 0.12 0.09 0.10 0.08 0.07 0.08 0.06 0.05 0.06 0.05 0.04
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 6.8 / Avg: 6.83 / Max: 6.85 Min: 3.64 / Avg: 3.65 / Max: 3.65 Min: 16.67 / Avg: 16.67 / Max: 16.67 Min: 16.95 / Avg: 16.95 / Max: 16.95 Min: 14.49 / Avg: 14.49 / Max: 14.49 Min: 14.29 / Avg: 14.29 / Max: 14.29 Min: 11.36 / Avg: 11.36 / Max: 11.36 Min: 10.42 / Avg: 10.42 / Max: 10.42 Min: 10.53 / Avg: 10.6 / Max: 10.64 Min: 8.77 / Avg: 8.77 / Max: 8.77 Min: 6.06 / Avg: 6.07 / Max: 6.1 Min: 5.65 / Avg: 5.65 / Max: 5.65 Min: 4.39 / Avg: 4.39 / Max: 4.39 Min: 2.82 / Avg: 2.82 / Max: 2.83
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 82.34, N = 3 SE +/- 31.62, N = 3 SE +/- 34.41, N = 3 SE +/- 298.03, N = 3 SE +/- 281.17, N = 3 SE +/- 82.31, N = 3 SE +/- 207.08, N = 3 SE +/- 26.26, N = 3 SE +/- 106.91, N = 3 SE +/- 81.46, N = 3 SE +/- 3.68, N = 3 SE +/- 22.56, N = 3 SE +/- 24.66, N = 3 SE +/- 12.95, N = 3 44891.69 43172.51 102548.95 103985.91 99557.23 95079.41 79644.88 86547.28 78846.96 74421.61 61041.27 49617.92 43321.38 33816.48 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 140 280 420 560 700 297.36 419.08 629.05 660.15 602.66 655.10 638.72 572.31 616.99 607.37 593.84 573.26 529.25 486.72
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K Min: 44754.21 / Avg: 44891.69 / Max: 45038.95 Min: 43122.35 / Avg: 43172.51 / Max: 43230.93 Min: 102490.22 / Avg: 102548.95 / Max: 102609.37 Min: 103567 / Avg: 103985.91 / Max: 104562.59 Min: 98994.91 / Avg: 99557.23 / Max: 99842.55 Min: 94922.44 / Avg: 95079.41 / Max: 95200.87 Min: 79399.38 / Avg: 79644.88 / Max: 80056.49 Min: 86498.25 / Avg: 86547.28 / Max: 86588.09 Min: 78710.27 / Avg: 78846.96 / Max: 79057.7 Min: 74315 / Avg: 74421.61 / Max: 74581.6 Min: 61034.85 / Avg: 61041.27 / Max: 61047.59 Min: 49572.87 / Avg: 49617.92 / Max: 49642.69 Min: 43285.83 / Avg: 43321.38 / Max: 43368.77 Min: 33792.21 / Avg: 33816.48 / Max: 33836.43 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.00961, N = 3 SE +/- 0.06606, N = 3 SE +/- 0.00813, N = 3 SE +/- 0.01627, N = 15 SE +/- 0.02012, N = 3 SE +/- 0.01668, N = 15 SE +/- 0.00152, N = 3 SE +/- 0.01295, N = 3 SE +/- 0.01343, N = 3 SE +/- 0.00832, N = 3 SE +/- 0.01740, N = 3 SE +/- 0.01071, N = 3 SE +/- 0.00923, N = 3 SE +/- 0.02020, N = 3 2.75743 5.03236 2.33519 2.02231 1.98726 2.15794 2.04866 2.22242 2.19727 2.48977 3.28377 3.50175 4.44077 6.12261 MIN: 2.63 MIN: 4.86 MIN: 2.08 MIN: 1.84 MIN: 1.87 MIN: 1.99 MIN: 1.99 MIN: 2.06 MIN: 2.12 MIN: 2.4 MIN: 3.08 MIN: 3.1 MIN: 4.2 MIN: 5.93 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 2.74 / Avg: 2.76 / Max: 2.77 Min: 4.94 / Avg: 5.03 / Max: 5.16 Min: 2.32 / Avg: 2.34 / Max: 2.35 Min: 1.93 / Avg: 2.02 / Max: 2.15 Min: 1.95 / Avg: 1.99 / Max: 2.02 Min: 2.07 / Avg: 2.16 / Max: 2.28 Min: 2.05 / Avg: 2.05 / Max: 2.05 Min: 2.21 / Avg: 2.22 / Max: 2.25 Min: 2.18 / Avg: 2.2 / Max: 2.22 Min: 2.48 / Avg: 2.49 / Max: 2.51 Min: 3.25 / Avg: 3.28 / Max: 3.3 Min: 3.49 / Avg: 3.5 / Max: 3.52 Min: 4.43 / Avg: 4.44 / Max: 4.46 Min: 6.1 / Avg: 6.12 / Max: 6.16 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
x265 This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 4K EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.12, N = 3 SE +/- 0.13, N = 4 SE +/- 0.08, N = 3 SE +/- 0.21, N = 3 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 20.00 10.20 25.04 25.65 25.23 25.36 25.21 23.71 25.13 23.84 20.44 20.03 17.09 9.20 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better x265 3.4 Video Input: Bosphorus 4K EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0518 0.1036 0.1554 0.2072 0.259 0.12 0.11 0.18 0.20 0.18 0.21 0.23 0.18 0.22 0.21 0.20 0.23 0.20 0.14
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 4K EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 19.78 / Avg: 20 / Max: 20.2 Min: 9.85 / Avg: 10.2 / Max: 10.45 Min: 24.94 / Avg: 25.04 / Max: 25.19 Min: 25.41 / Avg: 25.65 / Max: 26.06 Min: 25.14 / Avg: 25.23 / Max: 25.39 Min: 25.3 / Avg: 25.36 / Max: 25.48 Min: 25.09 / Avg: 25.21 / Max: 25.34 Min: 23.64 / Avg: 23.71 / Max: 23.81 Min: 25.08 / Avg: 25.13 / Max: 25.16 Min: 23.72 / Avg: 23.84 / Max: 23.98 Min: 20.34 / Avg: 20.44 / Max: 20.61 Min: 19.97 / Avg: 20.03 / Max: 20.1 Min: 16.96 / Avg: 17.09 / Max: 17.21 Min: 9.07 / Avg: 9.2 / Max: 9.37 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
AOM AV1 This is a simple test of the AOMedia AV1 encoder run on the CPU with a sample video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Two-Pass EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.8775 1.755 2.6325 3.51 4.3875 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.90 3.73 3.38 3.36 3.33 3.37 3.48 3.37 3.42 3.42 3.36 3.23 3.19 3.08 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Two-Pass EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0158 0.0316 0.0474 0.0632 0.079 0.05 0.06 0.04 0.05 0.05 0.05 0.06 0.05 0.06 0.06 0.06 0.07 0.07 0.07
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Two-Pass EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 3.89 / Avg: 3.9 / Max: 3.91 Min: 3.73 / Avg: 3.73 / Max: 3.74 Min: 3.38 / Avg: 3.38 / Max: 3.39 Min: 3.36 / Avg: 3.36 / Max: 3.36 Min: 3.32 / Avg: 3.33 / Max: 3.35 Min: 3.37 / Avg: 3.37 / Max: 3.38 Min: 3.48 / Avg: 3.48 / Max: 3.48 Min: 3.36 / Avg: 3.37 / Max: 3.37 Min: 3.41 / Avg: 3.42 / Max: 3.43 Min: 3.42 / Avg: 3.42 / Max: 3.43 Min: 3.33 / Avg: 3.36 / Max: 3.37 Min: 3.21 / Avg: 3.23 / Max: 3.27 Min: 3.18 / Avg: 3.19 / Max: 3.21 Min: 3.07 / Avg: 3.08 / Max: 3.1 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Pennant Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 15 30 45 60 75 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 4 SE +/- 0.03, N = 4 SE +/- 0.02, N = 4 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.11, N = 3 SE +/- 0.14, N = 3 25.88 52.79 11.94 11.80 15.35 22.23 18.06 23.03 25.89 34.03 41.69 49.11 66.42 1. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
Result Confidence
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 Min: 25.85 / Avg: 25.88 / Max: 25.92 Min: 52.75 / Avg: 52.79 / Max: 52.81 Min: 11.88 / Avg: 11.94 / Max: 12 Min: 11.75 / Avg: 11.8 / Max: 11.87 Min: 15.3 / Avg: 15.35 / Max: 15.39 Min: 22.07 / Avg: 22.23 / Max: 22.32 Min: 18.03 / Avg: 18.06 / Max: 18.09 Min: 22.87 / Avg: 23.03 / Max: 23.23 Min: 25.8 / Avg: 25.89 / Max: 25.96 Min: 33.93 / Avg: 34.03 / Max: 34.16 Min: 41.58 / Avg: 41.69 / Max: 41.79 Min: 48.99 / Avg: 49.11 / Max: 49.32 Min: 66.13 / Avg: 66.42 / Max: 66.58 1. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
ebizzy This is a test of ebizzy, a program to generate workloads resembling web server workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 600K 1200K 1800K 2400K 3000K SE +/- 7117.75, N = 3 SE +/- 6904.71, N = 7 SE +/- 25794.01, N = 15 SE +/- 34594.95, N = 12 SE +/- 37169.94, N = 3 SE +/- 24033.50, N = 3 SE +/- 8453.72, N = 3 SE +/- 15669.57, N = 3 SE +/- 10663.27, N = 3 SE +/- 15686.47, N = 3 SE +/- 8057.83, N = 3 SE +/- 8956.45, N = 3 SE +/- 906.56, N = 3 SE +/- 4933.88, N = 3 1475280 776880 2701767 2762647 2719388 2456511 2136850 1977325 1947836 1721854 1208466 1021990 883965 623272 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
Records/s Per Watt
OpenBenchmarking.org Records/s Per Watt, More Is Better ebizzy 0.3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K 7194.42 5966.03 16026.57 16672.76 15813.47 15696.63 13548.39 13352.11 14373.63 12779.06 10725.42 11127.27 9696.42 7399.84
Result Confidence
OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500K 1000K 1500K 2000K 2500K Min: 1461618 / Avg: 1475280 / Max: 1485575 Min: 747575 / Avg: 776880.43 / Max: 796470 Min: 2510210 / Avg: 2701767.07 / Max: 2845248 Min: 2508914 / Avg: 2762646.92 / Max: 2926207 Min: 2645052 / Avg: 2719388 / Max: 2757214 Min: 2420178 / Avg: 2456510.67 / Max: 2501931 Min: 2127912 / Avg: 2136850 / Max: 2153748 Min: 1951917 / Avg: 1977324.67 / Max: 2005917 Min: 1926759 / Avg: 1947835.67 / Max: 1961193 Min: 1691610 / Avg: 1721854 / Max: 1744199 Min: 1192406 / Avg: 1208466.33 / Max: 1217652 Min: 1009030 / Avg: 1021990 / Max: 1039179 Min: 882270 / Avg: 883965 / Max: 885370 Min: 616705 / Avg: 623272 / Max: 632934 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
LZ4 Compression This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 24.26, N = 3 SE +/- 42.46, N = 3 SE +/- 36.93, N = 3 SE +/- 27.42, N = 3 SE +/- 23.29, N = 3 SE +/- 47.44, N = 3 SE +/- 41.47, N = 3 SE +/- 5.92, N = 3 SE +/- 43.01, N = 3 SE +/- 18.80, N = 3 SE +/- 7.93, N = 3 SE +/- 27.35, N = 3 SE +/- 37.22, N = 3 10701.3 11210.0 10996.1 10926.3 10920.5 10993.2 10931.6 10968.6 10951.0 11006.6 10898.3 10929.9 10936.6 1. (CC) gcc options: -O3
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 139.96 189.50 142.24 147.64 172.19 202.37 146.91 200.00 202.51 201.84 227.91 234.75 237.48
Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Decompression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 10654.2 / Avg: 10701.33 / Max: 10734.9 Min: 11129.3 / Avg: 11209.97 / Max: 11273.3 Min: 10957.3 / Avg: 10996.07 / Max: 11069.9 Min: 10874.5 / Avg: 10926.3 / Max: 10967.8 Min: 10874.2 / Avg: 10920.47 / Max: 10948.2 Min: 10916.9 / Avg: 10993.2 / Max: 11080.2 Min: 10851.3 / Avg: 10931.63 / Max: 10989.7 Min: 10961.1 / Avg: 10968.63 / Max: 10980.3 Min: 10899.4 / Avg: 10951 / Max: 11036.4 Min: 10976.1 / Avg: 11006.63 / Max: 11040.9 Min: 10885.6 / Avg: 10898.33 / Max: 10912.9 Min: 10878.3 / Avg: 10929.9 / Max: 10971.4 Min: 10862.2 / Avg: 10936.63 / Max: 10975.1 1. (CC) gcc options: -O3
Result
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 19.46, N = 3 SE +/- 52.96, N = 3 SE +/- 50.78, N = 3 SE +/- 19.08, N = 3 SE +/- 18.08, N = 3 SE +/- 14.87, N = 3 SE +/- 55.86, N = 3 SE +/- 5.54, N = 3 SE +/- 57.96, N = 3 SE +/- 29.02, N = 3 SE +/- 8.54, N = 3 SE +/- 18.11, N = 3 SE +/- 52.82, N = 3 9461.09 9802.81 9496.00 9376.86 9369.35 9468.59 9434.31 9464.00 9472.86 9428.77 9360.35 9365.44 9421.69 1. (CC) gcc options: -O3
Result Confidence
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 9423.06 / Avg: 9461.09 / Max: 9487.27 Min: 9744.32 / Avg: 9802.81 / Max: 9908.52 Min: 9444.69 / Avg: 9496 / Max: 9597.56 Min: 9351.87 / Avg: 9376.86 / Max: 9414.33 Min: 9339.76 / Avg: 9369.35 / Max: 9402.14 Min: 9438.88 / Avg: 9468.59 / Max: 9484.45 Min: 9377.21 / Avg: 9434.31 / Max: 9546.03 Min: 9454.4 / Avg: 9464 / Max: 9473.6 Min: 9413.59 / Avg: 9472.86 / Max: 9588.77 Min: 9379.2 / Avg: 9428.77 / Max: 9479.71 Min: 9345.51 / Avg: 9360.35 / Max: 9375.08 Min: 9344.67 / Avg: 9365.44 / Max: 9401.53 Min: 9316.16 / Avg: 9421.69 / Max: 9478.74 1. (CC) gcc options: -O3
TTSIOD 3D Renderer A portable GPL 3D software renderer that supports OpenMP and Intel Threading Building Blocks with many different rendering modes. This version does not use OpenGL but is entirely CPU/software based. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 SE +/- 0.21, N = 3 SE +/- 0.88, N = 3 SE +/- 4.91, N = 3 SE +/- 3.57, N = 3 SE +/- 12.71, N = 3 SE +/- 6.72, N = 12 SE +/- 1.67, N = 3 SE +/- 3.72, N = 3 SE +/- 2.79, N = 3 SE +/- 2.84, N = 3 SE +/- 1.71, N = 3 SE +/- 1.34, N = 3 SE +/- 0.85, N = 3 SE +/- 0.73, N = 3 589.10 359.54 792.50 873.42 889.39 933.04 858.47 849.27 857.99 721.30 551.97 563.20 457.13 327.45 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 4.63 4.41 5.54 6.31 6.80 8.10 9.07 7.17 8.75 7.79 7.06 8.15 6.96 5.67
Result Confidence
OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160 320 480 640 800 Min: 588.69 / Avg: 589.1 / Max: 589.38 Min: 357.91 / Avg: 359.54 / Max: 360.93 Min: 782.72 / Avg: 792.5 / Max: 798.09 Min: 866.7 / Avg: 873.42 / Max: 878.89 Min: 864.6 / Avg: 889.39 / Max: 906.7 Min: 896.62 / Avg: 933.04 / Max: 961.45 Min: 855.29 / Avg: 858.47 / Max: 860.96 Min: 843.53 / Avg: 849.27 / Max: 856.24 Min: 853.39 / Avg: 857.99 / Max: 863.04 Min: 716.79 / Avg: 721.3 / Max: 726.53 Min: 548.55 / Avg: 551.97 / Max: 553.8 Min: 561.8 / Avg: 563.2 / Max: 565.87 Min: 455.79 / Avg: 457.13 / Max: 458.72 Min: 326.51 / Avg: 327.45 / Max: 328.88 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++
AOM AV1 This is a simple test of the AOMedia AV1 encoder run on the CPU with a sample video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Realtime EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.16, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 21.76 20.22 19.84 19.59 19.48 19.63 20.19 19.51 19.97 19.88 19.11 18.92 17.83 17.30 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Realtime EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.072 0.144 0.216 0.288 0.36 0.21 0.26 0.22 0.23 0.23 0.26 0.30 0.23 0.30 0.29 0.29 0.32 0.31 0.31
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Realtime EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 21.73 / Avg: 21.76 / Max: 21.79 Min: 20.16 / Avg: 20.22 / Max: 20.28 Min: 19.82 / Avg: 19.84 / Max: 19.86 Min: 19.58 / Avg: 19.59 / Max: 19.6 Min: 19.45 / Avg: 19.48 / Max: 19.52 Min: 19.61 / Avg: 19.63 / Max: 19.66 Min: 20.17 / Avg: 20.19 / Max: 20.22 Min: 19.46 / Avg: 19.51 / Max: 19.57 Min: 19.9 / Avg: 19.97 / Max: 20.03 Min: 19.75 / Avg: 19.88 / Max: 19.97 Min: 18.83 / Avg: 19.11 / Max: 19.38 Min: 18.8 / Avg: 18.92 / Max: 19.03 Min: 17.77 / Avg: 17.83 / Max: 17.92 Min: 17.21 / Avg: 17.3 / Max: 17.37 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 400 800 1200 1600 2000 SE +/- 9.24, N = 3 SE +/- 5.70, N = 3 SE +/- 4.04, N = 3 SE +/- 11.89, N = 3 SE +/- 5.74, N = 3 SE +/- 1.95, N = 3 SE +/- 3.36, N = 3 SE +/- 2.26, N = 3 SE +/- 1.26, N = 3 SE +/- 2.89, N = 3 934.00 1277.15 1971.50 2006.81 1885.98 1992.76 1884.31 1647.78 1422.07 1083.36 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 5 10 15 20 25 7.29 15.36 12.90 13.49 18.15 15.32 17.73 19.55 19.39 18.48
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 300 600 900 1200 1500 Min: 920.59 / Avg: 934 / Max: 951.73 Min: 1269.25 / Avg: 1277.15 / Max: 1288.21 Min: 1965.15 / Avg: 1971.5 / Max: 1979 Min: 1983.12 / Avg: 2006.81 / Max: 2020.44 Min: 1876.61 / Avg: 1885.98 / Max: 1896.4 Min: 1988.95 / Avg: 1992.76 / Max: 1995.4 Min: 1880.37 / Avg: 1884.31 / Max: 1890.99 Min: 1643.54 / Avg: 1647.78 / Max: 1651.27 Min: 1420.09 / Avg: 1422.07 / Max: 1424.4 Min: 1077.67 / Avg: 1083.36 / Max: 1087.08 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
rav1e Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 10 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.767 1.534 2.301 3.068 3.835 SE +/- 0.017, N = 3 SE +/- 0.010, N = 3 SE +/- 0.009, N = 3 SE +/- 0.009, N = 3 SE +/- 0.008, N = 3 SE +/- 0.005, N = 3 SE +/- 0.009, N = 3 SE +/- 0.006, N = 3 SE +/- 0.005, N = 3 SE +/- 0.004, N = 3 SE +/- 0.009, N = 3 SE +/- 0.008, N = 3 SE +/- 0.008, N = 3 SE +/- 0.006, N = 3 3.365 3.409 2.970 2.926 2.894 2.929 3.004 2.937 2.929 2.967 2.926 2.838 2.861 2.822
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better rav1e 0.4 Speed: 10 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0135 0.027 0.0405 0.054 0.0675 0.04 0.06 0.04 0.04 0.04 0.05 0.06 0.04 0.05 0.06 0.05 0.06 0.06 0.06
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 10 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 3.35 / Avg: 3.36 / Max: 3.4 Min: 3.39 / Avg: 3.41 / Max: 3.42 Min: 2.95 / Avg: 2.97 / Max: 2.98 Min: 2.91 / Avg: 2.93 / Max: 2.94 Min: 2.88 / Avg: 2.89 / Max: 2.9 Min: 2.92 / Avg: 2.93 / Max: 2.94 Min: 2.99 / Avg: 3 / Max: 3.02 Min: 2.93 / Avg: 2.94 / Max: 2.94 Min: 2.92 / Avg: 2.93 / Max: 2.94 Min: 2.96 / Avg: 2.97 / Max: 2.97 Min: 2.92 / Avg: 2.93 / Max: 2.94 Min: 2.82 / Avg: 2.84 / Max: 2.85 Min: 2.85 / Avg: 2.86 / Max: 2.87 Min: 2.81 / Avg: 2.82 / Max: 2.83
Zstd Compression This test measures the time needed to compress a sample file (an Ubuntu ISO) using Zstd compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 20.53, N = 3 SE +/- 4.75, N = 3 SE +/- 33.88, N = 3 SE +/- 8.45, N = 3 SE +/- 34.97, N = 3 SE +/- 25.58, N = 3 SE +/- 4.93, N = 3 SE +/- 14.29, N = 3 SE +/- 18.74, N = 3 SE +/- 21.24, N = 3 SE +/- 34.67, N = 3 SE +/- 3.19, N = 3 SE +/- 6.17, N = 3 SE +/- 5.69, N = 3 7899.7 6573.9 8248.3 8287.7 8499.2 8172.0 7885.0 8476.3 7903.6 8033.0 7699.4 6706.2 6498.4 5123.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Zstd Compression 1.4.5 Compression Level: 3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 86.38 96.70 94.05 98.16 102.10 108.55 121.69 100.70 121.39 124.95 124.18 123.70 122.04 101.96
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1500 3000 4500 6000 7500 Min: 7860.7 / Avg: 7899.7 / Max: 7930.3 Min: 6567.3 / Avg: 6573.87 / Max: 6583.1 Min: 8204.5 / Avg: 8248.33 / Max: 8315 Min: 8277.2 / Avg: 8287.67 / Max: 8304.4 Min: 8460.8 / Avg: 8499.17 / Max: 8569 Min: 8129 / Avg: 8172.03 / Max: 8217.5 Min: 7875.6 / Avg: 7885 / Max: 7892.3 Min: 8449.2 / Avg: 8476.3 / Max: 8497.7 Min: 7870.9 / Avg: 7903.63 / Max: 7935.8 Min: 7990.6 / Avg: 8033.03 / Max: 8056 Min: 7651 / Avg: 7699.4 / Max: 7766.6 Min: 6700.4 / Avg: 6706.17 / Max: 6711.4 Min: 6486.1 / Avg: 6498.4 / Max: 6505.4 Min: 5117.2 / Avg: 5123.23 / Max: 5134.6 1. (CC) gcc options: -O3 -pthread -lz -llzma
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: CPU Stress EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4K 8K 12K 16K 20K SE +/- 1.01, N = 3 SE +/- 4.39, N = 3 SE +/- 24.42, N = 3 SE +/- 45.79, N = 3 SE +/- 23.53, N = 3 SE +/- 39.53, N = 3 SE +/- 14.79, N = 3 SE +/- 8.85, N = 3 SE +/- 3.25, N = 3 SE +/- 1.11, N = 3 SE +/- 4.51, N = 3 SE +/- 5.85, N = 3 SE +/- 8.87, N = 3 6629.06 3306.39 19926.61 20174.76 15958.16 11234.60 10901.47 11121.59 8541.12 5622.82 5463.45 4078.21 2708.90 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: CPU Stress EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 47.44 41.61 119.94 125.49 108.92 94.22 82.69 96.72 79.28 68.61 75.08 61.95 48.74
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: CPU Stress EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 6628.02 / Avg: 6629.06 / Max: 6631.08 Min: 3301.92 / Avg: 3306.39 / Max: 3315.16 Min: 19890.13 / Avg: 19926.61 / Max: 19972.98 Min: 20083.31 / Avg: 20174.76 / Max: 20224.79 Min: 15919.41 / Avg: 15958.16 / Max: 16000.66 Min: 11159.1 / Avg: 11234.6 / Max: 11292.66 Min: 10874.37 / Avg: 10901.47 / Max: 10925.3 Min: 11109.84 / Avg: 11121.59 / Max: 11138.93 Min: 8535.73 / Avg: 8541.12 / Max: 8546.95 Min: 5620.68 / Avg: 5622.82 / Max: 5624.41 Min: 5456.78 / Avg: 5463.45 / Max: 5472.04 Min: 4067.37 / Avg: 4078.21 / Max: 4087.43 Min: 2691.46 / Avg: 2708.9 / Max: 2720.48 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16K 32K 48K 64K 80K SE +/- 3.33, N = 3 SE +/- 1.67, N = 3 SE +/- 15.77, N = 3 SE +/- 4.67, N = 3 SE +/- 3.00, N = 3 SE +/- 72.25, N = 3 SE +/- 17.06, N = 3 SE +/- 9.84, N = 3 SE +/- 8.37, N = 3 SE +/- 1.00, N = 3 SE +/- 3.06, N = 3 SE +/- 12.17, N = 3 SE +/- 4.10, N = 3 SE +/- 2.31, N = 3 26345 13183 70033 73579 61119 59963 44596 42800 43703 33755 22314 21454 16224 10825 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 156.27 144.00 420.61 439.24 337.12 379.39 317.83 273.00 316.71 260.68 235.34 255.06 213.21 177.43
Result Confidence
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13K 26K 39K 52K 65K Min: 26342 / Avg: 26345.33 / Max: 26352 Min: 13181 / Avg: 13182.67 / Max: 13186 Min: 70003 / Avg: 70033.33 / Max: 70056 Min: 73574 / Avg: 73578.67 / Max: 73588 Min: 61113 / Avg: 61119 / Max: 61122 Min: 59835 / Avg: 59963.33 / Max: 60085 Min: 44563 / Avg: 44596 / Max: 44620 Min: 42782 / Avg: 42799.67 / Max: 42816 Min: 43689 / Avg: 43703.33 / Max: 43718 Min: 33753 / Avg: 33755 / Max: 33756 Min: 22310 / Avg: 22314 / Max: 22320 Min: 21436 / Avg: 21453.67 / Max: 21477 Min: 16216 / Avg: 16223.67 / Max: 16230 Min: 10821 / Avg: 10825 / Max: 10829 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
Aircrack-ng Aircrack-ng is a tool for assessing WiFi/WLAN network security. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org k/s, More Is Better Aircrack-ng 1.5.2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K SE +/- 28.27, N = 3 SE +/- 28.30, N = 3 SE +/- 31.81, N = 3 SE +/- 32.87, N = 3 SE +/- 45.55, N = 3 SE +/- 87.07, N = 3 SE +/- 28.55, N = 3 SE +/- 7.45, N = 3 SE +/- 15.49, N = 3 SE +/- 16.32, N = 3 SE +/- 34.38, N = 3 SE +/- 30.64, N = 3 SE +/- 28.65, N = 3 SE +/- 6.91, N = 3 56973.85 28528.34 143169.12 143292.79 120520.68 119683.65 94237.65 85303.35 87391.79 71372.79 48284.83 46017.20 35086.95 23406.40 1. (CXX) g++ options: -O3 -fvisibility=hidden -masm=intel -fcommon -rdynamic -lpthread -lz -lcrypto -lhwloc -ldl -lm -pthread
k/s Per Watt
OpenBenchmarking.org k/s Per Watt, More Is Better Aircrack-ng 1.5.2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 275.97 262.96 798.57 819.37 669.52 725.65 584.98 551.30 628.13 481.81 426.14 473.01 387.56 334.92
Result Confidence
OpenBenchmarking.org k/s, More Is Better Aircrack-ng 1.5.2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K Min: 56938.61 / Avg: 56973.85 / Max: 57029.76 Min: 28484.88 / Avg: 28528.34 / Max: 28581.47 Min: 143136.28 / Avg: 143169.12 / Max: 143232.73 Min: 143233.45 / Avg: 143292.79 / Max: 143346.95 Min: 120433.36 / Avg: 120520.68 / Max: 120586.84 Min: 119543.75 / Avg: 119683.65 / Max: 119843.41 Min: 94189.9 / Avg: 94237.65 / Max: 94288.64 Min: 85294.1 / Avg: 85303.35 / Max: 85318.1 Min: 87370.61 / Avg: 87391.79 / Max: 87421.95 Min: 71350.26 / Avg: 71372.79 / Max: 71404.51 Min: 48218.5 / Avg: 48284.83 / Max: 48333.69 Min: 45958.39 / Avg: 46017.2 / Max: 46061.5 Min: 35057.87 / Avg: 35086.95 / Max: 35144.26 Min: 23398.23 / Avg: 23406.4 / Max: 23420.13 1. (CXX) g++ options: -O3 -fvisibility=hidden -masm=intel -fcommon -rdynamic -lpthread -lz -lcrypto -lhwloc -ldl -lm -pthread
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Crypto EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K SE +/- 2.53, N = 3 SE +/- 2.21, N = 3 SE +/- 5.99, N = 3 SE +/- 2.16, N = 3 SE +/- 6.05, N = 3 SE +/- 2.98, N = 3 SE +/- 2.43, N = 3 SE +/- 6.32, N = 3 SE +/- 0.38, N = 3 SE +/- 5.77, N = 3 SE +/- 0.60, N = 3 SE +/- 0.42, N = 3 SE +/- 0.98, N = 3 4580.08 2288.33 11996.59 12091.56 9992.15 7720.58 7105.73 7273.71 5847.70 3871.57 3760.66 2821.19 1878.74 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: Crypto EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 24.38 23.03 68.08 70.11 61.88 49.49 45.60 52.91 40.81 37.06 40.39 33.86 28.52
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Crypto EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K Min: 4577.39 / Avg: 4580.08 / Max: 4585.14 Min: 2285.96 / Avg: 2288.33 / Max: 2292.74 Min: 11984.94 / Avg: 11996.59 / Max: 12004.82 Min: 12087.35 / Avg: 12091.56 / Max: 12094.48 Min: 9980.12 / Avg: 9992.15 / Max: 9999.28 Min: 7715.51 / Avg: 7720.58 / Max: 7725.83 Min: 7100.89 / Avg: 7105.73 / Max: 7108.61 Min: 7261.15 / Avg: 7273.71 / Max: 7281.24 Min: 5846.99 / Avg: 5847.7 / Max: 5848.3 Min: 3864.76 / Avg: 3871.57 / Max: 3883.04 Min: 3759.79 / Avg: 3760.66 / Max: 3761.8 Min: 2820.54 / Avg: 2821.19 / Max: 2821.99 Min: 1877.16 / Avg: 1878.74 / Max: 1880.55 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Context Switching EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4M 8M 12M 16M 20M SE +/- 30731.73, N = 3 SE +/- 6120.44, N = 3 SE +/- 235328.27, N = 3 SE +/- 163606.09, N = 3 SE +/- 29075.41, N = 3 SE +/- 65922.96, N = 3 SE +/- 55451.20, N = 3 SE +/- 19374.10, N = 3 SE +/- 28222.94, N = 3 SE +/- 35625.01, N = 3 SE +/- 32954.83, N = 3 SE +/- 15626.15, N = 3 SE +/- 9437.39, N = 3 8225794.40 4265815.03 20917619.76 20981193.88 17727598.84 13858394.61 12627261.93 12896969.64 10443329.12 7003110.18 6612368.67 5095560.16 3438510.35 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: Context Switching EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K 41954.82 40925.63 117242.06 120719.38 107976.93 86092.82 80883.70 93564.11 70344.07 63978.31 68963.86 58318.22 50370.93
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Context Switching EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4M 8M 12M 16M 20M Min: 8193174.36 / Avg: 8225794.4 / Max: 8287218.36 Min: 4253702.33 / Avg: 4265815.03 / Max: 4273401.47 Min: 20467620.04 / Avg: 20917619.76 / Max: 21262048.98 Min: 20708784.06 / Avg: 20981193.88 / Max: 21274387.73 Min: 17694612.55 / Avg: 17727598.84 / Max: 17785565.68 Min: 13729117.41 / Avg: 13858394.61 / Max: 13945462.29 Min: 12516504.31 / Avg: 12627261.93 / Max: 12687546.78 Min: 12861868.91 / Avg: 12896969.64 / Max: 12928733.42 Min: 10405966.38 / Avg: 10443329.12 / Max: 10498652.3 Min: 6942183.09 / Avg: 7003110.18 / Max: 7065563.44 Min: 6549932.25 / Avg: 6612368.67 / Max: 6661871.66 Min: 5064455.01 / Avg: 5095560.16 / Max: 5113736.06 Min: 3425266.43 / Avg: 3438510.35 / Max: 3456778.84 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Matrix Math EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40K 80K 120K 160K 200K SE +/- 417.25, N = 3 SE +/- 382.53, N = 3 SE +/- 558.83, N = 3 SE +/- 1795.92, N = 3 SE +/- 457.19, N = 3 SE +/- 157.32, N = 3 SE +/- 270.71, N = 3 SE +/- 1288.21, N = 3 SE +/- 377.08, N = 3 SE +/- 123.36, N = 3 SE +/- 294.16, N = 3 SE +/- 183.06, N = 3 SE +/- 186.09, N = 3 76923.39 38378.08 179226.24 179496.78 152001.95 122762.96 110766.79 112205.01 92207.07 63930.72 59431.03 46780.94 31690.60 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: Matrix Math EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 348.06 324.00 987.36 1022.61 911.37 751.29 717.75 791.90 638.73 545.34 617.90 484.49 412.61
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Matrix Math EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K Min: 76194.59 / Avg: 76923.39 / Max: 77639.82 Min: 37633.18 / Avg: 38378.08 / Max: 38901.65 Min: 178344.47 / Avg: 179226.24 / Max: 180261.88 Min: 175929.7 / Avg: 179496.78 / Max: 181644.9 Min: 151506.09 / Avg: 152001.95 / Max: 152915.2 Min: 122497.64 / Avg: 122762.96 / Max: 123042.09 Min: 110323.78 / Avg: 110766.79 / Max: 111257.86 Min: 109640.44 / Avg: 112205.01 / Max: 113701 Min: 91633.53 / Avg: 92207.07 / Max: 92917.94 Min: 63684.02 / Avg: 63930.72 / Max: 64057.16 Min: 59024.95 / Avg: 59431.03 / Max: 60002.73 Min: 46417.91 / Avg: 46780.94 / Max: 47003.53 Min: 31367.04 / Avg: 31690.6 / Max: 32011.65 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Result
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Vector Math EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90K 180K 270K 360K 450K SE +/- 3.54, N = 3 SE +/- 4.12, N = 3 SE +/- 205.94, N = 3 SE +/- 137.59, N = 3 SE +/- 189.45, N = 3 SE +/- 32.95, N = 3 SE +/- 47.74, N = 3 SE +/- 66.32, N = 3 SE +/- 100.32, N = 3 SE +/- 3.30, N = 3 SE +/- 4.86, N = 3 SE +/- 1.61, N = 3 SE +/- 4.97, N = 3 142756.35 71366.17 397326.45 400773.38 328019.70 241722.09 232455.68 237596.66 182716.39 120826.81 117080.86 87851.58 58583.38 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Bogo Ops/s Per Watt
OpenBenchmarking.org Bogo Ops/s Per Watt, More Is Better Stress-NG 0.11.07 Test: Vector Math EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 500 1000 1500 2000 2500 850.73 783.29 2315.41 2387.46 2071.71 1741.29 1476.94 1705.78 1407.24 1278.65 1400.36 1163.29 961.66
Result Confidence
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Vector Math EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70K 140K 210K 280K 350K Min: 142752.27 / Avg: 142756.35 / Max: 142763.4 Min: 71360.29 / Avg: 71366.17 / Max: 71374.11 Min: 397034.87 / Avg: 397326.45 / Max: 397724.18 Min: 400607.54 / Avg: 400773.38 / Max: 401046.47 Min: 327828.63 / Avg: 328019.7 / Max: 328398.59 Min: 241657.95 / Avg: 241722.09 / Max: 241767.28 Min: 232394.78 / Avg: 232455.68 / Max: 232549.81 Min: 237474.07 / Avg: 237596.66 / Max: 237701.83 Min: 182574.85 / Avg: 182716.39 / Max: 182910.32 Min: 120822.19 / Avg: 120826.81 / Max: 120833.2 Min: 117074.78 / Avg: 117080.86 / Max: 117090.47 Min: 87849.93 / Avg: 87851.58 / Max: 87854.8 Min: 58573.44 / Avg: 58583.38 / Max: 58588.42 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
Google SynthMark SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Voices, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160 320 480 640 800 SE +/- 0.59, N = 3 SE +/- 0.17, N = 3 SE +/- 0.57, N = 3 SE +/- 0.57, N = 3 SE +/- 0.70, N = 3 SE +/- 0.70, N = 3 SE +/- 0.46, N = 3 SE +/- 0.11, N = 3 SE +/- 0.32, N = 3 SE +/- 0.13, N = 3 SE +/- 0.36, N = 3 SE +/- 0.12, N = 3 SE +/- 0.27, N = 3 739.72 740.58 635.92 624.73 626.57 645.87 627.06 637.38 636.43 627.74 605.83 608.05 608.61 1. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
Voices Per Watt
OpenBenchmarking.org Voices Per Watt, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 10.89 14.59 9.49 9.48 11.25 13.83 9.49 13.51 13.61 13.11 14.84 15.33 15.47
Result Confidence
OpenBenchmarking.org Voices, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 130 260 390 520 650 Min: 738.56 / Avg: 739.72 / Max: 740.43 Min: 740.39 / Avg: 740.58 / Max: 740.92 Min: 634.78 / Avg: 635.92 / Max: 636.53 Min: 623.79 / Avg: 624.73 / Max: 625.77 Min: 625.29 / Avg: 626.57 / Max: 627.72 Min: 644.48 / Avg: 645.87 / Max: 646.66 Min: 626.15 / Avg: 627.06 / Max: 627.62 Min: 637.17 / Avg: 637.38 / Max: 637.55 Min: 635.86 / Avg: 636.43 / Max: 636.98 Min: 627.49 / Avg: 627.74 / Max: 627.91 Min: 605.12 / Avg: 605.82 / Max: 606.33 Min: 607.84 / Avg: 608.05 / Max: 608.27 Min: 608.15 / Avg: 608.61 / Max: 609.07 1. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.084885, N = 6 SE +/- 0.288514, N = 15 SE +/- 0.025502, N = 9 SE +/- 0.021316, N = 15 SE +/- 0.024850, N = 7 SE +/- 0.017804, N = 7 SE +/- 0.076268, N = 15 SE +/- 0.046959, N = 15 SE +/- 0.076520, N = 15 SE +/- 0.017882, N = 6 SE +/- 0.040278, N = 15 SE +/- 0.040723, N = 15 SE +/- 0.172857, N = 15 SE +/- 0.312883, N = 14 8.683131 9.844102 3.086435 3.100122 3.194662 3.779568 4.395850 3.621507 4.246361 4.572591 4.176971 5.758925 7.631597 10.635023 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 8.26 / Avg: 8.68 / Max: 8.8 Min: 6.2 / Avg: 9.84 / Max: 10.6 Min: 2.96 / Avg: 3.09 / Max: 3.21 Min: 2.98 / Avg: 3.1 / Max: 3.25 Min: 3.06 / Avg: 3.19 / Max: 3.27 Min: 3.72 / Avg: 3.78 / Max: 3.86 Min: 3.64 / Avg: 4.4 / Max: 4.73 Min: 3.4 / Avg: 3.62 / Max: 3.93 Min: 3.7 / Avg: 4.25 / Max: 4.54 Min: 4.53 / Avg: 4.57 / Max: 4.63 Min: 4 / Avg: 4.18 / Max: 4.54 Min: 5.51 / Avg: 5.76 / Max: 6.05 Min: 5.38 / Avg: 7.63 / Max: 8.15 Min: 6.84 / Avg: 10.64 / Max: 11.74 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Darmstadt Automotive Parallel Heterogeneous Suite DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 SE +/- 5.28, N = 3 SE +/- 2.91, N = 3 SE +/- 3.26, N = 3 SE +/- 7.17, N = 3 SE +/- 1.66, N = 3 SE +/- 3.34, N = 3 SE +/- 3.23, N = 3 SE +/- 1.89, N = 3 SE +/- 3.41, N = 3 SE +/- 2.49, N = 3 SE +/- 1.97, N = 3 SE +/- 3.07, N = 3 SE +/- 3.25, N = 3 SE +/- 1.85, N = 3 773.86 967.37 861.35 897.59 903.30 895.52 908.46 895.33 898.84 899.45 898.21 858.54 864.62 861.07 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Test Cases Per Minute Per Watt
OpenBenchmarking.org Test Cases Per Minute Per Watt, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 9.53 16.68 10.75 11.61 12.13 13.37 16.61 11.94 16.03 16.53 16.79 18.48 19.36 19.69
Result Confidence
OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 763.33 / Avg: 773.86 / Max: 779.83 Min: 961.58 / Avg: 967.37 / Max: 970.7 Min: 854.83 / Avg: 861.35 / Max: 864.7 Min: 885.88 / Avg: 897.59 / Max: 910.61 Min: 900.01 / Avg: 903.3 / Max: 905.33 Min: 888.84 / Avg: 895.52 / Max: 899.09 Min: 902.91 / Avg: 908.46 / Max: 914.1 Min: 892.13 / Avg: 895.33 / Max: 898.66 Min: 892.02 / Avg: 898.84 / Max: 902.4 Min: 895.16 / Avg: 899.45 / Max: 903.77 Min: 894.42 / Avg: 898.21 / Max: 901.07 Min: 852.55 / Avg: 858.54 / Max: 862.73 Min: 858.29 / Avg: 864.62 / Max: 869.06 Min: 857.36 / Avg: 861.07 / Max: 863.04 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
CloverLeaf CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm.in input file (Problem 5). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 4 SE +/- 0.07, N = 4 SE +/- 0.08, N = 4 SE +/- 0.02, N = 4 SE +/- 0.03, N = 3 SE +/- 0.07, N = 4 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 22.83 36.80 14.70 14.03 13.66 15.28 18.61 15.29 19.07 20.35 25.21 45.46 48.29 69.91 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 14 28 42 56 70 Min: 22.76 / Avg: 22.83 / Max: 22.94 Min: 36.61 / Avg: 36.8 / Max: 36.9 Min: 14.59 / Avg: 14.7 / Max: 14.82 Min: 13.91 / Avg: 14.03 / Max: 14.23 Min: 13.55 / Avg: 13.66 / Max: 13.9 Min: 15.23 / Avg: 15.28 / Max: 15.32 Min: 18.57 / Avg: 18.61 / Max: 18.67 Min: 15.1 / Avg: 15.29 / Max: 15.4 Min: 19 / Avg: 19.07 / Max: 19.13 Min: 20.26 / Avg: 20.35 / Max: 20.46 Min: 25.12 / Avg: 25.21 / Max: 25.39 Min: 45.38 / Avg: 45.46 / Max: 45.52 Min: 48.21 / Avg: 48.29 / Max: 48.38 Min: 69.87 / Avg: 69.91 / Max: 69.96 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
Crafty This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2M 4M 6M 8M 10M SE +/- 3201.01, N = 3 SE +/- 9056.97, N = 3 SE +/- 4763.31, N = 3 SE +/- 6430.53, N = 3 SE +/- 15133.11, N = 3 SE +/- 15198.80, N = 3 SE +/- 8887.25, N = 3 SE +/- 2921.91, N = 3 SE +/- 5954.89, N = 3 SE +/- 16881.14, N = 3 SE +/- 18409.69, N = 3 SE +/- 4235.69, N = 3 SE +/- 8448.82, N = 3 7940047 7939883 6899948 6753902 6674881 6983945 6545211 6804305 6880760 6787692 6547690 6551351 6414609 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better Crafty 25.2 Elapsed Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K 114628.80 150336.84 99938.47 100655.93 116329.09 144587.96 97701.58 139718.79 142769.17 139072.47 156016.10 159454.66 158031.23
Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.4M 2.8M 4.2M 5.6M 7M Min: 7934298 / Avg: 7940047 / Max: 7945361 Min: 7924088 / Avg: 7939883 / Max: 7955460 Min: 6892591 / Avg: 6899948 / Max: 6908868 Min: 6743220 / Avg: 6753902.33 / Max: 6765446 Min: 6646698 / Avg: 6674880.67 / Max: 6698529 Min: 6962515 / Avg: 6983944.67 / Max: 7013330 Min: 6532471 / Avg: 6545211 / Max: 6562315 Min: 6798906 / Avg: 6804305.33 / Max: 6808941 Min: 6872751 / Avg: 6880759.67 / Max: 6892398 Min: 6763111 / Avg: 6787691.67 / Max: 6820026 Min: 6524614 / Avg: 6547690 / Max: 6584075 Min: 6543694 / Avg: 6551351.33 / Max: 6558318 Min: 6399116 / Avg: 6414609 / Max: 6428197 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
POV-Ray This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 4 SE +/- 0.03, N = 4 SE +/- 0.04, N = 4 SE +/- 0.04, N = 4 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.24, N = 3 24.18 45.47 11.50 11.50 13.27 13.27 15.94 17.67 17.14 20.46 28.70 30.51 38.41 55.78 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lSM -lICE -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11 22 33 44 55 Min: 24.15 / Avg: 24.18 / Max: 24.23 Min: 45.44 / Avg: 45.47 / Max: 45.5 Min: 11.45 / Avg: 11.5 / Max: 11.55 Min: 11.4 / Avg: 11.5 / Max: 11.56 Min: 13.19 / Avg: 13.27 / Max: 13.35 Min: 13.17 / Avg: 13.27 / Max: 13.35 Min: 15.92 / Avg: 15.94 / Max: 15.96 Min: 17.65 / Avg: 17.67 / Max: 17.7 Min: 17.12 / Avg: 17.14 / Max: 17.17 Min: 20.41 / Avg: 20.46 / Max: 20.56 Min: 28.69 / Avg: 28.7 / Max: 28.72 Min: 30.42 / Avg: 30.51 / Max: 30.57 Min: 38.36 / Avg: 38.41 / Max: 38.49 Min: 55.44 / Avg: 55.78 / Max: 56.24 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lSM -lICE -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: django_template EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 SE +/- 0.23, N = 3 SE +/- 0.45, N = 3 SE +/- 0.09, N = 3 SE +/- 0.19, N = 3 SE +/- 0.37, N = 3 SE +/- 0.52, N = 3 SE +/- 0.22, N = 3 SE +/- 0.26, N = 3 SE +/- 0.37, N = 3 SE +/- 0.03, N = 3 SE +/- 0.44, N = 3 SE +/- 0.27, N = 3 SE +/- 0.15, N = 3 SE +/- 0.40, N = 3 50.4 49.1 57.5 58.6 58.4 58.1 56.7 58.3 56.9 57.2 57.6 59.1 60.0 59.5
Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: django_template EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 Min: 50.2 / Avg: 50.43 / Max: 50.9 Min: 48.2 / Avg: 49.07 / Max: 49.7 Min: 57.3 / Avg: 57.47 / Max: 57.6 Min: 58.2 / Avg: 58.57 / Max: 58.8 Min: 57.9 / Avg: 58.37 / Max: 59.1 Min: 57.2 / Avg: 58.1 / Max: 59 Min: 56.4 / Avg: 56.67 / Max: 57.1 Min: 57.8 / Avg: 58.27 / Max: 58.7 Min: 56.2 / Avg: 56.93 / Max: 57.4 Min: 57.2 / Avg: 57.23 / Max: 57.3 Min: 56.8 / Avg: 57.63 / Max: 58.3 Min: 58.6 / Avg: 59.13 / Max: 59.4 Min: 59.8 / Avg: 60.03 / Max: 60.3 Min: 58.7 / Avg: 59.5 / Max: 59.9
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1100 2200 3300 4400 5500 SE +/- 17.61, N = 4 SE +/- 13.18, N = 4 SE +/- 23.53, N = 20 SE +/- 22.90, N = 5 SE +/- 25.46, N = 5 SE +/- 30.98, N = 5 SE +/- 38.48, N = 5 SE +/- 31.58, N = 20 SE +/- 36.08, N = 5 SE +/- 23.38, N = 5 SE +/- 23.50, N = 13 SE +/- 35.80, N = 6 SE +/- 47.23, N = 4 3288 4095 3626 3624 3550 3304 3497 3358 3310 3526 3492 3757 5021
Result Confidence
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 Min: 3254 / Avg: 3287.75 / Max: 3330 Min: 4078 / Avg: 4095 / Max: 4134 Min: 3427 / Avg: 3625.95 / Max: 3913 Min: 3562 / Avg: 3624.4 / Max: 3691 Min: 3490 / Avg: 3550 / Max: 3631 Min: 3195 / Avg: 3303.6 / Max: 3360 Min: 3410 / Avg: 3496.6 / Max: 3597 Min: 3220 / Avg: 3358.3 / Max: 3805 Min: 3228 / Avg: 3310.2 / Max: 3400 Min: 3448 / Avg: 3526.2 / Max: 3591 Min: 3310 / Avg: 3491.92 / Max: 3632 Min: 3621 / Avg: 3757.17 / Max: 3836 Min: 4937 / Avg: 5020.5 / Max: 5136
XZ Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using XZ compression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.19, N = 9 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 21.43 29.61 22.05 22.05 22.01 22.54 22.30 22.09 22.42 22.40 23.15 24.90 25.66 34.51 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 21.32 / Avg: 21.43 / Max: 21.65 Min: 29.47 / Avg: 29.61 / Max: 29.69 Min: 21.86 / Avg: 22.05 / Max: 22.28 Min: 22 / Avg: 22.05 / Max: 22.15 Min: 21.89 / Avg: 22.01 / Max: 22.17 Min: 22.49 / Avg: 22.54 / Max: 22.64 Min: 22.23 / Avg: 22.3 / Max: 22.43 Min: 22.03 / Avg: 22.09 / Max: 22.22 Min: 22.33 / Avg: 22.42 / Max: 22.59 Min: 22.31 / Avg: 22.4 / Max: 22.53 Min: 22.45 / Avg: 23.15 / Max: 23.78 Min: 24.75 / Avg: 24.9 / Max: 25.01 Min: 25.58 / Avg: 25.66 / Max: 25.79 Min: 34.46 / Avg: 34.51 / Max: 34.53 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
Tungsten Renderer Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Water Caustic EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.07, N = 3 SE +/- 0.22, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.25, N = 3 SE +/- 0.41, N = 3 24.42 28.10 21.97 22.13 22.55 22.65 22.64 23.55 23.11 23.87 26.02 26.79 29.95 33.92 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Water Caustic EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 24.31 / Avg: 24.42 / Max: 24.54 Min: 27.88 / Avg: 28.1 / Max: 28.53 Min: 21.94 / Avg: 21.97 / Max: 21.99 Min: 22.09 / Avg: 22.13 / Max: 22.15 Min: 22.48 / Avg: 22.55 / Max: 22.63 Min: 22.58 / Avg: 22.65 / Max: 22.7 Min: 22.58 / Avg: 22.64 / Max: 22.73 Min: 23.54 / Avg: 23.55 / Max: 23.58 Min: 23.07 / Avg: 23.11 / Max: 23.16 Min: 23.81 / Avg: 23.87 / Max: 23.93 Min: 25.95 / Avg: 26.02 / Max: 26.08 Min: 26.76 / Avg: 26.79 / Max: 26.84 Min: 29.5 / Avg: 29.95 / Max: 30.34 Min: 33.49 / Avg: 33.92 / Max: 34.73 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 SE +/- 5.46, N = 3 SE +/- 8.33, N = 3 SE +/- 6.84, N = 3 SE +/- 9.54, N = 3 SE +/- 7.55, N = 3 SE +/- 9.35, N = 3 SE +/- 1.20, N = 3 SE +/- 7.33, N = 3 SE +/- 10.82, N = 3 SE +/- 1.20, N = 3 SE +/- 1.86, N = 3 SE +/- 8.01, N = 3 SE +/- 9.68, N = 3 999 998 1166 1184 1184 1178 1137 1180 1173 1159 1172 1221 1208 1213
Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 992 / Avg: 999.33 / Max: 1010 Min: 990 / Avg: 998.33 / Max: 1015 Min: 1159 / Avg: 1166.33 / Max: 1180 Min: 1165 / Avg: 1184 / Max: 1195 Min: 1175 / Avg: 1184 / Max: 1199 Min: 1168 / Avg: 1178.33 / Max: 1197 Min: 1135 / Avg: 1137.33 / Max: 1139 Min: 1173 / Avg: 1180.33 / Max: 1195 Min: 1152 / Avg: 1173 / Max: 1188 Min: 1157 / Avg: 1159.33 / Max: 1161 Min: 1168 / Avg: 1171.67 / Max: 1174 Min: 1209 / Avg: 1220.67 / Max: 1236 Min: 1202 / Avg: 1212.67 / Max: 1232
LULESH LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K SE +/- 40.17, N = 6 SE +/- 12.34, N = 6 SE +/- 146.93, N = 12 SE +/- 142.13, N = 3 SE +/- 58.13, N = 4 SE +/- 60.62, N = 4 SE +/- 157.59, N = 4 SE +/- 43.36, N = 4 SE +/- 131.70, N = 5 SE +/- 9.05, N = 6 SE +/- 8.39, N = 6 SE +/- 15.52, N = 6 SE +/- 38.61, N = 6 SE +/- 11.79, N = 6 6839.17 8720.88 14612.21 14603.33 14015.66 13435.30 13099.62 13938.25 12840.64 8024.37 7954.15 6954.80 6871.93 6757.99 1. (CXX) g++ options: -O3 -fopenmp -lm -pthread -lmpi_cxx -lmpi
z/s Per Watt
OpenBenchmarking.org z/s Per Watt, More Is Better LULESH 2.0.3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 69.83 114.72 91.64 95.62 109.48 118.64 125.40 113.53 123.94 122.19 122.94 119.53 116.78 117.45
Result Confidence
OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3K 6K 9K 12K 15K Min: 6733.72 / Avg: 6839.17 / Max: 6954.28 Min: 8666.52 / Avg: 8720.88 / Max: 8753.04 Min: 13024.64 / Avg: 14612.21 / Max: 14895.69 Min: 14340.08 / Avg: 14603.33 / Max: 14827.83 Min: 13885.31 / Avg: 14015.66 / Max: 14137.76 Min: 13262.26 / Avg: 13435.3 / Max: 13545.43 Min: 12632.05 / Avg: 13099.62 / Max: 13300.62 Min: 13879.03 / Avg: 13938.25 / Max: 14067.14 Min: 12361.8 / Avg: 12840.64 / Max: 13103.71 Min: 7986.3 / Avg: 8024.37 / Max: 8053.74 Min: 7941.34 / Avg: 7954.15 / Max: 7995.56 Min: 6907.71 / Avg: 6954.8 / Max: 7005.82 Min: 6730.47 / Avg: 6871.93 / Max: 6957.51 Min: 6718.89 / Avg: 6757.99 / Max: 6804.97 1. (CXX) g++ options: -O3 -fopenmp -lm -pthread -lmpi_cxx -lmpi
Kvazaar This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Very Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 4 SE +/- 0.03, N = 4 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 26.31 15.56 36.07 39.15 37.39 35.81 35.06 32.72 33.25 26.61 22.94 21.70 17.83 12.98 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Very Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0608 0.1216 0.1824 0.2432 0.304 0.13 0.13 0.25 0.27 0.24 0.26 0.25 0.23 0.26 0.21 0.21 0.24 0.19 0.17
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Very Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 26.29 / Avg: 26.31 / Max: 26.33 Min: 15.54 / Avg: 15.56 / Max: 15.61 Min: 36 / Avg: 36.07 / Max: 36.11 Min: 39.12 / Avg: 39.15 / Max: 39.19 Min: 37.31 / Avg: 37.39 / Max: 37.44 Min: 35.77 / Avg: 35.81 / Max: 35.84 Min: 35.04 / Avg: 35.06 / Max: 35.09 Min: 32.69 / Avg: 32.72 / Max: 32.73 Min: 33.24 / Avg: 33.25 / Max: 33.26 Min: 26.56 / Avg: 26.61 / Max: 26.7 Min: 22.92 / Avg: 22.94 / Max: 22.97 Min: 21.66 / Avg: 21.7 / Max: 21.75 Min: 17.82 / Avg: 17.83 / Max: 17.84 Min: 12.97 / Avg: 12.98 / Max: 12.99 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Timed Apache Compilation This test times how long it takes to build the Apache HTTPD web server. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Apache Compilation 2.4.41 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 21.62 23.00 24.56 24.92 24.93 24.89 24.20 25.02 24.43 24.53 25.02 25.65 26.02 28.08
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Apache Compilation 2.4.41 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 21.59 / Avg: 21.62 / Max: 21.66 Min: 22.98 / Avg: 23 / Max: 23.04 Min: 24.54 / Avg: 24.56 / Max: 24.58 Min: 24.89 / Avg: 24.92 / Max: 24.95 Min: 24.92 / Avg: 24.93 / Max: 24.95 Min: 24.85 / Avg: 24.89 / Max: 24.92 Min: 24.17 / Avg: 24.2 / Max: 24.23 Min: 25 / Avg: 25.02 / Max: 25.06 Min: 24.41 / Avg: 24.43 / Max: 24.46 Min: 24.51 / Avg: 24.53 / Max: 24.56 Min: 24.97 / Avg: 25.02 / Max: 25.06 Min: 25.6 / Avg: 25.65 / Max: 25.68 Min: 26.01 / Avg: 26.02 / Max: 26.03 Min: 28.05 / Avg: 28.08 / Max: 28.12
Stream-Dynamic This is an open-source AMD modified copy of the Stream memory benchmark catered towards running the RAM benchmark on systems with the AMD Optimizing C/C++ Compiler (AOCC) among other by-default optimizations aiming for an easy and standardized deployment. This test profile though will attempt to fall-back to GCC / Clang for systems lacking AOCC, otherwise there is the existing "stream" test profile. Learn more via the OpenBenchmarking.org test page.
Swet Swet is a synthetic CPU/RAM benchmark, includes multi-processor test cases. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Operations Per Second, More Is Better Swet 1.5.16 Average EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 150M 300M 450M 600M 750M SE +/- 2209414.80, N = 3 SE +/- 2325206.90, N = 3 SE +/- 2469610.76, N = 3 SE +/- 5706451.14, N = 3 SE +/- 807930.96, N = 3 SE +/- 8506660.76, N = 3 SE +/- 1400113.05, N = 3 SE +/- 4052005.46, N = 3 SE +/- 2164773.96, N = 3 SE +/- 3828695.03, N = 3 SE +/- 5177314.46, N = 3 SE +/- 2920022.24, N = 3 SE +/- 6957648.38, N = 3 685188810 686395419 602434885 602203842 605644364 612218425 582684333 597361197 600065572 598589228 571825580 581194588 561816772 1. (CC) gcc options: -lm -lpthread -lcurses -lrt
Operations Per Second Per Watt
OpenBenchmarking.org Operations Per Second Per Watt, More Is Better Swet 1.5.16 Average EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3M 6M 9M 12M 15M 10268451.78 13794532.24 9098672.56 9228207.39 11032151.74 13255560.12 9008595.44 12873172.87 13079157.08 12821906.80 14256393.92 14898341.79 14546659.15
Result Confidence
OpenBenchmarking.org Operations Per Second, More Is Better Swet 1.5.16 Average EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 120M 240M 360M 480M 600M Min: 680933266 / Avg: 685188809.67 / Max: 688347262 Min: 682854921 / Avg: 686395419.33 / Max: 690776878 Min: 597611652 / Avg: 602434885.33 / Max: 605768048 Min: 590802604 / Avg: 602203842.33 / Max: 608351202 Min: 604227306 / Avg: 605644364.33 / Max: 607025364 Min: 596416785 / Avg: 612218425.33 / Max: 625580102 Min: 579892857 / Avg: 582684332.67 / Max: 584271636 Min: 589487365 / Avg: 597361197.33 / Max: 602958932 Min: 595836568 / Avg: 600065571.67 / Max: 602983432 Min: 593041731 / Avg: 598589227.67 / Max: 605934155 Min: 566389680 / Avg: 571825580.33 / Max: 582175830 Min: 575619977 / Avg: 581194588 / Max: 585489333 Min: 547924209 / Avg: 561816771.67 / Max: 569451631 1. (CC) gcc options: -lm -lpthread -lcurses -lrt
Pennant Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: leblancbig EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.007758, N = 4 SE +/- 0.037928, N = 3 SE +/- 0.058701, N = 6 SE +/- 0.054901, N = 6 SE +/- 0.095577, N = 5 SE +/- 0.043495, N = 4 SE +/- 0.014397, N = 5 SE +/- 0.026648, N = 4 SE +/- 0.019214, N = 3 SE +/- 0.036573, N = 3 SE +/- 0.019691, N = 3 SE +/- 0.292336, N = 3 SE +/- 0.039126, N = 3 16.139290 34.580030 7.033982 6.995321 8.988538 14.164820 10.885760 14.781570 16.653750 21.951880 28.993920 33.808520 45.154600 1. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
Result Confidence
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: leblancbig EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 16.12 / Avg: 16.14 / Max: 16.16 Min: 34.52 / Avg: 34.58 / Max: 34.65 Min: 6.86 / Avg: 7.03 / Max: 7.24 Min: 6.85 / Avg: 7 / Max: 7.2 Min: 8.8 / Avg: 8.99 / Max: 9.33 Min: 14.12 / Avg: 14.16 / Max: 14.3 Min: 10.85 / Avg: 10.89 / Max: 10.92 Min: 14.73 / Avg: 14.78 / Max: 14.84 Min: 16.63 / Avg: 16.65 / Max: 16.69 Min: 21.9 / Avg: 21.95 / Max: 22.02 Min: 28.96 / Avg: 28.99 / Max: 29.03 Min: 33.47 / Avg: 33.81 / Max: 34.39 Min: 45.1 / Avg: 45.15 / Max: 45.23 1. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
Algebraic Multi-Grid Benchmark AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200M 400M 600M 800M 1000M SE +/- 3149876.45, N = 4 SE +/- 322962.48, N = 6 SE +/- 507394.80, N = 3 SE +/- 229126.62, N = 3 SE +/- 989060.38, N = 3 SE +/- 142607.92, N = 3 SE +/- 422202.49, N = 3 SE +/- 998206.04, N = 3 SE +/- 643499.89, N = 3 SE +/- 405072.19, N = 4 SE +/- 314598.13, N = 3 SE +/- 252656.85, N = 4 SE +/- 295157.17, N = 5 643180925 809844583 878375333 883735200 856893400 774329467 909532667 774304800 778459433 788266875 455760633 457855350 449916460 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi
Figure Of Merit Per Watt
OpenBenchmarking.org Figure Of Merit Per Watt, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2M 4M 6M 8M 10M 4682548.38 9577991.03 5138262.51 5265803.49 5919717.78 6697956.10 6228384.62 6513520.25 7131421.00 8667240.71 5939960.67 6485573.10 7366263.78
Result Confidence
OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 160M 320M 480M 640M 800M Min: 634035200 / Avg: 643180925 / Max: 647428500 Min: 808702900 / Avg: 809844583.33 / Max: 810642700 Min: 877829000 / Avg: 878375333.33 / Max: 879389100 Min: 883287900 / Avg: 883735200 / Max: 884045100 Min: 854937100 / Avg: 856893400 / Max: 858125300 Min: 774119000 / Avg: 774329466.67 / Max: 774601400 Min: 908914700 / Avg: 909532666.67 / Max: 910340000 Min: 772483400 / Avg: 774304800 / Max: 775923400 Min: 777565100 / Avg: 778459433.33 / Max: 779708100 Min: 787578900 / Avg: 788266875 / Max: 789313600 Min: 455133300 / Avg: 455760633.33 / Max: 456116200 Min: 457151100 / Avg: 457855350 / Max: 458299300 Min: 448913300 / Avg: 449916460 / Max: 450629100 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi
librsvg RSVG/librsvg is an SVG vector graphics library. This test profile times how long it takes to complete various operations by rsvg-convert. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better librsvg Operation: SVG Files To PNG EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 20.48 20.03 24.34 24.73 24.53 23.35 24.09 23.65 23.66 23.78 24.46 24.09 25.02 1. rsvg-convert version 2.48.9
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better librsvg Operation: SVG Files To PNG EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 20.42 / Avg: 20.48 / Max: 20.58 Min: 19.96 / Avg: 20.03 / Max: 20.1 Min: 24.29 / Avg: 24.34 / Max: 24.4 Min: 24.63 / Avg: 24.73 / Max: 24.84 Min: 24.43 / Avg: 24.53 / Max: 24.6 Min: 23.26 / Avg: 23.35 / Max: 23.49 Min: 24 / Avg: 24.09 / Max: 24.2 Min: 23.57 / Avg: 23.65 / Max: 23.77 Min: 23.49 / Avg: 23.66 / Max: 23.78 Min: 23.65 / Avg: 23.78 / Max: 23.89 Min: 24.41 / Avg: 24.46 / Max: 24.53 Min: 23.94 / Avg: 24.09 / Max: 24.24 Min: 24.97 / Avg: 25.02 / Max: 25.1 1. rsvg-convert version 2.48.9
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: AES-256 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1100 2200 3300 4400 5500 SE +/- 1.30, N = 3 SE +/- 1.69, N = 3 SE +/- 1.00, N = 3 SE +/- 0.85, N = 3 SE +/- 3.58, N = 3 SE +/- 3.88, N = 3 SE +/- 2.98, N = 3 SE +/- 1.33, N = 3 SE +/- 3.31, N = 3 SE +/- 0.83, N = 3 SE +/- 0.26, N = 3 SE +/- 0.80, N = 3 SE +/- 7.02, N = 3 SE +/- 13.87, N = 3 5226.75 5238.76 4499.50 4429.19 4425.92 4421.41 4561.19 4429.47 4497.77 4487.90 4430.72 4293.87 4290.85 4311.06 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better Botan 2.13.0 Test: AES-256 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 76.11 101.50 66.83 67.17 70.28 79.86 97.57 67.57 95.12 96.36 94.08 105.14 108.21 109.56
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: AES-256 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 Min: 5224.38 / Avg: 5226.75 / Max: 5228.88 Min: 5235.4 / Avg: 5238.76 / Max: 5240.69 Min: 4497.86 / Avg: 4499.5 / Max: 4501.3 Min: 4427.63 / Avg: 4429.19 / Max: 4430.55 Min: 4421.28 / Avg: 4425.91 / Max: 4432.96 Min: 4413.66 / Avg: 4421.41 / Max: 4425.33 Min: 4555.36 / Avg: 4561.19 / Max: 4565.19 Min: 4426.94 / Avg: 4429.47 / Max: 4431.46 Min: 4492.53 / Avg: 4497.77 / Max: 4503.89 Min: 4486.88 / Avg: 4487.9 / Max: 4489.54 Min: 4430.21 / Avg: 4430.72 / Max: 4431.05 Min: 4292.82 / Avg: 4293.87 / Max: 4295.45 Min: 4276.91 / Avg: 4290.85 / Max: 4299.27 Min: 4293.42 / Avg: 4311.06 / Max: 4338.43 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
TNN TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 SE +/- 0.41, N = 3 SE +/- 0.35, N = 3 SE +/- 0.62, N = 3 SE +/- 0.29, N = 3 SE +/- 0.40, N = 3 SE +/- 0.46, N = 3 SE +/- 0.10, N = 3 SE +/- 0.88, N = 3 SE +/- 0.72, N = 3 SE +/- 0.03, N = 3 SE +/- 0.94, N = 3 SE +/- 0.21, N = 3 SE +/- 0.22, N = 3 SE +/- 0.14, N = 3 288.47 286.43 333.30 342.47 341.93 344.33 333.28 344.25 336.89 336.66 339.05 350.76 347.77 347.18 MIN: 285.34 / MAX: 304.15 MIN: 283.81 / MAX: 294.29 MIN: 330.49 / MAX: 343.73 MIN: 338.7 / MAX: 360.58 MIN: 337.09 / MAX: 359.76 MIN: 340.96 / MAX: 358.42 MIN: 329.87 / MAX: 349.8 MIN: 339.59 / MAX: 367.05 MIN: 332.6 / MAX: 347.08 MIN: 333.97 / MAX: 348.51 MIN: 335.68 / MAX: 355.89 MIN: 346.27 / MAX: 363.67 MIN: 345.44 / MAX: 366.04 MIN: 343.85 / MAX: 356.42 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 287.8 / Avg: 288.47 / Max: 289.23 Min: 285.73 / Avg: 286.43 / Max: 286.81 Min: 332.62 / Avg: 333.3 / Max: 334.54 Min: 342 / Avg: 342.47 / Max: 343.01 Min: 341.17 / Avg: 341.93 / Max: 342.54 Min: 343.4 / Avg: 344.33 / Max: 344.79 Min: 333.14 / Avg: 333.28 / Max: 333.46 Min: 342.92 / Avg: 344.25 / Max: 345.93 Min: 335.5 / Avg: 336.89 / Max: 337.88 Min: 336.61 / Avg: 336.66 / Max: 336.68 Min: 337.77 / Avg: 339.05 / Max: 340.89 Min: 350.35 / Avg: 350.76 / Max: 351.03 Min: 347.4 / Avg: 347.77 / Max: 348.16 Min: 347.01 / Avg: 347.18 / Max: 347.46 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
Timed MPlayer Compilation This test times how long it takes to build the MPlayer open-source media player program. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed MPlayer Compilation 1.4 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 4 SE +/- 0.03, N = 4 SE +/- 0.04, N = 4 SE +/- 0.03, N = 4 SE +/- 0.07, N = 4 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 20.57 35.25 12.99 12.96 13.71 13.82 15.69 16.38 16.46 18.59 24.06 26.27 31.14 45.59
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed MPlayer Compilation 1.4 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 Min: 20.43 / Avg: 20.57 / Max: 20.67 Min: 35.21 / Avg: 35.25 / Max: 35.28 Min: 12.91 / Avg: 12.99 / Max: 13.09 Min: 12.87 / Avg: 12.96 / Max: 13.01 Min: 13.64 / Avg: 13.71 / Max: 13.81 Min: 13.75 / Avg: 13.82 / Max: 13.87 Min: 15.6 / Avg: 15.69 / Max: 15.9 Min: 16.35 / Avg: 16.38 / Max: 16.42 Min: 16.42 / Avg: 16.46 / Max: 16.52 Min: 18.5 / Avg: 18.59 / Max: 18.68 Min: 24.05 / Avg: 24.06 / Max: 24.07 Min: 26.23 / Avg: 26.27 / Max: 26.29 Min: 31.1 / Avg: 31.14 / Max: 31.21 Min: 45.54 / Avg: 45.59 / Max: 45.62
Basis Universal Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 4 SE +/- 0.00, N = 4 SE +/- 0.01, N = 4 SE +/- 0.00, N = 4 SE +/- 0.01, N = 3 SE +/- 0.02, N = 4 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 19.86 32.63 13.30 13.21 14.24 15.64 16.70 16.31 18.17 23.34 24.24 29.24 39.67 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 19.86 / Avg: 19.86 / Max: 19.87 Min: 32.63 / Avg: 32.63 / Max: 32.64 Min: 13.28 / Avg: 13.3 / Max: 13.33 Min: 13.2 / Avg: 13.21 / Max: 13.22 Min: 14.23 / Avg: 14.24 / Max: 14.25 Min: 15.63 / Avg: 15.64 / Max: 15.64 Min: 16.69 / Avg: 16.7 / Max: 16.71 Min: 16.28 / Avg: 16.31 / Max: 16.38 Min: 18.16 / Avg: 18.17 / Max: 18.17 Min: 23.32 / Avg: 23.34 / Max: 23.38 Min: 24.19 / Avg: 24.24 / Max: 24.27 Min: 29.23 / Avg: 29.24 / Max: 29.27 Min: 39.66 / Avg: 39.67 / Max: 39.67 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
JPEG XL The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance. Learn more via the OpenBenchmarking.org test page.
Darmstadt Automotive Parallel Heterogeneous Suite DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 SE +/- 0.51, N = 3 SE +/- 1.50, N = 3 SE +/- 1.46, N = 3 SE +/- 0.53, N = 3 SE +/- 1.18, N = 3 SE +/- 0.82, N = 3 SE +/- 0.71, N = 3 SE +/- 0.79, N = 3 SE +/- 0.42, N = 3 SE +/- 0.40, N = 3 SE +/- 0.19, N = 3 SE +/- 1.22, N = 3 SE +/- 0.49, N = 3 SE +/- 0.40, N = 3 1062.32 1039.60 954.20 966.04 963.00 956.82 983.45 949.15 974.40 962.88 943.29 928.43 919.01 876.45 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Test Cases Per Minute Per Watt
OpenBenchmarking.org Test Cases Per Minute Per Watt, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 11.43 16.03 9.72 10.48 10.78 12.12 15.32 11.25 15.04 15.18 16.01 18.12 18.67 18.67
Result Confidence
OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 1061.29 / Avg: 1062.32 / Max: 1062.84 Min: 1037.63 / Avg: 1039.6 / Max: 1042.55 Min: 951.37 / Avg: 954.2 / Max: 956.21 Min: 965.08 / Avg: 966.04 / Max: 966.9 Min: 961.75 / Avg: 963 / Max: 965.37 Min: 955.86 / Avg: 956.82 / Max: 958.45 Min: 982.72 / Avg: 983.45 / Max: 984.87 Min: 947.85 / Avg: 949.15 / Max: 950.57 Min: 973.64 / Avg: 974.4 / Max: 975.08 Min: 962.13 / Avg: 962.88 / Max: 963.5 Min: 942.97 / Avg: 943.29 / Max: 943.63 Min: 926.64 / Avg: 928.43 / Max: 930.77 Min: 918.04 / Avg: 919.01 / Max: 919.66 Min: 876.05 / Avg: 876.45 / Max: 877.26 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
AOM AV1 This is a simple test of the AOMedia AV1 encoder run on the CPU with a sample video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 4 Two-Pass EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.5648 1.1296 1.6944 2.2592 2.824 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 2.51 2.38 2.17 2.16 2.12 2.16 2.22 2.15 2.20 2.19 2.14 2.08 2.05 1.98 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 2.0 Encoder Mode: Speed 4 Two-Pass EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.009 0.018 0.027 0.036 0.045 0.03 0.04 0.03 0.03 0.03 0.03 0.04 0.03 0.04 0.04 0.04 0.04 0.04 0.04
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 4 Two-Pass EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 2.5 / Avg: 2.51 / Max: 2.51 Min: 2.37 / Avg: 2.38 / Max: 2.39 Min: 2.17 / Avg: 2.17 / Max: 2.17 Min: 2.15 / Avg: 2.16 / Max: 2.16 Min: 2.12 / Avg: 2.12 / Max: 2.13 Min: 2.15 / Avg: 2.16 / Max: 2.16 Min: 2.21 / Avg: 2.22 / Max: 2.23 Min: 2.14 / Avg: 2.15 / Max: 2.15 Min: 2.2 / Avg: 2.2 / Max: 2.2 Min: 2.19 / Avg: 2.19 / Max: 2.19 Min: 2.14 / Avg: 2.14 / Max: 2.15 Min: 2.07 / Avg: 2.08 / Max: 2.09 Min: 2.04 / Avg: 2.05 / Max: 2.06 Min: 1.97 / Avg: 1.98 / Max: 1.98 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Thorough EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 4 SE +/- 0.01, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 14.02 27.16 6.35 6.38 7.42 9.08 9.99 9.73 11.71 16.68 17.73 22.40 33.15 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Thorough EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 14 / Avg: 14.02 / Max: 14.03 Min: 27.15 / Avg: 27.16 / Max: 27.17 Min: 6.35 / Avg: 6.35 / Max: 6.36 Min: 6.36 / Avg: 6.38 / Max: 6.4 Min: 7.41 / Avg: 7.42 / Max: 7.42 Min: 9.08 / Avg: 9.08 / Max: 9.08 Min: 9.99 / Avg: 9.99 / Max: 9.99 Min: 9.73 / Avg: 9.73 / Max: 9.74 Min: 11.71 / Avg: 11.71 / Max: 11.72 Min: 16.67 / Avg: 16.68 / Max: 16.69 Min: 17.72 / Avg: 17.73 / Max: 17.74 Min: 22.39 / Avg: 22.4 / Max: 22.42 Min: 33.13 / Avg: 33.15 / Max: 33.18 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Timed ImageMagick Compilation This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.15, N = 3 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 SE +/- 0.17, N = 3 21.63 28.60 17.12 17.19 17.67 18.06 18.94 19.27 19.41 20.43 23.55 25.46 28.22 37.72
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 21.4 / Avg: 21.63 / Max: 21.91 Min: 28.39 / Avg: 28.6 / Max: 28.73 Min: 16.99 / Avg: 17.11 / Max: 17.27 Min: 16.99 / Avg: 17.19 / Max: 17.31 Min: 17.49 / Avg: 17.67 / Max: 17.79 Min: 17.88 / Avg: 18.06 / Max: 18.19 Min: 18.83 / Avg: 18.93 / Max: 19.02 Min: 19.16 / Avg: 19.27 / Max: 19.35 Min: 19.38 / Avg: 19.4 / Max: 19.45 Min: 20.3 / Avg: 20.43 / Max: 20.52 Min: 23.47 / Avg: 23.55 / Max: 23.6 Min: 25.14 / Avg: 25.46 / Max: 25.64 Min: 28.17 / Avg: 28.22 / Max: 28.26 Min: 37.4 / Avg: 37.72 / Max: 38.01
OCRMyPDF OCRMyPDF is an optical character recognition (OCR) text layer to scanned PDF files, producing new PDFs with the text now selectable/searchable/copy-paste capable. OCRMyPDF leverages the Tesseract OCR engine and is written in Python. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better OCRMyPDF 9.6.0+dfsg Processing 60 Page PDF Document EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 SE +/- 0.20, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.01, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 19.86 27.20 18.73 18.79 18.86 18.66 19.94 19.38 20.87 23.39 24.23 26.90 33.02
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better OCRMyPDF 9.6.0+dfsg Processing 60 Page PDF Document EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 19.76 / Avg: 19.86 / Max: 19.94 Min: 26.98 / Avg: 27.2 / Max: 27.33 Min: 18.39 / Avg: 18.73 / Max: 19.1 Min: 18.7 / Avg: 18.79 / Max: 18.88 Min: 18.73 / Avg: 18.85 / Max: 18.96 Min: 18.45 / Avg: 18.66 / Max: 18.82 Min: 19.92 / Avg: 19.94 / Max: 19.97 Min: 19.15 / Avg: 19.38 / Max: 19.51 Min: 20.74 / Avg: 20.87 / Max: 21.08 Min: 23.21 / Avg: 23.39 / Max: 23.54 Min: 24.11 / Avg: 24.23 / Max: 24.33 Min: 26.81 / Avg: 26.9 / Max: 26.96 Min: 32.94 / Avg: 33.02 / Max: 33.14
Scikit-Learn Scikit-learn is a Python module for machine learning Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.22.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.031, N = 5 SE +/- 0.004, N = 5 SE +/- 0.051, N = 5 SE +/- 0.006, N = 5 SE +/- 0.016, N = 5 SE +/- 0.009, N = 5 SE +/- 0.158, N = 15 SE +/- 0.004, N = 5 SE +/- 0.003, N = 5 SE +/- 0.006, N = 5 SE +/- 0.012, N = 5 SE +/- 0.010, N = 5 SE +/- 0.007, N = 5 SE +/- 0.168, N = 15 9.240 9.293 10.131 10.130 10.337 10.358 10.074 10.325 10.083 10.154 10.281 10.539 10.746 11.327
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.22.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 9.14 / Avg: 9.24 / Max: 9.33 Min: 9.28 / Avg: 9.29 / Max: 9.3 Min: 10.04 / Avg: 10.13 / Max: 10.33 Min: 10.11 / Avg: 10.13 / Max: 10.14 Min: 10.28 / Avg: 10.34 / Max: 10.37 Min: 10.34 / Avg: 10.36 / Max: 10.39 Min: 9.88 / Avg: 10.07 / Max: 12.28 Min: 10.32 / Avg: 10.33 / Max: 10.34 Min: 10.08 / Avg: 10.08 / Max: 10.09 Min: 10.13 / Avg: 10.15 / Max: 10.17 Min: 10.25 / Avg: 10.28 / Max: 10.32 Min: 10.52 / Avg: 10.54 / Max: 10.57 Min: 10.72 / Avg: 10.75 / Max: 10.76 Min: 11.13 / Avg: 11.33 / Max: 13.67
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Streamcluster EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.049, N = 4 SE +/- 0.017, N = 3 SE +/- 0.017, N = 5 SE +/- 0.030, N = 5 SE +/- 0.074, N = 15 SE +/- 0.119, N = 4 SE +/- 0.022, N = 4 SE +/- 0.016, N = 5 SE +/- 0.007, N = 4 SE +/- 0.042, N = 4 SE +/- 0.034, N = 3 SE +/- 0.159, N = 8 SE +/- 0.079, N = 3 SE +/- 0.040, N = 4 15.049 19.707 8.928 8.905 9.185 12.027 14.316 9.913 14.320 14.761 17.729 18.189 16.895 15.540 1. (CXX) g++ options: -O2 -lOpenCL
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Streamcluster EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 14.94 / Avg: 15.05 / Max: 15.15 Min: 19.68 / Avg: 19.71 / Max: 19.74 Min: 8.9 / Avg: 8.93 / Max: 8.99 Min: 8.85 / Avg: 8.91 / Max: 9.02 Min: 9.06 / Avg: 9.19 / Max: 10.22 Min: 11.69 / Avg: 12.03 / Max: 12.26 Min: 14.25 / Avg: 14.32 / Max: 14.36 Min: 9.88 / Avg: 9.91 / Max: 9.97 Min: 14.31 / Avg: 14.32 / Max: 14.34 Min: 14.65 / Avg: 14.76 / Max: 14.83 Min: 17.66 / Avg: 17.73 / Max: 17.77 Min: 17.99 / Avg: 18.19 / Max: 19.3 Min: 16.76 / Avg: 16.9 / Max: 17.03 Min: 15.47 / Avg: 15.54 / Max: 15.65 1. (CXX) g++ options: -O2 -lOpenCL
Etcpak Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: ETC1 + Dithering EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 256.42 257.01 220.80 217.45 217.50 223.49 216.79 220.76 220.71 217.50 210.72 211.03 210.93 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
Mpx/s Per Watt
OpenBenchmarking.org Mpx/s Per Watt, More Is Better Etcpak 0.7 Configuration: ETC1 + Dithering EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.2038 2.4076 3.6114 4.8152 6.019 3.74 5.05 3.28 3.28 3.92 4.78 3.32 4.65 4.70 4.57 5.14 5.28 5.35
Result Confidence
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: ETC1 + Dithering EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 256.36 / Avg: 256.42 / Max: 256.5 Min: 257 / Avg: 257.01 / Max: 257.02 Min: 220.64 / Avg: 220.8 / Max: 220.91 Min: 217.36 / Avg: 217.45 / Max: 217.6 Min: 217.48 / Avg: 217.5 / Max: 217.52 Min: 223.43 / Avg: 223.49 / Max: 223.58 Min: 216.67 / Avg: 216.79 / Max: 216.93 Min: 220.68 / Avg: 220.76 / Max: 220.9 Min: 220.68 / Avg: 220.71 / Max: 220.76 Min: 217.45 / Avg: 217.5 / Max: 217.57 Min: 210.52 / Avg: 210.72 / Max: 210.94 Min: 210.87 / Avg: 211.03 / Max: 211.13 Min: 210.74 / Avg: 210.93 / Max: 211.03 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
TNN TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 70 140 210 280 350 SE +/- 0.12, N = 3 SE +/- 0.36, N = 3 SE +/- 0.80, N = 3 SE +/- 0.83, N = 3 SE +/- 1.03, N = 3 SE +/- 1.22, N = 3 SE +/- 0.29, N = 3 SE +/- 0.26, N = 3 SE +/- 0.71, N = 3 SE +/- 0.19, N = 3 SE +/- 0.82, N = 3 SE +/- 1.44, N = 3 SE +/- 0.11, N = 3 SE +/- 0.77, N = 3 269.92 269.64 313.01 317.55 318.67 318.67 308.94 318.58 312.89 314.31 318.37 328.64 328.36 327.99 MIN: 267.18 / MAX: 272.47 MIN: 267.19 / MAX: 279.76 MIN: 310.72 / MAX: 314.66 MIN: 315.59 / MAX: 319.4 MIN: 315.68 / MAX: 321.91 MIN: 315.63 / MAX: 323.01 MIN: 306.29 / MAX: 322.41 MIN: 315.09 / MAX: 320.21 MIN: 310.9 / MAX: 315.08 MIN: 311.3 / MAX: 317.52 MIN: 315.66 / MAX: 323.53 MIN: 325.19 / MAX: 342.8 MIN: 325.02 / MAX: 339.93 MIN: 325.21 / MAX: 329.77 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 269.68 / Avg: 269.92 / Max: 270.07 Min: 269.01 / Avg: 269.64 / Max: 270.25 Min: 311.43 / Avg: 313.01 / Max: 313.95 Min: 316.01 / Avg: 317.55 / Max: 318.87 Min: 316.65 / Avg: 318.67 / Max: 319.98 Min: 316.23 / Avg: 318.67 / Max: 320.01 Min: 308.42 / Avg: 308.94 / Max: 309.44 Min: 318.08 / Avg: 318.58 / Max: 318.92 Min: 311.52 / Avg: 312.89 / Max: 313.94 Min: 313.92 / Avg: 314.31 / Max: 314.52 Min: 317.19 / Avg: 318.37 / Max: 319.96 Min: 325.78 / Avg: 328.64 / Max: 330.41 Min: 328.14 / Avg: 328.36 / Max: 328.49 Min: 326.49 / Avg: 327.99 / Max: 329.05 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.02104, N = 3 SE +/- 0.00611, N = 3 SE +/- 0.00645, N = 3 SE +/- 0.00798, N = 3 SE +/- 0.00608, N = 3 SE +/- 0.00201, N = 3 SE +/- 0.00531, N = 3 SE +/- 0.00309, N = 3 SE +/- 0.00464, N = 3 SE +/- 0.02587, N = 3 SE +/- 0.00946, N = 3 SE +/- 0.02527, N = 3 SE +/- 0.00535, N = 3 SE +/- 0.02294, N = 3 5.49605 7.23763 1.88842 1.79874 1.93042 1.99707 2.06983 2.20543 2.15194 5.92204 6.03497 6.06333 5.25331 8.67154 MIN: 5.32 MIN: 7.01 MIN: 1.8 MIN: 1.69 MIN: 1.82 MIN: 1.91 MIN: 2 MIN: 2.05 MIN: 2.02 MIN: 5.79 MIN: 5.92 MIN: 5.87 MIN: 5.07 MIN: 8.41 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 5.46 / Avg: 5.5 / Max: 5.54 Min: 7.23 / Avg: 7.24 / Max: 7.25 Min: 1.88 / Avg: 1.89 / Max: 1.9 Min: 1.78 / Avg: 1.8 / Max: 1.81 Min: 1.92 / Avg: 1.93 / Max: 1.94 Min: 1.99 / Avg: 2 / Max: 2 Min: 2.06 / Avg: 2.07 / Max: 2.08 Min: 2.2 / Avg: 2.21 / Max: 2.21 Min: 2.15 / Avg: 2.15 / Max: 2.16 Min: 5.89 / Avg: 5.92 / Max: 5.97 Min: 6.02 / Avg: 6.03 / Max: 6.04 Min: 6.02 / Avg: 6.06 / Max: 6.1 Min: 5.25 / Avg: 5.25 / Max: 5.26 Min: 8.63 / Avg: 8.67 / Max: 8.71 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 4 - Input: 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.035, N = 4 SE +/- 0.023, N = 3 SE +/- 0.003, N = 3 SE +/- 0.060, N = 3 SE +/- 0.039, N = 4 SE +/- 0.012, N = 3 SE +/- 0.032, N = 4 SE +/- 0.048, N = 4 SE +/- 0.032, N = 3 SE +/- 0.043, N = 3 SE +/- 0.017, N = 3 SE +/- 0.006, N = 3 SE +/- 0.008, N = 3 SE +/- 0.008, N = 3 5.333 3.184 6.752 6.966 6.678 6.494 6.091 5.958 5.866 5.563 4.506 4.225 3.593 2.536 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 4 - Input: 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0135 0.027 0.0405 0.054 0.0675 0.03 0.03 0.06 0.06 0.06 0.06 0.06 0.05 0.06 0.05 0.05 0.05 0.04 0.04
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 4 - Input: 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 5.23 / Avg: 5.33 / Max: 5.39 Min: 3.14 / Avg: 3.18 / Max: 3.21 Min: 6.75 / Avg: 6.75 / Max: 6.76 Min: 6.88 / Avg: 6.97 / Max: 7.08 Min: 6.57 / Avg: 6.68 / Max: 6.75 Min: 6.47 / Avg: 6.49 / Max: 6.51 Min: 6.04 / Avg: 6.09 / Max: 6.18 Min: 5.88 / Avg: 5.96 / Max: 6.09 Min: 5.8 / Avg: 5.87 / Max: 5.91 Min: 5.48 / Avg: 5.56 / Max: 5.61 Min: 4.47 / Avg: 4.51 / Max: 4.53 Min: 4.22 / Avg: 4.23 / Max: 4.24 Min: 3.58 / Avg: 3.59 / Max: 3.61 Min: 2.52 / Avg: 2.54 / Max: 2.55 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
Etcpak Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: ETC1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 SE +/- 0.18, N = 3 SE +/- 0.19, N = 3 SE +/- 0.12, N = 3 SE +/- 0.16, N = 3 SE +/- 0.00, N = 3 SE +/- 0.18, N = 3 SE +/- 0.10, N = 3 SE +/- 0.11, N = 3 SE +/- 0.16, N = 3 SE +/- 0.08, N = 3 SE +/- 0.21, N = 3 SE +/- 0.03, N = 3 SE +/- 0.14, N = 3 271.07 271.32 233.35 229.86 229.46 236.91 229.73 233.51 233.39 229.56 222.36 222.62 223.03 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
Mpx/s Per Watt
OpenBenchmarking.org Mpx/s Per Watt, More Is Better Etcpak 0.7 Configuration: ETC1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.2713 2.5426 3.8139 5.0852 6.3565 3.89 5.29 3.48 3.46 4.14 5.04 3.48 4.95 4.94 4.81 5.44 5.60 5.65
Result Confidence
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: ETC1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 Min: 270.72 / Avg: 271.07 / Max: 271.3 Min: 271.12 / Avg: 271.32 / Max: 271.69 Min: 233.22 / Avg: 233.35 / Max: 233.6 Min: 229.54 / Avg: 229.86 / Max: 230.03 Min: 229.45 / Avg: 229.46 / Max: 229.47 Min: 236.57 / Avg: 236.91 / Max: 237.18 Min: 229.59 / Avg: 229.73 / Max: 229.93 Min: 233.3 / Avg: 233.51 / Max: 233.66 Min: 233.07 / Avg: 233.39 / Max: 233.6 Min: 229.4 / Avg: 229.56 / Max: 229.64 Min: 221.94 / Avg: 222.36 / Max: 222.64 Min: 222.56 / Avg: 222.62 / Max: 222.66 Min: 222.74 / Avg: 223.03 / Max: 223.2 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.14, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.17, N = 3 SE +/- 0.09, N = 3 SE +/- 0.13, N = 3 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 SE +/- 0.03, N = 3 17.67 17.70 20.39 20.90 20.86 20.85 20.10 20.86 20.29 20.50 20.87 21.38 21.27 22.45 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 17.4 / Avg: 17.67 / Max: 17.81 Min: 17.51 / Avg: 17.7 / Max: 17.84 Min: 20.19 / Avg: 20.39 / Max: 20.52 Min: 20.76 / Avg: 20.9 / Max: 21.07 Min: 20.76 / Avg: 20.86 / Max: 20.97 Min: 20.67 / Avg: 20.85 / Max: 20.97 Min: 19.77 / Avg: 20.1 / Max: 20.29 Min: 20.71 / Avg: 20.86 / Max: 21.01 Min: 20.1 / Avg: 20.29 / Max: 20.55 Min: 20.34 / Avg: 20.5 / Max: 20.64 Min: 20.48 / Avg: 20.87 / Max: 21.12 Min: 21.02 / Avg: 21.38 / Max: 21.69 Min: 21.11 / Avg: 21.27 / Max: 21.58 Min: 22.41 / Avg: 22.45 / Max: 22.5 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9 18 27 36 45 SE +/- 0.06, N = 3 SE +/- 0.08, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 5 SE +/- 0.17, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 13.45 8.16 40.00 40.00 34.48 34.48 27.78 25.00 26.05 20.83 14.08 13.89 10.42 6.99 MIN: 12.82 / MAX: 13.89 MIN: 6.85 / MAX: 8.4 MIN: 34.48 / MAX: 41.67 MIN: 38.46 / MAX: 41.67 MIN: 33.33 / MAX: 35.71 MIN: 33.33 / MAX: 35.71 MIN: 27.03 MIN: 23.81 / MAX: 25.64 MIN: 25.64 / MAX: 26.32 MIN: 20.41 / MAX: 21.28 MIN: 12.99 / MAX: 14.29 MIN: 13.33 / MAX: 14.08 MIN: 10.1 / MAX: 10.64 MIN: 5.95 / MAX: 7.04
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0698 0.1396 0.2094 0.2792 0.349 0.09 0.08 0.30 0.31 0.25 0.28 0.22 0.19 0.23 0.17 0.15 0.16 0.13 0.11
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 13.33 / Avg: 13.45 / Max: 13.51 Min: 7.75 / Avg: 8.16 / Max: 8.26 Min: 34.48 / Avg: 34.48 / Max: 34.48 Min: 34.48 / Avg: 34.48 / Max: 34.48 Min: 27.78 / Avg: 27.78 / Max: 27.78 Min: 25.64 / Avg: 26.05 / Max: 26.32 Min: 20.83 / Avg: 20.83 / Max: 20.83 Min: 14.08 / Avg: 14.08 / Max: 14.08 Min: 13.89 / Avg: 13.89 / Max: 13.89 Min: 10.42 / Avg: 10.42 / Max: 10.42 Min: 6.99 / Avg: 6.99 / Max: 6.99
Numenta Anomaly Benchmark Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Relative Entropy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.08, N = 4 SE +/- 0.09, N = 3 SE +/- 0.03, N = 4 SE +/- 0.10, N = 4 SE +/- 0.13, N = 4 SE +/- 0.08, N = 4 SE +/- 0.06, N = 4 SE +/- 0.08, N = 4 SE +/- 0.15, N = 4 SE +/- 0.05, N = 4 SE +/- 0.17, N = 3 SE +/- 0.10, N = 3 SE +/- 0.17, N = 3 SE +/- 0.17, N = 3 14.60 18.88 15.94 15.97 15.89 15.97 15.72 16.33 15.85 16.12 17.18 18.10 19.11 23.56
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Relative Entropy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 14.38 / Avg: 14.6 / Max: 14.74 Min: 18.71 / Avg: 18.88 / Max: 19.01 Min: 15.88 / Avg: 15.94 / Max: 16.04 Min: 15.81 / Avg: 15.97 / Max: 16.23 Min: 15.65 / Avg: 15.89 / Max: 16.17 Min: 15.82 / Avg: 15.97 / Max: 16.13 Min: 15.6 / Avg: 15.72 / Max: 15.88 Min: 16.13 / Avg: 16.33 / Max: 16.51 Min: 15.61 / Avg: 15.85 / Max: 16.23 Min: 16.01 / Avg: 16.12 / Max: 16.23 Min: 17 / Avg: 17.18 / Max: 17.52 Min: 17.96 / Avg: 18.1 / Max: 18.29 Min: 18.77 / Avg: 19.11 / Max: 19.31 Min: 23.35 / Avg: 23.56 / Max: 23.91
Dolfyn Dolfyn is a Computational Fluid Dynamics (CFD) code of modern numerical simulation techniques. The Dolfyn test profile measures the execution time of the bundled computational fluid dynamics demos that are bundled with Dolfyn. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Dolfyn 0.527 Computational Fluid Dynamics EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 17.60 17.56 20.56 20.78 20.74 20.81 20.11 20.79 20.69 20.48 20.79 21.42 21.39 21.55
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Dolfyn 0.527 Computational Fluid Dynamics EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 17.53 / Avg: 17.6 / Max: 17.67 Min: 17.53 / Avg: 17.56 / Max: 17.6 Min: 20.54 / Avg: 20.56 / Max: 20.59 Min: 20.77 / Avg: 20.78 / Max: 20.8 Min: 20.73 / Avg: 20.74 / Max: 20.77 Min: 20.79 / Avg: 20.81 / Max: 20.83 Min: 20.1 / Avg: 20.11 / Max: 20.15 Min: 20.71 / Avg: 20.79 / Max: 20.84 Min: 20.68 / Avg: 20.69 / Max: 20.7 Min: 20.46 / Avg: 20.48 / Max: 20.5 Min: 20.67 / Avg: 20.79 / Max: 20.9 Min: 21.35 / Avg: 21.42 / Max: 21.51 Min: 21.37 / Avg: 21.39 / Max: 21.42 Min: 21.5 / Avg: 21.55 / Max: 21.61
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: Blowfish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 90 180 270 360 450 SE +/- 0.07, N = 3 SE +/- 0.15, N = 3 SE +/- 0.16, N = 3 SE +/- 0.16, N = 3 SE +/- 0.20, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.13, N = 3 SE +/- 0.14, N = 3 SE +/- 0.40, N = 3 SE +/- 0.15, N = 3 SE +/- 0.25, N = 3 SE +/- 0.16, N = 3 SE +/- 0.13, N = 3 429.14 429.41 368.93 362.99 361.80 363.50 374.52 363.10 368.69 368.38 362.20 351.14 352.07 352.29 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better Botan 2.13.0 Test: Blowfish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 6.21 8.25 5.49 5.51 5.83 6.50 8.01 5.52 7.85 7.92 7.70 8.62 8.88 8.96
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: Blowfish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 Min: 429.06 / Avg: 429.14 / Max: 429.27 Min: 429.18 / Avg: 429.41 / Max: 429.7 Min: 368.71 / Avg: 368.93 / Max: 369.25 Min: 362.68 / Avg: 362.99 / Max: 363.15 Min: 361.43 / Avg: 361.8 / Max: 362.1 Min: 363.36 / Avg: 363.5 / Max: 363.69 Min: 374.32 / Avg: 374.52 / Max: 374.67 Min: 362.85 / Avg: 363.1 / Max: 363.25 Min: 368.54 / Avg: 368.69 / Max: 368.97 Min: 367.96 / Avg: 368.38 / Max: 369.17 Min: 362.03 / Avg: 362.2 / Max: 362.49 Min: 350.66 / Avg: 351.14 / Max: 351.45 Min: 351.75 / Avg: 352.07 / Max: 352.26 Min: 352.1 / Avg: 352.29 / Max: 352.53 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.4541 0.9082 1.3623 1.8164 2.2705 SE +/- 0.001140, N = 4 SE +/- 0.001580, N = 4 SE +/- 0.002887, N = 4 SE +/- 0.005359, N = 5 SE +/- 0.003528, N = 4 SE +/- 0.004962, N = 15 SE +/- 0.000738, N = 4 SE +/- 0.005528, N = 6 SE +/- 0.000902, N = 4 SE +/- 0.000660, N = 4 SE +/- 0.000863, N = 4 SE +/- 0.000258, N = 4 SE +/- 0.000447, N = 4 SE +/- 0.005714, N = 4 0.660419 1.160170 0.587443 0.517023 0.484219 0.533961 0.500521 0.553444 0.537636 0.592308 0.802272 0.875749 1.093440 2.018380 MIN: 0.62 MIN: 1.13 MIN: 0.54 MIN: 0.47 MIN: 0.46 MIN: 0.49 MIN: 0.48 MIN: 0.49 MIN: 0.51 MIN: 0.56 MIN: 0.72 MIN: 0.74 MIN: 0.96 MIN: 1.95 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.66 / Avg: 0.66 / Max: 0.66 Min: 1.16 / Avg: 1.16 / Max: 1.16 Min: 0.58 / Avg: 0.59 / Max: 0.6 Min: 0.5 / Avg: 0.52 / Max: 0.53 Min: 0.48 / Avg: 0.48 / Max: 0.49 Min: 0.51 / Avg: 0.53 / Max: 0.57 Min: 0.5 / Avg: 0.5 / Max: 0.5 Min: 0.54 / Avg: 0.55 / Max: 0.58 Min: 0.54 / Avg: 0.54 / Max: 0.54 Min: 0.59 / Avg: 0.59 / Max: 0.59 Min: 0.8 / Avg: 0.8 / Max: 0.8 Min: 0.88 / Avg: 0.88 / Max: 0.88 Min: 1.09 / Avg: 1.09 / Max: 1.09 Min: 2 / Avg: 2.02 / Max: 2.03 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: Twofish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.37, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 351.63 351.99 302.14 297.81 296.91 297.10 306.62 297.71 302.25 302.12 297.65 287.93 288.62 288.70 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better Botan 2.13.0 Test: Twofish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 5.10 6.82 4.50 4.53 4.77 5.41 6.60 4.57 6.42 6.50 6.28 7.11 7.30 7.37
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: Twofish EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 351.56 / Avg: 351.63 / Max: 351.71 Min: 351.85 / Avg: 351.99 / Max: 352.11 Min: 302.1 / Avg: 302.14 / Max: 302.21 Min: 297.64 / Avg: 297.81 / Max: 297.92 Min: 296.81 / Avg: 296.91 / Max: 296.96 Min: 297.02 / Avg: 297.1 / Max: 297.2 Min: 305.88 / Avg: 306.62 / Max: 307 Min: 297.66 / Avg: 297.7 / Max: 297.77 Min: 302.19 / Avg: 302.25 / Max: 302.32 Min: 301.76 / Avg: 302.12 / Max: 302.35 Min: 297.5 / Avg: 297.65 / Max: 297.86 Min: 287.77 / Avg: 287.93 / Max: 288.1 Min: 288.49 / Avg: 288.62 / Max: 288.69 Min: 288.49 / Avg: 288.7 / Max: 288.83 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
AOM AV1 This is a simple test of the AOMedia AV1 encoder run on the CPU with a sample video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 8 Realtime EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.34, N = 5 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 35.56 35.48 32.78 32.54 32.31 32.60 33.51 32.35 32.86 32.99 32.62 32.03 31.63 31.08 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better AOM AV1 2.0 Encoder Mode: Speed 8 Realtime EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1508 0.3016 0.4524 0.6032 0.754 0.42 0.58 0.42 0.44 0.44 0.51 0.60 0.45 0.59 0.59 0.59 0.66 0.67 0.67
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 8 Realtime EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 Min: 35.51 / Avg: 35.56 / Max: 35.62 Min: 35.37 / Avg: 35.48 / Max: 35.54 Min: 32.72 / Avg: 32.78 / Max: 32.82 Min: 32.54 / Avg: 32.54 / Max: 32.55 Min: 32.25 / Avg: 32.31 / Max: 32.36 Min: 32.57 / Avg: 32.6 / Max: 32.62 Min: 33.4 / Avg: 33.51 / Max: 33.6 Min: 32.3 / Avg: 32.35 / Max: 32.38 Min: 31.51 / Avg: 32.86 / Max: 33.21 Min: 32.85 / Avg: 32.99 / Max: 33.06 Min: 32.42 / Avg: 32.62 / Max: 32.79 Min: 31.98 / Avg: 32.03 / Max: 32.11 Min: 31.58 / Avg: 31.63 / Max: 31.69 Min: 31.04 / Avg: 31.08 / Max: 31.13 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.0662 2.1324 3.1986 4.2648 5.331 SE +/- 0.01053, N = 4 SE +/- 0.00382, N = 4 SE +/- 0.00178, N = 4 SE +/- 0.01406, N = 4 SE +/- 0.00906, N = 4 SE +/- 0.00439, N = 4 SE +/- 0.00341, N = 4 SE +/- 0.00891, N = 4 SE +/- 0.00185, N = 4 SE +/- 0.00090, N = 4 SE +/- 0.00144, N = 4 SE +/- 0.00488, N = 4 SE +/- 0.00316, N = 4 SE +/- 0.00331, N = 4 1.98035 3.56020 1.70360 1.48981 1.46775 1.59938 1.47222 1.61502 1.59300 1.82951 2.41082 3.44668 3.80955 4.73866 MIN: 1.86 MIN: 3.44 MIN: 1.48 MIN: 1.33 MIN: 1.39 MIN: 1.51 MIN: 1.41 MIN: 1.5 MIN: 1.51 MIN: 1.74 MIN: 2.2 MIN: 2.92 MIN: 3.44 MIN: 4.55 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.96 / Avg: 1.98 / Max: 2.01 Min: 3.55 / Avg: 3.56 / Max: 3.57 Min: 1.7 / Avg: 1.7 / Max: 1.71 Min: 1.45 / Avg: 1.49 / Max: 1.52 Min: 1.45 / Avg: 1.47 / Max: 1.49 Min: 1.59 / Avg: 1.6 / Max: 1.61 Min: 1.46 / Avg: 1.47 / Max: 1.48 Min: 1.6 / Avg: 1.62 / Max: 1.64 Min: 1.59 / Avg: 1.59 / Max: 1.6 Min: 1.83 / Avg: 1.83 / Max: 1.83 Min: 2.41 / Avg: 2.41 / Max: 2.41 Min: 3.43 / Avg: 3.45 / Max: 3.45 Min: 3.8 / Avg: 3.81 / Max: 3.81 Min: 4.73 / Avg: 4.74 / Max: 4.74 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.7576 1.5152 2.2728 3.0304 3.788 SE +/- 0.00643, N = 4 SE +/- 0.00055, N = 4 SE +/- 0.01081, N = 4 SE +/- 0.01452, N = 4 SE +/- 0.00473, N = 4 SE +/- 0.01048, N = 4 SE +/- 0.00302, N = 4 SE +/- 0.00172, N = 4 SE +/- 0.00119, N = 4 SE +/- 0.00089, N = 4 SE +/- 0.00055, N = 4 SE +/- 0.00098, N = 4 SE +/- 0.00157, N = 4 SE +/- 0.00108, N = 4 1.52143 2.79034 1.80598 1.74199 1.66750 1.73007 1.16402 1.21230 1.18204 1.43487 1.74975 1.80647 2.43760 3.36727 MIN: 1.48 MIN: 2.71 MIN: 1.54 MIN: 1.45 MIN: 1.4 MIN: 1.5 MIN: 1.12 MIN: 1.14 MIN: 1.13 MIN: 1.4 MIN: 1.72 MIN: 1.75 MIN: 2.4 MIN: 3.3 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.51 / Avg: 1.52 / Max: 1.54 Min: 2.79 / Avg: 2.79 / Max: 2.79 Min: 1.78 / Avg: 1.81 / Max: 1.83 Min: 1.72 / Avg: 1.74 / Max: 1.78 Min: 1.65 / Avg: 1.67 / Max: 1.67 Min: 1.72 / Avg: 1.73 / Max: 1.76 Min: 1.16 / Avg: 1.16 / Max: 1.17 Min: 1.21 / Avg: 1.21 / Max: 1.22 Min: 1.18 / Avg: 1.18 / Max: 1.19 Min: 1.43 / Avg: 1.43 / Max: 1.44 Min: 1.75 / Avg: 1.75 / Max: 1.75 Min: 1.8 / Avg: 1.81 / Max: 1.81 Min: 2.43 / Avg: 2.44 / Max: 2.44 Min: 3.37 / Avg: 3.37 / Max: 3.37 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Botan Botan is a cross-platform open-source C++ crypto library that supports most all publicly known cryptographic algorithms. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: CAST-256 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.11, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.28, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 139.71 139.83 120.00 118.23 117.99 118.32 121.92 118.29 120.06 119.99 117.88 114.47 114.69 114.59 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better Botan 2.13.0 Test: CAST-256 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.675 1.35 2.025 2.7 3.375 2.09 2.83 1.83 1.82 1.93 2.16 2.67 1.83 2.61 2.65 2.56 2.88 2.98 3.00
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: CAST-256 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30 60 90 120 150 Min: 139.63 / Avg: 139.71 / Max: 139.74 Min: 139.8 / Avg: 139.83 / Max: 139.86 Min: 119.99 / Avg: 119.99 / Max: 120.01 Min: 118.06 / Avg: 118.23 / Max: 118.33 Min: 117.81 / Avg: 117.99 / Max: 118.19 Min: 118.29 / Avg: 118.32 / Max: 118.36 Min: 121.89 / Avg: 121.92 / Max: 121.94 Min: 118.22 / Avg: 118.29 / Max: 118.36 Min: 119.97 / Avg: 120.06 / Max: 120.12 Min: 119.95 / Avg: 119.99 / Max: 120.04 Min: 117.32 / Avg: 117.88 / Max: 118.21 Min: 114.34 / Avg: 114.47 / Max: 114.63 Min: 114.63 / Avg: 114.69 / Max: 114.8 Min: 114.51 / Avg: 114.59 / Max: 114.64 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Result
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: KASUMI EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 90.42 90.46 77.70 76.55 76.22 76.58 78.88 76.53 77.69 77.57 76.51 74.06 74.21 74.27 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
MiB/s Per Watt
OpenBenchmarking.org MiB/s Per Watt, More Is Better Botan 2.13.0 Test: KASUMI EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.4365 0.873 1.3095 1.746 2.1825 1.33 1.82 1.18 1.18 1.25 1.40 1.73 1.18 1.69 1.70 1.66 1.86 1.92 1.94
Result Confidence
OpenBenchmarking.org MiB/s, More Is Better Botan 2.13.0 Test: KASUMI EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 Min: 90.37 / Avg: 90.42 / Max: 90.45 Min: 90.45 / Avg: 90.46 / Max: 90.47 Min: 77.66 / Avg: 77.7 / Max: 77.77 Min: 76.54 / Avg: 76.55 / Max: 76.55 Min: 76.12 / Avg: 76.22 / Max: 76.31 Min: 76.57 / Avg: 76.57 / Max: 76.58 Min: 78.84 / Avg: 78.88 / Max: 78.91 Min: 76.42 / Avg: 76.53 / Max: 76.6 Min: 77.68 / Avg: 77.69 / Max: 77.7 Min: 77.5 / Avg: 77.57 / Max: 77.66 Min: 76.49 / Avg: 76.51 / Max: 76.52 Min: 73.98 / Avg: 74.06 / Max: 74.21 Min: 74.18 / Avg: 74.21 / Max: 74.22 Min: 74.23 / Avg: 74.27 / Max: 74.29 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP CFD Solver EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 SE +/- 0.022, N = 4 SE +/- 0.078, N = 3 SE +/- 0.066, N = 8 SE +/- 0.051, N = 15 SE +/- 0.073, N = 5 SE +/- 0.083, N = 6 SE +/- 0.005, N = 5 SE +/- 0.057, N = 5 SE +/- 0.030, N = 5 SE +/- 0.007, N = 4 SE +/- 0.022, N = 4 SE +/- 0.009, N = 4 SE +/- 0.065, N = 3 SE +/- 0.030, N = 3 13.530 22.218 7.819 7.747 8.434 8.405 9.806 9.858 9.892 11.577 15.445 14.792 18.134 25.374 1. (CXX) g++ options: -O2 -lOpenCL
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP CFD Solver EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 Min: 13.49 / Avg: 13.53 / Max: 13.59 Min: 22.12 / Avg: 22.22 / Max: 22.37 Min: 7.45 / Avg: 7.82 / Max: 8.05 Min: 7.37 / Avg: 7.75 / Max: 8.02 Min: 8.16 / Avg: 8.43 / Max: 8.59 Min: 8.14 / Avg: 8.41 / Max: 8.71 Min: 9.79 / Avg: 9.81 / Max: 9.82 Min: 9.76 / Avg: 9.86 / Max: 10.08 Min: 9.81 / Avg: 9.89 / Max: 9.96 Min: 11.57 / Avg: 11.58 / Max: 11.6 Min: 15.41 / Avg: 15.44 / Max: 15.51 Min: 14.76 / Avg: 14.79 / Max: 14.8 Min: 18.02 / Avg: 18.13 / Max: 18.24 Min: 25.33 / Avg: 25.37 / Max: 25.43 1. (CXX) g++ options: -O2 -lOpenCL
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.2.0 Scene: Memorial EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 SE +/- 0.02, N = 4 SE +/- 0.01, N = 3 SE +/- 0.03, N = 6 SE +/- 0.04, N = 6 SE +/- 0.02, N = 6 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 4 SE +/- 0.12, N = 6 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 14.24 7.94 27.67 29.41 26.13 21.55 20.14 20.25 17.17 12.18 11.34 9.03 6.55
Images / Sec Per Watt
OpenBenchmarking.org Images / Sec Per Watt, More Is Better Intel Open Image Denoise 1.2.0 Scene: Memorial EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0473 0.0946 0.1419 0.1892 0.2365 0.07 0.06 0.20 0.21 0.20 0.16 0.15 0.16 0.13 0.11 0.12 0.10 0.08
Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.2.0 Scene: Memorial EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 14.2 / Avg: 14.24 / Max: 14.27 Min: 7.92 / Avg: 7.94 / Max: 7.96 Min: 27.55 / Avg: 27.67 / Max: 27.73 Min: 29.29 / Avg: 29.41 / Max: 29.54 Min: 26.05 / Avg: 26.13 / Max: 26.17 Min: 21.52 / Avg: 21.55 / Max: 21.6 Min: 20.11 / Avg: 20.14 / Max: 20.18 Min: 20.23 / Avg: 20.25 / Max: 20.28 Min: 17.16 / Avg: 17.17 / Max: 17.19 Min: 11.56 / Avg: 12.18 / Max: 12.32 Min: 11.34 / Avg: 11.34 / Max: 11.36 Min: 9.02 / Avg: 9.03 / Max: 9.03 Min: 6.54 / Avg: 6.55 / Max: 6.55
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Second, More Is Better ASKAP 1.0 Test: Hogbom Clean OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 SE +/- 0.57, N = 3 SE +/- 1.74, N = 3 SE +/- 2.16, N = 3 SE +/- 1.71, N = 3 SE +/- 0.00, N = 3 SE +/- 0.98, N = 3 SE +/- 1.71, N = 3 SE +/- 0.64, N = 3 SE +/- 1.71, N = 3 SE +/- 3.44, N = 3 SE +/- 0.62, N = 3 SE +/- 2.18, N = 3 SE +/- 2.09, N = 3 SE +/- 1.09, N = 3 140.92 380.73 432.92 439.25 432.90 411.53 439.25 439.24 439.25 432.33 432.28 378.81 370.39 351.71 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Iterations Per Second Per Watt
OpenBenchmarking.org Iterations Per Second Per Watt, More Is Better ASKAP 1.0 Test: Hogbom Clean OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 1.59 6.05 5.10 5.45 5.46 5.89 7.49 5.38 7.30 7.39 7.45 7.57 7.49 7.29
Result Confidence
OpenBenchmarking.org Iterations Per Second, More Is Better ASKAP 1.0 Test: Hogbom Clean OpenMP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 Min: 140.25 / Avg: 140.92 / Max: 142.05 Min: 377.36 / Avg: 380.73 / Max: 383.14 Min: 429.19 / Avg: 432.92 / Max: 436.68 Min: 436.68 / Avg: 439.25 / Max: 442.48 Min: 432.9 / Avg: 432.9 / Max: 432.9 Min: 409.84 / Avg: 411.53 / Max: 413.22 Min: 436.68 / Avg: 439.25 / Max: 442.48 Min: 438.6 / Avg: 439.24 / Max: 440.53 Min: 436.68 / Avg: 439.25 / Max: 442.48 Min: 425.53 / Avg: 432.33 / Max: 436.68 Min: 431.03 / Avg: 432.28 / Max: 432.9 Min: 374.53 / Avg: 378.81 / Max: 381.68 Min: 366.3 / Avg: 370.39 / Max: 373.13 Min: 349.65 / Avg: 351.71 / Max: 353.36 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.8063 1.6126 2.4189 3.2252 4.0315 SE +/- 0.001361, N = 4 SE +/- 0.000640, N = 4 SE +/- 0.002127, N = 4 SE +/- 0.000537, N = 4 SE +/- 0.001659, N = 4 SE +/- 0.000788, N = 4 SE +/- 0.005244, N = 4 SE +/- 0.001948, N = 4 SE +/- 0.003967, N = 4 SE +/- 0.009261, N = 15 SE +/- 0.001219, N = 4 SE +/- 0.001269, N = 4 SE +/- 0.001298, N = 4 SE +/- 0.001594, N = 4 1.565030 3.012920 0.800643 0.780803 0.863146 0.868024 1.015820 1.104860 1.073990 1.295990 1.841780 1.913050 2.493720 3.583630 MIN: 1.54 MIN: 2.73 MIN: 0.76 MIN: 0.7 MIN: 0.77 MIN: 0.81 MIN: 0.98 MIN: 0.99 MIN: 0.98 MIN: 1.25 MIN: 1.82 MIN: 1.87 MIN: 2.45 MIN: 3.54 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.56 / Avg: 1.57 / Max: 1.57 Min: 3.01 / Avg: 3.01 / Max: 3.01 Min: 0.79 / Avg: 0.8 / Max: 0.8 Min: 0.78 / Avg: 0.78 / Max: 0.78 Min: 0.86 / Avg: 0.86 / Max: 0.87 Min: 0.87 / Avg: 0.87 / Max: 0.87 Min: 1 / Avg: 1.02 / Max: 1.03 Min: 1.1 / Avg: 1.1 / Max: 1.11 Min: 1.07 / Avg: 1.07 / Max: 1.08 Min: 1.27 / Avg: 1.3 / Max: 1.37 Min: 1.84 / Avg: 1.84 / Max: 1.85 Min: 1.91 / Avg: 1.91 / Max: 1.92 Min: 2.49 / Avg: 2.49 / Max: 2.5 Min: 3.58 / Avg: 3.58 / Max: 3.59 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Sysbench This is a benchmark of Sysbench with CPU and memory sub-tests. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: Memory EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2M 4M 6M 8M 10M SE +/- 17904.29, N = 5 SE +/- 717.93, N = 5 SE +/- 11992.02, N = 5 SE +/- 35042.66, N = 5 SE +/- 14871.56, N = 5 SE +/- 8561.16, N = 5 SE +/- 61191.25, N = 15 SE +/- 5458.69, N = 5 SE +/- 6709.34, N = 5 SE +/- 780.60, N = 5 SE +/- 11575.58, N = 5 SE +/- 1918.40, N = 5 SE +/- 12382.78, N = 5 3087955.63 4656182.26 6302136.65 6374595.12 6380400.35 6618850.24 4720934.26 6612746.50 5614019.49 4502591.45 7880986.59 7232205.30 5942474.37 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=amdfam10 -rdynamic -ldl -laio -lm
Events Per Second Per Watt
OpenBenchmarking.org Events Per Second Per Watt, More Is Better Sysbench 2018-07-28 Test: Memory EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K 37471.23 84636.76 62605.70 66221.01 80342.22 105418.49 57649.40 103730.05 93633.65 83787.81 162496.00 160055.91 141909.12
Result Confidence
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: Memory EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.4M 2.8M 4.2M 5.6M 7M Min: 3067897.69 / Avg: 3087955.63 / Max: 3159400.31 Min: 4654731.01 / Avg: 4656182.26 / Max: 4658759.57 Min: 6269344.44 / Avg: 6302136.65 / Max: 6328806.52 Min: 6319346.34 / Avg: 6374595.12 / Max: 6505001.22 Min: 6349678.61 / Avg: 6380400.35 / Max: 6428305.03 Min: 6607713.8 / Avg: 6618850.24 / Max: 6652865.09 Min: 4407088.34 / Avg: 4720934.26 / Max: 5255661.79 Min: 6604337.31 / Avg: 6612746.5 / Max: 6634214.25 Min: 5602311.72 / Avg: 5614019.49 / Max: 5638809.43 Min: 4500022.71 / Avg: 4502591.45 / Max: 4504708.71 Min: 7856776.88 / Avg: 7880986.59 / Max: 7920212.37 Min: 7225906.31 / Avg: 7232205.3 / Max: 7236184.18 Min: 5895950.69 / Avg: 5942474.37 / Max: 5962905.57 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=amdfam10 -rdynamic -ldl -laio -lm
Tungsten Renderer Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Hair EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8 16 24 32 40 SE +/- 0.03705, N = 4 SE +/- 0.01111, N = 3 SE +/- 0.02278, N = 6 SE +/- 0.02077, N = 6 SE +/- 0.03455, N = 5 SE +/- 0.04134, N = 5 SE +/- 0.02895, N = 5 SE +/- 0.03618, N = 5 SE +/- 0.00983, N = 5 SE +/- 0.02622, N = 4 SE +/- 0.02745, N = 3 SE +/- 0.03890, N = 3 SE +/- 0.03918, N = 3 SE +/- 0.03115, N = 3 14.13270 26.51610 7.50706 7.55093 8.56828 8.71953 9.92849 10.81310 10.57400 12.26230 16.61750 17.65420 21.99580 32.53880 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Hair EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7 14 21 28 35 Min: 14.08 / Avg: 14.13 / Max: 14.23 Min: 26.5 / Avg: 26.52 / Max: 26.54 Min: 7.46 / Avg: 7.51 / Max: 7.59 Min: 7.5 / Avg: 7.55 / Max: 7.63 Min: 8.43 / Avg: 8.57 / Max: 8.61 Min: 8.63 / Avg: 8.72 / Max: 8.86 Min: 9.86 / Avg: 9.93 / Max: 10.01 Min: 10.71 / Avg: 10.81 / Max: 10.91 Min: 10.56 / Avg: 10.57 / Max: 10.61 Min: 12.21 / Avg: 12.26 / Max: 12.33 Min: 16.56 / Avg: 16.62 / Max: 16.65 Min: 17.6 / Avg: 17.65 / Max: 17.73 Min: 21.93 / Avg: 22 / Max: 22.07 Min: 32.48 / Avg: 32.54 / Max: 32.59 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Degridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1500 3000 4500 6000 7500 SE +/- 3.57, N = 4 SE +/- 30.11, N = 2 SE +/- 47.34, N = 4 SE +/- 42.67, N = 4 SE +/- 40.59, N = 4 SE +/- 36.86, N = 4 SE +/- 0.00, N = 4 SE +/- 38.66, N = 4 SE +/- 0.00, N = 4 SE +/- 35.55, N = 4 SE +/- 37.73, N = 3 SE +/- 10.27, N = 4 SE +/- 10.02, N = 4 SE +/- 6.46, N = 4 1406.93 4004.08 7148.77 6699.07 6615.81 6302.57 5788.17 6455.40 5788.17 5726.60 5509.30 3317.93 3257.04 2616.81 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Million Grid Points Per Second Per Watt
OpenBenchmarking.org Million Grid Points Per Second Per Watt, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Degridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 17.31 70.19 96.86 92.78 93.63 100.05 108.46 89.44 108.49 109.78 103.76 72.91 72.53 58.97
Result Confidence
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Degridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1200 2400 3600 4800 6000 Min: 1401.35 / Avg: 1406.93 / Max: 1416.26 Min: 3973.97 / Avg: 4004.08 / Max: 4034.18 Min: 7006.74 / Avg: 7148.77 / Max: 7196.11 Min: 6656.4 / Avg: 6699.07 / Max: 6827.08 Min: 6494.05 / Avg: 6615.81 / Max: 6656.4 Min: 6192 / Avg: 6302.57 / Max: 6339.43 Min: 5788.17 / Avg: 5788.17 / Max: 5788.17 Min: 6339.43 / Avg: 6455.4 / Max: 6494.05 Min: 5788.17 / Avg: 5788.17 / Max: 5788.17 Min: 5665.02 / Avg: 5726.6 / Max: 5788.17 Min: 5433.8 / Avg: 5509.27 / Max: 5547 Min: 3287.11 / Avg: 3317.93 / Max: 3328.2 Min: 3247.02 / Avg: 3257.04 / Max: 3287.11 Min: 2610.35 / Avg: 2616.81 / Max: 2636.2 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Result
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2K 4K 6K 8K 10K SE +/- 0.00, N = 4 SE +/- 35.30, N = 4 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 81.97, N = 4 SE +/- 77.48, N = 4 SE +/- 67.99, N = 4 SE +/- 63.04, N = 4 SE +/- 79.54, N = 4 SE +/- 72.46, N = 4 SE +/- 57.68, N = 5 SE +/- 0.00, N = 4 SE +/- 17.38, N = 4 SE +/- 7.61, N = 4 2113.14 4035.11 9509.14 9509.14 9427.17 8454.70 6658.48 8257.47 6617.90 6379.88 5503.59 4294.45 4004.08 2840.13 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Result Confidence
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 1.0 Test: tConvolve OpenMP - Gridding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1700 3400 5100 6800 8500 Min: 2113.14 / Avg: 2113.14 / Max: 2113.14 Min: 3973.97 / Avg: 4035.11 / Max: 4096.25 Min: 9509.14 / Avg: 9509.14 / Max: 9509.14 Min: 9509.14 / Avg: 9509.14 / Max: 9509.14 Min: 9181.24 / Avg: 9427.17 / Max: 9509.14 Min: 8320.5 / Avg: 8454.7 / Max: 8588.9 Min: 6494.05 / Avg: 6658.48 / Max: 6827.08 Min: 8068.36 / Avg: 8257.47 / Max: 8320.5 Min: 6494.05 / Avg: 6617.9 / Max: 6827.08 Min: 6192 / Avg: 6379.88 / Max: 6494.05 Min: 5325.12 / Avg: 5503.59 / Max: 5665.02 Min: 4294.45 / Avg: 4294.45 / Max: 4294.45 Min: 3973.97 / Avg: 4004.08 / Max: 4034.18 Min: 2832.51 / Avg: 2840.13 / Max: 2862.97 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
JPEG XL The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance. Learn more via the OpenBenchmarking.org test page.
Kvazaar This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Ultra Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 13 26 39 52 65 SE +/- 0.02, N = 4 SE +/- 0.04, N = 3 SE +/- 0.54, N = 5 SE +/- 0.44, N = 5 SE +/- 0.16, N = 5 SE +/- 0.48, N = 5 SE +/- 0.05, N = 5 SE +/- 0.03, N = 5 SE +/- 0.05, N = 5 SE +/- 0.05, N = 4 SE +/- 0.03, N = 4 SE +/- 0.03, N = 4 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 47.17 27.16 53.52 58.64 59.59 55.99 56.98 54.07 54.92 48.33 41.52 39.65 31.39 23.11 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Ultra Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1058 0.2116 0.3174 0.4232 0.529 0.26 0.24 0.41 0.42 0.40 0.43 0.47 0.40 0.47 0.40 0.40 0.45 0.35 0.31
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Ultra Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 12 24 36 48 60 Min: 47.11 / Avg: 47.17 / Max: 47.21 Min: 27.11 / Avg: 27.16 / Max: 27.23 Min: 52.03 / Avg: 53.52 / Max: 54.9 Min: 57.87 / Avg: 58.64 / Max: 60.27 Min: 59.13 / Avg: 59.59 / Max: 59.95 Min: 55.17 / Avg: 55.99 / Max: 57.85 Min: 56.85 / Avg: 56.98 / Max: 57.12 Min: 54 / Avg: 54.07 / Max: 54.14 Min: 54.78 / Avg: 54.92 / Max: 55.03 Min: 48.24 / Avg: 48.33 / Max: 48.45 Min: 41.47 / Avg: 41.52 / Max: 41.57 Min: 39.58 / Avg: 39.65 / Max: 39.69 Min: 31.34 / Avg: 31.39 / Max: 31.49 Min: 23.08 / Avg: 23.11 / Max: 23.14 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
JPEG XL The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance. Learn more via the OpenBenchmarking.org test page.
Nebular Empirical Analysis Tool NEAT is the Nebular Empirical Analysis Tool for empirical analysis of ionised nebulae, with uncertainty propagation. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Nebular Empirical Analysis Tool 2020-02-29 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.01, N = 4 SE +/- 0.01, N = 3 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.16, N = 3 SE +/- 0.08, N = 3 SE +/- 0.01, N = 4 SE +/- 0.00, N = 3 SE +/- 0.03, N = 4 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 16.20 17.44 17.37 16.89 17.03 16.67 16.37 17.21 16.55 16.92 17.75 17.39 18.10 21.35 1. (F9X) gfortran options: -cpp -ffree-line-length-0 -Jsource/ -fopenmp -O3 -fno-backtrace
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Nebular Empirical Analysis Tool 2020-02-29 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 16.19 / Avg: 16.2 / Max: 16.23 Min: 17.42 / Avg: 17.43 / Max: 17.46 Min: 17.08 / Avg: 17.37 / Max: 17.52 Min: 16.79 / Avg: 16.89 / Max: 16.96 Min: 16.72 / Avg: 17.03 / Max: 17.24 Min: 16.52 / Avg: 16.67 / Max: 16.78 Min: 16.34 / Avg: 16.37 / Max: 16.39 Min: 17.21 / Avg: 17.21 / Max: 17.22 Min: 16.49 / Avg: 16.55 / Max: 16.65 Min: 16.91 / Avg: 16.92 / Max: 16.93 Min: 17.73 / Avg: 17.75 / Max: 17.77 Min: 17.38 / Avg: 17.39 / Max: 17.4 Min: 18.08 / Avg: 18.1 / Max: 18.11 Min: 21.31 / Avg: 21.35 / Max: 21.39 1. (F9X) gfortran options: -cpp -ffree-line-length-0 -Jsource/ -fopenmp -O3 -fno-backtrace
Stream This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 330.72, N = 5 SE +/- 52.59, N = 5 SE +/- 31.65, N = 5 SE +/- 13.27, N = 5 SE +/- 15.41, N = 5 SE +/- 5.66, N = 5 SE +/- 10.31, N = 5 SE +/- 14.85, N = 5 SE +/- 5.59, N = 5 SE +/- 8.02, N = 5 SE +/- 2.37, N = 5 SE +/- 7.06, N = 5 SE +/- 9.94, N = 5 66901.1 82399.7 90511.8 90296.5 88717.2 79674.1 90663.2 79342.9 79677.2 80140.1 51093.1 51714.4 52390.0 1. (CC) gcc options: -O3 -march=native -fopenmp
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Stream 2013-01-17 Type: Copy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 560.85 973.24 648.19 678.70 790.62 873.56 762.36 844.44 902.05 1003.43 734.63 785.09 854.90
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16K 32K 48K 64K 80K Min: 65648.8 / Avg: 66901.12 / Max: 67581.9 Min: 82189.9 / Avg: 82399.72 / Max: 82465.6 Min: 90390.8 / Avg: 90511.82 / Max: 90564 Min: 90262 / Avg: 90296.46 / Max: 90328.8 Min: 88676.8 / Avg: 88717.18 / Max: 88755.4 Min: 79657.3 / Avg: 79674.1 / Max: 79689.4 Min: 90631.3 / Avg: 90663.16 / Max: 90687.7 Min: 79306.1 / Avg: 79342.9 / Max: 79388.7 Min: 79657.3 / Avg: 79677.16 / Max: 79688.5 Min: 80119.5 / Avg: 80140.14 / Max: 80160.6 Min: 51087.4 / Avg: 51093.14 / Max: 51100.2 Min: 51688.2 / Avg: 51714.44 / Max: 51726.5 Min: 52373.2 / Avg: 52390.04 / Max: 52428 1. (CC) gcc options: -O3 -march=native -fopenmp
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 4K 8K 12K 16K 20K SE +/- 43.06, N = 3 SE +/- 18.04, N = 4 SE +/- 15.75, N = 5 SE +/- 27.38, N = 5 SE +/- 30.70, N = 5 SE +/- 25.61, N = 5 SE +/- 69.28, N = 5 SE +/- 15.81, N = 5 SE +/- 18.07, N = 4 SE +/- 22.89, N = 4 7779.49 12350.87 15079.08 15230.23 14977.29 18019.01 14759.40 15688.57 9697.83 9844.95 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 40 80 120 160 200 56.39 120.79 103.87 107.24 137.33 137.30 129.34 159.36 119.50 136.77
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 3K 6K 9K 12K 15K Min: 7733.94 / Avg: 7779.49 / Max: 7865.57 Min: 12309.98 / Avg: 12350.87 / Max: 12395.51 Min: 15027.23 / Avg: 15079.08 / Max: 15122.17 Min: 15146.06 / Avg: 15230.23 / Max: 15316.01 Min: 14873.66 / Avg: 14977.29 / Max: 15062.42 Min: 17953.88 / Avg: 18019.01 / Max: 18084.24 Min: 14524.99 / Avg: 14759.4 / Max: 14889.39 Min: 15654.67 / Avg: 15688.57 / Max: 15744.56 Min: 9644.91 / Avg: 9697.83 / Max: 9725.86 Min: 9819.42 / Avg: 9844.95 / Max: 9913.54 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 12K 24K 36K 48K 60K SE +/- 44.41, N = 3 SE +/- 10.84, N = 4 SE +/- 40.27, N = 5 SE +/- 50.60, N = 5 SE +/- 17.91, N = 5 SE +/- 17.78, N = 5 SE +/- 36.50, N = 5 SE +/- 19.15, N = 4 SE +/- 23.10, N = 4 SE +/- 23.87, N = 3 22117.17 27284.69 55136.50 55668.63 47098.22 49235.92 46206.33 37175.20 29526.97 21822.06 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 100 200 300 400 500 153.51 260.77 399.60 419.28 436.08 395.37 440.80 398.01 373.22 297.55
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 10K 20K 30K 40K 50K Min: 22040.65 / Avg: 22117.17 / Max: 22194.48 Min: 27272.21 / Avg: 27284.69 / Max: 27317.11 Min: 55002.69 / Avg: 55136.5 / Max: 55211.78 Min: 55491.76 / Avg: 55668.63 / Max: 55799.49 Min: 47048.85 / Avg: 47098.22 / Max: 47153.11 Min: 49168.13 / Avg: 49235.92 / Max: 49272.59 Min: 46062.44 / Avg: 46206.33 / Max: 46260.89 Min: 37140.1 / Avg: 37175.2 / Max: 37216.01 Min: 29462.38 / Avg: 29526.97 / Max: 29571 Min: 21777.28 / Avg: 21822.06 / Max: 21858.76 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 100, Compression Effort 5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.012, N = 6 SE +/- 0.031, N = 4 SE +/- 0.012, N = 6 SE +/- 0.009, N = 6 SE +/- 0.006, N = 6 SE +/- 0.013, N = 6 SE +/- 0.015, N = 6 SE +/- 0.010, N = 6 SE +/- 0.027, N = 6 SE +/- 0.009, N = 6 SE +/- 0.020, N = 5 SE +/- 0.012, N = 5 SE +/- 0.018, N = 5 SE +/- 0.032, N = 4 7.973 12.788 8.125 8.052 8.074 8.162 7.949 8.055 7.972 7.974 9.354 9.654 11.131 15.547 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better WebP2 Image Encode 20210126 Encode Settings: Quality 100, Compression Effort 5 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 7.94 / Avg: 7.97 / Max: 8.02 Min: 12.74 / Avg: 12.79 / Max: 12.88 Min: 8.1 / Avg: 8.13 / Max: 8.17 Min: 8.04 / Avg: 8.05 / Max: 8.09 Min: 8.05 / Avg: 8.07 / Max: 8.09 Min: 8.12 / Avg: 8.16 / Max: 8.21 Min: 7.92 / Avg: 7.95 / Max: 8.02 Min: 8.01 / Avg: 8.06 / Max: 8.08 Min: 7.91 / Avg: 7.97 / Max: 8.08 Min: 7.94 / Avg: 7.97 / Max: 8 Min: 9.31 / Avg: 9.35 / Max: 9.43 Min: 9.63 / Avg: 9.65 / Max: 9.69 Min: 11.09 / Avg: 11.13 / Max: 11.19 Min: 15.46 / Avg: 15.55 / Max: 15.62 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lwebp -lwebpdemux
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1000 2000 3000 4000 5000 SE +/- 35.77, N = 5 SE +/- 35.92, N = 5 SE +/- 33.26, N = 5 SE +/- 25.14, N = 5 SE +/- 22.36, N = 5 SE +/- 12.58, N = 5 SE +/- 15.90, N = 5 SE +/- 28.80, N = 5 SE +/- 9.28, N = 5 SE +/- 20.03, N = 5 SE +/- 32.99, N = 5 SE +/- 33.20, N = 5 SE +/- 22.80, N = 5 4085 3724 4407 4457 4218 3886 4303 3912 3993 4138 3881 3971 3894
Result Confidence
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 3969 / Avg: 4084.8 / Max: 4158 Min: 3653 / Avg: 3723.6 / Max: 3860 Min: 4330 / Avg: 4406.6 / Max: 4494 Min: 4392 / Avg: 4456.8 / Max: 4536 Min: 4158 / Avg: 4217.6 / Max: 4273 Min: 3845 / Avg: 3886.4 / Max: 3916 Min: 4246 / Avg: 4302.6 / Max: 4341 Min: 3827 / Avg: 3912.2 / Max: 3971 Min: 3975 / Avg: 3993.2 / Max: 4028 Min: 4074 / Avg: 4138.2 / Max: 4180 Min: 3765 / Avg: 3881.4 / Max: 3963 Min: 3886 / Avg: 3970.6 / Max: 4055 Min: 3832 / Avg: 3893.8 / Max: 3948
Sysbench This is a benchmark of Sysbench with CPU and memory sub-tests. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 7.33, N = 5 SE +/- 2.64, N = 5 SE +/- 97.15, N = 5 SE +/- 52.65, N = 5 SE +/- 23.00, N = 5 SE +/- 20.39, N = 5 SE +/- 7.32, N = 5 SE +/- 9.32, N = 5 SE +/- 12.89, N = 5 SE +/- 9.93, N = 5 SE +/- 10.65, N = 5 SE +/- 5.53, N = 5 SE +/- 1.09, N = 5 32649.36 16323.92 106421.95 108892.22 80743.89 56066.79 55254.66 56038.72 42008.11 27632.19 26778.62 20091.96 13398.10 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=amdfam10 -rdynamic -ldl -laio -lm
Events Per Second Per Watt
OpenBenchmarking.org Events Per Second Per Watt, More Is Better Sysbench 2018-07-28 Test: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 313.31 261.75 754.73 783.68 733.57 667.05 521.63 638.23 539.88 432.15 474.66 391.60 297.08
Result Confidence
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 2018-07-28 Test: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K Min: 32621.61 / Avg: 32649.36 / Max: 32663.3 Min: 16316.25 / Avg: 16323.92 / Max: 16331.8 Min: 106076.46 / Avg: 106421.95 / Max: 106643.26 Min: 108721.1 / Avg: 108892.22 / Max: 109052.86 Min: 80668.55 / Avg: 80743.89 / Max: 80805.58 Min: 55989.07 / Avg: 56066.79 / Max: 56109.25 Min: 55238.42 / Avg: 55254.66 / Max: 55276.83 Min: 56017.18 / Avg: 56038.72 / Max: 56065.58 Min: 41974.26 / Avg: 42008.11 / Max: 42044 Min: 27594.22 / Avg: 27632.19 / Max: 27651.61 Min: 26746.06 / Avg: 26778.62 / Max: 26804.47 Min: 20074.61 / Avg: 20091.96 / Max: 20104.41 Min: 13395.23 / Avg: 13398.1 / Max: 13400.85 1. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=amdfam10 -rdynamic -ldl -laio -lm
Timed MAFFT Alignment This test performs an alignment of 100 pyruvate decarboxylase sequences. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.471 Multiple Sequence Alignment - LSU RNA EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.037, N = 5 SE +/- 0.030, N = 5 SE +/- 0.027, N = 5 SE +/- 0.024, N = 5 SE +/- 0.037, N = 5 SE +/- 0.033, N = 5 SE +/- 0.014, N = 5 SE +/- 0.021, N = 5 SE +/- 0.030, N = 5 SE +/- 0.027, N = 5 SE +/- 0.004, N = 5 SE +/- 0.028, N = 5 SE +/- 0.023, N = 5 SE +/- 0.035, N = 5 8.868 8.598 10.097 9.989 10.040 9.925 9.602 9.938 9.735 9.725 9.892 10.221 10.236 11.112 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.471 Multiple Sequence Alignment - LSU RNA EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 8.78 / Avg: 8.87 / Max: 8.95 Min: 8.55 / Avg: 8.6 / Max: 8.69 Min: 10.02 / Avg: 10.1 / Max: 10.18 Min: 9.92 / Avg: 9.99 / Max: 10.06 Min: 9.91 / Avg: 10.04 / Max: 10.12 Min: 9.82 / Avg: 9.92 / Max: 10 Min: 9.57 / Avg: 9.6 / Max: 9.64 Min: 9.89 / Avg: 9.94 / Max: 10.01 Min: 9.66 / Avg: 9.74 / Max: 9.84 Min: 9.65 / Avg: 9.73 / Max: 9.8 Min: 9.88 / Avg: 9.89 / Max: 9.9 Min: 10.15 / Avg: 10.22 / Max: 10.32 Min: 10.16 / Avg: 10.24 / Max: 10.3 Min: 11.04 / Avg: 11.11 / Max: 11.22 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.21, N = 5 SE +/- 0.03, N = 4 SE +/- 0.27, N = 5 SE +/- 0.32, N = 5 SE +/- 0.16, N = 5 SE +/- 0.20, N = 5 SE +/- 0.18, N = 6 SE +/- 0.17, N = 6 SE +/- 0.25, N = 6 SE +/- 0.03, N = 6 SE +/- 0.17, N = 5 SE +/- 0.09, N = 5 SE +/- 0.11, N = 4 SE +/- 0.07, N = 4 42.84 26.13 79.82 83.40 63.21 62.40 59.94 56.53 57.32 54.87 37.04 35.38 30.38 21.76 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.2003 0.4006 0.6009 0.8012 1.0015 0.32 0.28 0.83 0.89 0.62 0.69 0.68 0.56 0.68 0.61 0.44 0.49 0.41 0.33
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16 32 48 64 80 Min: 42.28 / Avg: 42.84 / Max: 43.39 Min: 26.07 / Avg: 26.13 / Max: 26.22 Min: 79.13 / Avg: 79.82 / Max: 80.5 Min: 82.2 / Avg: 83.4 / Max: 84.06 Min: 62.71 / Avg: 63.21 / Max: 63.66 Min: 61.9 / Avg: 62.4 / Max: 62.94 Min: 59.31 / Avg: 59.94 / Max: 60.53 Min: 56.08 / Avg: 56.53 / Max: 57.07 Min: 56.32 / Avg: 57.32 / Max: 57.97 Min: 54.76 / Avg: 54.87 / Max: 54.97 Min: 36.48 / Avg: 37.04 / Max: 37.5 Min: 35.07 / Avg: 35.38 / Max: 35.57 Min: 30.14 / Avg: 30.38 / Max: 30.6 Min: 21.59 / Avg: 21.76 / Max: 21.91 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
rays1bench This is a test of rays1bench, a simple path-tracer / ray-tracing that supports SSE and AVX instructions, multi-threading, and other features. This test profile is measuring the performance of the "large scene" in rays1bench. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org mrays/s, More Is Better rays1bench 2020-01-09 Large Scene EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 SE +/- 0.10, N = 5 SE +/- 0.01, N = 4 SE +/- 0.20, N = 7 SE +/- 0.21, N = 7 SE +/- 0.11, N = 7 SE +/- 0.08, N = 7 SE +/- 0.03, N = 6 SE +/- 0.06, N = 6 SE +/- 0.06, N = 6 SE +/- 0.04, N = 6 SE +/- 0.08, N = 5 SE +/- 0.06, N = 4 SE +/- 0.04, N = 4 SE +/- 0.08, N = 3 109.91 59.60 243.57 243.25 218.73 217.70 182.59 163.00 167.75 134.19 90.37 84.54 68.51 48.61
mrays/s Per Watt
OpenBenchmarking.org mrays/s Per Watt, More Is Better rays1bench 2020-01-09 Large Scene EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.5378 1.0756 1.6134 2.1512 2.689 0.69 0.54 2.33 2.39 2.04 2.24 1.85 1.54 1.81 1.34 0.97 1.09 0.85 0.68
Result Confidence
OpenBenchmarking.org mrays/s, More Is Better rays1bench 2020-01-09 Large Scene EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 109.61 / Avg: 109.91 / Max: 110.19 Min: 59.57 / Avg: 59.6 / Max: 59.62 Min: 242.82 / Avg: 243.57 / Max: 244.37 Min: 242.31 / Avg: 243.25 / Max: 243.89 Min: 218.36 / Avg: 218.73 / Max: 219.13 Min: 217.39 / Avg: 217.7 / Max: 217.92 Min: 182.51 / Avg: 182.59 / Max: 182.66 Min: 162.8 / Avg: 163 / Max: 163.17 Min: 167.52 / Avg: 167.75 / Max: 167.95 Min: 134.04 / Avg: 134.19 / Max: 134.3 Min: 90.09 / Avg: 90.37 / Max: 90.53 Min: 84.42 / Avg: 84.54 / Max: 84.72 Min: 68.41 / Avg: 68.51 / Max: 68.59 Min: 48.46 / Avg: 48.61 / Max: 48.73
Basis Universal Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.005, N = 6 SE +/- 0.002, N = 6 SE +/- 0.006, N = 6 SE +/- 0.002, N = 6 SE +/- 0.003, N = 6 SE +/- 0.005, N = 6 SE +/- 0.003, N = 5 SE +/- 0.002, N = 6 SE +/- 0.009, N = 6 SE +/- 0.005, N = 5 SE +/- 0.007, N = 5 SE +/- 0.002, N = 5 SE +/- 0.005, N = 5 7.453 8.077 8.095 8.132 8.198 8.038 8.357 8.149 8.267 8.660 8.923 9.190 9.756 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 0 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 7.44 / Avg: 7.45 / Max: 7.48 Min: 8.07 / Avg: 8.08 / Max: 8.08 Min: 8.06 / Avg: 8.09 / Max: 8.11 Min: 8.13 / Avg: 8.13 / Max: 8.14 Min: 8.19 / Avg: 8.2 / Max: 8.21 Min: 8.02 / Avg: 8.04 / Max: 8.05 Min: 8.35 / Avg: 8.36 / Max: 8.37 Min: 8.14 / Avg: 8.15 / Max: 8.16 Min: 8.24 / Avg: 8.27 / Max: 8.3 Min: 8.65 / Avg: 8.66 / Max: 8.67 Min: 8.9 / Avg: 8.92 / Max: 8.94 Min: 9.18 / Avg: 9.19 / Max: 9.19 Min: 9.74 / Avg: 9.76 / Max: 9.77 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.003, N = 6 SE +/- 0.010, N = 6 SE +/- 0.005, N = 5 SE +/- 0.025, N = 5 SE +/- 0.009, N = 5 SE +/- 0.001, N = 5 SE +/- 0.004, N = 5 SE +/- 0.005, N = 5 SE +/- 0.025, N = 5 SE +/- 0.014, N = 5 SE +/- 0.009, N = 5 SE +/- 0.025, N = 5 SE +/- 0.009, N = 5 SE +/- 0.004, N = 5 7.739 7.745 9.003 9.154 9.145 9.126 8.861 9.140 9.014 9.014 9.151 9.481 9.419 9.414 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 7.73 / Avg: 7.74 / Max: 7.75 Min: 7.72 / Avg: 7.74 / Max: 7.79 Min: 8.99 / Avg: 9 / Max: 9.02 Min: 9.12 / Avg: 9.15 / Max: 9.25 Min: 9.13 / Avg: 9.14 / Max: 9.17 Min: 9.12 / Avg: 9.13 / Max: 9.13 Min: 8.85 / Avg: 8.86 / Max: 8.87 Min: 9.13 / Avg: 9.14 / Max: 9.15 Min: 8.98 / Avg: 9.01 / Max: 9.12 Min: 8.99 / Avg: 9.01 / Max: 9.06 Min: 9.12 / Avg: 9.15 / Max: 9.17 Min: 9.43 / Avg: 9.48 / Max: 9.58 Min: 9.41 / Avg: 9.42 / Max: 9.46 Min: 9.4 / Avg: 9.41 / Max: 9.42 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.4004 0.8008 1.2012 1.6016 2.002 SE +/- 0.002832, N = 5 SE +/- 0.006135, N = 5 SE +/- 0.001522, N = 5 SE +/- 0.000841, N = 5 SE +/- 0.002582, N = 5 SE +/- 0.005510, N = 5 SE +/- 0.003216, N = 5 SE +/- 0.002578, N = 5 SE +/- 0.001432, N = 5 SE +/- 0.003852, N = 5 SE +/- 0.001293, N = 5 SE +/- 0.003097, N = 5 SE +/- 0.003339, N = 5 SE +/- 0.002555, N = 5 0.489983 0.798217 1.162730 1.146010 0.813280 1.132370 1.000426 0.641197 0.990316 0.711317 0.601288 0.816826 1.038950 1.779470 MIN: 0.45 MIN: 0.75 MIN: 1.1 MIN: 1.09 MIN: 0.76 MIN: 1.07 MIN: 0.94 MIN: 0.58 MIN: 0.95 MIN: 0.66 MIN: 0.56 MIN: 0.74 MIN: 0.95 MIN: 1.6 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 0.48 / Avg: 0.49 / Max: 0.5 Min: 0.79 / Avg: 0.8 / Max: 0.82 Min: 1.16 / Avg: 1.16 / Max: 1.17 Min: 1.14 / Avg: 1.15 / Max: 1.15 Min: 0.81 / Avg: 0.81 / Max: 0.82 Min: 1.11 / Avg: 1.13 / Max: 1.14 Min: 0.99 / Avg: 1 / Max: 1.01 Min: 0.63 / Avg: 0.64 / Max: 0.65 Min: 0.99 / Avg: 0.99 / Max: 1 Min: 0.7 / Avg: 0.71 / Max: 0.72 Min: 0.6 / Avg: 0.6 / Max: 0.6 Min: 0.81 / Avg: 0.82 / Max: 0.83 Min: 1.03 / Avg: 1.04 / Max: 1.05 Min: 1.78 / Avg: 1.78 / Max: 1.79 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: H2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1000 2000 3000 4000 5000 SE +/- 31.15, N = 7 SE +/- 31.43, N = 6 SE +/- 47.69, N = 5 SE +/- 39.23, N = 5 SE +/- 34.34, N = 5 SE +/- 15.11, N = 5 SE +/- 44.06, N = 5 SE +/- 22.77, N = 5 SE +/- 13.74, N = 5 SE +/- 38.29, N = 5 SE +/- 31.79, N = 5 SE +/- 22.48, N = 6 SE +/- 29.37, N = 5 3323 3316 4686 4773 4463 3861 4026 3917 3778 3633 3671 3552 3793
Result Confidence
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: H2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 800 1600 2400 3200 4000 Min: 3208 / Avg: 3323.43 / Max: 3442 Min: 3220 / Avg: 3315.5 / Max: 3444 Min: 4536 / Avg: 4686 / Max: 4799 Min: 4691 / Avg: 4773.4 / Max: 4896 Min: 4370 / Avg: 4463.2 / Max: 4566 Min: 3817 / Avg: 3860.6 / Max: 3906 Min: 3901 / Avg: 4025.6 / Max: 4157 Min: 3832 / Avg: 3917.4 / Max: 3957 Min: 3743 / Avg: 3778.2 / Max: 3810 Min: 3534 / Avg: 3633.2 / Max: 3722 Min: 3564 / Avg: 3671.2 / Max: 3763 Min: 3511 / Avg: 3551.67 / Max: 3657 Min: 3701 / Avg: 3792.8 / Max: 3886
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.01324, N = 5 SE +/- 0.00869, N = 5 SE +/- 0.00630, N = 5 SE +/- 0.01221, N = 5 SE +/- 0.01041, N = 5 SE +/- 0.00795, N = 5 SE +/- 0.00208, N = 5 SE +/- 0.00475, N = 5 SE +/- 0.00256, N = 5 SE +/- 0.01609, N = 5 SE +/- 0.00471, N = 5 SE +/- 0.00189, N = 5 SE +/- 0.00853, N = 5 SE +/- 0.01715, N = 5 2.77826 6.68209 1.15943 1.12502 1.22011 2.28611 2.95830 1.25154 2.96283 4.13382 3.76395 4.47287 4.97512 7.15030 MIN: 2.68 MIN: 6.52 MIN: 2.2 MIN: 2.88 MIN: 1.2 MIN: 2.89 MIN: 3.91 MIN: 3.68 MIN: 4.3 MIN: 4.75 MIN: 6.45 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 2.75 / Avg: 2.78 / Max: 2.83 Min: 6.66 / Avg: 6.68 / Max: 6.71 Min: 1.14 / Avg: 1.16 / Max: 1.17 Min: 1.1 / Avg: 1.13 / Max: 1.17 Min: 1.19 / Avg: 1.22 / Max: 1.25 Min: 2.26 / Avg: 2.29 / Max: 2.3 Min: 2.95 / Avg: 2.96 / Max: 2.97 Min: 1.24 / Avg: 1.25 / Max: 1.26 Min: 2.96 / Avg: 2.96 / Max: 2.97 Min: 4.08 / Avg: 4.13 / Max: 4.18 Min: 3.75 / Avg: 3.76 / Max: 3.78 Min: 4.47 / Avg: 4.47 / Max: 4.48 Min: 4.94 / Avg: 4.98 / Max: 4.99 Min: 7.11 / Avg: 7.15 / Max: 7.19 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OSPray Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20 40 60 80 100 SE +/- 0.18, N = 5 SE +/- 0.00, N = 4 SE +/- 0.00, N = 7 SE +/- 0.00, N = 7 SE +/- 0.00, N = 7 SE +/- 0.00, N = 7 SE +/- 0.49, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 6 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 4 SE +/- 0.00, N = 3 29.59 17.24 76.92 76.92 66.67 66.67 55.07 47.62 50.00 41.67 29.41 27.78 21.74 14.29 MIN: 27.78 / MAX: 31.25 MIN: 16.67 / MAX: 17.86 MIN: 71.43 / MAX: 83.33 MIN: 62.5 / MAX: 83.33 MIN: 62.5 / MAX: 71.43 MIN: 58.82 / MAX: 71.43 MIN: 50 / MAX: 55.56 MIN: 45.45 / MAX: 50 MIN: 45.45 / MAX: 52.63 MIN: 38.46 / MAX: 43.48 MIN: 27.78 MIN: 26.32 MIN: 20.83 / MAX: 22.22 MIN: 13.89 / MAX: 14.49
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.1755 0.351 0.5265 0.702 0.8775 0.21 0.18 0.76 0.78 0.65 0.71 0.59 0.47 0.57 0.44 0.34 0.38 0.29 0.22
Result Confidence
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 15 30 45 60 75 Min: 29.41 / Avg: 29.59 / Max: 30.3 Min: 17.24 / Avg: 17.24 / Max: 17.24 Min: 76.92 / Avg: 76.92 / Max: 76.92 Min: 76.92 / Avg: 76.92 / Max: 76.92 Min: 66.67 / Avg: 66.67 / Max: 66.67 Min: 66.67 / Avg: 66.67 / Max: 66.67 Min: 52.63 / Avg: 55.07 / Max: 55.56 Min: 47.62 / Avg: 47.62 / Max: 47.62 Min: 41.67 / Avg: 41.67 / Max: 41.67 Min: 29.41 / Avg: 29.41 / Max: 29.41 Min: 27.78 / Avg: 27.78 / Max: 27.78 Min: 21.74 / Avg: 21.74 / Max: 21.74 Min: 14.29 / Avg: 14.29 / Max: 14.29
JPEG XL The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance. Learn more via the OpenBenchmarking.org test page.
dav1d Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 SE +/- 1.86, N = 3 SE +/- 0.38, N = 3 SE +/- 3.47, N = 3 SE +/- 2.82, N = 3 SE +/- 0.45, N = 3 SE +/- 2.26, N = 3 SE +/- 0.52, N = 3 SE +/- 2.20, N = 3 SE +/- 0.95, N = 3 SE +/- 1.22, N = 3 SE +/- 0.72, N = 3 SE +/- 0.66, N = 3 SE +/- 1.45, N = 3 SE +/- 0.34, N = 3 671.07 496.11 983.61 1158.08 1116.83 1095.40 939.90 892.42 937.73 839.06 698.08 712.26 624.79 473.82 MIN: 524.48 / MAX: 829.64 MIN: 390.79 / MAX: 704.31 MIN: 616.84 / MAX: 1260.07 MIN: 644.06 / MAX: 1489.95 MIN: 672.49 / MAX: 1434.66 MIN: 671.87 / MAX: 1406.03 MIN: 689.36 / MAX: 1207.51 MIN: 641.23 / MAX: 1144.96 MIN: 687.68 / MAX: 1200.87 MIN: 646.27 / MAX: 1067.76 MIN: 539.13 / MAX: 873.36 MIN: 546.18 / MAX: 898.36 MIN: 483.27 / MAX: 784.91 MIN: 365.77 / MAX: 674.59 1. (CC) gcc options: -pthread
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 8.07 7.83 12.64 15.25 15.55 16.67 17.37 11.84 16.80 14.96 12.61 15.03 13.08 10.24
Result Confidence
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 667.49 / Avg: 671.07 / Max: 673.75 Min: 495.72 / Avg: 496.11 / Max: 496.87 Min: 977.12 / Avg: 983.61 / Max: 988.98 Min: 1153.84 / Avg: 1158.08 / Max: 1163.42 Min: 1116.14 / Avg: 1116.83 / Max: 1117.68 Min: 1091.59 / Avg: 1095.4 / Max: 1099.4 Min: 938.86 / Avg: 939.9 / Max: 940.47 Min: 888.17 / Avg: 892.42 / Max: 895.56 Min: 935.96 / Avg: 937.73 / Max: 939.19 Min: 837.5 / Avg: 839.06 / Max: 841.47 Min: 697.03 / Avg: 698.08 / Max: 699.47 Min: 711 / Avg: 712.26 / Max: 713.24 Min: 622.12 / Avg: 624.79 / Max: 627.09 Min: 473.15 / Avg: 473.82 / Max: 474.25 1. (CC) gcc options: -pthread
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Medium EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.01, N = 6 SE +/- 0.01, N = 5 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 SE +/- 0.02, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 7.10 9.04 6.68 6.78 6.93 7.02 7.33 7.17 7.47 8.30 8.63 9.38 5.82 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Medium EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 7.07 / Avg: 7.1 / Max: 7.13 Min: 9.02 / Avg: 9.04 / Max: 9.06 Min: 6.64 / Avg: 6.68 / Max: 6.71 Min: 6.76 / Avg: 6.78 / Max: 6.8 Min: 6.89 / Avg: 6.93 / Max: 6.96 Min: 7.01 / Avg: 7.02 / Max: 7.04 Min: 7.31 / Avg: 7.33 / Max: 7.35 Min: 7.15 / Avg: 7.17 / Max: 7.18 Min: 7.43 / Avg: 7.47 / Max: 7.49 Min: 8.25 / Avg: 8.3 / Max: 8.34 Min: 8.59 / Avg: 8.63 / Max: 8.68 Min: 9.37 / Avg: 9.38 / Max: 9.39 Min: 5.81 / Avg: 5.82 / Max: 5.83 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.03514, N = 7 SE +/- 0.04127, N = 7 SE +/- 0.00551, N = 7 SE +/- 0.01107, N = 7 SE +/- 0.00965, N = 7 SE +/- 0.00723, N = 7 SE +/- 0.00813, N = 7 SE +/- 0.01167, N = 7 SE +/- 0.01750, N = 7 SE +/- 0.00525, N = 7 SE +/- 0.00742, N = 7 SE +/- 0.03193, N = 7 SE +/- 0.00984, N = 7 SE +/- 0.04608, N = 7 6.91561 9.15124 3.55892 3.50412 3.54316 3.84018 3.97699 3.61268 3.97043 4.25111 5.45400 6.76242 7.60505 10.95820 MIN: 6.67 MIN: 8.56 MIN: 3.47 MIN: 3.39 MIN: 3.43 MIN: 3.75 MIN: 3.87 MIN: 3.49 MIN: 3.86 MIN: 4.06 MIN: 5.12 MIN: 6.41 MIN: 7.15 MIN: 10.24 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 6.79 / Avg: 6.92 / Max: 7.05 Min: 9.06 / Avg: 9.15 / Max: 9.39 Min: 3.54 / Avg: 3.56 / Max: 3.58 Min: 3.47 / Avg: 3.5 / Max: 3.54 Min: 3.52 / Avg: 3.54 / Max: 3.59 Min: 3.81 / Avg: 3.84 / Max: 3.86 Min: 3.94 / Avg: 3.98 / Max: 4 Min: 3.55 / Avg: 3.61 / Max: 3.64 Min: 3.92 / Avg: 3.97 / Max: 4.03 Min: 4.23 / Avg: 4.25 / Max: 4.27 Min: 5.41 / Avg: 5.45 / Max: 5.47 Min: 6.64 / Avg: 6.76 / Max: 6.85 Min: 7.56 / Avg: 7.61 / Max: 7.64 Min: 10.75 / Avg: 10.96 / Max: 11.12 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Numenta Anomaly Benchmark Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Windowed Gaussian EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.016, N = 6 SE +/- 0.034, N = 5 SE +/- 0.029, N = 6 SE +/- 0.038, N = 6 SE +/- 0.036, N = 6 SE +/- 0.025, N = 6 SE +/- 0.016, N = 6 SE +/- 0.044, N = 6 SE +/- 0.020, N = 6 SE +/- 0.041, N = 6 SE +/- 0.018, N = 6 SE +/- 0.033, N = 6 SE +/- 0.043, N = 5 SE +/- 0.047, N = 4 6.586 9.790 6.930 6.971 7.063 7.033 6.900 7.220 6.989 7.121 7.645 8.105 8.798 11.912
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Numenta Anomaly Benchmark 1.1 Detector: Windowed Gaussian EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 6.53 / Avg: 6.59 / Max: 6.62 Min: 9.66 / Avg: 9.79 / Max: 9.84 Min: 6.86 / Avg: 6.93 / Max: 7.04 Min: 6.85 / Avg: 6.97 / Max: 7.12 Min: 6.98 / Avg: 7.06 / Max: 7.22 Min: 6.95 / Avg: 7.03 / Max: 7.12 Min: 6.84 / Avg: 6.9 / Max: 6.95 Min: 7.06 / Avg: 7.22 / Max: 7.34 Min: 6.93 / Avg: 6.99 / Max: 7.06 Min: 6.99 / Avg: 7.12 / Max: 7.23 Min: 7.59 / Avg: 7.65 / Max: 7.7 Min: 8 / Avg: 8.1 / Max: 8.23 Min: 8.7 / Avg: 8.8 / Max: 8.93 Min: 11.83 / Avg: 11.91 / Max: 12.03
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.002179, N = 7 SE +/- 0.007190, N = 7 SE +/- 0.002719, N = 7 SE +/- 0.002146, N = 7 SE +/- 0.002747, N = 7 SE +/- 0.003838, N = 7 SE +/- 0.011195, N = 7 SE +/- 0.001701, N = 7 SE +/- 0.008223, N = 7 SE +/- 0.005689, N = 7 SE +/- 0.006442, N = 7 SE +/- 0.008434, N = 7 SE +/- 0.005350, N = 7 SE +/- 0.004801, N = 7 2.001830 4.301300 1.078440 0.993732 1.108050 1.719670 3.461740 1.483740 3.462360 3.543470 3.706400 5.864580 5.953480 6.376680 MIN: 1.91 MIN: 4.22 MIN: 1.63 MIN: 3.38 MIN: 1.28 MIN: 3.38 MIN: 3.47 MIN: 3.6 MIN: 5.63 MIN: 5.82 MIN: 6.27 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 1.99 / Avg: 2 / Max: 2.01 Min: 4.26 / Avg: 4.3 / Max: 4.32 Min: 1.07 / Avg: 1.08 / Max: 1.09 Min: 0.99 / Avg: 0.99 / Max: 1 Min: 1.1 / Avg: 1.11 / Max: 1.12 Min: 1.7 / Avg: 1.72 / Max: 1.73 Min: 3.44 / Avg: 3.46 / Max: 3.53 Min: 1.47 / Avg: 1.48 / Max: 1.49 Min: 3.43 / Avg: 3.46 / Max: 3.5 Min: 3.52 / Avg: 3.54 / Max: 3.56 Min: 3.67 / Avg: 3.71 / Max: 3.73 Min: 5.83 / Avg: 5.86 / Max: 5.9 Min: 5.94 / Avg: 5.95 / Max: 5.98 Min: 6.36 / Avg: 6.38 / Max: 6.4 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
libjpeg-turbo tjbench tjbench is a JPEG decompression/compression benchmark part of libjpeg-turbo. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 SE +/- 0.04, N = 7 SE +/- 0.02, N = 7 SE +/- 0.04, N = 7 SE +/- 0.04, N = 7 SE +/- 0.03, N = 7 SE +/- 0.02, N = 7 SE +/- 0.07, N = 7 SE +/- 0.07, N = 7 SE +/- 0.05, N = 7 SE +/- 0.44, N = 7 SE +/- 0.04, N = 7 SE +/- 0.10, N = 7 SE +/- 0.26, N = 7 SE +/- 0.03, N = 7 197.79 197.77 169.54 167.62 167.57 167.61 172.55 167.52 170.11 169.70 167.63 162.07 162.37 162.56 1. (CC) gcc options: -O3 -rdynamic
Megapixels/sec Per Watt
OpenBenchmarking.org Megapixels/sec Per Watt, More Is Better libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.0148 2.0296 3.0444 4.0592 5.074 3.16 4.25 2.78 2.80 2.88 3.31 4.03 2.85 3.96 3.95 3.87 4.30 4.45 4.51
Result Confidence
OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 197.62 / Avg: 197.79 / Max: 197.87 Min: 197.66 / Avg: 197.77 / Max: 197.86 Min: 169.39 / Avg: 169.54 / Max: 169.69 Min: 167.5 / Avg: 167.62 / Max: 167.77 Min: 167.44 / Avg: 167.57 / Max: 167.69 Min: 167.49 / Avg: 167.61 / Max: 167.66 Min: 172.16 / Avg: 172.55 / Max: 172.69 Min: 167.15 / Avg: 167.52 / Max: 167.68 Min: 169.86 / Avg: 170.11 / Max: 170.26 Min: 167.04 / Avg: 169.7 / Max: 170.26 Min: 167.45 / Avg: 167.63 / Max: 167.79 Min: 161.84 / Avg: 162.07 / Max: 162.46 Min: 160.82 / Avg: 162.37 / Max: 162.74 Min: 162.5 / Avg: 162.56 / Max: 162.7 1. (CC) gcc options: -O3 -rdynamic
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 11K 22K 33K 44K 55K SE +/- 131.48, N = 6 SE +/- 204.79, N = 6 SE +/- 126.64, N = 6 SE +/- 399.17, N = 6 SE +/- 392.71, N = 6 SE +/- 152.42, N = 6 SE +/- 307.98, N = 6 SE +/- 312.60, N = 6 SE +/- 288.40, N = 6 SE +/- 343.41, N = 6 SE +/- 385.00, N = 7 SE +/- 244.70, N = 6 SE +/- 304.76, N = 6 SE +/- 369.07, N = 6 50969 50675 44328 42730 42836 43404 43936 43275 43903 43414 42761 42375 41912 41704 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Mflops Per Watt
OpenBenchmarking.org Mflops Per Watt, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 775.34 1037.75 696.81 683.14 736.70 820.23 1000.27 709.62 984.98 988.44 963.62 1101.31 1109.61 1123.76
Result Confidence
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 9K 18K 27K 36K 45K Min: 50577 / Avg: 50968.83 / Max: 51456 Min: 49987 / Avg: 50674.5 / Max: 51435 Min: 43820 / Avg: 44328.17 / Max: 44688 Min: 40860 / Avg: 42729.67 / Max: 43619 Min: 41412 / Avg: 42836.17 / Max: 44004 Min: 42872 / Avg: 43404.33 / Max: 43831 Min: 43041 / Avg: 43936.17 / Max: 44903 Min: 41954 / Avg: 43274.83 / Max: 44008 Min: 42938 / Avg: 43902.83 / Max: 44700 Min: 42288 / Avg: 43414 / Max: 44416 Min: 40864 / Avg: 42761.14 / Max: 44031 Min: 41289 / Avg: 42374.5 / Max: 42923 Min: 40544 / Avg: 41912 / Max: 42672 Min: 40672 / Avg: 41703.5 / Max: 42890 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
ECP-CANDLE The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ECP-CANDLE 0.3 Benchmark: P1B2 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 40.55 36.98 43.16 42.35 41.85 41.91 40.89 41.52 41.04 41.33 41.15 42.78 43.38 44.49
dav1d Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Summer Nature 4K EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 SE +/- 0.46, N = 3 SE +/- 0.27, N = 3 SE +/- 0.34, N = 3 SE +/- 0.74, N = 3 SE +/- 0.17, N = 3 SE +/- 0.66, N = 3 SE +/- 0.75, N = 3 SE +/- 0.29, N = 3 SE +/- 0.16, N = 3 SE +/- 0.61, N = 3 SE +/- 0.23, N = 3 SE +/- 0.22, N = 3 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 244.17 168.17 437.31 457.27 416.96 406.93 367.19 349.99 361.72 328.54 245.60 239.84 203.40 150.43 MIN: 200.55 / MAX: 270.42 MIN: 157.91 / MAX: 191.26 MIN: 217.67 / MAX: 473.18 MIN: 225.33 / MAX: 494.73 MIN: 251.44 / MAX: 465 MIN: 257.69 / MAX: 456.39 MIN: 288.65 / MAX: 424.55 MIN: 266.02 / MAX: 399.72 MIN: 287.95 / MAX: 416.23 MIN: 249.05 / MAX: 376.65 MIN: 210.68 / MAX: 281.84 MIN: 211 / MAX: 275.93 MIN: 187.66 / MAX: 232.21 MIN: 141.09 / MAX: 170.16 1. (CC) gcc options: -pthread
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better dav1d 0.8.1 Video Input: Summer Nature 4K EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.107 2.214 3.321 4.428 5.535 2.06 2.00 4.45 4.92 4.61 4.65 4.91 3.54 4.62 4.14 3.32 3.73 3.13 2.51
Result Confidence
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Summer Nature 4K EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 Min: 243.51 / Avg: 244.17 / Max: 245.05 Min: 167.67 / Avg: 168.17 / Max: 168.6 Min: 436.66 / Avg: 437.31 / Max: 437.79 Min: 455.84 / Avg: 457.27 / Max: 458.3 Min: 416.77 / Avg: 416.96 / Max: 417.29 Min: 405.7 / Avg: 406.93 / Max: 407.97 Min: 365.83 / Avg: 367.19 / Max: 368.43 Min: 349.42 / Avg: 349.99 / Max: 350.39 Min: 361.4 / Avg: 361.72 / Max: 361.91 Min: 327.42 / Avg: 328.54 / Max: 329.53 Min: 245.22 / Avg: 245.6 / Max: 246 Min: 239.57 / Avg: 239.84 / Max: 240.27 Min: 203.1 / Avg: 203.4 / Max: 203.66 Min: 150.28 / Avg: 150.43 / Max: 150.7 1. (CC) gcc options: -pthread
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.01, N = 7 SE +/- 0.01, N = 7 SE +/- 0.01, N = 7 SE +/- 0.00, N = 7 SE +/- 0.01, N = 7 SE +/- 0.01, N = 7 SE +/- 0.01, N = 7 SE +/- 0.01, N = 7 SE +/- 0.00, N = 7 SE +/- 0.01, N = 6 SE +/- 0.00, N = 6 SE +/- 0.01, N = 6 SE +/- 0.01, N = 6 5.51 6.28 5.76 5.83 5.89 5.83 6.05 5.93 6.06 6.47 6.68 6.97 7.65 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Fast EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 5.49 / Avg: 5.51 / Max: 5.54 Min: 6.24 / Avg: 6.28 / Max: 6.31 Min: 5.75 / Avg: 5.76 / Max: 5.8 Min: 5.82 / Avg: 5.83 / Max: 5.84 Min: 5.87 / Avg: 5.89 / Max: 5.91 Min: 5.81 / Avg: 5.83 / Max: 5.86 Min: 6.01 / Avg: 6.05 / Max: 6.07 Min: 5.92 / Avg: 5.93 / Max: 5.96 Min: 6.04 / Avg: 6.06 / Max: 6.07 Min: 6.45 / Avg: 6.47 / Max: 6.5 Min: 6.66 / Avg: 6.68 / Max: 6.69 Min: 6.93 / Avg: 6.97 / Max: 7.01 Min: 7.6 / Avg: 7.65 / Max: 7.69 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Tungsten Renderer Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Volumetric Caustic EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.05079, N = 6 SE +/- 0.08977, N = 4 SE +/- 0.00214, N = 8 SE +/- 0.00245, N = 8 SE +/- 0.00169, N = 8 SE +/- 0.00349, N = 8 SE +/- 0.00749, N = 8 SE +/- 0.00562, N = 7 SE +/- 0.00220, N = 8 SE +/- 0.01885, N = 7 SE +/- 0.03030, N = 6 SE +/- 0.04707, N = 6 SE +/- 0.06509, N = 5 SE +/- 0.01019, N = 4 6.56276 12.41410 4.46261 4.32609 4.36145 4.36372 4.54904 4.84256 4.69550 5.27499 7.46871 7.40074 9.81098 13.92660 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Volumetric Caustic EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 6.4 / Avg: 6.56 / Max: 6.75 Min: 12.18 / Avg: 12.41 / Max: 12.6 Min: 4.46 / Avg: 4.46 / Max: 4.47 Min: 4.32 / Avg: 4.33 / Max: 4.34 Min: 4.35 / Avg: 4.36 / Max: 4.37 Min: 4.35 / Avg: 4.36 / Max: 4.38 Min: 4.51 / Avg: 4.55 / Max: 4.57 Min: 4.83 / Avg: 4.84 / Max: 4.87 Min: 4.68 / Avg: 4.7 / Max: 4.7 Min: 5.21 / Avg: 5.27 / Max: 5.33 Min: 7.35 / Avg: 7.47 / Max: 7.56 Min: 7.21 / Avg: 7.4 / Max: 7.56 Min: 9.7 / Avg: 9.81 / Max: 10.01 Min: 13.91 / Avg: 13.93 / Max: 13.96 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1200 2400 3600 4800 6000 SE +/- 11.37, N = 7 SE +/- 22.07, N = 7 SE +/- 22.61, N = 6 SE +/- 12.22, N = 6 SE +/- 9.85, N = 6 SE +/- 17.79, N = 6 SE +/- 21.30, N = 6 SE +/- 12.59, N = 6 SE +/- 17.40, N = 6 SE +/- 20.02, N = 6 SE +/- 22.94, N = 6 SE +/- 17.45, N = 6 SE +/- 49.86, N = 6 4375 4544 5012 5021 5112 4935 5070 4916 5053 5145 5280 5353 5580
Result Confidence
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1000 2000 3000 4000 5000 Min: 4328 / Avg: 4374.57 / Max: 4420 Min: 4475 / Avg: 4544.29 / Max: 4659 Min: 4957 / Avg: 5011.67 / Max: 5119 Min: 4981 / Avg: 5020.83 / Max: 5072 Min: 5077 / Avg: 5112 / Max: 5138 Min: 4891 / Avg: 4934.83 / Max: 4996 Min: 5014 / Avg: 5070.17 / Max: 5141 Min: 4883 / Avg: 4916 / Max: 4959 Min: 5000 / Avg: 5053.33 / Max: 5104 Min: 5069 / Avg: 5145.33 / Max: 5205 Min: 5219 / Avg: 5279.5 / Max: 5358 Min: 5275 / Avg: 5352.5 / Max: 5392 Min: 5502 / Avg: 5579.5 / Max: 5827
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 SE +/- 2.97, N = 15 SE +/- 0.53, N = 7 SE +/- 4.45, N = 15 SE +/- 5.77, N = 15 SE +/- 6.69, N = 15 SE +/- 6.16, N = 15 SE +/- 6.75, N = 15 SE +/- 5.85, N = 15 SE +/- 5.89, N = 15 SE +/- 5.48, N = 15 SE +/- 2.50, N = 15 SE +/- 1.82, N = 15 SE +/- 1.04, N = 9 SE +/- 0.51, N = 7 230.31 124.35 305.10 332.56 346.63 336.24 350.63 325.84 334.18 319.05 242.24 229.82 184.83 107.11 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 1.1565 2.313 3.4695 4.626 5.7825 2.19 1.67 3.67 4.04 4.11 4.48 5.14 4.00 5.00 4.59 3.55 3.89 3.02 1.90
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 60 120 180 240 300 Min: 188.8 / Avg: 230.31 / Max: 234.56 Min: 121.26 / Avg: 124.35 / Max: 125.39 Min: 248.76 / Avg: 305.1 / Max: 317.63 Min: 253.16 / Avg: 332.56 / Max: 350.88 Min: 253.27 / Avg: 346.63 / Max: 357.36 Min: 250.31 / Avg: 336.24 / Max: 345.82 Min: 256.63 / Avg: 350.63 / Max: 362.54 Min: 244.6 / Avg: 325.84 / Max: 336.51 Min: 252.42 / Avg: 334.18 / Max: 348.23 Min: 242.72 / Avg: 319.05 / Max: 326.26 Min: 207.33 / Avg: 242.24 / Max: 245.7 Min: 204.43 / Avg: 229.82 / Max: 232.47 Min: 176.68 / Avg: 184.83 / Max: 187.03 Min: 104.11 / Avg: 107.11 / Max: 107.86 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
N-Queens This is a test of the OpenMP version of a test that solves the N-queens problem. The board problem size is 18. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better N-Queens 1.0 Elapsed Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 SE +/- 0.000, N = 6 SE +/- 0.000, N = 4 SE +/- 0.001, N = 10 SE +/- 0.001, N = 10 SE +/- 0.001, N = 9 SE +/- 0.000, N = 8 SE +/- 0.001, N = 8 SE +/- 0.001, N = 8 SE +/- 0.001, N = 7 SE +/- 0.000, N = 5 SE +/- 0.004, N = 5 SE +/- 0.000, N = 4 SE +/- 0.001, N = 3 7.152 14.298 2.802 2.787 3.352 4.273 4.710 4.599 5.645 8.449 8.785 11.618 17.421 1. (CC) gcc options: -static -fopenmp -O3 -march=native
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better N-Queens 1.0 Elapsed Time EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 Min: 7.15 / Avg: 7.15 / Max: 7.15 Min: 14.3 / Avg: 14.3 / Max: 14.3 Min: 2.8 / Avg: 2.8 / Max: 2.81 Min: 2.78 / Avg: 2.79 / Max: 2.79 Min: 3.35 / Avg: 3.35 / Max: 3.36 Min: 4.27 / Avg: 4.27 / Max: 4.27 Min: 4.71 / Avg: 4.71 / Max: 4.71 Min: 4.6 / Avg: 4.6 / Max: 4.6 Min: 5.64 / Avg: 5.65 / Max: 5.65 Min: 8.45 / Avg: 8.45 / Max: 8.45 Min: 8.78 / Avg: 8.79 / Max: 8.8 Min: 11.62 / Avg: 11.62 / Max: 11.62 Min: 17.42 / Avg: 17.42 / Max: 17.42 1. (CC) gcc options: -static -fopenmp -O3 -march=native
Tungsten Renderer Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Non-Exponential EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 SE +/- 0.01335, N = 7 SE +/- 0.02852, N = 5 SE +/- 0.00885, N = 9 SE +/- 0.01113, N = 9 SE +/- 0.06689, N = 15 SE +/- 0.01129, N = 9 SE +/- 0.00498, N = 9 SE +/- 0.00744, N = 8 SE +/- 0.00918, N = 9 SE +/- 0.01244, N = 7 SE +/- 0.00686, N = 6 SE +/- 0.01670, N = 7 SE +/- 0.01455, N = 5 SE +/- 0.02902, N = 4 5.84607 10.90090 2.88175 2.80142 3.19889 3.31262 3.47096 3.70234 3.56555 5.71426 6.36513 6.21815 9.83106 11.96540 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Non-Exponential EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 5.8 / Avg: 5.85 / Max: 5.89 Min: 10.84 / Avg: 10.9 / Max: 11 Min: 2.85 / Avg: 2.88 / Max: 2.93 Min: 2.77 / Avg: 2.8 / Max: 2.88 Min: 2.55 / Avg: 3.2 / Max: 3.35 Min: 3.27 / Avg: 3.31 / Max: 3.37 Min: 3.44 / Avg: 3.47 / Max: 3.49 Min: 3.66 / Avg: 3.7 / Max: 3.73 Min: 3.5 / Avg: 3.57 / Max: 3.61 Min: 5.67 / Avg: 5.71 / Max: 5.76 Min: 6.35 / Avg: 6.37 / Max: 6.39 Min: 6.17 / Avg: 6.22 / Max: 6.29 Min: 9.79 / Avg: 9.83 / Max: 9.88 Min: 11.89 / Avg: 11.97 / Max: 12.03 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz -ljpeg -lpthread -ldl
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 900 1800 2700 3600 4500 SE +/- 0.69, N = 6 SE +/- 0.27, N = 4 SE +/- 13.78, N = 10 SE +/- 10.02, N = 10 SE +/- 6.29, N = 9 SE +/- 9.01, N = 9 SE +/- 4.60, N = 8 SE +/- 5.08, N = 8 SE +/- 3.24, N = 8 SE +/- 1.48, N = 7 SE +/- 0.45, N = 6 SE +/- 0.78, N = 6 SE +/- 0.17, N = 5 SE +/- 0.60, N = 4 1407.62 704.69 3908.23 3967.28 3310.34 3252.13 2375.48 2318.44 2368.69 1806.54 1191.90 1155.88 867.81 578.12 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10 20 30 40 50 11.81 9.02 41.11 42.58 32.45 36.28 28.10 22.61 27.29 21.02 16.06 17.46 13.77 10.52
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 700 1400 2100 2800 3500 Min: 1404.92 / Avg: 1407.62 / Max: 1409.44 Min: 703.93 / Avg: 704.69 / Max: 705.19 Min: 3840.34 / Avg: 3908.23 / Max: 3966.84 Min: 3922.41 / Avg: 3967.28 / Max: 4021.83 Min: 3286.05 / Avg: 3310.34 / Max: 3336.05 Min: 3210.93 / Avg: 3252.13 / Max: 3290.8 Min: 2361.46 / Avg: 2375.48 / Max: 2394.12 Min: 2296.55 / Avg: 2318.44 / Max: 2338.82 Min: 2349.79 / Avg: 2368.69 / Max: 2379.63 Min: 1799.65 / Avg: 1806.54 / Max: 1812.17 Min: 1190.44 / Avg: 1191.9 / Max: 1193.16 Min: 1152.71 / Avg: 1155.88 / Max: 1157.48 Min: 867.17 / Avg: 867.81 / Max: 868.12 Min: 576.33 / Avg: 578.12 / Max: 578.84 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
x264 This is a simple test of the x264 encoder run on the CPU (OpenCL support disabled) with a sample video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better x264 2019-12-17 H.264 Video Encoding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 50 100 150 200 250 SE +/- 0.76, N = 9 SE +/- 0.33, N = 7 SE +/- 1.78, N = 15 SE +/- 1.83, N = 15 SE +/- 1.24, N = 15 SE +/- 1.18, N = 15 SE +/- 0.83, N = 9 SE +/- 0.75, N = 9 SE +/- 1.00, N = 9 SE +/- 0.75, N = 9 SE +/- 0.59, N = 8 SE +/- 0.71, N = 8 SE +/- 0.60, N = 7 SE +/- 0.37, N = 6 173.50 95.50 198.58 210.69 203.74 198.00 188.45 181.15 182.42 174.63 150.45 142.37 117.14 78.21 1. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better x264 2019-12-17 H.264 Video Encoding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.5355 1.071 1.6065 2.142 2.6775 1.47 1.14 2.06 2.22 2.08 2.30 2.38 1.93 2.32 2.12 2.01 2.16 1.74 1.31
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better x264 2019-12-17 H.264 Video Encoding EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 40 80 120 160 200 Min: 167.66 / Avg: 173.5 / Max: 175.1 Min: 94.18 / Avg: 95.5 / Max: 96.44 Min: 179.11 / Avg: 198.58 / Max: 206.01 Min: 187.36 / Avg: 210.69 / Max: 218.31 Min: 187.17 / Avg: 203.74 / Max: 207.43 Min: 182.68 / Avg: 198 / Max: 201.65 Min: 183.08 / Avg: 188.45 / Max: 190.79 Min: 175.43 / Avg: 181.15 / Max: 183.15 Min: 175.43 / Avg: 182.42 / Max: 186.12 Min: 169.05 / Avg: 174.63 / Max: 176.72 Min: 146.52 / Avg: 150.45 / Max: 151.9 Min: 137.77 / Avg: 142.37 / Max: 143.62 Min: 113.91 / Avg: 117.14 / Max: 119.03 Min: 76.63 / Avg: 78.21 / Max: 79.16 1. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 11K 22K 33K 44K 55K SE +/- 80.46, N = 5 SE +/- 20.16, N = 8 SE +/- 224.56, N = 8 SE +/- 122.43, N = 8 SE +/- 86.13, N = 8 SE +/- 218.93, N = 8 SE +/- 76.06, N = 8 SE +/- 17.34, N = 8 SE +/- 23.57, N = 6 SE +/- 37.32, N = 7 21714.65 45336.89 51795.61 52245.23 44205.80 52022.76 44082.95 47349.29 29776.81 30311.94 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Total Mop/s Per Watt
OpenBenchmarking.org Total Mop/s Per Watt, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 140 280 420 560 700 197.31 592.89 476.11 490.40 555.62 520.26 538.87 660.02 463.71 525.61
Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7302P EPYC 7282 EPYC 7232P 9K 18K 27K 36K 45K Min: 21536.77 / Avg: 21714.65 / Max: 22018.31 Min: 45233.69 / Avg: 45336.89 / Max: 45424.87 Min: 50825.97 / Avg: 51795.61 / Max: 52696.05 Min: 51776.39 / Avg: 52245.23 / Max: 52860.72 Min: 43844.26 / Avg: 44205.8 / Max: 44502.94 Min: 50688.93 / Avg: 52022.76 / Max: 52586.47 Min: 43772.91 / Avg: 44082.95 / Max: 44328.66 Min: 47278.3 / Avg: 47349.29 / Max: 47425.88 Min: 29684.13 / Avg: 29776.81 / Max: 29837.83 Min: 30136.45 / Avg: 30311.94 / Max: 30404.24 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.3
Etcpak Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: DXT1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 SE +/- 2.05, N = 8 SE +/- 0.24, N = 8 SE +/- 1.34, N = 7 SE +/- 2.20, N = 7 SE +/- 1.97, N = 7 SE +/- 1.94, N = 7 SE +/- 2.27, N = 7 SE +/- 1.96, N = 7 SE +/- 1.99, N = 7 SE +/- 1.74, N = 7 SE +/- 2.23, N = 7 SE +/- 1.94, N = 7 SE +/- 1.98, N = 7 1179.02 1179.30 1019.21 1007.22 1008.24 1035.23 1006.92 1022.48 1018.97 1007.95 975.22 978.85 976.59 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
Mpx/s Per Watt
OpenBenchmarking.org Mpx/s Per Watt, More Is Better Etcpak 0.7 Configuration: DXT1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 6 12 18 24 30 19.82 26.64 17.07 17.06 20.49 24.75 17.36 24.28 24.45 23.75 26.39 27.10 27.45
Result Confidence
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 0.7 Configuration: DXT1 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 1173.37 / Avg: 1179.02 / Max: 1186.77 Min: 1178.67 / Avg: 1179.3 / Max: 1180.64 Min: 1017.42 / Avg: 1019.21 / Max: 1027.18 Min: 1002.3 / Avg: 1007.22 / Max: 1013.81 Min: 1002.46 / Avg: 1008.24 / Max: 1013.3 Min: 1031.59 / Avg: 1035.22 / Max: 1042.84 Min: 999.92 / Avg: 1006.92 / Max: 1013.6 Min: 1017.68 / Avg: 1022.48 / Max: 1028.23 Min: 1012.27 / Avg: 1018.97 / Max: 1026.96 Min: 1001.21 / Avg: 1007.95 / Max: 1011.76 Min: 970.09 / Avg: 975.22 / Max: 983.12 Min: 972.88 / Avg: 978.85 / Max: 983.32 Min: 970.13 / Avg: 976.59 / Max: 981.11 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
FFTE FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K SE +/- 308.97, N = 9 SE +/- 37.94, N = 10 SE +/- 1033.87, N = 15 SE +/- 1366.04, N = 15 SE +/- 641.51, N = 9 SE +/- 375.36, N = 8 SE +/- 143.74, N = 9 SE +/- 154.09, N = 8 SE +/- 160.95, N = 8 SE +/- 134.09, N = 10 SE +/- 80.44, N = 9 SE +/- 71.59, N = 9 SE +/- 22.37, N = 11 SE +/- 139.55, N = 9 71767.96 54800.99 145868.29 155375.98 149967.74 135838.12 144583.87 135786.94 136276.50 111563.84 84316.23 76796.04 61838.52 44242.98 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
MFLOPS Per Watt
OpenBenchmarking.org MFLOPS Per Watt, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 604.59 805.56 1260.43 1389.25 1309.22 1263.45 1396.95 1272.00 1457.28 1403.88 1116.04 1200.32 1139.68 839.58
Result Confidence
OpenBenchmarking.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 30K 60K 90K 120K 150K Min: 70448.73 / Avg: 71767.96 / Max: 73371.09 Min: 54543.54 / Avg: 54800.99 / Max: 54955.54 Min: 143378.84 / Avg: 145868.29 / Max: 159924 Min: 151616.85 / Avg: 155375.98 / Max: 173605.93 Min: 148789.93 / Avg: 149967.74 / Max: 154980.74 Min: 133822.4 / Avg: 135838.12 / Max: 137193.41 Min: 143957.47 / Avg: 144583.87 / Max: 145193.47 Min: 135207.86 / Avg: 135786.94 / Max: 136435.92 Min: 135695.57 / Avg: 136276.5 / Max: 136949.63 Min: 110544.87 / Avg: 111563.84 / Max: 112037.3 Min: 83921.86 / Avg: 84316.23 / Max: 84643.19 Min: 76508.11 / Avg: 76796.04 / Max: 77062.73 Min: 61730.3 / Avg: 61838.52 / Max: 61950.09 Min: 43253.52 / Avg: 44242.98 / Max: 44735.98 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.00511, N = 9 SE +/- 0.00118, N = 9 SE +/- 0.03302, N = 15 SE +/- 0.02912, N = 15 SE +/- 0.01930, N = 15 SE +/- 0.00950, N = 9 SE +/- 0.00582, N = 9 SE +/- 0.00881, N = 9 SE +/- 0.00414, N = 9 SE +/- 0.00636, N = 9 SE +/- 0.00474, N = 9 SE +/- 0.00482, N = 9 SE +/- 0.00724, N = 9 SE +/- 0.00735, N = 9 3.92737 6.46398 2.82015 2.52363 2.40380 2.55335 3.44118 3.60130 3.68833 3.34633 4.72608 5.08328 6.29135 8.27184 MIN: 3.81 MIN: 6.38 MIN: 2.49 MIN: 2.25 MIN: 2.23 MIN: 2.41 MIN: 3.03 MIN: 3.22 MIN: 3.26 MIN: 3.15 MIN: 4.52 MIN: 4.73 MIN: 6.07 MIN: 8 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 3.89 / Avg: 3.93 / Max: 3.94 Min: 6.46 / Avg: 6.46 / Max: 6.47 Min: 2.71 / Avg: 2.82 / Max: 3.12 Min: 2.43 / Avg: 2.52 / Max: 2.85 Min: 2.29 / Avg: 2.4 / Max: 2.52 Min: 2.52 / Avg: 2.55 / Max: 2.62 Min: 3.42 / Avg: 3.44 / Max: 3.47 Min: 3.57 / Avg: 3.6 / Max: 3.64 Min: 3.67 / Avg: 3.69 / Max: 3.7 Min: 3.33 / Avg: 3.35 / Max: 3.39 Min: 4.71 / Avg: 4.73 / Max: 4.74 Min: 5.06 / Avg: 5.08 / Max: 5.11 Min: 6.26 / Avg: 6.29 / Max: 6.32 Min: 8.23 / Avg: 8.27 / Max: 8.29 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.6833 1.3666 2.0499 2.7332 3.4165 SE +/- 0.002, N = 10 SE +/- 0.001, N = 10 SE +/- 0.003, N = 9 SE +/- 0.003, N = 9 SE +/- 0.003, N = 9 SE +/- 0.002, N = 9 SE +/- 0.001, N = 9 SE +/- 0.001, N = 9 SE +/- 0.003, N = 9 SE +/- 0.002, N = 9 SE +/- 0.002, N = 9 SE +/- 0.002, N = 9 SE +/- 0.001, N = 9 SE +/- 0.002, N = 9 2.500 2.495 2.905 2.946 2.952 2.945 2.858 2.945 2.899 2.901 2.945 3.037 3.029 3.035 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Result Confidence
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 2.49 / Avg: 2.5 / Max: 2.51 Min: 2.49 / Avg: 2.49 / Max: 2.5 Min: 2.89 / Avg: 2.9 / Max: 2.92 Min: 2.94 / Avg: 2.95 / Max: 2.97 Min: 2.94 / Avg: 2.95 / Max: 2.97 Min: 2.94 / Avg: 2.94 / Max: 2.95 Min: 2.85 / Avg: 2.86 / Max: 2.87 Min: 2.94 / Avg: 2.94 / Max: 2.95 Min: 2.89 / Avg: 2.9 / Max: 2.92 Min: 2.89 / Avg: 2.9 / Max: 2.91 Min: 2.94 / Avg: 2.94 / Max: 2.95 Min: 3.03 / Avg: 3.04 / Max: 3.05 Min: 3.02 / Avg: 3.03 / Max: 3.03 Min: 3.03 / Avg: 3.04 / Max: 3.05 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 SE +/- 0.00628, N = 9 SE +/- 0.00176, N = 9 SE +/- 0.00637, N = 9 SE +/- 0.00834, N = 9 SE +/- 0.00625, N = 9 SE +/- 0.00157, N = 9 SE +/- 0.00215, N = 9 SE +/- 0.00104, N = 9 SE +/- 0.00338, N = 9 SE +/- 0.00168, N = 9 SE +/- 0.00304, N = 9 SE +/- 0.00324, N = 9 SE +/- 0.00336, N = 9 SE +/- 0.00235, N = 9 3.02897 5.58260 1.49446 1.43124 1.53915 1.55494 1.89624 1.98809 1.94845 2.38306 3.34663 3.45290 4.52758 6.79445 MIN: 2.92 MIN: 5.52 MIN: 1.36 MIN: 1.26 MIN: 1.37 MIN: 1.42 MIN: 1.77 MIN: 1.83 MIN: 1.82 MIN: 2.33 MIN: 3.25 MIN: 3.34 MIN: 4.47 MIN: 6.75 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 3 6 9 12 15 Min: 2.99 / Avg: 3.03 / Max: 3.05 Min: 5.57 / Avg: 5.58 / Max: 5.59 Min: 1.48 / Avg: 1.49 / Max: 1.53 Min: 1.41 / Avg: 1.43 / Max: 1.48 Min: 1.52 / Avg: 1.54 / Max: 1.57 Min: 1.55 / Avg: 1.55 / Max: 1.56 Min: 1.89 / Avg: 1.9 / Max: 1.91 Min: 1.98 / Avg: 1.99 / Max: 1.99 Min: 1.94 / Avg: 1.95 / Max: 1.97 Min: 2.38 / Avg: 2.38 / Max: 2.39 Min: 3.33 / Avg: 3.35 / Max: 3.36 Min: 3.44 / Avg: 3.45 / Max: 3.47 Min: 4.51 / Avg: 4.53 / Max: 4.54 Min: 6.79 / Avg: 6.79 / Max: 6.81 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 SE +/- 0.028, N = 11 SE +/- 0.003, N = 9 SE +/- 0.257, N = 15 SE +/- 0.301, N = 15 SE +/- 0.291, N = 15 SE +/- 0.286, N = 15 SE +/- 0.252, N = 15 SE +/- 0.178, N = 15 SE +/- 0.286, N = 15 SE +/- 0.089, N = 15 SE +/- 0.024, N = 10 SE +/- 0.026, N = 10 SE +/- 0.008, N = 9 SE +/- 0.002, N = 8 11.521 6.470 19.997 21.763 19.328 18.506 16.564 16.265 15.685 14.038 10.374 9.685 7.582 5.237 1. (CXX) g++ options: -O3 -pthread -lm
ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.0653 0.1306 0.1959 0.2612 0.3265 0.14 0.10 0.25 0.29 0.25 0.26 0.26 0.22 0.25 0.23 0.18 0.18 0.15 0.11
Result Confidence
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 5 10 15 20 25 Min: 11.4 / Avg: 11.52 / Max: 11.65 Min: 6.45 / Avg: 6.47 / Max: 6.49 Min: 19.07 / Avg: 20 / Max: 22.21 Min: 20.67 / Avg: 21.76 / Max: 24.47 Min: 18.26 / Avg: 19.33 / Max: 21.71 Min: 17.37 / Avg: 18.51 / Max: 20.7 Min: 14.32 / Avg: 16.56 / Max: 18.32 Min: 14.97 / Avg: 16.27 / Max: 17.08 Min: 13.34 / Avg: 15.69 / Max: 17.35 Min: 13.39 / Avg: 14.04 / Max: 14.46 Min: 10.19 / Avg: 10.37 / Max: 10.46 Min: 9.59 / Avg: 9.69 / Max: 9.82 Min: 7.55 / Avg: 7.58 / Max: 7.61 Min: 5.23 / Avg: 5.24 / Max: 5.25 1. (CXX) g++ options: -O3 -pthread -lm
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 100 200 300 400 500 SE +/- 0.36, N = 10 SE +/- 0.26, N = 8 SE +/- 3.60, N = 15 SE +/- 0.96, N = 10 SE +/- 0.61, N = 10 SE +/- 0.97, N = 10 SE +/- 0.74, N = 11 SE +/- 1.03, N = 11 SE +/- 0.76, N = 11 SE +/- 0.86, N = 11 SE +/- 0.29, N = 10 SE +/- 0.35, N = 10 SE +/- 0.45, N = 9 SE +/- 0.15, N = 8 263.13 167.08 409.36 448.08 458.24 445.92 459.14 426.52 437.88 413.09 316.49 295.80 239.47 144.32 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 2.76 2.34 5.28 5.97 6.14 6.57 7.38 5.70 7.27 6.59 5.07 5.47 4.25 2.72
Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 80 160 240 320 400 Min: 260.87 / Avg: 263.13 / Max: 264.67 Min: 165.84 / Avg: 167.08 / Max: 168.07 Min: 379.51 / Avg: 409.36 / Max: 426.74 Min: 442.15 / Avg: 448.08 / Max: 452.83 Min: 456.27 / Avg: 458.24 / Max: 462.61 Min: 441.83 / Avg: 445.92 / Max: 451.13 Min: 454.55 / Avg: 459.14 / Max: 462.25 Min: 419.29 / Avg: 426.52 / Max: 430.73 Min: 434.47 / Avg: 437.88 / Max: 443.13 Min: 407.89 / Avg: 413.09 / Max: 416.67 Min: 315.62 / Avg: 316.49 / Max: 317.97 Min: 293.54 / Avg: 295.8 / Max: 297.47 Min: 237.06 / Avg: 239.47 / Max: 241.16 Min: 143.85 / Avg: 144.32 / Max: 145.1 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 0.7261 1.4522 2.1783 2.9044 3.6305 SE +/- 0.007208, N = 11 SE +/- 0.019546, N = 10 SE +/- 0.002056, N = 13 SE +/- 0.001641, N = 13 SE +/- 0.001666, N = 13 SE +/- 0.001391, N = 13 SE +/- 0.005531, N = 12 SE +/- 0.006800, N = 12 SE +/- 0.006410, N = 12 SE +/- 0.003909, N = 12 SE +/- 0.010317, N = 11 SE +/- 0.005522, N = 11 SE +/- 0.006375, N = 9 SE +/- 0.016438, N = 9 1.715944 2.796277 0.773474 0.762238 0.902611 0.890726 1.134961 1.256882 1.149282 1.561688 1.873804 1.783589 2.965136 3.227103 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 2 4 6 8 10 Min: 1.66 / Avg: 1.72 / Max: 1.74 Min: 2.71 / Avg: 2.8 / Max: 2.88 Min: 0.76 / Avg: 0.77 / Max: 0.79 Min: 0.75 / Avg: 0.76 / Max: 0.77 Min: 0.89 / Avg: 0.9 / Max: 0.91 Min: 0.88 / Avg: 0.89 / Max: 0.9 Min: 1.11 / Avg: 1.13 / Max: 1.17 Min: 1.22 / Avg: 1.26 / Max: 1.3 Min: 1.12 / Avg: 1.15 / Max: 1.18 Min: 1.54 / Avg: 1.56 / Max: 1.59 Min: 1.83 / Avg: 1.87 / Max: 1.93 Min: 1.76 / Avg: 1.78 / Max: 1.82 Min: 2.92 / Avg: 2.97 / Max: 2.99 Min: 3.17 / Avg: 3.23 / Max: 3.29 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
TSCP This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300K 600K 900K 1200K 1500K SE +/- 791.31, N = 12 SE +/- 1531.45, N = 12 SE +/- 658.02, N = 11 SE +/- 1511.23, N = 11 SE +/- 587.46, N = 11 SE +/- 372.53, N = 12 SE +/- 986.25, N = 11 SE +/- 1366.07, N = 11 SE +/- 1181.68, N = 11 SE +/- 609.53, N = 11 SE +/- 1355.09, N = 11 SE +/- 863.61, N = 11 SE +/- 504.67, N = 11 1178140 1180682 1015858 997809 997461 1030295 999442 1014342 1013320 997133 968149 969066 969371 1. (CC) gcc options: -O3 -march=native
Nodes Per Second Per Watt
OpenBenchmarking.org Nodes Per Second Per Watt, More Is Better TSCP 1.81 AI Chess Performance EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 7K 14K 21K 28K 35K 23615.12 31397.39 19805.30 19457.25 23213.22 28093.43 20016.67 27717.32 27950.92 26682.94 29161.11 30478.57 30915.01
Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200K 400K 600K 800K 1000K Min: 1174366 / Avg: 1178140.42 / Max: 1181927 Min: 1166902 / Avg: 1180681.92 / Max: 1187021 Min: 1010601 / Avg: 1015858.45 / Max: 1018073 Min: 987057 / Avg: 997809 / Max: 1001414 Min: 994184 / Avg: 997460.73 / Max: 999597 Min: 1029491 / Avg: 1030294.92 / Max: 1033354 Min: 990607 / Avg: 999441.64 / Max: 1003238 Min: 1001414 / Avg: 1014342.27 / Max: 1018073 Min: 1003238 / Avg: 1013319.55 / Max: 1016195 Min: 994184 / Avg: 997132.73 / Max: 999597 Min: 956211 / Avg: 968148.55 / Max: 971389 Min: 961218 / Avg: 969066.36 / Max: 971389 Min: 966277 / Avg: 969371.36 / Max: 971389 1. (CC) gcc options: -O3 -march=native
dav1d Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Summer Nature 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 300 600 900 1200 1500 SE +/- 1.44, N = 3 SE +/- 0.28, N = 3 SE +/- 1.60, N = 3 SE +/- 3.75, N = 3 SE +/- 1.10, N = 3 SE +/- 3.10, N = 3 SE +/- 0.47, N = 3 SE +/- 1.15, N = 3 SE +/- 1.87, N = 3 SE +/- 2.68, N = 3 SE +/- 0.73, N = 3 SE +/- 0.46, N = 3 SE +/- 0.90, N = 3 SE +/- 0.82, N = 3 651.36 453.36 1050.99 1193.66 1070.09 1044.25 937.44 880.57 932.75 847.40 634.51 644.23 555.11 419.19 MIN: 480.87 / MAX: 706.63 MIN: 400.76 / MAX: 488.59 MIN: 478.51 / MAX: 1166.67 MIN: 500.62 / MAX: 1329.43 MIN: 574.58 / MAX: 1186.8 MIN: 554.63 / MAX: 1157.48 MIN: 659.51 / MAX: 1052.41 MIN: 567.79 / MAX: 985.57 MIN: 659.65 / MAX: 1047.61 MIN: 566.75 / MAX: 939.62 MIN: 482.67 / MAX: 693.38 MIN: 503.48 / MAX: 704.1 MIN: 463.4 / MAX: 605.59 MIN: 370.7 / MAX: 456.84 1. (CC) gcc options: -pthread
FPS Per Watt
OpenBenchmarking.org FPS Per Watt, More Is Better dav1d 0.8.1 Video Input: Summer Nature 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 4 8 12 16 20 7.40 7.06 13.08 15.54 14.70 15.42 16.36 11.30 15.80 14.37 10.93 12.88 11.35 8.74
Result Confidence
OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Summer Nature 1080p EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 200 400 600 800 1000 Min: 648.52 / Avg: 651.36 / Max: 653.25 Min: 452.94 / Avg: 453.36 / Max: 453.9 Min: 1048.15 / Avg: 1050.99 / Max: 1053.7 Min: 1186.16 / Avg: 1193.66 / Max: 1197.47 Min: 1068.71 / Avg: 1070.09 / Max: 1072.27 Min: 1038.12 / Avg: 1044.25 / Max: 1048.12 Min: 936.52 / Avg: 937.44 / Max: 938.02 Min: 879.23 / Avg: 880.57 / Max: 882.86 Min: 929.58 / Avg: 932.75 / Max: 936.06 Min: 842.07 / Avg: 847.4 / Max: 850.64 Min: 633.06 / Avg: 634.51 / Max: 635.27 Min: 643.31 / Avg: 644.23 / Max: 644.7 Min: 553.68 / Avg: 555.11 / Max: 556.76 Min: 417.79 / Avg: 419.19 / Max: 420.62 1. (CC) gcc options: -pthread
Stream OpenBenchmarking.org MB/s Per Watt, More Is Better Stream 2013-01-17 Type: Triad EPYC 7402P EPYC 7302P EPYC 7232P 600 1200 1800 2400 3000 2788.51 2743.30 2060.05
Result
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 194.23, N = 5 SE +/- 59.32, N = 5 SE +/- 26.33, N = 5 SE +/- 21.10, N = 5 SE +/- 11.91, N = 5 SE +/- 36.22, N = 5 SE +/- 16.47, N = 5 SE +/- 19.43, N = 5 SE +/- 7.48, N = 5 SE +/- 5.88, N = 5 SE +/- 8.84, N = 5 SE +/- 2.27, N = 5 SE +/- 65.26, N = 5 72315.6 89567.2 98034.8 98343.2 96497.1 87057.0 99248.2 87239.8 87308.4 88105.7 55596.3 56066.6 56788.6 1. (CC) gcc options: -O3 -march=native -fopenmp
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Stream 2013-01-17 Type: Triad EPYC 7402P EPYC 7302P EPYC 7232P 600 1200 1800 2400 3000 2788.51 2743.30 2060.05
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K Min: 71942.4 / Avg: 72315.56 / Max: 73068.3 Min: 89468.9 / Avg: 89567.18 / Max: 89800.2 Min: 97939.6 / Avg: 98034.8 / Max: 98096.1 Min: 98280 / Avg: 98343.22 / Max: 98393.4 Min: 96474.4 / Avg: 96497.14 / Max: 96541 Min: 86972.9 / Avg: 87057.04 / Max: 87161.9 Min: 99194.2 / Avg: 99248.22 / Max: 99293.1 Min: 87168 / Avg: 87239.76 / Max: 87282.1 Min: 87291.9 / Avg: 87308.42 / Max: 87333.6 Min: 88086.3 / Avg: 88105.74 / Max: 88121.8 Min: 55564.4 / Avg: 55596.28 / Max: 55617.5 Min: 56061.4 / Avg: 56066.58 / Max: 56074.5 Min: 56528 / Avg: 56788.62 / Max: 56863.9 1. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.org MB/s Per Watt, More Is Better Stream 2013-01-17 Type: Add EPYC 7F52 EPYC 7702 500 1000 1500 2000 2500 1709.94 2144.89
Result
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 39.00, N = 5 SE +/- 18.23, N = 5 SE +/- 25.81, N = 5 SE +/- 29.12, N = 5 SE +/- 16.17, N = 5 SE +/- 16.89, N = 5 SE +/- 7.96, N = 5 SE +/- 13.36, N = 5 SE +/- 9.19, N = 5 SE +/- 17.97, N = 5 SE +/- 6.11, N = 5 SE +/- 9.00, N = 5 SE +/- 61.95, N = 5 72752.2 89255.4 97206.3 97180.9 95807.0 86737.7 97912.5 86677.2 86805.6 87568.8 55437.2 55911.6 56795.1 1. (CC) gcc options: -O3 -march=native -fopenmp
MB/s Per Watt
OpenBenchmarking.org MB/s Per Watt, More Is Better Stream 2013-01-17 Type: Add EPYC 7F52 EPYC 7702 500 1000 1500 2000 2500 1709.94 2144.89
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K Min: 72639.1 / Avg: 72752.2 / Max: 72844.7 Min: 89196.2 / Avg: 89255.38 / Max: 89299.1 Min: 97146.6 / Avg: 97206.28 / Max: 97299.6 Min: 97111 / Avg: 97180.94 / Max: 97268.6 Min: 95762.2 / Avg: 95807.04 / Max: 95857.9 Min: 86695.7 / Avg: 86737.72 / Max: 86780.2 Min: 97891 / Avg: 97912.52 / Max: 97938.6 Min: 86645.7 / Avg: 86677.24 / Max: 86711.4 Min: 86780.2 / Avg: 86805.64 / Max: 86837.1 Min: 87527.2 / Avg: 87568.78 / Max: 87617.1 Min: 55417 / Avg: 55437.24 / Max: 55454.2 Min: 55883.1 / Avg: 55911.58 / Max: 55937.4 Min: 56547.6 / Avg: 56795.1 / Max: 56866.8 1. (CC) gcc options: -O3 -march=native -fopenmp
Result
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 20K 40K 60K 80K 100K SE +/- 31.64, N = 5 SE +/- 30.51, N = 5 SE +/- 27.01, N = 5 SE +/- 21.55, N = 5 SE +/- 32.31, N = 5 SE +/- 14.00, N = 5 SE +/- 18.59, N = 5 SE +/- 12.22, N = 5 SE +/- 7.47, N = 5 SE +/- 10.74, N = 5 SE +/- 3.08, N = 5 SE +/- 4.61, N = 5 SE +/- 61.32, N = 5 67011.9 81753.0 87714.6 87848.1 86805.6 78399.9 89487.6 78638.6 79147.3 79703.5 51044.2 51978.2 52630.1 1. (CC) gcc options: -O3 -march=native -fopenmp
Result Confidence
OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 16K 32K 48K 64K 80K Min: 66926.2 / Avg: 67011.88 / Max: 67120.3 Min: 81720.5 / Avg: 81753 / Max: 81875 Min: 87618.6 / Avg: 87714.62 / Max: 87786 Min: 87786 / Avg: 87848.06 / Max: 87907.9 Min: 86725.2 / Avg: 86805.6 / Max: 86880.2 Min: 78366.2 / Avg: 78399.86 / Max: 78439.4 Min: 89445.1 / Avg: 89487.56 / Max: 89550.1 Min: 78592.9 / Avg: 78638.56 / Max: 78666.6 Min: 79125.7 / Avg: 79147.34 / Max: 79168.6 Min: 79669.6 / Avg: 79703.46 / Max: 79733 Min: 51038.4 / Avg: 51044.22 / Max: 51055.9 Min: 51965.2 / Avg: 51978.22 / Max: 51992.1 Min: 52385 / Avg: 52630.06 / Max: 52697.2 1. (CC) gcc options: -O3 -march=native -fopenmp
BlogBench OpenBenchmarking.org Final Score Per Watt, More Is Better BlogBench 1.1 Test: Write EPYC 7702 EPYC 7302P 200 400 600 800 1000 984.94 1143.48
Result
OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 10K 20K 30K 40K 50K SE +/- 379.22, N = 3 SE +/- 648.32, N = 3 SE +/- 312.03, N = 3 SE +/- 249.84, N = 3 SE +/- 2455.63, N = 3 SE +/- 2893.83, N = 3 SE +/- 1923.72, N = 3 SE +/- 1222.34, N = 3 SE +/- 1515.78, N = 3 SE +/- 1363.32, N = 3 SE +/- 448.51, N = 3 SE +/- 361.51, N = 3 SE +/- 66.78, N = 3 SE +/- 510.12, N = 3 36834 27443 44434 45477 39319 41586 43408 39412 43733 42491 37026 33627 28663 21916 1. (CC) gcc options: -O2 -pthread
Final Score Per Watt
OpenBenchmarking.org Final Score Per Watt, More Is Better BlogBench 1.1 Test: Write EPYC 7702 EPYC 7302P 200 400 600 800 1000 984.94 1143.48
Result Confidence
OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write EPYC 7F52 EPYC 7F32 EPYC 7702 EPYC 7662 EPYC 7642 EPYC 7552 EPYC 7542 EPYC 7532 EPYC 7502P EPYC 7402P EPYC 7302P EPYC 7282 EPYC 7272 EPYC 7232P 8K 16K 24K 32K 40K Min: 36386 / Avg: 36834 / Max: 37588 Min: 26499 / Avg: 27443.33 / Max: 28685 Min: 43811 / Avg: 44434.33 / Max: 44772 Min: 44984 / Avg: 45477 / Max: 45794 Min: 34451 / Avg: 39318.67 / Max: 42318 Min: 35861 / Avg: 41586 / Max: 45184 Min: 39566 / Avg: 43408.33 / Max: 45501 Min: 36967 / Avg: 39411.67 / Max: 40640 Min: 40707 / Avg: 43733 / Max: 45405 Min: 39807 / Avg: 42491 / Max: 44249 Min: 36148 / Avg: 37025.67 / Max: 37625 Min: 33112 / Avg: 33627 / Max: 34324 Min: 28551 / Avg: 28663 / Max: 28782 Min: 20947 / Avg: 21916 / Max: 22677 1. (CC) gcc options: -O2 -pthread
EPYC 7702 Processor: AMD EPYC 7702 64-Core @ 2.00GHz (64 Cores / 128 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 1 February 2021 14:42 by user root.
EPYC 7402P Processor: AMD EPYC 7402P 24-Core @ 2.80GHz (24 Cores / 48 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 3 February 2021 10:08 by user root.
EPYC 7302P Processor: AMD EPYC 7302P 16-Core @ 3.00GHz (16 Cores / 32 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 4 February 2021 15:16 by user root.
EPYC 7232P Processor: AMD EPYC 7232P 8-Core @ 3.10GHz (8 Cores / 16 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 6 February 2021 05:20 by user root.
EPYC 7552 Processor: AMD EPYC 7552 48-Core @ 2.20GHz (48 Cores / 96 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 7 February 2021 14:57 by user root.
EPYC 7272 Processor: AMD EPYC 7272 12-Core @ 2.90GHz (12 Cores / 24 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 8 February 2021 18:51 by user root.
EPYC 7662 Processor: AMD EPYC 7662 64-Core @ 2.00GHz (64 Cores / 128 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: llvmpipe, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 10 February 2021 06:13 by user root.
EPYC 7502P Processor: AMD EPYC 7502P 32-Core @ 2.50GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 February 2021 11:19 by user root.
EPYC 7F52 Processor: AMD EPYC 7F52 16-Core @ 3.50GHz (16 Cores / 32 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 7 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 February 2021 15:29 by user root.
EPYC 7542 Processor: AMD EPYC 7542 32-Core @ 2.90GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 13 February 2021 20:39 by user root.
EPYC 7282 Processor: AMD EPYC 7282 16-Core @ 2.80GHz (16 Cores / 32 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 15 February 2021 05:59 by user root.
EPYC 7F32 Processor: AMD EPYC 7F32 8-Core @ 3.70GHz (8 Cores / 16 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 16 February 2021 12:26 by user root.
EPYC 7532 Processor: AMD EPYC 7532 32-Core @ 2.40GHz (32 Cores / 64 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 17 February 2021 20:08 by user root.
EPYC 7642 Processor: AMD EPYC 7642 48-Core @ 2.30GHz (48 Cores / 96 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 8 x 16384 MB DDR4-3200MT/s 18ASF2G72PDZ-3G2E1, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I350
OS: Ubuntu 20.04, Kernel: 5.11.0-051100rc6daily20210201-generic (x86_64) 20210131, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, OpenGL: 4.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034Java Notes: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 19 February 2021 08:25 by user root.