epyc-march

2 x AMD EPYC 7742 64-Core testing with a Supermicro H11DSi-NT v2.00 (2.1 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2103122-HA-EPYCMARCH14
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

Audio Encoding 5 Tests
AV1 6 Tests
Bioinformatics 5 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 5 Tests
C++ Boost Tests 6 Tests
Chess Test Suite 7 Tests
Timed Code Compilation 14 Tests
C/C++ Compiler Tests 36 Tests
Compression Tests 10 Tests
CPU Massive 46 Tests
Creator Workloads 49 Tests
Encoding 16 Tests
Finance 2 Tests
Fortran Tests 11 Tests
Game Development 6 Tests
HPC - High Performance Computing 34 Tests
Imaging 10 Tests
LAPACK (Linear Algebra Pack) Tests 4 Tests
Linear Algebra 2 Tests
Machine Learning 5 Tests
Molecular Dynamics 10 Tests
MPI Benchmarks 11 Tests
Multi-Core 58 Tests
NVIDIA GPU Compute 6 Tests
Intel oneAPI 6 Tests
OpenCL 2 Tests
OpenMPI Tests 21 Tests
Programmer / Developer System Benchmarks 18 Tests
Python Tests 7 Tests
Raytracing 6 Tests
Renderers 12 Tests
Rust Tests 2 Tests
Scientific Computing 22 Tests
Software Defined Radio 4 Tests
Server CPU Tests 26 Tests
Single-Threaded 12 Tests
Speech 4 Tests
Telephony 4 Tests
Video Encoding 11 Tests
Common Workstation Benchmarks 5 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 7742 2P
March 09 2021
  1 Day, 11 Hours, 43 Minutes
2P
March 11 2021
  6 Hours, 49 Minutes
2 x AMD EPYC 7742 64-Core
March 11 2021
  2 Hours, 16 Minutes
7742 2P Repeat
March 11 2021
  22 Hours, 50 Minutes
Invert Hiding All Results Option
  16 Hours, 55 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


epyc-march OpenBenchmarking.orgPhoronix Test Suite2 x AMD EPYC 7742 64-Core @ 2.25GHz (128 Cores / 256 Threads)Supermicro H11DSi-NT v2.00 (2.1 BIOS)AMD Starship/Matisse16 x 8192 MB DDR4-3200MT/s HMA81GR7CJR8N-XN3841GB Micron_9300_MTFDHAL3T8TDPASPEEDVGA HDMI2 x Intel 10G X550TUbuntu 20.045.8.0-44-generic (x86_64)X Server 1.20.8GCC 9.3.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDisplay ServerCompilerFile-SystemScreen ResolutionEpyc-march BenchmarksSystem Logs- Transparent Huge Pages: madvise- --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 -v - EPYC 7742 2P, 7742 2P Repeat: NONE / errors=remount-ro,relatime,rw / Block Size: 4096 - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x8301034 - EPYC 7742 2P, 7742 2P Repeat: OpenJDK Runtime Environment (build 11.0.10+9-Ubuntu-0ubuntu1.20.04) - EPYC 7742 2P, 2P, 7742 2P Repeat: Python 2.7.18 + Python 3.8.5- 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

epyc-march hpl: askap: tConvolve MT - Degriddingaskap: tConvolve MT - Griddingcloverleaf: Lagrangian-Eulerian Hydrodynamicsqe: AUSURF112blosc: blosclzior: 32MB - Default Test Directorygnuradio: Hilbert Transformgnuradio: FM Deemphasis Filtergnuradio: IIR Filtergnuradio: FIR Filtergnuradio: Signal Source (Cosine)gnuradio: Five Back to Back FIR Filtersrelion: Basic - CPUlczero: BLASlczero: Eigenbuild-gcc: Time To Compilelibgav1: Chimera 1080pnwchem: C240 Buckyballopenvkl: vklBenchmarkcompress-7zip: Compress Speed Testhpcg: ior: 16MB - Default Test Directoryblender: Barbershop - CPU-Onlywebp2: Quality 100, Lossless Compressionhmmer: Pfam Database Searchonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUbuild-erlang: Time To Compileonednn: Recurrent Neural Network Training - u8s8f32 - CPUneat: fftw: Float + SSE - 2D FFT Size 4096rodinia: OpenMP HotSpot3Dincompact3d: Cylinderjpegxl: PNG - 8yafaray: Total Time For Sample Scenedeepspeech: CPUnumpy: lammps: 20k Atomsluaradio: Complex Phaseluaradio: Hilbert Transformluaradio: FM Deemphasis Filterluaradio: Five Back to Back FIR Filterslibgav1: Summer Nature 1080pwebp2: Quality 95, Compression Effort 7onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - f32 - CPUmocassin: Dust 2D tau100.0fftw: Float + SSE - 2D FFT Size 2048ior: 8MB - Default Test Directoryonednn: Recurrent Neural Network Inference - u8s8f32 - CPUcompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedminife: Smallrodinia: OpenMP Streamclusterfftw: Stock - 2D FFT Size 4096parboil: OpenMP MRI Griddinglibgav1: Summer Nature 4Kbuild-llvm: Time To Compileopenfoam: Motorbike 60Mcompress-zstd: 8 - Decompression Speedcompress-zstd: 8 - Compression Speedgraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spaceparboil: OpenMP LBMcompress-zstd: 3, Long Mode - Decompression Speedcompress-zstd: 3, Long Mode - Compression Speedngspice: C2670graphics-magick: Rotatetungsten: Water Causticrodinia: OpenMP Leukocytecompress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedvpxenc: Speed 5asmfish: 1024 Hash Memory, 26 Depthqmcpack: simple-H2Ottsiod-renderer: Phong Rendering With Soft-Shadow Mappingwebp2: Quality 75, Compression Effort 7ngspice: C7552compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9kvazaar: Bosphorus 4K - Mediumavifenc: 2ior: 4MB - Default Test Directoryaskap: Hogbom Clean OpenMProdinia: OpenMP CFD Solvercouchdb: 100 - 1000 - 24jpegxl: PNG - 5mrbayes: Primate Phylogeny Analysisior: 2MB - Default Test Directoryospray: San Miguel - Path Tracerfinancebench: Bonds OpenMPbuild-linux-kernel: Time To Compilejpegxl: PNG - 7build-eigen: Time To Compilemontage: Mosaic of M17, K band, 1.5 deg x 1.5 degaskap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddingbuild-gdb: Time To Compilecompress-zstd: 8, Long Mode - Decompression Speedcompress-zstd: 8, Long Mode - Compression Speedluajit: Compositejpegxl-decode: 1jpegxl-decode: Alldacapobench: H2kvazaar: Bosphorus 4K - Very Fasttachyon: Total Timeaskap: tConvolve OpenMP - Degriddingaskap: tConvolve OpenMP - Griddingradiance: SMP Paralleldacapobench: Tradebeansbuild-wasmer: Time To Compilecompress-lz4: 9 - Decompression Speedcompress-lz4: 9 - Compression Speedcompress-lz4: 3 - Decompression Speedcompress-lz4: 3 - Compression Speedblender: Pabellon Barcelona - CPU-Onlybuild2: Time To Compileluxcorerender: DLSCbuild-godot: Time To Compileluxcorerender: Rainbow Colors and Prismfinancebench: Repo OpenMPgraphics-magick: Resizingavifenc: 0jpegxl: JPEG - 5himeno: Poisson Pressure Solvergraphics-magick: Sharpengraphics-magick: Enhancedgraphics-magick: Swirljpegxl: JPEG - 7amg: srslte: OFDM_Testkvazaar: Bosphorus 4K - Ultra Fastjpegxl: JPEG - 8liquid-dsp: 128 - 256 - 57srslte: PHY_DL_Testsrslte: PHY_DL_Testfftw: Stock - 2D FFT Size 2048dcraw: RAW To PPM Image Conversionblender: Classroom - CPU-Onlyx265: Bosphorus 1080plzbench: XZ 0 - Decompressionlzbench: XZ 0 - Compressionlibraw: Post-Processing Benchmarkespeak: Text-To-Speech Synthesiswebp: Quality 100, Lossless, Highest Compressiononednn: IP Shapes 3D - f32 - CPUcompress-gzip: Linux Source Tree Archiving To .tar.gzbuild-php: Time To Compilerav1e: 6aobench: 2048 x 2048 - Total Timeavifenc: 6quantlib: blender: Fishy Cat - CPU-Onlycompress-zstd: 3 - Decompression Speedcompress-zstd: 3 - Compression Speedgromacs: water_GMX50_bareavifenc: 6, Losslessetcpak: ETC2aom-av1: Speed 6 Realtimeospray: San Miguel - SciVisc-ray: Total Time - 4K, 16 Rays Per Pixelaom-av1: Speed 6 Two-Passx265: Bosphorus 4Ktungsten: Haircompress-lz4: 1 - Decompression Speedcompress-lz4: 1 - Compression Speedembree: Pathtracer ISPC - Asian Dragon Objembree: Pathtracer - Asian Dragon Objrodinia: OpenMP LavaMDmafft: Multiple Sequence Alignment - LSU RNAstockfish: Total Timesynthmark: VoiceMark_100onednn: Convolution Batch Shapes Auto - u8s8f32 - CPUrav1e: 10crafty: Elapsed Timeospray: XFrog Forest - Path Traceropenfoam: Motorbike 30Mtoybrot: OpenMPlzbench: Libdeflate 1 - Decompressionlzbench: Libdeflate 1 - Compressionblender: BMW27 - CPU-Onlybuild-apache: Time To Compileencode-ape: WAV To APElzbench: Brotli 2 - Decompressionlzbench: Brotli 2 - Compressionencode-ogg: WAV To Ogglzbench: Brotli 0 - Decompressionlzbench: Brotli 0 - Compressionrnnoise: lzbench: Zstd 8 - Decompressionlzbench: Zstd 8 - Compressionavifenc: 10, Losslesslzbench: Zstd 1 - Decompressionlzbench: Zstd 1 - Compressionlzbench: Crush 0 - Decompressionlzbench: Crush 0 - Compressionetcpak: ETC1 + Ditheringonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUsvt-av1: Enc Mode 4 - 1080petcpak: ETC1webp: Quality 100, Losslessnamd: ATPase Simulation - 327,506 Atomsdolfyn: Computational Fluid Dynamicsliquid-dsp: 256 - 256 - 57liquid-dsp: 8 - 256 - 57liquid-dsp: 64 - 256 - 57liquid-dsp: 32 - 256 - 57liquid-dsp: 16 - 256 - 57liquid-dsp: 4 - 256 - 57liquid-dsp: 2 - 256 - 57liquid-dsp: 1 - 256 - 57build-ffmpeg: Time To Compileaom-av1: Speed 8 Realtimeffte: N=256, 3D Complex FFT Routinenpb: EP.Dwebp2: Defaultkvazaar: Bosphorus 1080p - Ultra Fastencode-flac: WAV To FLACsvt-vp9: VMAF Optimized - Bosphorus 1080pospray: XFrog Forest - SciVisbuild-imagemagick: Time To Compileembree: Pathtracer ISPC - Asian Dragonencode-opus: WAV To Opus Encodex264: H.264 Video Encodingonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUembree: Pathtracer - Asian Dragonsvt-av1: Enc Mode 8 - 1080pdav1d: Summer Nature 1080ponednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUnpb: LU.Ctungsten: Non-Exponentialonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUembree: Pathtracer ISPC - Crownnpb: IS.Dpovray: Trace Timeospray: NASA Streamlines - Path Tracertoybrot: TBBbuild-mplayer: Time To Compileembree: Pathtracer - Crownkvazaar: Bosphorus 1080p - Mediumonednn: IP Shapes 3D - u8s8f32 - CPUwebp: Quality 100, Highest Compressiondacapobench: Jythonencode-mp3: WAV To MP3n-queens: Elapsed Timedav1d: Summer Nature 4Kmt-dgemm: Sustained Floating-Point Ratewebp2: Quality 100, Compression Effort 5npb: FT.Conednn: Deconvolution Batch shapes_3d - f32 - CPUparboil: OpenMP Stencilfftw: Float + SSE - 1D FFT Size 4096system-decompress-gzip: oidn: Memorialm-queens: Time To Solvesystem-decompress-zlib: pennant: sedovbigfftw: Stock - 1D FFT Size 4096onednn: Convolution Batch Shapes Auto - f32 - CPUlammps: Rhodopsin Proteinnpb: CG.Ctjbench: Decompression Throughputospray: Magnetic Reconnection - SciVissvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080petcpak: DXT1ospray: NASA Streamlines - SciVistungsten: Volumetric Causticpennant: leblancbigkvazaar: Bosphorus 1080p - Very Fastsvt-vp9: Visual Quality Optimized - Bosphorus 1080ptoybrot: C++ Taskssystem-decompress-xz: avifenc: 10toybrot: C++ Threadsrays1bench: Large Scenenpb: MG.Cwebp: Quality 100parboil: OpenMP CUTCPtscp: AI Chess Performancewebp: Defaultospray: Magnetic Reconnection - Path Tracernpb: EP.Cluajit: Jacobi Successive Over-Relaxationluajit: Dense LU Matrix Factorizationluajit: Sparse Matrix Multiplyluajit: Fast Fourier Transformluajit: Monte CarloEPYC 7742 2P2P2 x AMD EPYC 7742 64-Core7742 2P Repeat153.597117.005224.9823.361219.363491.5436.5744.6506.9554.93032.8400.8542.04839364198715.1131963.347333831525.6364480.63440.928398.6351245.811267.20190.7892910.8862.75218663112.829345.7868850.7065.66078.03710305.7231.977532.784.4346.8643.972.74251.4342923.302948.0623926509489.401281.862828.532.811312.669.9705387.5194.08074418.97200.816112.802975.71989.965092951.0181873092.0629.2169.46254323.600850.8502792.570.720.8523609311344.322581.879136.405130.44126.58122.7032.717485.74217.82310.627112.55863.73108.673445.106.7689585.74218721.5009.7794.90993.07737599.738291.891.3093205.6587.61178.9599.54594740.159.89093991.994826.83213.866486668.21910418.344.8610188.345.3664.21315.0361.40716.8952054.5937506960.11251.713962.2148198331199172151.3612474276679833333344.4522.91313560000083.0197.76160.750.51661.36983230.9935.07941.9761.5465041.88841.6681.46039.90212.1032015.934.934139.50118.4783.337.7543.3618.775.6169611001.59527.7236.328339.047430.2110.266190042987646.9054.680783.102675778710.1014.12517995920524.67714.36957116523.69948241523.1421497827.587134643538189224.1642.205392.869377.456236.72220.3600.2795220.2065525100000427203333270393333316165666678322766672132800001071433335359466719.74831.89150712.968196998426.043.272181.149.830340.0219.7415.66742.11859.150204.012.073472.1332944.972785.7861245.910.7129360.812990194294.651.722561.2103259.33233268.828.02830.30391010.26967.586864.433.089178.90450279.1031.770387.0528.4941217.72176051.362.884375.389561449393.60028.727.0832025.9847125.8728486873.90.72439529.00041060.19172.30442345.45363.961039.8131254.461363.949192136.84274.5642954.3114.2284039492.7872806.592.8630.83187510318131.855333.338223.34436.2751.2505.0555.83040.5433.2532.584.4343.0653.0101100000321813846287.5207.65550733333427276667269376666716180000008316133332132866671071666675357966781.79185.8242824.933.02982.62243.52781.569.232.69721.5523199.6561.668.43364.5660.62649.0612.95836.028.06435.432514124.227.400388643074.18940502994.271225.263788.1461.38437.6747.2506.9555.93090.2423.3541.4713333351251.201933.825.9558468.11408.8981355.403152.1685.37317541150.965348.5163880.7031.833534.584.6347.1643.83209.713125.8723926795485.131285.172825.833.98169.1796.1605306.8208.037842112.702977.71992.565488074.0686773090.0620.552355.8612791.970.946.283480.55219.37864.40109.158452.039.773205.8566.11200.4132.9799.326295499110278.843.4910373.045.166853.118291203173051.1712461386679796000022.7286.8204.56086.5993229.7841.9362.498662009.22908.55053.7139.56110868.79289.0732.51110.4685.59055677860814.15961206569165481416148783133443738289224.5992.199372.86260237.24520.4320.2830620.201148613.891462097885.522.065642.137580.714842176465.671.201633313.122.888228.864509671133.342.708985.544628442945.8757146890.61.9546128.18339810.961037.4603.92644673254.922.8640.85252910306551.8548108.431644.112811.621008.92210.57412.34OpenBenchmarking.org

HPL Linpack

HPL is a well known portable Linpack implementation for distributed memory systems. This test profile is testing HPL upstream directly, outside the scope of the HPC Challenge test profile also available through the Phoronix Test Suite (hpcc). The test profile attempts to generate an optimized HPL.dat input file based on the CPU/memory under test. The automated HPL.dat input generation is still being tuned and thus for now this test profile remains "experimental". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPL Linpack 2.3EPYC 7742 2P306090120150SE +/- 0.42, N = 3153.591. (CC) gcc options: -O2 -lopenblas -lm -pthread -lmpi

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.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingEPYC 7742 2P15003000450060007500SE +/- 208.57, N = 97117.001. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingEPYC 7742 2P11002200330044005500SE +/- 48.52, N = 95224.981. (CXX) g++ options: -O3 -fstrict-aliasing -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.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsEPYC 7742 2P7742 2P Repeat6001200180024003000SE +/- 0.38, N = 15SE +/- 0.03, N = 323.362994.271. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsEPYC 7742 2P7742 2P Repeat5001000150020002500Min: 20.3 / Avg: 23.36 / Max: 25.05Min: 2994.21 / Avg: 2994.27 / Max: 2994.321. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterQuantum ESPRESSO 6.7Input: AUSURF112EPYC 7742 2P7742 2P Repeat30060090012001500SE +/- 4.15, N = 3SE +/- 3.23, N = 31219.361225.261. (F9X) gfortran options: -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterQuantum ESPRESSO 6.7Input: AUSURF112EPYC 7742 2P7742 2P Repeat2004006008001000Min: 1212.69 / Avg: 1219.36 / Max: 1226.96Min: 1220.48 / Avg: 1225.26 / Max: 1231.411. (F9X) gfortran options: -lopenblas -lFoX_dom -lFoX_sax -lFoX_wxml -lFoX_common -lFoX_utils -lFoX_fsys -lfftw3 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

C-Blosc

A simple, compressed, fast and persistent data store library for C. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterC-Blosc 2.0 Beta 5Compressor: blosclz7742 2P RepeatEPYC 7742 2P8001600240032004000SE +/- 39.54, N = 3SE +/- 23.71, N = 33788.13491.51. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgMB/s, More Is BetterC-Blosc 2.0 Beta 5Compressor: blosclz7742 2P RepeatEPYC 7742 2P7001400210028003500Min: 3709.4 / Avg: 3788.07 / Max: 3834.4Min: 3445.9 / Avg: 3491.47 / Max: 3525.61. (CXX) g++ options: -rdynamic

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 32MB - Disk Target: Default Test Directory7742 2P Repeat100200300400500SE +/- 2.10, N = 3461.38MIN: 406.14 / MAX: 1020.591. (CC) gcc options: -O2 -lm -pthread -lmpi

GNU Radio

GNU Radio is a free software development toolkit providing signal processing blocks to implement software-defined radios (SDR) and signal processing systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Hilbert Transform7742 2P RepeatEPYC 7742 2P2P90180270360450SE +/- 0.91, N = 5SE +/- 0.61, N = 9SE +/- 0.53, N = 9437.6436.5436.21. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Hilbert Transform7742 2P RepeatEPYC 7742 2P2P80160240320400Min: 434.9 / Avg: 437.62 / Max: 440.2Min: 433.4 / Avg: 436.46 / Max: 438.9Min: 434 / Avg: 436.21 / Max: 438.21. 3.8.1.0

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FM Deemphasis Filter2P7742 2P RepeatEPYC 7742 2P160320480640800SE +/- 8.43, N = 9SE +/- 17.28, N = 5SE +/- 8.87, N = 9751.2747.2744.61. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FM Deemphasis Filter2P7742 2P RepeatEPYC 7742 2P130260390520650Min: 686 / Avg: 751.19 / Max: 770.6Min: 678.3 / Avg: 747.24 / Max: 768.7Min: 699.2 / Avg: 744.58 / Max: 769.11. 3.8.1.0

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: IIR Filter7742 2P RepeatEPYC 7742 2P2P110220330440550SE +/- 0.31, N = 5SE +/- 0.97, N = 9SE +/- 1.12, N = 9506.9506.9505.01. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: IIR Filter7742 2P RepeatEPYC 7742 2P2P90180270360450Min: 506.2 / Avg: 506.88 / Max: 507.9Min: 500.6 / Avg: 506.89 / Max: 510.5Min: 501.3 / Avg: 505 / Max: 510.81. 3.8.1.0

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FIR Filter7742 2P Repeat2PEPYC 7742 2P120240360480600SE +/- 0.41, N = 5SE +/- 1.19, N = 9SE +/- 0.84, N = 9555.9555.8554.91. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: FIR Filter7742 2P Repeat2PEPYC 7742 2P100200300400500Min: 554.4 / Avg: 555.94 / Max: 556.9Min: 551.7 / Avg: 555.8 / Max: 561.6Min: 549.3 / Avg: 554.89 / Max: 557.51. 3.8.1.0

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Signal Source (Cosine)7742 2P Repeat2PEPYC 7742 2P7001400210028003500SE +/- 29.08, N = 5SE +/- 19.01, N = 9SE +/- 23.25, N = 93090.23040.53032.81. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Signal Source (Cosine)7742 2P Repeat2PEPYC 7742 2P5001000150020002500Min: 2974.2 / Avg: 3090.24 / Max: 3126.7Min: 2941.8 / Avg: 3040.48 / Max: 3109.2Min: 2903.6 / Avg: 3032.81 / Max: 3150.31. 3.8.1.0

OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Five Back to Back FIR Filters2P7742 2P RepeatEPYC 7742 2P90180270360450SE +/- 12.19, N = 9SE +/- 4.29, N = 5SE +/- 11.37, N = 9433.2423.3400.81. 3.8.1.0
OpenBenchmarking.orgMiB/s, More Is BetterGNU RadioTest: Five Back to Back FIR Filters2P7742 2P RepeatEPYC 7742 2P80160240320400Min: 383.1 / Avg: 433.18 / Max: 498.1Min: 408.7 / Avg: 423.32 / Max: 432Min: 345.5 / Avg: 400.76 / Max: 445.61. 3.8.1.0

RELION

RELION - REgularised LIkelihood OptimisatioN - is a stand-alone computer program for Maximum A Posteriori refinement of (multiple) 3D reconstructions or 2D class averages in cryo-electron microscopy (cryo-EM). It is developed in the research group of Sjors Scheres at the MRC Laboratory of Molecular Biology. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 3.1.1Test: Basic - Device: CPU7742 2P RepeatEPYC 7742 2P120240360480600SE +/- 5.12, N = 6SE +/- 4.40, N = 3541.47542.051. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 3.1.1Test: Basic - Device: CPU7742 2P RepeatEPYC 7742 2P100200300400500Min: 535.98 / Avg: 541.47 / Max: 567.04Min: 537.48 / Avg: 542.05 / Max: 550.851. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: BLASEPYC 7742 2P7742 2P Repeat8001600240032004000SE +/- 49.55, N = 9SE +/- 37.92, N = 4393633331. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: BLASEPYC 7742 2P7742 2P Repeat7001400210028003500Min: 3628 / Avg: 3936.44 / Max: 4170Min: 3250 / Avg: 3333 / Max: 34131. (CXX) g++ options: -flto -pthread

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: EigenEPYC 7742 2P7742 2P Repeat9001800270036004500SE +/- 41.70, N = 3SE +/- 49.56, N = 9419835121. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.26Backend: EigenEPYC 7742 2P7742 2P Repeat7001400210028003500Min: 4116 / Avg: 4198.33 / Max: 4251Min: 3302 / Avg: 3512.11 / Max: 37401. (CXX) g++ options: -flto -pthread

Timed GCC Compilation

This test times how long it takes to build the GNU Compiler Collection (GCC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 9.3.0Time To CompileEPYC 7742 2P150300450600750SE +/- 0.46, N = 3715.11

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Chimera 1080p7742 2P Repeat1224364860SE +/- 0.37, N = 1151.201. (CXX) g++ options: -O3 -lpthread

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.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 Buckyball7742 2P RepeatEPYC 7742 2P4008001200160020001933.81963.31. (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

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkEPYC 7742 2P100200300400500473MIN: 1 / MAX: 1361

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.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 16.02Compress Speed TestEPYC 7742 2P70K140K210K280K350KSE +/- 3983.57, N = 153383151. (CXX) g++ options: -pipe -lpthread

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.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.17742 2P RepeatEPYC 7742 2P612182430SE +/- 0.20, N = 3SE +/- 0.28, N = 525.9625.641. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.17742 2P RepeatEPYC 7742 2P612182430Min: 25.56 / Avg: 25.96 / Max: 26.23Min: 24.94 / Avg: 25.64 / Max: 26.661. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 16MB - Disk Target: Default Test DirectoryEPYC 7742 2P7742 2P Repeat100200300400500SE +/- 1.21, N = 3SE +/- 5.31, N = 3480.63468.11MIN: 410.37 / MAX: 1034.82MIN: 411 / MAX: 1031.921. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 16MB - Disk Target: Default Test DirectoryEPYC 7742 2P7742 2P Repeat90180270360450Min: 478.62 / Avg: 480.63 / Max: 482.81Min: 458.5 / Avg: 468.11 / Max: 476.841. (CC) gcc options: -O2 -lm -pthread -lmpi

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CPU-Only2 x AMD EPYC 7742 64-Core20406080100SE +/- 0.90, N = 1581.79

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Lossless CompressionEPYC 7742 2P100200300400500SE +/- 0.44, N = 3440.931. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lgif

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchEPYC 7742 2P7742 2P Repeat90180270360450SE +/- 4.02, N = 3SE +/- 5.75, N = 3398.64408.901. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchEPYC 7742 2P7742 2P Repeat70140210280350Min: 390.6 / Avg: 398.64 / Max: 402.72Min: 397.54 / Avg: 408.9 / Max: 416.181. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUEPYC 7742 2P30060090012001500SE +/- 17.20, N = 151245.81MIN: 1111.621. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat30060090012001500SE +/- 15.39, N = 15SE +/- 34.94, N = 151267.201355.40MIN: 1101.26MIN: 1136.71. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat2004006008001000Min: 1159.63 / Avg: 1267.2 / Max: 1368.11Min: 1223.33 / Avg: 1355.4 / Max: 1655.671. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Timed Erlang/OTP Compilation

This test times how long it takes to compile Erlang/OTP. Erlang is a programming language and run-time for massively scalable soft real-time systems with high availability requirements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To Compile2 x AMD EPYC 7742 64-CoreEPYC 7742 2P4080120160200SE +/- 0.13, N = 3SE +/- 1.54, N = 9185.82190.79
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To Compile2 x AMD EPYC 7742 64-CoreEPYC 7742 2P306090120150Min: 185.67 / Avg: 185.82 / Max: 186.08Min: 186.01 / Avg: 190.79 / Max: 199.06

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat7001400210028003500SE +/- 63.06, N = 15SE +/- 92.93, N = 122910.883152.16MIN: 2232.96MIN: 2386.061. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat5001000150020002500Min: 2576.69 / Avg: 2910.88 / Max: 3298.24Min: 2598.8 / Avg: 3152.16 / Max: 3860.91. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterNebular Empirical Analysis Tool 2020-02-29EPYC 7742 2P7742 2P Repeat20406080100SE +/- 8.47, N = 15SE +/- 6.74, N = 1562.7585.371. (F9X) gfortran options: -cpp -ffree-line-length-0 -Jsource/ -fopenmp -O3 -fno-backtrace
OpenBenchmarking.orgSeconds, Fewer Is BetterNebular Empirical Analysis Tool 2020-02-29EPYC 7742 2P7742 2P Repeat1632486480Min: 30.59 / Avg: 62.75 / Max: 122.66Min: 28.35 / Avg: 85.37 / Max: 125.381. (F9X) gfortran options: -cpp -ffree-line-length-0 -Jsource/ -fopenmp -O3 -fno-backtrace

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.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096EPYC 7742 2P7742 2P Repeat4K8K12K16K20KSE +/- 230.17, N = 318663175411. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096EPYC 7742 2P7742 2P Repeat3K6K9K12K15KMin: 17085 / Avg: 17541.33 / Max: 178221. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DEPYC 7742 2P7742 2P Repeat306090120150SE +/- 1.17, N = 15SE +/- 0.67, N = 3112.83150.971. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DEPYC 7742 2P7742 2P Repeat306090120150Min: 103.77 / Avg: 112.83 / Max: 116.54Min: 149.69 / Avg: 150.97 / Max: 151.981. (CXX) g++ options: -O2 -lOpenCL

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterIncompact3D 2020-09-17Input: CylinderEPYC 7742 2P7742 2P Repeat80160240320400SE +/- 0.88, N = 3SE +/- 1.44, N = 3345.79348.521. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterIncompact3D 2020-09-17Input: CylinderEPYC 7742 2P7742 2P Repeat60120180240300Min: 344.21 / Avg: 345.79 / Max: 347.24Min: 346.98 / Avg: 348.52 / Max: 351.391. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: PNG - Encode Speed: 87742 2P RepeatEPYC 7742 2P0.15750.3150.47250.630.7875SE +/- 0.00, N = 3SE +/- 0.00, N = 30.700.701. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: PNG - Encode Speed: 87742 2P RepeatEPYC 7742 2P246810Min: 0.69 / Avg: 0.7 / Max: 0.7Min: 0.69 / Avg: 0.7 / Max: 0.71. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl

YafaRay

YafaRay is an open-source physically based montecarlo ray-tracing engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterYafaRay 3.4.1Total Time For Sample SceneEPYC 7742 2P1530456075SE +/- 0.61, N = 1565.661. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterDeepSpeech 0.6Acceleration: CPUEPYC 7742 2P20406080100SE +/- 2.69, N = 1578.04

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkEPYC 7742 2P70140210280350SE +/- 0.81, N = 3305.72

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.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsEPYC 7742 2P7742 2P Repeat714212835SE +/- 0.06, N = 3SE +/- 0.06, N = 331.9831.831. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsEPYC 7742 2P7742 2P Repeat714212835Min: 31.92 / Avg: 31.98 / Max: 32.09Min: 31.72 / Avg: 31.83 / Max: 31.931. (CXX) g++ options: -O3 -pthread -lm

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex Phase7742 2P RepeatEPYC 7742 2P2P120240360480600SE +/- 0.76, N = 3SE +/- 0.59, N = 3SE +/- 0.72, N = 4534.5532.7532.5
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex Phase7742 2P RepeatEPYC 7742 2P2P90180270360450Min: 533.1 / Avg: 534.53 / Max: 535.7Min: 531.6 / Avg: 532.73 / Max: 533.6Min: 531.5 / Avg: 532.48 / Max: 534.6

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert Transform7742 2P RepeatEPYC 7742 2P2P20406080100SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.11, N = 484.684.484.4
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert Transform7742 2P RepeatEPYC 7742 2P2P1632486480Min: 84.5 / Avg: 84.57 / Max: 84.6Min: 84.4 / Avg: 84.4 / Max: 84.4Min: 84.1 / Avg: 84.4 / Max: 84.6

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis Filter7742 2P RepeatEPYC 7742 2P2P80160240320400SE +/- 0.12, N = 3SE +/- 0.15, N = 3SE +/- 2.83, N = 4347.1346.8343.0
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis Filter7742 2P RepeatEPYC 7742 2P2P60120180240300Min: 346.9 / Avg: 347.07 / Max: 347.3Min: 346.5 / Avg: 346.77 / Max: 347Min: 334.9 / Avg: 342.95 / Max: 347.2

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR Filters2PEPYC 7742 2P7742 2P Repeat140280420560700SE +/- 7.57, N = 4SE +/- 4.40, N = 3SE +/- 3.58, N = 3653.0643.9643.8
OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR Filters2PEPYC 7742 2P7742 2P Repeat120240360480600Min: 638.4 / Avg: 653 / Max: 671.3Min: 635.2 / Avg: 643.9 / Max: 649.4Min: 638.1 / Avg: 643.83 / Max: 650.4

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Summer Nature 1080pEPYC 7742 2P1632486480SE +/- 0.95, N = 1572.741. (CXX) g++ options: -O3 -lpthread

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 95, Compression Effort 7EPYC 7742 2P50100150200250SE +/- 0.06, N = 3251.431. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lgif

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUEPYC 7742 2P7742 2P Repeat7001400210028003500SE +/- 22.66, N = 3SE +/- 91.94, N = 152923.303209.71MIN: 2522.76MIN: 2404.691. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUEPYC 7742 2P7742 2P Repeat6001200180024003000Min: 2881.24 / Avg: 2923.3 / Max: 2958.95Min: 2739.71 / Avg: 3209.71 / Max: 3854.261. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat7001400210028003500SE +/- 36.76, N = 3SE +/- 60.50, N = 152948.063125.87MIN: 2599.44MIN: 2374.921. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat5001000150020002500Min: 2882.52 / Avg: 2948.06 / Max: 3009.68Min: 2681.49 / Avg: 3125.87 / Max: 3495.911. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterMonte Carlo Simulations of Ionised Nebulae 2019-03-24Input: Dust 2D tau100.0EPYC 7742 2P7742 2P Repeat50100150200250SE +/- 0.33, N = 32392391. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterMonte Carlo Simulations of Ionised Nebulae 2019-03-24Input: Dust 2D tau100.0EPYC 7742 2P7742 2P Repeat4080120160200Min: 239 / Avg: 239.33 / Max: 2401. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

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.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 20487742 2P RepeatEPYC 7742 2P6K12K18K24K30KSE +/- 139.43, N = 3SE +/- 200.43, N = 1226795265091. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 20487742 2P RepeatEPYC 7742 2P5K10K15K20K25KMin: 26536 / Avg: 26795 / Max: 27014Min: 24506 / Avg: 26509.17 / Max: 273071. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 8MB - Disk Target: Default Test DirectoryEPYC 7742 2P7742 2P Repeat110220330440550SE +/- 1.12, N = 3SE +/- 3.76, N = 3489.40485.13MIN: 410.57 / MAX: 920.3MIN: 411.08 / MAX: 1041.221. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 8MB - Disk Target: Default Test DirectoryEPYC 7742 2P7742 2P Repeat90180270360450Min: 487.16 / Avg: 489.4 / Max: 490.59Min: 478.63 / Avg: 485.13 / Max: 491.651. (CC) gcc options: -O2 -lm -pthread -lmpi

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat30060090012001500SE +/- 19.98, N = 15SE +/- 14.75, N = 31281.861285.17MIN: 1116.9MIN: 1199.951. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat2004006008001000Min: 1169.05 / Avg: 1281.86 / Max: 1421.67Min: 1256.66 / Avg: 1285.17 / Max: 1305.991. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedEPYC 7742 2P7742 2P Repeat2 x AMD EPYC 7742 64-Core6001200180024003000SE +/- 3.12, N = 15SE +/- 4.52, N = 12SE +/- 2.77, N = 122828.52825.82824.91. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedEPYC 7742 2P7742 2P Repeat2 x AMD EPYC 7742 64-Core5001000150020002500Min: 2802.3 / Avg: 2828.51 / Max: 2848.3Min: 2792.5 / Avg: 2825.84 / Max: 2842.4Min: 2804.7 / Avg: 2824.91 / Max: 2834.81. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression Speed7742 2P Repeat2 x AMD EPYC 7742 64-CoreEPYC 7742 2P816243240SE +/- 0.66, N = 12SE +/- 0.52, N = 12SE +/- 0.58, N = 1533.933.032.81. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression Speed7742 2P Repeat2 x AMD EPYC 7742 64-CoreEPYC 7742 2P714212835Min: 30.6 / Avg: 33.94 / Max: 37.3Min: 30.7 / Avg: 32.97 / Max: 36.4Min: 28.7 / Avg: 32.77 / Max: 36.71. (CC) gcc options: -O3 -pthread -lz -llzma

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallEPYC 7742 2P7742 2P Repeat2K4K6K8K10KSE +/- 414.96, N = 15SE +/- 70.94, N = 311312.668169.171. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallEPYC 7742 2P7742 2P Repeat2K4K6K8K10KMin: 9177.96 / Avg: 11312.66 / Max: 14223.4Min: 8096.33 / Avg: 8169.17 / Max: 8311.041. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -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.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP StreamclusterEPYC 7742 2P7742 2P Repeat20406080100SE +/- 0.138, N = 15SE +/- 4.228, N = 129.97096.1601. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP StreamclusterEPYC 7742 2P7742 2P Repeat20406080100Min: 9.33 / Avg: 9.97 / Max: 10.73Min: 52.46 / Avg: 96.16 / Max: 103.311. (CXX) g++ options: -O2 -lOpenCL

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.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096EPYC 7742 2P7742 2P Repeat12002400360048006000SE +/- 13.54, N = 3SE +/- 54.29, N = 35387.55306.81. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096EPYC 7742 2P7742 2P Repeat9001800270036004500Min: 5373.1 / Avg: 5387.53 / Max: 5414.6Min: 5198.9 / Avg: 5306.77 / Max: 5371.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -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.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingEPYC 7742 2P7742 2P Repeat50100150200250SE +/- 1.09, N = 3SE +/- 1.80, N = 3194.08208.041. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingEPYC 7742 2P7742 2P Repeat4080120160200Min: 192 / Avg: 194.08 / Max: 195.68Min: 204.49 / Avg: 208.04 / Max: 210.321. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 2019-10-05Video Input: Summer Nature 4KEPYC 7742 2P510152025SE +/- 0.04, N = 318.971. (CXX) g++ options: -O3 -lpthread

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileEPYC 7742 2P4080120160200SE +/- 1.00, N = 3200.82

OpenFOAM

OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60M7742 2P RepeatEPYC 7742 2P306090120150SE +/- 0.16, N = 3SE +/- 0.06, N = 3112.70112.801. (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
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60M7742 2P RepeatEPYC 7742 2P20406080100Min: 112.45 / Avg: 112.7 / Max: 113Min: 112.69 / Avg: 112.8 / Max: 112.91. (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

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Decompression Speed2 x AMD EPYC 7742 64-Core7742 2P RepeatEPYC 7742 2P6001200180024003000SE +/- 2.98, N = 15SE +/- 3.43, N = 15SE +/- 3.84, N = 112982.62977.72975.71. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Decompression Speed2 x AMD EPYC 7742 64-Core7742 2P RepeatEPYC 7742 2P5001000150020002500Min: 2961.4 / Avg: 2982.61 / Max: 3000.9Min: 2960.3 / Avg: 2977.73 / Max: 2999Min: 2958 / Avg: 2975.7 / Max: 2998.61. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Compression Speed2 x AMD EPYC 7742 64-Core7742 2P RepeatEPYC 7742 2P5001000150020002500SE +/- 63.80, N = 15SE +/- 73.14, N = 15SE +/- 78.60, N = 122243.51992.51989.91. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8 - Compression Speed2 x AMD EPYC 7742 64-Core7742 2P RepeatEPYC 7742 2P400800120016002000Min: 1594 / Avg: 2243.51 / Max: 2499.7Min: 1312.9 / Avg: 1992.55 / Max: 2385.1Min: 1596.2 / Avg: 1989.9 / Max: 2595.31. (CC) gcc options: -O3 -pthread -lz -llzma

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-Gaussian7742 2P RepeatEPYC 7742 2P140280420560700SE +/- 4.16, N = 3SE +/- 6.04, N = 156546501. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Noise-Gaussian7742 2P RepeatEPYC 7742 2P120240360480600Min: 648 / Avg: 654 / Max: 662Min: 622 / Avg: 649.87 / Max: 6871. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color SpaceEPYC 7742 2P7742 2P Repeat2004006008001000SE +/- 13.26, N = 15SE +/- 4.06, N = 39298801. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: HWB Color SpaceEPYC 7742 2P7742 2P Repeat160320480640800Min: 868 / Avg: 929.47 / Max: 997Min: 873 / Avg: 879.67 / Max: 8871. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMEPYC 7742 2P7742 2P Repeat1632486480SE +/- 1.30, N = 15SE +/- 0.77, N = 351.0274.071. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMEPYC 7742 2P7742 2P Repeat1428425670Min: 38.02 / Avg: 51.02 / Max: 57.33Min: 73.25 / Avg: 74.07 / Max: 75.61. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3, Long Mode - Decompression SpeedEPYC 7742 2P7742 2P Repeat7001400210028003500SE +/- 2.93, N = 15SE +/- 2.35, N = 133092.03090.01. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3, Long Mode - Decompression SpeedEPYC 7742 2P7742 2P Repeat5001000150020002500Min: 3054.3 / Avg: 3092.05 / Max: 3101.7Min: 3063.7 / Avg: 3090.02 / Max: 30991. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3, Long Mode - Compression SpeedEPYC 7742 2P7742 2P Repeat140280420560700SE +/- 13.86, N = 15SE +/- 14.50, N = 13629.2620.51. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3, Long Mode - Compression SpeedEPYC 7742 2P7742 2P Repeat110220330440550Min: 528.4 / Avg: 629.21 / Max: 682.3Min: 507.2 / Avg: 620.52 / Max: 6771. (CC) gcc options: -O3 -pthread -lz -llzma

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670EPYC 7742 2P4080120160200SE +/- 0.82, N = 3169.461. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotateEPYC 7742 2P7742 2P Repeat120240360480600SE +/- 4.66, N = 8SE +/- 4.25, N = 95435231. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: RotateEPYC 7742 2P7742 2P Repeat100200300400500Min: 511 / Avg: 543 / Max: 551Min: 490 / Avg: 523 / Max: 5321. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Water CausticEPYC 7742 2P612182430SE +/- 0.32, N = 1523.601. (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

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteEPYC 7742 2P7742 2P Repeat1326395265SE +/- 0.49, N = 3SE +/- 0.43, N = 1550.8555.861. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteEPYC 7742 2P7742 2P Repeat1122334455Min: 49.91 / Avg: 50.85 / Max: 51.55Min: 53.64 / Avg: 55.86 / Max: 59.161. (CXX) g++ options: -O2 -lOpenCL

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedEPYC 7742 2P7742 2P Repeat2 x AMD EPYC 7742 64-Core6001200180024003000SE +/- 6.97, N = 3SE +/- 2.86, N = 15SE +/- 2.65, N = 152792.52791.92781.51. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedEPYC 7742 2P7742 2P Repeat2 x AMD EPYC 7742 64-Core5001000150020002500Min: 2781.6 / Avg: 2792.53 / Max: 2805.5Min: 2767.5 / Avg: 2791.92 / Max: 2811.9Min: 2768.1 / Avg: 2781.47 / Max: 27961. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression Speed7742 2P RepeatEPYC 7742 2P2 x AMD EPYC 7742 64-Core1632486480SE +/- 1.10, N = 15SE +/- 0.71, N = 3SE +/- 1.15, N = 1570.970.769.21. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression Speed7742 2P RepeatEPYC 7742 2P2 x AMD EPYC 7742 64-Core1428425670Min: 64.8 / Avg: 70.95 / Max: 78Min: 69.8 / Avg: 70.7 / Max: 72.1Min: 58.9 / Avg: 69.15 / Max: 761. (CC) gcc options: -O3 -pthread -lz -llzma

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format using a sample 1080p video. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.8.2Speed: Speed 5EPYC 7742 2P510152025SE +/- 0.23, N = 1520.851. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthEPYC 7742 2P50M100M150M200M250MSE +/- 1860963.69, N = 3236093113

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.10Input: simple-H2OEPYC 7742 2P7742 2P Repeat1020304050SE +/- 0.29, N = 3SE +/- 1.61, N = 1544.3246.281. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm
OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.10Input: simple-H2OEPYC 7742 2P7742 2P Repeat918273645Min: 43.76 / Avg: 44.32 / Max: 44.76Min: 43.66 / Avg: 46.28 / Max: 68.781. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm

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.

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3bPhong Rendering With Soft-Shadow MappingEPYC 7742 2P130260390520650SE +/- 9.68, N = 15581.881. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 75, Compression Effort 7EPYC 7742 2P306090120150SE +/- 0.15, N = 3136.411. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lgif

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552EPYC 7742 2P306090120150SE +/- 0.03, N = 3130.441. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterXZ Compression 5.2.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9EPYC 7742 2P612182430SE +/- 0.32, N = 1526.581. (CC) gcc options: -pthread -fvisibility=hidden -O2

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: MediumEPYC 7742 2P510152025SE +/- 0.19, N = 1522.701. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 22 x AMD EPYC 7742 64-CoreEPYC 7742 2P816243240SE +/- 0.22, N = 15SE +/- 0.28, N = 832.7032.721. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 22 x AMD EPYC 7742 64-CoreEPYC 7742 2P714212835Min: 31.69 / Avg: 32.7 / Max: 34.09Min: 31.85 / Avg: 32.72 / Max: 33.91. (CXX) g++ options: -O3 -fPIC -lm

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 4MB - Disk Target: Default Test DirectoryEPYC 7742 2P7742 2P Repeat110220330440550SE +/- 6.27, N = 3SE +/- 2.94, N = 3485.74480.55MIN: 400.81 / MAX: 1055.7MIN: 405.25 / MAX: 1028.471. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 4MB - Disk Target: Default Test DirectoryEPYC 7742 2P7742 2P Repeat90180270360450Min: 473.98 / Avg: 485.74 / Max: 495.41Min: 474.68 / Avg: 480.55 / Max: 483.771. (CC) gcc options: -O2 -lm -pthread -lmpi

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.

OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMPEPYC 7742 2P50100150200250SE +/- 2.90, N = 15217.821. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD SolverEPYC 7742 2P7742 2P Repeat50100150200250SE +/- 0.12, N = 3SE +/- 0.14, N = 310.63219.381. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD SolverEPYC 7742 2P7742 2P Repeat4080120160200Min: 10.39 / Avg: 10.63 / Max: 10.8Min: 219.13 / Avg: 219.38 / Max: 219.61. (CXX) g++ options: -O2 -lOpenCL

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.1.1Bulk Size: 100 - Inserts: 1000 - Rounds: 24EPYC 7742 2P306090120150SE +/- 0.97, N = 3112.561. (CXX) g++ options: -std=c++14 -lmozjs-68 -lm -lerl_interface -lei -fPIC -MMD

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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: PNG - Encode Speed: 57742 2P RepeatEPYC 7742 2P1428425670SE +/- 0.52, N = 15SE +/- 0.58, N = 1564.4063.731. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: PNG - Encode Speed: 57742 2P RepeatEPYC 7742 2P1326395265Min: 60.19 / Avg: 64.4 / Max: 67.71Min: 59.87 / Avg: 63.73 / Max: 67.21. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisEPYC 7742 2P7742 2P Repeat20406080100SE +/- 0.20, N = 3SE +/- 0.21, N = 3108.67109.161. (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 -lreadline
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisEPYC 7742 2P7742 2P Repeat20406080100Min: 108.28 / Avg: 108.67 / Max: 108.91Min: 108.91 / Avg: 109.16 / Max: 109.571. (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 -lreadline

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 2MB - Disk Target: Default Test Directory7742 2P RepeatEPYC 7742 2P100200300400500SE +/- 4.26, N = 7SE +/- 1.26, N = 3452.03445.10MIN: 378.75 / MAX: 1032.92MIN: 379.87 / MAX: 836.631. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 2MB - Disk Target: Default Test Directory7742 2P RepeatEPYC 7742 2P80160240320400Min: 438.2 / Avg: 452.03 / Max: 468.25Min: 442.85 / Avg: 445.1 / Max: 447.191. (CC) gcc options: -O2 -lm -pthread -lmpi

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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerEPYC 7742 2P246810SE +/- 0.07, N = 56.76MIN: 3.98 / MAX: 7.19

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.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMPEPYC 7742 2P20K40K60K80K100KSE +/- 634.52, N = 389585.741. (CXX) g++ options: -O3 -march=native -fopenmp

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileEPYC 7742 2P2 x AMD EPYC 7742 64-Core510152025SE +/- 0.15, N = 14SE +/- 0.13, N = 1421.5021.55
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileEPYC 7742 2P2 x AMD EPYC 7742 64-Core510152025Min: 21.17 / Avg: 21.5 / Max: 23.46Min: 21.27 / Avg: 21.55 / Max: 23.27

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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: PNG - Encode Speed: 77742 2P RepeatEPYC 7742 2P3691215SE +/- 0.03, N = 3SE +/- 0.02, N = 39.779.771. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: PNG - Encode Speed: 77742 2P RepeatEPYC 7742 2P3691215Min: 9.72 / Avg: 9.77 / Max: 9.81Min: 9.73 / Avg: 9.77 / Max: 9.81. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To CompileEPYC 7742 2P20406080100SE +/- 0.01, N = 394.91

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterMontage Astronomical Image Mosaic Engine 6.0Mosaic of M17, K band, 1.5 deg x 1.5 degEPYC 7742 2P20406080100SE +/- 0.05, N = 393.081. (CC) gcc options: -std=gnu99 -lcfitsio -lm -O2

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.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingEPYC 7742 2P8K16K24K32K40KSE +/- 223.13, N = 337599.71. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingEPYC 7742 2P8K16K24K32K40KSE +/- 416.26, N = 338291.81. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileEPYC 7742 2P20406080100SE +/- 0.21, N = 391.31

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Decompression Speed7742 2P RepeatEPYC 7742 2P2 x AMD EPYC 7742 64-Core7001400210028003500SE +/- 10.95, N = 3SE +/- 3.61, N = 15SE +/- 9.37, N = 33205.83205.63199.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Decompression Speed7742 2P RepeatEPYC 7742 2P2 x AMD EPYC 7742 64-Core6001200180024003000Min: 3186 / Avg: 3205.77 / Max: 3223.8Min: 3180.9 / Avg: 3205.6 / Max: 3221.1Min: 3182.9 / Avg: 3199.63 / Max: 3215.31. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Compression SpeedEPYC 7742 2P7742 2P Repeat2 x AMD EPYC 7742 64-Core130260390520650SE +/- 10.51, N = 15SE +/- 6.00, N = 3SE +/- 4.02, N = 3587.6566.1561.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Compression SpeedEPYC 7742 2P7742 2P Repeat2 x AMD EPYC 7742 64-Core100200300400500Min: 538.3 / Avg: 587.62 / Max: 657.9Min: 558.4 / Avg: 566.07 / Max: 577.9Min: 554 / Avg: 561.57 / Max: 567.71. (CC) gcc options: -O3 -pthread -lz -llzma

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Composite7742 2P RepeatEPYC 7742 2P30060090012001500SE +/- 9.48, N = 15SE +/- 14.63, N = 41200.411178.951. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Composite7742 2P RepeatEPYC 7742 2P2004006008001000Min: 1181.61 / Avg: 1200.41 / Max: 1295.07Min: 1153.31 / Avg: 1178.95 / Max: 1220.641. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.1CPU Threads: 17742 2P Repeat816243240SE +/- 0.11, N = 332.97

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.1CPU Threads: AllEPYC 7742 2P7742 2P Repeat20406080100SE +/- 0.77, N = 3SE +/- 0.25, N = 399.5499.32
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding 0.3.1CPU Threads: AllEPYC 7742 2P7742 2P Repeat20406080100Min: 98 / Avg: 99.54 / Max: 100.42Min: 99.02 / Avg: 99.32 / Max: 99.83

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.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2EPYC 7742 2P7742 2P Repeat13002600390052006500SE +/- 109.01, N = 20SE +/- 160.58, N = 2059476295
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2EPYC 7742 2P7742 2P Repeat11002200330044005500Min: 5324 / Avg: 5946.85 / Max: 6791Min: 5110 / Avg: 6295 / Max: 7860

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastEPYC 7742 2P918273645SE +/- 0.92, N = 1540.151. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.99b6Total TimeEPYC 7742 2P3691215SE +/- 0.0989, N = 159.89091. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

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.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingEPYC 7742 2P9001800270036004500SE +/- 55.12, N = 123991.991. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - GriddingEPYC 7742 2P10002000300040005000SE +/- 78.34, N = 124826.831. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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.orgSeconds, Fewer Is BetterRadiance Benchmark 5.0Test: SMP ParallelEPYC 7742 2P50100150200250213.87

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.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansEPYC 7742 2P7742 2P Repeat11002200330044005500SE +/- 98.30, N = 17SE +/- 116.87, N = 2048664991
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansEPYC 7742 2P7742 2P Repeat9001800270036004500Min: 4384 / Avg: 4866.24 / Max: 5741Min: 4394 / Avg: 4991.25 / Max: 6010

Timed Wasmer Compilation

This test times how long it takes to compile Wasmer. Wasmer is written in the Rust programming language and is a WebAssembly runtime implementation that supports WASI and EmScripten. This test profile builds Wasmer with the Cranelift and Singlepast compiler features enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileEPYC 7742 2P2 x AMD EPYC 7742 64-Core1530456075SE +/- 0.38, N = 3SE +/- 0.18, N = 368.2268.431. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileEPYC 7742 2P2 x AMD EPYC 7742 64-Core1326395265Min: 67.51 / Avg: 68.22 / Max: 68.79Min: 68.13 / Avg: 68.43 / Max: 68.751. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

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.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedEPYC 7742 2P7742 2P Repeat2K4K6K8K10KSE +/- 44.54, N = 3SE +/- 45.56, N = 310418.310278.81. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedEPYC 7742 2P7742 2P Repeat2K4K6K8K10KMin: 10329.3 / Avg: 10418.33 / Max: 10465.5Min: 10225.9 / Avg: 10278.8 / Max: 10369.51. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedEPYC 7742 2P7742 2P Repeat1020304050SE +/- 0.35, N = 3SE +/- 0.03, N = 344.8643.491. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedEPYC 7742 2P7742 2P Repeat918273645Min: 44.17 / Avg: 44.86 / Max: 45.22Min: 43.45 / Avg: 43.49 / Max: 43.541. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression Speed7742 2P RepeatEPYC 7742 2P2K4K6K8K10KSE +/- 28.45, N = 3SE +/- 80.01, N = 310373.010188.31. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression Speed7742 2P RepeatEPYC 7742 2P2K4K6K8K10KMin: 10344.1 / Avg: 10373 / Max: 10429.9Min: 10054 / Avg: 10188.27 / Max: 10330.81. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedEPYC 7742 2P7742 2P Repeat1020304050SE +/- 0.39, N = 3SE +/- 0.26, N = 345.3645.161. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedEPYC 7742 2P7742 2P Repeat918273645Min: 44.88 / Avg: 45.36 / Max: 46.14Min: 44.64 / Avg: 45.16 / Max: 45.431. (CC) gcc options: -O3

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: CPU-Only2 x AMD EPYC 7742 64-Core1428425670SE +/- 0.17, N = 364.56

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To CompileEPYC 7742 2P1428425670SE +/- 0.18, N = 364.21

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.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: DLSCEPYC 7742 2P48121620SE +/- 0.04, N = 315.03MIN: 14.9 / MAX: 15.86

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompileEPYC 7742 2P1428425670SE +/- 0.42, N = 361.41

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.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: Rainbow Colors and PrismEPYC 7742 2P48121620SE +/- 0.05, N = 316.89MIN: 16.01 / MAX: 17.19

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.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMPEPYC 7742 2P11K22K33K44K55KSE +/- 187.65, N = 352054.591. (CXX) g++ options: -O3 -march=native -fopenmp

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingEPYC 7742 2P7742 2P Repeat1530456075SE +/- 0.88, N = 3SE +/- 0.88, N = 369681. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: ResizingEPYC 7742 2P7742 2P Repeat1326395265Min: 68 / Avg: 69.33 / Max: 71Min: 67 / Avg: 68.33 / Max: 701. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 0EPYC 7742 2P2 x AMD EPYC 7742 64-Core1428425670SE +/- 0.20, N = 3SE +/- 0.76, N = 360.1160.631. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 0EPYC 7742 2P2 x AMD EPYC 7742 64-Core1224364860Min: 59.76 / Avg: 60.11 / Max: 60.45Min: 59.12 / Avg: 60.63 / Max: 61.561. (CXX) g++ options: -O3 -fPIC -lm

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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: JPEG - Encode Speed: 57742 2P RepeatEPYC 7742 2P1224364860SE +/- 0.58, N = 3SE +/- 0.39, N = 1153.1151.711. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: JPEG - Encode Speed: 57742 2P RepeatEPYC 7742 2P1122334455Min: 52.33 / Avg: 53.11 / Max: 54.25Min: 50.18 / Avg: 51.71 / Max: 54.211. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl

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.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverEPYC 7742 2P8001600240032004000SE +/- 19.77, N = 33962.211. (CC) gcc options: -O3 -mavx2

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenEPYC 7742 2P7742 2P Repeat2004006008001000SE +/- 5.33, N = 38338291. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: SharpenEPYC 7742 2P7742 2P Repeat150300450600750Min: 818 / Avg: 828.67 / Max: 8341. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Enhanced7742 2P RepeatEPYC 7742 2P30060090012001500SE +/- 0.88, N = 3SE +/- 2.40, N = 3120311991. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Enhanced7742 2P RepeatEPYC 7742 2P2004006008001000Min: 1202 / Avg: 1203.33 / Max: 1205Min: 1196 / Avg: 1199.33 / Max: 12041. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Swirl7742 2P RepeatEPYC 7742 2P400800120016002000SE +/- 14.38, N = 3SE +/- 11.05, N = 3173017211. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.33Operation: Swirl7742 2P RepeatEPYC 7742 2P30060090012001500Min: 1701 / Avg: 1729.67 / Max: 1746Min: 1705 / Avg: 1720.67 / Max: 17421. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: JPEG - Encode Speed: 7EPYC 7742 2P7742 2P Repeat1224364860SE +/- 0.17, N = 3SE +/- 0.33, N = 1551.3651.171. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: JPEG - Encode Speed: 7EPYC 7742 2P7742 2P Repeat1020304050Min: 51.02 / Avg: 51.36 / Max: 51.54Min: 49.61 / Avg: 51.17 / Max: 53.461. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl

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.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2EPYC 7742 2P7742 2P Repeat300M600M900M1200M1500MSE +/- 1663713.95, N = 3SE +/- 737641.36, N = 3124742766712461386671. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi
OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2EPYC 7742 2P7742 2P Repeat200M400M600M800M1000MMin: 1244398000 / Avg: 1247427666.67 / Max: 1250134000Min: 1244946000 / Avg: 1246138666.67 / Max: 12474870001. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi

srsLTE

srsLTE is an open-source LTE software radio suite created by Software Radio Systems (SRS). srsLTE can be used for building your own software defined (SDR) LTE mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples / Second, More Is BettersrsLTE 20.10.1Test: OFDM_Test2PEPYC 7742 2P7742 2P Repeat20M40M60M80M100MSE +/- 404145.19, N = 3SE +/- 520683.31, N = 3SE +/- 1084711.94, N = 510110000098333333979600001. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
OpenBenchmarking.orgSamples / Second, More Is BettersrsLTE 20.10.1Test: OFDM_Test2PEPYC 7742 2P7742 2P Repeat20M40M60M80M100MMin: 100600000 / Avg: 101100000 / Max: 101900000Min: 97400000 / Avg: 98333333.33 / Max: 99200000Min: 95700000 / Avg: 97960000 / Max: 1018000001. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastEPYC 7742 2P1020304050SE +/- 1.77, N = 1244.451. (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.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: JPEG - Encode Speed: 8EPYC 7742 2P7742 2P Repeat510152025SE +/- 0.17, N = 15SE +/- 0.26, N = 1522.9122.721. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL 0.3.1Input: JPEG - Encode Speed: 8EPYC 7742 2P7742 2P Repeat510152025Min: 21.91 / Avg: 22.91 / Max: 23.99Min: 20.58 / Avg: 22.72 / Max: 24.161. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie -pthread -ldl

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P700M1400M2100M2800M3500MSE +/- 79162966.07, N = 13SE +/- 13159153.97, N = 3321813846231356000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P600M1200M1800M2400M3000MMin: 3063800000 / Avg: 3218138461.54 / Max: 4125500000Min: 3121600000 / Avg: 3135600000 / Max: 31619000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

srsLTE

srsLTE is an open-source LTE software radio suite created by Software Radio Systems (SRS). srsLTE can be used for building your own software defined (SDR) LTE mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test2P7742 2P RepeatEPYC 7742 2P20406080100SE +/- 0.03, N = 3SE +/- 0.21, N = 3SE +/- 0.33, N = 387.586.883.01. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
OpenBenchmarking.orgUE Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test2P7742 2P RepeatEPYC 7742 2P20406080100Min: 87.4 / Avg: 87.47 / Max: 87.5Min: 86.4 / Avg: 86.8 / Max: 87.1Min: 82.7 / Avg: 83.03 / Max: 83.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test2P7742 2P RepeatEPYC 7742 2P50100150200250SE +/- 0.19, N = 3SE +/- 0.92, N = 3SE +/- 0.38, N = 3207.6204.5197.71. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
OpenBenchmarking.orgeNb Mb/s, More Is BettersrsLTE 20.10.1Test: PHY_DL_Test2P7742 2P RepeatEPYC 7742 2P4080120160200Min: 207.4 / Avg: 207.63 / Max: 208Min: 202.9 / Avg: 204.53 / Max: 206.1Min: 197.1 / Avg: 197.7 / Max: 198.41. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f

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.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 2048EPYC 7742 2P7742 2P Repeat13002600390052006500SE +/- 3.97, N = 3SE +/- 47.11, N = 36160.76086.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 2048EPYC 7742 2P7742 2P Repeat11002200330044005500Min: 6152.8 / Avg: 6160.7 / Max: 6165.3Min: 5992.5 / Avg: 6086.47 / Max: 6139.51. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

dcraw

This test times how long it takes to convert several high-resolution RAW NEF image files to PPM image format using dcraw. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterdcrawRAW To PPM Image ConversionEPYC 7742 2P1122334455SE +/- 0.03, N = 350.521. (CC) gcc options: -lm

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-Only2 x AMD EPYC 7742 64-Core1122334455SE +/- 0.29, N = 349.06

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.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pEPYC 7742 2P1428425670SE +/- 0.62, N = 1561.361. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: XZ 0 - Process: Decompression7742 2P RepeatEPYC 7742 2P20406080100SE +/- 0.33, N = 399981. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: XZ 0 - Process: Decompression7742 2P RepeatEPYC 7742 2P20406080100Min: 98 / Avg: 98.33 / Max: 991. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: XZ 0 - Process: Compression7742 2P RepeatEPYC 7742 2P71421283532321. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

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.

OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkEPYC 7742 2P7742 2P Repeat714212835SE +/- 0.37, N = 4SE +/- 0.10, N = 330.9929.781. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
OpenBenchmarking.orgMpix/sec, More Is BetterLibRaw 0.20Post-Processing BenchmarkEPYC 7742 2P7742 2P Repeat714212835Min: 30.44 / Avg: 30.99 / Max: 32.07Min: 29.59 / Avg: 29.78 / Max: 29.911. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm

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.

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisEPYC 7742 2P816243240SE +/- 0.10, N = 435.081. (CC) gcc options: -O2 -std=c99

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.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest Compression7742 2P RepeatEPYC 7742 2P1020304050SE +/- 0.03, N = 3SE +/- 0.05, N = 341.9441.981. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest Compression7742 2P RepeatEPYC 7742 2P918273645Min: 41.88 / Avg: 41.94 / Max: 41.98Min: 41.91 / Avg: 41.98 / Max: 42.071. (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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat0.56221.12441.68662.24882.811SE +/- 0.02717, N = 15SE +/- 0.13368, N = 121.546502.49866MIN: 1.18MIN: 1.591. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat246810Min: 1.29 / Avg: 1.55 / Max: 1.68Min: 1.73 / Avg: 2.5 / Max: 2.91. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Gzip Compression

This test measures the time needed to archive/compress two copies of the Linux 4.13 kernel source tree using Gzip compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGzip CompressionLinux Source Tree Archiving To .tar.gzEPYC 7742 2P1020304050SE +/- 0.07, N = 341.89

Timed PHP Compilation

This test times how long it takes to build PHP 7. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To CompileEPYC 7742 2P1020304050SE +/- 0.19, N = 341.67

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 6EPYC 7742 2P0.32850.6570.98551.3141.6425SE +/- 0.019, N = 31.460

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimeEPYC 7742 2P918273645SE +/- 0.01, N = 339.901. (CC) gcc options: -lm -O3

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6EPYC 7742 2P2 x AMD EPYC 7742 64-Core3691215SE +/- 0.15, N = 3SE +/- 0.26, N = 1512.1012.961. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6EPYC 7742 2P2 x AMD EPYC 7742 64-Core48121620Min: 11.91 / Avg: 12.1 / Max: 12.41Min: 11.92 / Avg: 12.96 / Max: 14.661. (CXX) g++ options: -O3 -fPIC -lm

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.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21EPYC 7742 2P7742 2P Repeat400800120016002000SE +/- 14.66, N = 3SE +/- 13.17, N = 32015.92009.21. (CXX) g++ options: -O3 -march=native -rdynamic
OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21EPYC 7742 2P7742 2P Repeat400800120016002000Min: 1986.7 / Avg: 2015.87 / Max: 2033Min: 1983.1 / Avg: 2009.2 / Max: 2025.31. (CXX) g++ options: -O3 -march=native -rdynamic

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CPU-Only2 x AMD EPYC 7742 64-Core816243240SE +/- 0.30, N = 336.02

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3 - Decompression Speed7742 2P Repeat6001200180024003000SE +/- 3.38, N = 32908.51. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 3 - Compression Speed7742 2P Repeat11002200330044005500SE +/- 62.74, N = 35053.71. (CC) gcc options: -O3 -pthread -lz -llzma

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.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021Input: water_GMX50_bare2 x AMD EPYC 7742 64-Core246810SE +/- 0.019, N = 38.0641. (CXX) g++ options: -O3 -pthread

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessEPYC 7742 2P2 x AMD EPYC 7742 64-Core816243240SE +/- 0.07, N = 3SE +/- 0.42, N = 334.9335.431. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessEPYC 7742 2P2 x AMD EPYC 7742 64-Core816243240Min: 34.85 / Avg: 34.93 / Max: 35.08Min: 34.8 / Avg: 35.43 / Max: 36.211. (CXX) g++ options: -O3 -fPIC -lm

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.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC27742 2P RepeatEPYC 7742 2P306090120150SE +/- 0.07, N = 3SE +/- 0.00, N = 3139.56139.501. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC27742 2P RepeatEPYC 7742 2P306090120150Min: 139.42 / Avg: 139.56 / Max: 139.65Min: 139.5 / Avg: 139.5 / Max: 139.511. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 6 RealtimeEPYC 7742 2P510152025SE +/- 0.12, N = 318.471. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisEPYC 7742 2P20406080100SE +/- 0.00, N = 383.33MIN: 26.32 / MAX: 90.91

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelEPYC 7742 2P246810SE +/- 0.165, N = 127.7541. (CC) gcc options: -lm -lpthread -O3

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 6 Two-PassEPYC 7742 2P0.7561.5122.2683.0243.78SE +/- 0.01, N = 33.361. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -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.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KEPYC 7742 2P510152025SE +/- 0.09, N = 318.771. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: HairEPYC 7742 2P1.26382.52763.79145.05526.319SE +/- 0.05055, N = 155.616961. (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

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.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedEPYC 7742 2P7742 2P Repeat2K4K6K8K10KSE +/- 91.80, N = 3SE +/- 21.34, N = 311001.510868.71. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedEPYC 7742 2P7742 2P Repeat2K4K6K8K10KMin: 10893.1 / Avg: 11001.47 / Max: 11184Min: 10834.6 / Avg: 10868.73 / Max: 109081. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedEPYC 7742 2P7742 2P Repeat2K4K6K8K10KSE +/- 65.80, N = 3SE +/- 29.92, N = 39527.729289.071. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedEPYC 7742 2P7742 2P Repeat17003400510068008500Min: 9396.7 / Avg: 9527.72 / Max: 9603.95Min: 9249.26 / Avg: 9289.07 / Max: 9347.661. (CC) gcc options: -O3

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian Dragon ObjEPYC 7742 2P816243240SE +/- 0.31, N = 336.33MIN: 34.33 / MAX: 39.39

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian Dragon ObjEPYC 7742 2P918273645SE +/- 0.11, N = 339.05MIN: 37.1 / MAX: 42.47

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDEPYC 7742 2P7742 2P Repeat816243240SE +/- 0.27, N = 3SE +/- 0.11, N = 330.2132.511. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDEPYC 7742 2P7742 2P Repeat714212835Min: 29.92 / Avg: 30.21 / Max: 30.74Min: 32.29 / Avg: 32.51 / Max: 32.651. (CXX) g++ options: -O2 -lOpenCL

Timed MAFFT Alignment

This test performs an alignment of 100 pyruvate decarboxylase sequences. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.471Multiple Sequence Alignment - LSU RNAEPYC 7742 2P7742 2P Repeat3691215SE +/- 0.14, N = 3SE +/- 0.11, N = 1510.2710.471. (CC) gcc options: -std=c99 -O3 -lm -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.471Multiple Sequence Alignment - LSU RNAEPYC 7742 2P7742 2P Repeat3691215Min: 10 / Avg: 10.27 / Max: 10.45Min: 9.96 / Avg: 10.47 / Max: 11.591. (CC) gcc options: -std=c99 -O3 -lm -lpthread

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.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total TimeEPYC 7742 2P40M80M120M160M200MSE +/- 2342263.11, N = 31900429871. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver

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.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100EPYC 7742 2P140280420560700SE +/- 0.61, N = 3646.911. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat1.25792.51583.77375.03166.2895SE +/- 0.15852, N = 13SE +/- 0.12818, N = 154.680785.59055MIN: 2.73MIN: 4.051. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat246810Min: 3.5 / Avg: 4.68 / Max: 5.44Min: 4.67 / Avg: 5.59 / Max: 6.171. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 10EPYC 7742 2P0.6981.3962.0942.7923.49SE +/- 0.019, N = 33.102

Crafty

This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed Time7742 2P RepeatEPYC 7742 2P1.5M3M4.5M6M7.5MSE +/- 3617.62, N = 3SE +/- 11969.14, N = 3677860867577871. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed Time7742 2P RepeatEPYC 7742 2P1.2M2.4M3.6M4.8M6MMin: 6771408 / Avg: 6778607.67 / Max: 6782828Min: 6733857 / Avg: 6757787 / Max: 67702971. (CC) gcc options: -pthread -lstdc++ -fprofile-use -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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerEPYC 7742 2P3691215SE +/- 0.06, N = 310.10MIN: 7.81 / MAX: 10.64

OpenFOAM

OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MEPYC 7742 2P7742 2P Repeat48121620SE +/- 0.07, N = 3SE +/- 0.08, N = 314.1214.151. (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
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MEPYC 7742 2P7742 2P Repeat48121620Min: 14.01 / Avg: 14.12 / Max: 14.26Min: 14.05 / Avg: 14.15 / Max: 14.311. (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

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: OpenMP2 x AMD EPYC 7742 64-CoreEPYC 7742 2P11002200330044005500SE +/- 99.81, N = 15SE +/- 104.40, N = 15514151791. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: OpenMP2 x AMD EPYC 7742 64-CoreEPYC 7742 2P9001800270036004500Min: 4665 / Avg: 5141 / Max: 5788Min: 4672 / Avg: 5179.27 / Max: 58841. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Decompression7742 2P RepeatEPYC 7742 2P2004006008001000SE +/- 0.67, N = 3SE +/- 5.51, N = 39619591. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Decompression7742 2P RepeatEPYC 7742 2P2004006008001000Min: 960 / Avg: 961.33 / Max: 962Min: 953 / Avg: 959 / Max: 9701. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Compression7742 2P RepeatEPYC 7742 2P50100150200250SE +/- 0.33, N = 32062051. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Compression7742 2P RepeatEPYC 7742 2P4080120160200Min: 206 / Avg: 206.33 / Max: 2071. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-Only2 x AMD EPYC 7742 64-Core612182430SE +/- 0.22, N = 324.22

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To CompileEPYC 7742 2P612182430SE +/- 0.02, N = 324.68

Monkey Audio Encoding

This test times how long it takes to encode a sample WAV file to Monkey's Audio APE format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APEEPYC 7742 2P48121620SE +/- 0.02, N = 514.371. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: DecompressionEPYC 7742 2P7742 2P Repeat120240360480600SE +/- 1.67, N = 35715691. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: DecompressionEPYC 7742 2P7742 2P Repeat100200300400500Min: 568 / Avg: 571.33 / Max: 5731. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: Compression7742 2P RepeatEPYC 7742 2P4080120160200SE +/- 0.67, N = 3SE +/- 0.58, N = 31651651. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: Compression7742 2P RepeatEPYC 7742 2P306090120150Min: 164 / Avg: 164.67 / Max: 166Min: 164 / Avg: 165 / Max: 1661. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Ogg Audio Encoding

This test times how long it takes to encode a sample WAV file to Ogg format using the reference Xiph.org tools/libraries. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOgg Audio Encoding 1.3.4WAV To OggEPYC 7742 2P612182430SE +/- 0.04, N = 323.701. (CC) gcc options: -O2 -ffast-math -fsigned-char

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: DecompressionEPYC 7742 2P7742 2P Repeat100200300400500SE +/- 1.53, N = 3SE +/- 1.86, N = 34824811. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: DecompressionEPYC 7742 2P7742 2P Repeat90180270360450Min: 479 / Avg: 482 / Max: 484Min: 479 / Avg: 481.33 / Max: 4851. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: Compression7742 2P RepeatEPYC 7742 2P90180270360450SE +/- 0.67, N = 3SE +/- 1.53, N = 34164151. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: Compression7742 2P RepeatEPYC 7742 2P70140210280350Min: 415 / Avg: 415.67 / Max: 417Min: 413 / Avg: 415 / Max: 4181. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28EPYC 7742 2P612182430SE +/- 0.04, N = 323.141. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: DecompressionEPYC 7742 2P7742 2P Repeat30060090012001500SE +/- 0.67, N = 3SE +/- 0.88, N = 3149714871. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: DecompressionEPYC 7742 2P7742 2P Repeat30060090012001500Min: 1496 / Avg: 1497.33 / Max: 1498Min: 1485 / Avg: 1486.67 / Max: 14881. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: Compression7742 2P RepeatEPYC 7742 2P20406080100SE +/- 0.67, N = 383821. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: Compression7742 2P RepeatEPYC 7742 2P1632486480Min: 82 / Avg: 82.67 / Max: 841. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, Lossless2 x AMD EPYC 7742 64-CoreEPYC 7742 2P246810SE +/- 0.037, N = 3SE +/- 0.073, N = 157.4007.5871. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, Lossless2 x AMD EPYC 7742 64-CoreEPYC 7742 2P3691215Min: 7.34 / Avg: 7.4 / Max: 7.46Min: 7.25 / Avg: 7.59 / Max: 8.011. (CXX) g++ options: -O3 -fPIC -lm

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: DecompressionEPYC 7742 2P7742 2P Repeat30060090012001500SE +/- 1.45, N = 3SE +/- 4.36, N = 3134613341. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: DecompressionEPYC 7742 2P7742 2P Repeat2004006008001000Min: 1343 / Avg: 1345.67 / Max: 1348Min: 1326 / Avg: 1334 / Max: 13411. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: Compression7742 2P RepeatEPYC 7742 2P90180270360450SE +/- 0.67, N = 3SE +/- 0.88, N = 34374351. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: Compression7742 2P RepeatEPYC 7742 2P80160240320400Min: 436 / Avg: 437.33 / Max: 438Min: 434 / Avg: 435.33 / Max: 4371. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: Decompression7742 2P RepeatEPYC 7742 2P801602403204003823811. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: Compression7742 2P RepeatEPYC 7742 2P20406080100SE +/- 0.67, N = 389891. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: Compression7742 2P RepeatEPYC 7742 2P20406080100Min: 88 / Avg: 88.67 / Max: 901. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

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.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1 + Dithering7742 2P RepeatEPYC 7742 2P50100150200250SE +/- 0.03, N = 3SE +/- 0.05, N = 3224.60224.161. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1 + Dithering7742 2P RepeatEPYC 7742 2P4080120160200Min: 224.55 / Avg: 224.6 / Max: 224.64Min: 224.09 / Avg: 224.16 / Max: 224.261. (CXX) g++ options: -O3 -march=native -std=c++11 -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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P0.49620.99241.48861.98482.481SE +/- 0.00319, N = 3SE +/- 0.00840, N = 32.199372.20539MIN: 2.03MIN: 2.031. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P246810Min: 2.19 / Avg: 2.2 / Max: 2.2Min: 2.19 / Avg: 2.21 / Max: 2.221. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P0.64561.29121.93682.58243.228SE +/- 0.01034, N = 3SE +/- 0.01849, N = 32.862602.86937MIN: 2.65MIN: 2.621. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P246810Min: 2.85 / Avg: 2.86 / Max: 2.88Min: 2.84 / Avg: 2.87 / Max: 2.911. (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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pEPYC 7742 2P246810SE +/- 0.065, N = 37.4561. (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.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC17742 2P RepeatEPYC 7742 2P50100150200250SE +/- 0.11, N = 3SE +/- 0.10, N = 3237.25236.721. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC17742 2P RepeatEPYC 7742 2P4080120160200Min: 237.04 / Avg: 237.24 / Max: 237.41Min: 236.52 / Avg: 236.72 / Max: 236.871. (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.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessEPYC 7742 2P7742 2P Repeat510152025SE +/- 0.02, N = 3SE +/- 0.04, N = 320.3620.431. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, LosslessEPYC 7742 2P7742 2P Repeat510152025Min: 20.34 / Avg: 20.36 / Max: 20.4Min: 20.37 / Avg: 20.43 / Max: 20.491. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

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.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsEPYC 7742 2P7742 2P Repeat0.06370.12740.19110.25480.3185SE +/- 0.00196, N = 3SE +/- 0.00247, N = 30.279520.28306
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsEPYC 7742 2P7742 2P Repeat12345Min: 0.28 / Avg: 0.28 / Max: 0.28Min: 0.28 / Avg: 0.28 / Max: 0.29

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid Dynamics7742 2P RepeatEPYC 7742 2P510152025SE +/- 0.04, N = 3SE +/- 0.04, N = 320.2020.21
OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid Dynamics7742 2P RepeatEPYC 7742 2P510152025Min: 20.12 / Avg: 20.2 / Max: 20.25Min: 20.14 / Avg: 20.21 / Max: 20.28

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 256 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P1200M2400M3600M4800M6000MSE +/- 16339556.64, N = 3SE +/- 29512765.60, N = 3555073333355251000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 256 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P1000M2000M3000M4000M5000MMin: 5522300000 / Avg: 5550733333.33 / Max: 5578900000Min: 5468000000 / Avg: 5525100000 / Max: 55666000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P90M180M270M360M450MSE +/- 127322.86, N = 3SE +/- 101707.64, N = 34272766674272033331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P70M140M210M280M350MMin: 427090000 / Avg: 427276666.67 / Max: 427520000Min: 427040000 / Avg: 427203333.33 / Max: 4273900001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 64 - Buffer Length: 256 - Filter Length: 57EPYC 7742 2P2P600M1200M1800M2400M3000MSE +/- 8434123.81, N = 3SE +/- 16574813.56, N = 3270393333326937666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 64 - Buffer Length: 256 - Filter Length: 57EPYC 7742 2P2P500M1000M1500M2000M2500MMin: 2695300000 / Avg: 2703933333.33 / Max: 2720800000Min: 2676300000 / Avg: 2693766666.67 / Max: 27269000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P300M600M900M1200M1500MSE +/- 1153256.26, N = 3SE +/- 1822391.59, N = 3161800000016165666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P300M600M900M1200M1500MMin: 1615700000 / Avg: 1618000000 / Max: 1619300000Min: 1613300000 / Avg: 1616566666.67 / Max: 16196000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57EPYC 7742 2P2P200M400M600M800M1000MSE +/- 1082240.47, N = 3SE +/- 620358.32, N = 38322766678316133331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57EPYC 7742 2P2P140M280M420M560M700MMin: 831010000 / Avg: 832276666.67 / Max: 834430000Min: 830600000 / Avg: 831613333.33 / Max: 8327400001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 4 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P50M100M150M200M250MSE +/- 107445.08, N = 3SE +/- 80829.04, N = 32132866672132800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 4 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P40M80M120M160M200MMin: 213090000 / Avg: 213286666.67 / Max: 213460000Min: 213180000 / Avg: 213280000 / Max: 2134400001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 2 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P20M40M60M80M100MSE +/- 18559.21, N = 3SE +/- 61191.87, N = 31071666671071433331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 2 - Buffer Length: 256 - Filter Length: 572PEPYC 7742 2P20M40M60M80M100MMin: 107130000 / Avg: 107166666.67 / Max: 107190000Min: 107030000 / Avg: 107143333.33 / Max: 1072400001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 1 - Buffer Length: 256 - Filter Length: 57EPYC 7742 2P2P11M22M33M44M55MSE +/- 22980.67, N = 3SE +/- 10477.49, N = 353594667535796671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 1 - Buffer Length: 256 - Filter Length: 57EPYC 7742 2P2P9M18M27M36M45MMin: 53549000 / Avg: 53594666.67 / Max: 53622000Min: 53563000 / Avg: 53579666.67 / Max: 535990001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Timed FFmpeg Compilation

This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 4.2.2Time To CompileEPYC 7742 2P510152025SE +/- 0.09, N = 319.75

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 8 RealtimeEPYC 7742 2P714212835SE +/- 0.37, N = 331.891. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -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.

OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 7.0N=256, 3D Complex FFT RoutineEPYC 7742 2P7742 2P Repeat30K60K90K120K150KSE +/- 3136.68, N = 12SE +/- 3906.26, N = 15150712.97148613.891. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 7.0N=256, 3D Complex FFT RoutineEPYC 7742 2P7742 2P Repeat30K60K90K120K150KMin: 129331.54 / Avg: 150712.97 / Max: 172903.22Min: 125117.64 / Avg: 148613.89 / Max: 182817.871. (F9X) gfortran options: -O3 -fomit-frame-pointer -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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DEPYC 7742 2P7742 2P Repeat2K4K6K8K10KSE +/- 14.28, N = 3SE +/- 8.01, N = 38426.047885.521. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DEPYC 7742 2P7742 2P Repeat15003000450060007500Min: 8401.91 / Avg: 8426.04 / Max: 8451.33Min: 7871.33 / Avg: 7885.52 / Max: 7899.041. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: DefaultEPYC 7742 2P0.73621.47242.20862.94483.681SE +/- 0.035, N = 153.2721. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lgif

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Ultra FastEPYC 7742 2P4080120160200SE +/- 2.71, N = 15181.141. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

FLAC Audio Encoding

This test times how long it takes to encode a sample WAV file to FLAC format five times. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACEPYC 7742 2P3691215SE +/- 0.009, N = 59.8301. (CXX) g++ options: -O2 -fvisibility=hidden -logg -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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: VMAF Optimized - Input: Bosphorus 1080pEPYC 7742 2P70140210280350SE +/- 14.04, N = 12340.021. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisEPYC 7742 2P510152025SE +/- 0.13, N = 319.74MIN: 12.82 / MAX: 20.83

Timed ImageMagick Compilation

This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To CompileEPYC 7742 2P48121620SE +/- 0.12, N = 315.67

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian DragonEPYC 7742 2P1020304050SE +/- 0.14, N = 342.12MIN: 39.85 / MAX: 44.87

Opus Codec Encoding

Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeEPYC 7742 2P3691215SE +/- 0.019, N = 59.1501. (CXX) g++ options: -fvisibility=hidden -logg -lm

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.

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2019-12-17H.264 Video EncodingEPYC 7742 2P4080120160200SE +/- 2.66, N = 15204.011. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P0.46650.9331.39951.8662.3325SE +/- 0.00722, N = 3SE +/- 0.00986, N = 32.065642.07347MIN: 1.87MIN: 1.861. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P246810Min: 2.05 / Avg: 2.07 / Max: 2.08Min: 2.06 / Avg: 2.07 / Max: 2.091. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat0.4810.9621.4431.9242.405SE +/- 0.01302, N = 3SE +/- 0.00603, N = 32.133292.13758MIN: 1.93MIN: 1.931. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat246810Min: 2.12 / Avg: 2.13 / Max: 2.16Min: 2.13 / Avg: 2.14 / Max: 2.141. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian DragonEPYC 7742 2P1020304050SE +/- 0.32, N = 344.97MIN: 41.97 / MAX: 48.41

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pEPYC 7742 2P20406080100SE +/- 0.35, N = 385.791. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

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.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 1080pEPYC 7742 2P30060090012001500SE +/- 12.08, N = 141245.91MIN: 153.88 / MAX: 1618.671. (CC) gcc options: -pthread -lm

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat0.16080.32160.48240.64320.804SE +/- 0.003926, N = 3SE +/- 0.008252, N = 30.7129360.714842MIN: 0.65MIN: 0.651. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat246810Min: 0.71 / Avg: 0.71 / Max: 0.72Min: 0.7 / Avg: 0.71 / Max: 0.731. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUEPYC 7742 2P0.18290.36580.54870.73160.9145SE +/- 0.002167, N = 30.812990MIN: 0.761. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CEPYC 7742 2P7742 2P Repeat40K80K120K160K200KSE +/- 770.77, N = 3SE +/- 1600.26, N = 3194294.65176465.671. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CEPYC 7742 2P7742 2P Repeat30K60K90K120K150KMin: 193269.3 / Avg: 194294.65 / Max: 195804.2Min: 174555.81 / Avg: 176465.67 / Max: 179644.751. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Non-ExponentialEPYC 7742 2P0.38760.77521.16281.55041.938SE +/- 0.03014, N = 151.722561. (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

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P0.27230.54460.81691.08921.3615SE +/- 0.01177, N = 6SE +/- 0.00941, N = 151.201631.21032MIN: 1.06MIN: 1.071. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P246810Min: 1.17 / Avg: 1.2 / Max: 1.25Min: 1.17 / Avg: 1.21 / Max: 1.311. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: CrownEPYC 7742 2P1326395265SE +/- 0.51, N = 359.33MIN: 55.07 / MAX: 65.52

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.D7742 2P RepeatEPYC 7742 2P7001400210028003500SE +/- 10.37, N = 3SE +/- 30.48, N = 33313.123268.821. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.D7742 2P RepeatEPYC 7742 2P6001200180024003000Min: 3295.98 / Avg: 3313.12 / Max: 3331.79Min: 3232.27 / Avg: 3268.82 / Max: 3329.351. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeEPYC 7742 2P246810SE +/- 0.028, N = 38.0281. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lXpm -lSM -lICE -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerEPYC 7742 2P714212835SE +/- 0.00, N = 330.30MIN: 11.24 / MAX: 31.25

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: TBB2 x AMD EPYC 7742 64-CoreEPYC 7742 2P8001600240032004000SE +/- 31.51, N = 3SE +/- 27.56, N = 12388639101. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: TBB2 x AMD EPYC 7742 64-CoreEPYC 7742 2P7001400210028003500Min: 3823 / Avg: 3886 / Max: 3919Min: 3805 / Avg: 3910.08 / Max: 40991. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MPlayer Compilation 1.4Time To CompileEPYC 7742 2P3691215SE +/- 0.02, N = 310.27

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: CrownEPYC 7742 2P1530456075SE +/- 0.29, N = 367.59MIN: 62.78 / MAX: 72.69

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: MediumEPYC 7742 2P1428425670SE +/- 0.46, N = 364.431. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P0.69511.39022.08532.78043.4755SE +/- 0.01047, N = 3SE +/- 0.03597, N = 32.888223.08917MIN: 2.66MIN: 2.761. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P246810Min: 2.87 / Avg: 2.89 / Max: 2.9Min: 3.02 / Avg: 3.09 / Max: 3.141. (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.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest Compression7742 2P RepeatEPYC 7742 2P246810SE +/- 0.002, N = 3SE +/- 0.002, N = 38.8648.9041. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest Compression7742 2P RepeatEPYC 7742 2P3691215Min: 8.86 / Avg: 8.86 / Max: 8.87Min: 8.9 / Avg: 8.9 / Max: 8.911. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

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.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonEPYC 7742 2P7742 2P Repeat11002200330044005500SE +/- 60.94, N = 4SE +/- 52.30, N = 450275096
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonEPYC 7742 2P7742 2P Repeat9001800270036004500Min: 4900 / Avg: 5026.5 / Max: 5183Min: 5002 / Avg: 5096 / Max: 5200

LAME MP3 Encoding

LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3EPYC 7742 2P3691215SE +/- 0.009, N = 39.1031. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lncurses -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.

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed TimeEPYC 7742 2P0.39830.79661.19491.59321.9915SE +/- 0.068, N = 151.7701. (CC) gcc options: -static -fopenmp -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.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 4KEPYC 7742 2P80160240320400SE +/- 2.34, N = 3387.05MIN: 84.5 / MAX: 472.331. (CC) gcc options: -pthread -lm

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateEPYC 7742 2P714212835SE +/- 0.14, N = 328.491. (CC) gcc options: -O3 -march=native -fopenmp

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Compression Effort 5EPYC 7742 2P246810SE +/- 0.027, N = 37.7211. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg -lgif

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CEPYC 7742 2P7742 2P Repeat16K32K48K64K80KSE +/- 839.98, N = 4SE +/- 262.47, N = 376051.3671133.341. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CEPYC 7742 2P7742 2P Repeat13K26K39K52K65KMin: 73899.67 / Avg: 76051.36 / Max: 77952.09Min: 70608.85 / Avg: 71133.34 / Max: 71414.321. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P0.6491.2981.9472.5963.245SE +/- 0.00982, N = 3SE +/- 0.04670, N = 122.708982.88437MIN: 2.55MIN: 2.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU7742 2P RepeatEPYC 7742 2P246810Min: 2.7 / Avg: 2.71 / Max: 2.73Min: 2.7 / Avg: 2.88 / Max: 3.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilEPYC 7742 2P7742 2P Repeat1.24752.4953.74254.996.2375SE +/- 0.018548, N = 3SE +/- 0.042058, N = 35.3895615.5446281. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilEPYC 7742 2P7742 2P Repeat246810Min: 5.35 / Avg: 5.39 / Max: 5.42Min: 5.46 / Avg: 5.54 / Max: 5.61. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -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.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 4096EPYC 7742 2P7742 2P Repeat10K20K30K40K50KSE +/- 343.27, N = 3SE +/- 125.07, N = 344939442941. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 1D FFT Size 4096EPYC 7742 2P7742 2P Repeat8K16K24K32K40KMin: 44255 / Avg: 44939 / Max: 45332Min: 44073 / Avg: 44293.67 / Max: 445061. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

System GZIP Decompression

This simple test measures the time to decompress a gzipped tarball (the Qt5 toolkit source package). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSystem GZIP DecompressionEPYC 7742 2P0.811.622.433.244.05SE +/- 0.034, N = 63.600

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.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialEPYC 7742 2P714212835SE +/- 0.20, N = 328.72

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.

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.2Time To SolveEPYC 7742 2P246810SE +/- 0.092, N = 37.0831. (CXX) g++ options: -fopenmp -O2 -march=native

System ZLIB Decompression

This test measures the time to decompress a Linux kernel tarball using ZLIB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSystem ZLIB Decompression 1.2.7EPYC 7742 2P400800120016002000SE +/- 6.97, N = 102025.98

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigEPYC 7742 2P7742 2P Repeat1.3222.6443.9665.2886.61SE +/- 0.011805, N = 3SE +/- 0.025529, N = 35.8728485.8757141. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigEPYC 7742 2P7742 2P Repeat246810Min: 5.85 / Avg: 5.87 / Max: 5.89Min: 5.82 / Avg: 5.88 / Max: 5.91. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi

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.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 40967742 2P RepeatEPYC 7742 2P15003000450060007500SE +/- 9.28, N = 3SE +/- 6.26, N = 36890.66873.91. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 40967742 2P RepeatEPYC 7742 2P12002400360048006000Min: 6872.5 / Avg: 6890.63 / Max: 6903.1Min: 6866.9 / Avg: 6873.9 / Max: 6886.41. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

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.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat0.43980.87961.31941.75922.199SE +/- 0.008019, N = 3SE +/- 0.027365, N = 30.7243951.954610MIN: 0.67MIN: 1.851. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUEPYC 7742 2P7742 2P Repeat246810Min: 0.71 / Avg: 0.72 / Max: 0.73Min: 1.92 / Avg: 1.95 / Max: 2.011. (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.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinEPYC 7742 2P7742 2P Repeat714212835SE +/- 0.29, N = 6SE +/- 0.20, N = 1529.0028.181. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinEPYC 7742 2P7742 2P Repeat612182430Min: 27.96 / Avg: 29 / Max: 29.71Min: 26.71 / Avg: 28.18 / Max: 29.71. (CXX) g++ options: -O3 -pthread -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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CEPYC 7742 2P7742 2P Repeat9K18K27K36K45KSE +/- 380.32, N = 3SE +/- 428.23, N = 541060.1939810.961. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CEPYC 7742 2P7742 2P Repeat7K14K21K28K35KMin: 40299.57 / Avg: 41060.19 / Max: 41443.93Min: 38468.72 / Avg: 39810.96 / Max: 41087.71. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

libjpeg-turbo tjbench

tjbench is a JPEG decompression/compression benchmark part of libjpeg-turbo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.0.2Test: Decompression ThroughputEPYC 7742 2P4080120160200SE +/- 0.03, N = 3172.301. (CC) gcc options: -O3 -rdynamic

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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisEPYC 7742 2P1020304050SE +/- 0.00, N = 345.45MIN: 9.09 / MAX: 47.62

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pEPYC 7742 2P80160240320400SE +/- 4.38, N = 4363.961. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: DXT1EPYC 7742 2P7742 2P Repeat2004006008001000SE +/- 3.71, N = 3SE +/- 3.65, N = 31039.811037.461. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: DXT1EPYC 7742 2P7742 2P Repeat2004006008001000Min: 1032.39 / Avg: 1039.81 / Max: 1043.57Min: 1033.77 / Avg: 1037.46 / Max: 1044.751. (CXX) g++ options: -O3 -march=native -std=c++11 -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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisEPYC 7742 2P306090120150125MIN: 16.13

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Volumetric CausticEPYC 7742 2P1.00382.00763.01144.01525.019SE +/- 0.00391, N = 34.461361. (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

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbig7742 2P RepeatEPYC 7742 2P0.88861.77722.66583.55444.443SE +/- 0.047128, N = 3SE +/- 0.028494, N = 33.9264463.9491921. (CXX) g++ options: -fopenmp -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbig7742 2P RepeatEPYC 7742 2P246810Min: 3.84 / Avg: 3.93 / Max: 4Min: 3.91 / Avg: 3.95 / Max: 41. (CXX) g++ options: -fopenmp -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.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 1080p - Video Preset: Very FastEPYC 7742 2P306090120150SE +/- 0.43, N = 3136.841. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: Visual Quality Optimized - Input: Bosphorus 1080pEPYC 7742 2P60120180240300SE +/- 0.09, N = 3274.561. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ TasksEPYC 7742 2P2 x AMD EPYC 7742 64-Core9001800270036004500SE +/- 24.69, N = 3SE +/- 28.75, N = 3429543071. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ TasksEPYC 7742 2P2 x AMD EPYC 7742 64-Core7001400210028003500Min: 4258 / Avg: 4295.33 / Max: 4342Min: 4254 / Avg: 4306.67 / Max: 43531. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc

System XZ Decompression

This test measures the time to decompress a Linux kernel tarball using XZ. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSystem XZ DecompressionEPYC 7742 2P0.971.942.913.884.85SE +/- 0.004, N = 34.311

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 102 x AMD EPYC 7742 64-CoreEPYC 7742 2P0.95131.90262.85393.80524.7565SE +/- 0.048, N = 3SE +/- 0.028, N = 34.1894.2281. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 102 x AMD EPYC 7742 64-CoreEPYC 7742 2P246810Min: 4.12 / Avg: 4.19 / Max: 4.28Min: 4.17 / Avg: 4.23 / Max: 4.261. (CXX) g++ options: -O3 -fPIC -lm

toyBrot Fractal Generator

ToyBrot is a Mandelbrot fractal generator supporting C++ threads/tasks, OpenMP, Intel Threaded Building Blocks (TBB), and other targets. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ ThreadsEPYC 7742 2P2 x AMD EPYC 7742 64-Core9001800270036004500SE +/- 25.44, N = 3SE +/- 23.63, N = 3403940501. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc
OpenBenchmarking.orgms, Fewer Is BettertoyBrot Fractal Generator 2020-11-18Implementation: C++ ThreadsEPYC 7742 2P2 x AMD EPYC 7742 64-Core7001400210028003500Min: 3996 / Avg: 4038.67 / Max: 4084Min: 4005 / Avg: 4050 / Max: 40851. (CXX) g++ options: -O3 -lpthread -lm -lgcc -lgcc_s -lc

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.

OpenBenchmarking.orgmrays/s, More Is Betterrays1bench 2020-01-09Large SceneEPYC 7742 2P110220330440550SE +/- 2.26, N = 3492.78

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.C7742 2P RepeatEPYC 7742 2P16K32K48K64K80KSE +/- 149.89, N = 3SE +/- 474.90, N = 373254.9272806.591. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.C7742 2P RepeatEPYC 7742 2P13K26K39K52K65KMin: 72958.71 / Avg: 73254.92 / Max: 73442.98Min: 72129.29 / Avg: 72806.59 / Max: 73721.91. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

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.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100EPYC 7742 2P7742 2P Repeat0.64441.28881.93322.57763.222SE +/- 0.001, N = 3SE +/- 0.003, N = 32.8632.8641. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100EPYC 7742 2P7742 2P Repeat246810Min: 2.86 / Avg: 2.86 / Max: 2.87Min: 2.86 / Avg: 2.86 / Max: 2.871. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPEPYC 7742 2P7742 2P Repeat0.19180.38360.57540.76720.959SE +/- 0.009850, N = 3SE +/- 0.008717, N = 150.8318750.8525291. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPEPYC 7742 2P7742 2P Repeat246810Min: 0.82 / Avg: 0.83 / Max: 0.85Min: 0.78 / Avg: 0.85 / Max: 0.921. (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.

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceEPYC 7742 2P7742 2P Repeat200K400K600K800K1000KSE +/- 1419.07, N = 5SE +/- 1440.27, N = 5103181310306551. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceEPYC 7742 2P7742 2P Repeat200K400K600K800K1000KMin: 1027570 / Avg: 1031813 / Max: 1035296Min: 1025657 / Avg: 1030655 / Max: 10333541. (CC) gcc options: -O3 -march=native

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.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Default7742 2P RepeatEPYC 7742 2P0.41740.83481.25221.66962.087SE +/- 0.002, N = 3SE +/- 0.003, N = 31.8541.8551. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Default7742 2P RepeatEPYC 7742 2P246810Min: 1.85 / Avg: 1.85 / Max: 1.86Min: 1.85 / Avg: 1.86 / Max: 1.861. (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.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerEPYC 7742 2P70140210280350SE +/- 0.00, N = 3333.33MIN: 37.04 / MAX: 500

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.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CEPYC 7742 2P7742 2P Repeat2K4K6K8K10KSE +/- 21.18, N = 3SE +/- 9.74, N = 38223.348108.431. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CEPYC 7742 2P7742 2P Repeat14002800420056007000Min: 8183.48 / Avg: 8223.34 / Max: 8255.69Min: 8096.22 / Avg: 8108.43 / Max: 8127.681. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

LuaJIT

This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Jacobi Successive Over-Relaxation7742 2P Repeat400800120016002000SE +/- 0.26, N = 31644.111. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Dense LU Matrix Factorization7742 2P Repeat6001200180024003000SE +/- 173.31, N = 32811.621. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Sparse Matrix Multiply7742 2P Repeat2004006008001000SE +/- 6.13, N = 31008.921. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Fast Fourier Transform7742 2P Repeat50100150200250SE +/- 0.69, N = 3210.571. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

OpenBenchmarking.orgMflops, More Is BetterLuaJIT 2.1-gitTest: Monte Carlo7742 2P Repeat90180270360450SE +/- 0.31, N = 3412.341. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector

290 Results Shown

HPL Linpack
ASKAP:
  tConvolve MT - Degridding
  tConvolve MT - Gridding
CloverLeaf
Quantum ESPRESSO
C-Blosc
IOR
GNU Radio:
  Hilbert Transform
  FM Deemphasis Filter
  IIR Filter
  FIR Filter
  Signal Source (Cosine)
  Five Back to Back FIR Filters
RELION
LeelaChessZero:
  BLAS
  Eigen
Timed GCC Compilation
libgav1
NWChem
OpenVKL
7-Zip Compression
High Performance Conjugate Gradient
IOR
Blender
WebP2 Image Encode
Timed HMMer Search
oneDNN:
  Recurrent Neural Network Inference - bf16bf16bf16 - CPU
  Recurrent Neural Network Inference - f32 - CPU
Timed Erlang/OTP Compilation
oneDNN
Nebular Empirical Analysis Tool
FFTW
Rodinia
Incompact3D
JPEG XL
YafaRay
DeepSpeech
Numpy Benchmark
LAMMPS Molecular Dynamics Simulator
LuaRadio:
  Complex Phase
  Hilbert Transform
  FM Deemphasis Filter
  Five Back to Back FIR Filters
libgav1
WebP2 Image Encode
oneDNN:
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
  Recurrent Neural Network Training - f32 - CPU
Monte Carlo Simulations of Ionised Nebulae
FFTW
IOR
oneDNN
Zstd Compression:
  19, Long Mode - Decompression Speed
  19, Long Mode - Compression Speed
miniFE
Rodinia
FFTW
Parboil
libgav1
Timed LLVM Compilation
OpenFOAM
Zstd Compression:
  8 - Decompression Speed
  8 - Compression Speed
GraphicsMagick:
  Noise-Gaussian
  HWB Color Space
Parboil
Zstd Compression:
  3, Long Mode - Decompression Speed
  3, Long Mode - Compression Speed
Ngspice
GraphicsMagick
Tungsten Renderer
Rodinia
Zstd Compression:
  19 - Decompression Speed
  19 - Compression Speed
VP9 libvpx Encoding
asmFish
QMCPACK
TTSIOD 3D Renderer
WebP2 Image Encode
Ngspice
XZ Compression
Kvazaar
libavif avifenc
IOR
ASKAP
Rodinia
Apache CouchDB
JPEG XL
Timed MrBayes Analysis
IOR
OSPray
FinanceBench
Timed Linux Kernel Compilation
JPEG XL
Timed Eigen Compilation
Montage Astronomical Image Mosaic Engine
ASKAP:
  tConvolve MPI - Gridding
  tConvolve MPI - Degridding
Timed GDB GNU Debugger Compilation
Zstd Compression:
  8, Long Mode - Decompression Speed
  8, Long Mode - Compression Speed
LuaJIT
JPEG XL Decoding:
  1
  All
DaCapo Benchmark
Kvazaar
Tachyon
ASKAP:
  tConvolve OpenMP - Degridding
  tConvolve OpenMP - Gridding
Radiance Benchmark
DaCapo Benchmark
Timed Wasmer Compilation
LZ4 Compression:
  9 - Decompression Speed
  9 - Compression Speed
  3 - Decompression Speed
  3 - Compression Speed
Blender
Build2
LuxCoreRender
Timed Godot Game Engine Compilation
LuxCoreRender
FinanceBench
GraphicsMagick
libavif avifenc
JPEG XL
Himeno Benchmark
GraphicsMagick:
  Sharpen
  Enhanced
  Swirl
JPEG XL
Algebraic Multi-Grid Benchmark
srsLTE
Kvazaar
JPEG XL
Liquid-DSP
srsLTE:
  PHY_DL_Test:
    UE Mb/s
    eNb Mb/s
FFTW
dcraw
Blender
x265
lzbench:
  XZ 0 - Decompression
  XZ 0 - Compression
LibRaw
eSpeak-NG Speech Engine
WebP Image Encode
oneDNN
Gzip Compression
Timed PHP Compilation
rav1e
AOBench
libavif avifenc
QuantLib
Blender
Zstd Compression:
  3 - Decompression Speed
  3 - Compression Speed
GROMACS
libavif avifenc
Etcpak
AOM AV1
OSPray
C-Ray
AOM AV1
x265
Tungsten Renderer
LZ4 Compression:
  1 - Decompression Speed
  1 - Compression Speed
Embree:
  Pathtracer ISPC - Asian Dragon Obj
  Pathtracer - Asian Dragon Obj
Rodinia
Timed MAFFT Alignment
Stockfish
Google SynthMark
oneDNN
rav1e
Crafty
OSPray
OpenFOAM
toyBrot Fractal Generator
lzbench:
  Libdeflate 1 - Decompression
  Libdeflate 1 - Compression
Blender
Timed Apache Compilation
Monkey Audio Encoding
lzbench:
  Brotli 2 - Decompression
  Brotli 2 - Compression
Ogg Audio Encoding
lzbench:
  Brotli 0 - Decompression
  Brotli 0 - Compression
RNNoise
lzbench:
  Zstd 8 - Decompression
  Zstd 8 - Compression
libavif avifenc
lzbench:
  Zstd 1 - Decompression
  Zstd 1 - Compression
  Crush 0 - Decompression
  Crush 0 - Compression
Etcpak
oneDNN:
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  Deconvolution Batch shapes_1d - f32 - CPU
SVT-AV1
Etcpak
WebP Image Encode
NAMD
Dolfyn
Liquid-DSP:
  256 - 256 - 57
  8 - 256 - 57
  64 - 256 - 57
  32 - 256 - 57
  16 - 256 - 57
  4 - 256 - 57
  2 - 256 - 57
  1 - 256 - 57
Timed FFmpeg Compilation
AOM AV1
FFTE
NAS Parallel Benchmarks
WebP2 Image Encode
Kvazaar
FLAC Audio Encoding
SVT-VP9
OSPray
Timed ImageMagick Compilation
Embree
Opus Codec Encoding
x264
oneDNN:
  IP Shapes 1D - f32 - CPU
  IP Shapes 1D - u8s8f32 - CPU
Embree
SVT-AV1
dav1d
oneDNN:
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
  Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU
NAS Parallel Benchmarks
Tungsten Renderer
oneDNN
Embree
NAS Parallel Benchmarks
POV-Ray
OSPray
toyBrot Fractal Generator
Timed MPlayer Compilation
Embree
Kvazaar
oneDNN
WebP Image Encode
DaCapo Benchmark
LAME MP3 Encoding
N-Queens
dav1d
ACES DGEMM
WebP2 Image Encode
NAS Parallel Benchmarks
oneDNN
Parboil
FFTW
System GZIP Decompression
Intel Open Image Denoise
m-queens
System ZLIB Decompression
Pennant
FFTW
oneDNN
LAMMPS Molecular Dynamics Simulator
NAS Parallel Benchmarks
libjpeg-turbo tjbench
OSPray
SVT-VP9
Etcpak
OSPray
Tungsten Renderer
Pennant
Kvazaar
SVT-VP9
toyBrot Fractal Generator
System XZ Decompression
libavif avifenc
toyBrot Fractal Generator
rays1bench
NAS Parallel Benchmarks
WebP Image Encode
Parboil
TSCP
WebP Image Encode
OSPray
NAS Parallel Benchmarks
LuaJIT:
  Jacobi Successive Over-Relaxation
  Dense LU Matrix Factorization
  Sparse Matrix Multiply
  Fast Fourier Transform
  Monte Carlo