Ryzen 9 3950X Ubuntu 20.04 LTS AMD Ryzen Threadripper 3970X 32-Core testing with a ASUS ROG ZENITH II EXTREME (0702 BIOS) and AMD Radeon VII 16GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2002026-VE-2002024PT53&grr&sor .
Ryzen 9 3950X Ubuntu 20.04 LTS Processor Motherboard Chipset Memory Disk Graphics Audio Network Monitor OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution Ryzen 9 3950X Threadripper 3970X AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR VIII HERO (WI-FI) (1201 BIOS) AMD Starship/Matisse 16GB 2000GB Force MP600 + 2000GB SABRENT llvmpipe 16GB AMD Navi 10 HDMI Audio Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 20.04 5.4.0-12-generic (x86_64) GNOME Shell 3.34.3 X Server 1.20.6 modesetting 1.20.6 3.3 Mesa 20.0.0-devel (git-d3fa18a 2020-01-30 focal-oibaf-ppa) (LLVM 9.0.1 128 bits) GCC 9.2.1 20200123 ext4 3840x2160 AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads) ASUS ROG ZENITH II EXTREME (0702 BIOS) 64GB 1000GB Force MP600 AMD Radeon VII 16GB (1801/1000MHz) AMD Vega 20 HDMI Audio ASUS MG28U Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 X Server 1.20.7 amdgpu 19.1.0 4.5 Mesa 19.2.4 (LLVM 9.0.0) OpenBenchmarking.org Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,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-build-config=bootstrap-lto-lean --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 Processor Details - Ryzen 9 3950X: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x8701013 - Threadripper 3970X: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025 Python Details - + Python 3.7.6 Security Details - 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 STIBP: conditional RSB filling + tsx_async_abort: Not affected
Ryzen 9 3950X Ubuntu 20.04 LTS octave-benchmark: qmcpack: mkl-dnn: Convolution Batch conv_all - f32 build-gcc: Time To Compile blender: Barbershop - CPU-Only hpcg: numpy: blender: Pabellon Barcelona - CPU-Only luxcorerender: DLSC luxcorerender: Rainbow Colors and Prism blender: Classroom - CPU-Only asmfish: 1024 Hash Memory, 26 Depth dav1d: Chimera 1080p 10-bit ospray: San Miguel - Path Tracer appleseed: Emily ospray: XFrog Forest - Path Tracer parboil: OpenMP LBM gromacs: Water Benchmark ospray: San Miguel - SciVis gegl: Cartoon askap: tConvolve MT - Degridding askap: tConvolve MT - Gridding npb: EP.D appleseed: Material Tester npb: BT.C v-ray: CPU aom-av1: Speed 4 Realtime mt-dgemm: Sustained Floating-Point Rate mrbayes: Primate Phylogeny Analysis ospray: NASA Streamlines - Path Tracer blender: BMW27 - CPU-Only appleseed: Disney Material indigobench: Bedroom indigobench: Supercar gegl: Wavelet Blur graphics-magick: Sharpen graphics-magick: Enhanced graphics-magick: Noise-Gaussian graphics-magick: Resizing graphics-magick: Rotate graphics-magick: Swirl graphics-magick: HWB Color Space himeno: Poisson Pressure Solver radiance: SMP Parallel build-llvm: Time To Compile stockfish: Total Time npb: LU.C rav1e: 9 gegl: Color Enhance build2: Time To Compile namd: ATPase Simulation - 327,506 Atoms minife: Small parboil: OpenMP MRI Gridding mkl-dnn: Recurrent Neural Network Training - f32 build-linux-kernel: Time To Compile mkl-dnn: Convolution Batch conv_3d - f32 compress-7zip: Compress Speed Test gegl: Antialias ospray: XFrog Forest - SciVis gegl: Rotate 90 Degrees aom-av1: Speed 5 Two-Pass embree: Pathtracer ISPC - Asian Dragon Obj embree: Pathtracer - Asian Dragon Obj aom-av1: Speed 6 Realtime gegl: Tile Glass askap: tConvolve OpenMP - Degridding askap: tConvolve OpenMP - Gridding gegl: Reflect coremark: CoreMark Size 666 - Iterations Per Second npb: SP.B ospray: NASA Streamlines - SciVis npb: FT.C embree: Pathtracer ISPC - Crown c-ray: Total Time - 4K, 16 Rays Per Pixel embree: Pathtracer - Crown ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping ospray: Magnetic Reconnection - SciVis vpxenc: Speed 5 inkscape: SVG Files To PNG embree: Pathtracer ISPC - Asian Dragon embree: Pathtracer - Asian Dragon rodinia: OpenMP Streamcluster crafty: Elapsed Time compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 tungsten: Water Caustic mkl-dnn: Deconvolution Batch deconv_1d - f32 npb: CG.C dav1d: Summer Nature 4K gimp: unsharp-mask mlpack: scikit_svm scikit-learn: parboil: OpenMP Stencil aom-av1: Speed 8 Realtime compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 gimp: auto-levels rodinia: OpenMP LavaMD tungsten: Hair gimp: rotate cloverleaf: Lagrangian-Eulerian Hydrodynamics rodinia: OpenMP CFD Solver darktable: Boat - CPU-only x265: H.265 1080p Video Encoding lammps: Rhodopsin Protein gegl: Crop svt-av1: Enc Mode 8 - 1080p gimp: resize rays1bench: Large Scene oidn: Memorial gegl: Scale darktable: Masskrug - CPU-only tungsten: Volumetric Caustic smallpt: Global Illumination Renderer; 128 Samples tungsten: Non-Exponential darktable: Server Room - CPU-only x264: H.264 Video Encoding mkl-dnn: Deconvolution Batch deconv_3d - f32 svt-vp9: Visual Quality Optimized - Bosphorus 1080p tscp: AI Chess Performance darktable: Server Rack - CPU-only ospray: Magnetic Reconnection - Path Tracer parboil: OpenMP CUTCP ctx-clock: Context Switch Time Ryzen 9 3950X Threadripper 3970X 3491.910 1239.3 1924.63 552.973 336.32 4.39961 427.65 282.89 2.90 2.80 223.93 53529390 92.66 1.94 209.453163 2.48 168.503128 1.219 25.64 88.168 1575.48 859.295 1081.43 128.980739 22719.47 28435 0.82 5.453255 71.532 7.30 84.09 119.752861 2.644 5.658 60.901 250 398 463 1743 789 1002 1603 5151.911554 170.808 216.743 52431679 26185.31 1.738 54.489 61.166 1.12262 3614.85 32.023638 276.215 39.667 19.8186 87868 36.648 4.75 36.633 1.27 18.2652 18.8993 20.90 28.767 2074.75 2226.07 28.907 723526.762660 8376.69 35.71 11232.55 19.3810 31.457 20.1765 656.144 16.88 29.59 21.075 21.1389 21.1958 29.473 9238828 24.551 21.4772 3.86247 5684.44 183.58 13.924 11.15 8.295 16.693681 62.32 15.819 10.780 15.213 13.6295 9.619 4.88 12.142 13.446 74.32 10.955 7.465 45.250 5.798 109.46 11.57 4.759 4.840 5.56177 5.297 5.26490 3.976 182.51 4.97893 201.76 1383174 0.164 250 1.382408 140 8.529 2601.7 778.937 576.160 187.72 10.9613 425.22 152.68 2.19 2.33 118.88 98209423 96.92 3.72 135.884696 4.77 42.063107 3.304 50 89.878 3834.47 2683.55 2903.61 129.256446 76011.81 46725 0.79 11.159777 73.296 13.83 46.86 67.91895 5.108 9.634 61.484 460 722 650 2683 718 1686 1524 4732.261828 171.095 116.545 92051339 67403.73 1.668 55.030 47.920 0.56254 11576.9 59.979718 187.895 23.643 4.79339 199448 38.088 9.15 37.295 1.19 35.8190 37.3960 15.70 30.233 4563.03 5620.81 29.088 1385656.062117 34674.00 66.67 35255.47 36.6547 16.328 38.6521 1101.22 33.33 25.29 24.109 41.5189 41.8189 14.588 8910970 18.353 18.5005 1.83938 10971.80 324.27 16.228 11.59 21.343 4.670667 40.63 8.463 13.007 8.063 8.13851 10.774 0.57 7.634 4.601 65.68 18.921 7.156 71.140 6.595 228.00 24.03 5.093 2.828 3.51755 2.804 2.60739 1.733 210.42 2.69438 368.40 1317021 0.068 500 0.982801 148 OpenBenchmarking.org
GNU Octave Benchmark OpenBenchmarking.org Seconds, Fewer Is Better GNU Octave Benchmark 5.1.0 Threadripper 3970X Ryzen 9 3950X 700 1400 2100 2800 3500 SE +/- 0.051, N = 3 SE +/- 3485.627, N = 3 8.529 3491.910
QMCPACK OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.8 Ryzen 9 3950X Threadripper 3970X 600 1200 1800 2400 3000 1239.3 2601.7 1. (CXX) g++ options: -fopenmp -fomit-frame-pointer -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -ffast-math -lm
MKL-DNN DNNL Harness: Convolution Batch conv_all - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_all - Data Type: f32 Threadripper 3970X Ryzen 9 3950X 400 800 1200 1600 2000 SE +/- 4.23, N = 3 SE +/- 0.75, N = 3 778.94 1924.63 MIN: 768.93 MIN: 1787.24 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
Timed GCC Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GCC Compilation 8.2 Time To Compile Ryzen 9 3950X Threadripper 3970X 120 240 360 480 600 SE +/- 0.88, N = 3 SE +/- 1.18, N = 3 552.97 576.16
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.81 Blend File: Barbershop - Compute: CPU-Only Threadripper 3970X Ryzen 9 3950X 70 140 210 280 350 SE +/- 0.15, N = 3 SE +/- 0.38, N = 3 187.72 336.32
High Performance Conjugate Gradient OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 Threadripper 3970X Ryzen 9 3950X 3 6 9 12 15 SE +/- 0.01350, N = 3 SE +/- 0.00848, N = 3 10.96130 4.39961 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
Numpy Benchmark OpenBenchmarking.org Score, More Is Better Numpy Benchmark Ryzen 9 3950X Threadripper 3970X 90 180 270 360 450 SE +/- 4.18, N = 3 SE +/- 0.60, N = 3 427.65 425.22
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.81 Blend File: Pabellon Barcelona - Compute: CPU-Only Threadripper 3970X Ryzen 9 3950X 60 120 180 240 300 SE +/- 0.32, N = 3 SE +/- 0.80, N = 3 152.68 282.89
LuxCoreRender Scene: DLSC OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.2 Scene: DLSC Ryzen 9 3950X Threadripper 3970X 0.6525 1.305 1.9575 2.61 3.2625 SE +/- 0.04, N = 3 SE +/- 0.22, N = 15 2.90 2.19 MIN: 2.74 / MAX: 3.04 MIN: 1.27 / MAX: 3.01
LuxCoreRender Scene: Rainbow Colors and Prism OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.2 Scene: Rainbow Colors and Prism Ryzen 9 3950X Threadripper 3970X 0.63 1.26 1.89 2.52 3.15 SE +/- 0.05, N = 15 SE +/- 0.01, N = 3 2.80 2.33 MIN: 2.39 / MAX: 3.09 MIN: 2.22 / MAX: 2.37
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.81 Blend File: Classroom - Compute: CPU-Only Threadripper 3970X Ryzen 9 3950X 50 100 150 200 250 SE +/- 0.19, N = 3 SE +/- 0.26, N = 3 118.88 223.93
asmFish 1024 Hash Memory, 26 Depth OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth Threadripper 3970X Ryzen 9 3950X 20M 40M 60M 80M 100M SE +/- 1021889.01, N = 3 SE +/- 721437.39, N = 4 98209423 53529390
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 0.5.0 Video Input: Chimera 1080p 10-bit Threadripper 3970X Ryzen 9 3950X 20 40 60 80 100 SE +/- 0.14, N = 3 SE +/- 0.22, N = 3 96.92 92.66 MIN: 61.24 / MAX: 179.55 MIN: 54.57 / MAX: 198.73 1. (CC) gcc options: -pthread
OSPray Demo: San Miguel - Renderer: Path Tracer OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer Threadripper 3970X Ryzen 9 3950X 0.837 1.674 2.511 3.348 4.185 SE +/- 0.01, N = 3 SE +/- 0.00, N = 5 3.72 1.94 MIN: 3.69 / MAX: 3.76 MIN: 1.9 / MAX: 1.96
Appleseed Scene: Emily OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Emily Threadripper 3970X Ryzen 9 3950X 50 100 150 200 250 135.88 209.45
OSPray Demo: XFrog Forest - Renderer: Path Tracer OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer Threadripper 3970X Ryzen 9 3950X 1.0733 2.1466 3.2199 4.2932 5.3665 SE +/- 0.01, N = 3 SE +/- 0.00, N = 6 4.77 2.48 MIN: 4.72 / MAX: 4.83 MIN: 2.46 / MAX: 2.51
Parboil Test: OpenMP LBM OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM Threadripper 3970X Ryzen 9 3950X 40 80 120 160 200 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 42.06 168.50 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
GROMACS Water Benchmark OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2019.4 Water Benchmark Threadripper 3970X Ryzen 9 3950X 0.7434 1.4868 2.2302 2.9736 3.717 SE +/- 0.003, N = 3 SE +/- 0.002, N = 3 3.304 1.219 1. (CXX) g++ options: -mavx2 -mfma -pthread -std=c++11 -O3 -funroll-all-loops -lrt -lpthread -lm
OSPray Demo: San Miguel - Renderer: SciVis OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: SciVis Threadripper 3970X Ryzen 9 3950X 11 22 33 44 55 SE +/- 0.00, N = 15 50.00 25.64 MIN: 47.62 / MAX: 52.63 MIN: 23.81 / MAX: 27.03
GEGL Operation: Cartoon OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Cartoon Ryzen 9 3950X Threadripper 3970X 20 40 60 80 100 SE +/- 0.26, N = 3 SE +/- 0.11, N = 3 88.17 89.88
ASKAP Test: tConvolve MT - Degridding OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MT - Degridding Threadripper 3970X Ryzen 9 3950X 800 1600 2400 3200 4000 SE +/- 1.73, N = 3 SE +/- 0.89, N = 3 3834.47 1575.48 1. (CXX) g++ options: -lpthread
ASKAP Test: tConvolve MT - Gridding OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MT - Gridding Threadripper 3970X Ryzen 9 3950X 600 1200 1800 2400 3000 SE +/- 5.06, N = 3 SE +/- 0.60, N = 3 2683.55 859.30 1. (CXX) g++ options: -lpthread
NAS Parallel Benchmarks Test / Class: EP.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D Threadripper 3970X Ryzen 9 3950X 600 1200 1800 2400 3000 SE +/- 1.66, N = 3 SE +/- 1.71, N = 3 2903.61 1081.43 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.2
Appleseed Scene: Material Tester OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Material Tester Ryzen 9 3950X Threadripper 3970X 30 60 90 120 150 128.98 129.26
NAS Parallel Benchmarks Test / Class: BT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C Threadripper 3970X Ryzen 9 3950X 16K 32K 48K 64K 80K SE +/- 727.43, N = 3 SE +/- 54.75, N = 3 76011.81 22719.47 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.2
Chaos Group V-RAY Mode: CPU OpenBenchmarking.org Ksamples, More Is Better Chaos Group V-RAY 4.10.07 Mode: CPU Threadripper 3970X Ryzen 9 3950X 10K 20K 30K 40K 50K SE +/- 646.47, N = 4 SE +/- 130.33, N = 3 46725 28435
AOM AV1 Encoder Mode: Speed 4 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 4 Realtime Ryzen 9 3950X Threadripper 3970X 0.1845 0.369 0.5535 0.738 0.9225 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.82 0.79 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
ACES DGEMM Sustained Floating-Point Rate OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate Threadripper 3970X Ryzen 9 3950X 3 6 9 12 15 SE +/- 0.111325, N = 15 SE +/- 0.041732, N = 3 11.159777 5.453255 1. (CC) gcc options: -O3 -march=native -fopenmp
Timed MrBayes Analysis Primate Phylogeny Analysis OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis Ryzen 9 3950X Threadripper 3970X 16 32 48 64 80 SE +/- 0.47, N = 3 SE +/- 0.01, N = 3 71.53 73.30 -lreadline 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
OSPray Demo: NASA Streamlines - Renderer: Path Tracer OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer Threadripper 3970X Ryzen 9 3950X 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.00, N = 12 13.83 7.30 MIN: 13.51 / MAX: 14.08 MIN: 6.94 / MAX: 7.46
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.81 Blend File: BMW27 - Compute: CPU-Only Threadripper 3970X Ryzen 9 3950X 20 40 60 80 100 SE +/- 0.16, N = 3 SE +/- 0.06, N = 3 46.86 84.09
Appleseed Scene: Disney Material OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Disney Material Threadripper 3970X Ryzen 9 3950X 30 60 90 120 150 67.92 119.75
IndigoBench Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.0.64 Scene: Bedroom Threadripper 3970X Ryzen 9 3950X 1.1493 2.2986 3.4479 4.5972 5.7465 SE +/- 0.006, N = 3 SE +/- 0.004, N = 3 5.108 2.644
IndigoBench Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.0.64 Scene: Supercar Threadripper 3970X Ryzen 9 3950X 3 6 9 12 15 SE +/- 0.008, N = 3 SE +/- 0.011, N = 3 9.634 5.658
GEGL Operation: Wavelet Blur OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Wavelet Blur Ryzen 9 3950X Threadripper 3970X 14 28 42 56 70 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 60.90 61.48
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Sharpen Threadripper 3970X Ryzen 9 3950X 100 200 300 400 500 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 460 250 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Enhanced Threadripper 3970X Ryzen 9 3950X 160 320 480 640 800 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 722 398 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: Noise-Gaussian OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Noise-Gaussian Threadripper 3970X Ryzen 9 3950X 140 280 420 560 700 SE +/- 1.20, N = 3 650 463 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Resizing Threadripper 3970X Ryzen 9 3950X 600 1200 1800 2400 3000 SE +/- 9.29, N = 3 SE +/- 5.21, N = 3 2683 1743 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Rotate Ryzen 9 3950X Threadripper 3970X 200 400 600 800 1000 SE +/- 3.46, N = 3 789 718 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Swirl Threadripper 3970X Ryzen 9 3950X 400 800 1200 1600 2000 SE +/- 3.06, N = 3 SE +/- 3.33, N = 3 1686 1002 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: HWB Color Space Ryzen 9 3950X Threadripper 3970X 300 600 900 1200 1500 SE +/- 0.58, N = 3 1603 1524 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver Ryzen 9 3950X Threadripper 3970X 1100 2200 3300 4400 5500 SE +/- 71.26, N = 3 SE +/- 29.55, N = 3 5151.91 4732.26 1. (CC) gcc options: -O3 -mavx2
Radiance Benchmark Test: SMP Parallel OpenBenchmarking.org Seconds, Fewer Is Better Radiance Benchmark 5.0 Test: SMP Parallel Ryzen 9 3950X Threadripper 3970X 40 80 120 160 200 170.81 171.10
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 6.0.1 Time To Compile Threadripper 3970X Ryzen 9 3950X 50 100 150 200 250 116.55 216.74
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time Threadripper 3970X Ryzen 9 3950X 20M 40M 60M 80M 100M SE +/- 404576.49, N = 3 SE +/- 221711.10, N = 3 92051339 52431679 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C Threadripper 3970X Ryzen 9 3950X 14K 28K 42K 56K 70K SE +/- 42.48, N = 3 SE +/- 9.01, N = 3 67403.73 26185.31 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.2
rav1e Speed: 9 OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.2.0 Speed: 9 Ryzen 9 3950X Threadripper 3970X 0.3911 0.7822 1.1733 1.5644 1.9555 SE +/- 0.001, N = 3 SE +/- 0.001, N = 4 1.738 1.668
GEGL Operation: Color Enhance OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Color Enhance Ryzen 9 3950X Threadripper 3970X 12 24 36 48 60 SE +/- 0.12, N = 3 SE +/- 0.07, N = 3 54.49 55.03
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.12 Time To Compile Threadripper 3970X Ryzen 9 3950X 14 28 42 56 70 SE +/- 0.14, N = 3 SE +/- 0.24, N = 3 47.92 61.17
NAMD ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms Threadripper 3970X Ryzen 9 3950X 0.2526 0.5052 0.7578 1.0104 1.263 SE +/- 0.00112, N = 3 SE +/- 0.00594, N = 3 0.56254 1.12262
miniFE Problem Size: Small OpenBenchmarking.org CG Mflops, More Is Better miniFE 2.2 Problem Size: Small Threadripper 3970X Ryzen 9 3950X 2K 4K 6K 8K 10K SE +/- 6.44, N = 3 SE +/- 4.80, N = 3 11576.90 3614.85 1. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi
Parboil Test: OpenMP MRI Gridding OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding Ryzen 9 3950X Threadripper 3970X 13 26 39 52 65 SE +/- 0.05, N = 3 SE +/- 0.16, N = 3 32.02 59.98 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
MKL-DNN DNNL Harness: Recurrent Neural Network Training - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Recurrent Neural Network Training - Data Type: f32 Threadripper 3970X Ryzen 9 3950X 60 120 180 240 300 SE +/- 0.07, N = 3 SE +/- 4.06, N = 4 187.90 276.22 MIN: 186.73 MIN: 230.49 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.4 Time To Compile Threadripper 3970X Ryzen 9 3950X 9 18 27 36 45 SE +/- 0.30, N = 5 SE +/- 0.34, N = 3 23.64 39.67
MKL-DNN DNNL Harness: Convolution Batch conv_3d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Convolution Batch conv_3d - Data Type: f32 Threadripper 3970X Ryzen 9 3950X 5 10 15 20 25 SE +/- 0.03709, N = 3 SE +/- 0.05904, N = 3 4.79339 19.81860 MIN: 4.6 MIN: 16.92 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test Threadripper 3970X Ryzen 9 3950X 40K 80K 120K 160K 200K SE +/- 665.62, N = 3 SE +/- 367.81, N = 3 199448 87868 1. (CXX) g++ options: -pipe -lpthread
GEGL Operation: Antialias OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Antialias Ryzen 9 3950X Threadripper 3970X 9 18 27 36 45 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 36.65 38.09
OSPray Demo: XFrog Forest - Renderer: SciVis OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: SciVis Threadripper 3970X Ryzen 9 3950X 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 9.15 4.75 MIN: 9.01 / MAX: 9.35 MIN: 4.67 / MAX: 4.78
GEGL Operation: Rotate 90 Degrees OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Rotate 90 Degrees Ryzen 9 3950X Threadripper 3970X 9 18 27 36 45 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 36.63 37.30
AOM AV1 Encoder Mode: Speed 5 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 5 Two-Pass Ryzen 9 3950X Threadripper 3970X 0.2858 0.5716 0.8574 1.1432 1.429 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 1.27 1.19 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer ISPC - Model: Asian Dragon Obj Threadripper 3970X Ryzen 9 3950X 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 35.82 18.27 MIN: 35.56 / MAX: 36.5 MIN: 17.98 / MAX: 18.87
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer - Model: Asian Dragon Obj Threadripper 3970X Ryzen 9 3950X 9 18 27 36 45 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 37.40 18.90 MIN: 37.05 / MAX: 38.06 MIN: 18.57 / MAX: 19.48
AOM AV1 Encoder Mode: Speed 6 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 6 Realtime Ryzen 9 3950X Threadripper 3970X 5 10 15 20 25 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 20.90 15.70 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
GEGL Operation: Tile Glass OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Tile Glass Ryzen 9 3950X Threadripper 3970X 7 14 21 28 35 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 28.77 30.23
ASKAP Test: tConvolve OpenMP - Degridding OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve OpenMP - Degridding Threadripper 3970X Ryzen 9 3950X 1000 2000 3000 4000 5000 SE +/- 13.29, N = 14 SE +/- 5.37, N = 3 4563.03 2074.75 1. (CXX) g++ options: -lpthread
ASKAP Test: tConvolve OpenMP - Gridding OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve OpenMP - Gridding Threadripper 3970X Ryzen 9 3950X 1200 2400 3600 4800 6000 SE +/- 57.72, N = 14 SE +/- 35.01, N = 3 5620.81 2226.07 1. (CXX) g++ options: -lpthread
GEGL Operation: Reflect OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Reflect Ryzen 9 3950X Threadripper 3970X 7 14 21 28 35 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 28.91 29.09
Coremark CoreMark Size 666 - Iterations Per Second OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second Threadripper 3970X Ryzen 9 3950X 300K 600K 900K 1200K 1500K SE +/- 2371.45, N = 3 SE +/- 1768.75, N = 3 1385656.06 723526.76 1. (CC) gcc options: -O2 -lrt" -lrt
NAS Parallel Benchmarks Test / Class: SP.B OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B Threadripper 3970X Ryzen 9 3950X 7K 14K 21K 28K 35K SE +/- 335.31, N = 3 SE +/- 18.14, N = 3 34674.00 8376.69 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.2
OSPray Demo: NASA Streamlines - Renderer: SciVis OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis Threadripper 3970X Ryzen 9 3950X 15 30 45 60 75 SE +/- 0.00, N = 12 SE +/- 0.00, N = 12 66.67 35.71 MIN: 62.5 / MAX: 71.43 MIN: 27.78 / MAX: 37.04
NAS Parallel Benchmarks Test / Class: FT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C Threadripper 3970X Ryzen 9 3950X 8K 16K 24K 32K 40K SE +/- 5.50, N = 3 SE +/- 16.75, N = 3 35255.47 11232.55 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.2
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer ISPC - Model: Crown Threadripper 3970X Ryzen 9 3950X 8 16 24 32 40 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 36.65 19.38 MIN: 36.21 / MAX: 37.4 MIN: 19.06 / MAX: 19.92
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Threadripper 3970X Ryzen 9 3950X 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 16.33 31.46 1. (CC) gcc options: -lm -lpthread -O3
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer - Model: Crown Threadripper 3970X Ryzen 9 3950X 9 18 27 36 45 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 38.65 20.18 MIN: 38.19 / MAX: 39.56 MIN: 19.42 / MAX: 20.75
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping Threadripper 3970X Ryzen 9 3950X 200 400 600 800 1000 SE +/- 5.63, N = 3 SE +/- 2.49, N = 3 1101.22 656.14 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++
OSPray Demo: Magnetic Reconnection - Renderer: SciVis OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: SciVis Threadripper 3970X Ryzen 9 3950X 8 16 24 32 40 SE +/- 0.00, N = 12 SE +/- 0.07, N = 4 33.33 16.88 MIN: 32.26 MIN: 15.15 / MAX: 17.24
VP9 libvpx Encoding Speed: Speed 5 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.2 Speed: Speed 5 Ryzen 9 3950X Threadripper 3970X 7 14 21 28 35 SE +/- 0.14, N = 3 SE +/- 0.15, N = 3 29.59 25.29 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
Inkscape Operation: SVG Files To PNG OpenBenchmarking.org Seconds, Fewer Is Better Inkscape Operation: SVG Files To PNG Ryzen 9 3950X Threadripper 3970X 6 12 18 24 30 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 21.08 24.11 1. Inkscape 0.92.4 (5da689c313, 2019-01-14)
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer ISPC - Model: Asian Dragon Threadripper 3970X Ryzen 9 3950X 9 18 27 36 45 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 41.52 21.14 MIN: 41.2 / MAX: 42.16 MIN: 20.71 / MAX: 21.87
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.6.1 Binary: Pathtracer - Model: Asian Dragon Threadripper 3970X Ryzen 9 3950X 10 20 30 40 50 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 41.82 21.20 MIN: 41.49 / MAX: 42.49 MIN: 20.7 / MAX: 21.98
Rodinia Test: OpenMP Streamcluster OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster Threadripper 3970X Ryzen 9 3950X 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.24, N = 3 14.59 29.47 1. (CXX) g++ options: -O2 -lOpenCL
Crafty Elapsed Time OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time Ryzen 9 3950X Threadripper 3970X 2M 4M 6M 8M 10M SE +/- 20277.84, N = 3 SE +/- 21294.36, N = 3 9238828 8910970 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Threadripper 3970X Ryzen 9 3950X 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 18.35 24.55 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
Tungsten Renderer Scene: Water Caustic OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Water Caustic Threadripper 3970X Ryzen 9 3950X 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 18.50 21.48 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -ljpeg -lpthread -ldl
MKL-DNN DNNL Harness: Deconvolution Batch deconv_1d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_1d - Data Type: f32 Threadripper 3970X Ryzen 9 3950X 0.8691 1.7382 2.6073 3.4764 4.3455 SE +/- 0.00225, N = 3 SE +/- 0.05128, N = 3 1.83938 3.86247 MIN: 1.79 MIN: 3.27 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
NAS Parallel Benchmarks Test / Class: CG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C Threadripper 3970X Ryzen 9 3950X 2K 4K 6K 8K 10K SE +/- 17.02, N = 3 SE +/- 12.88, N = 3 10971.80 5684.44 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 4.0.2
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.5.0 Video Input: Summer Nature 4K Threadripper 3970X Ryzen 9 3950X 70 140 210 280 350 SE +/- 1.03, N = 3 SE +/- 0.91, N = 3 324.27 183.58 MIN: 189.04 / MAX: 345.75 MIN: 128.65 / MAX: 191.05 1. (CC) gcc options: -pthread
GIMP Test: unsharp-mask OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.14 Test: unsharp-mask Ryzen 9 3950X Threadripper 3970X 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 13.92 16.23
Mlpack Benchmark Benchmark: scikit_svm OpenBenchmarking.org Seconds, Fewer Is Better Mlpack Benchmark Benchmark: scikit_svm Ryzen 9 3950X Threadripper 3970X 3 6 9 12 15 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 11.15 11.59
Scikit-Learn OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.22.1 Ryzen 9 3950X Threadripper 3970X 5 10 15 20 25 SE +/- 0.083, N = 3 SE +/- 0.137, N = 3 8.295 21.343
Parboil Test: OpenMP Stencil OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil Threadripper 3970X Ryzen 9 3950X 4 8 12 16 20 SE +/- 0.060228, N = 5 SE +/- 0.074790, N = 3 4.670667 16.693681 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
AOM AV1 Encoder Mode: Speed 8 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 8 Realtime Ryzen 9 3950X Threadripper 3970X 14 28 42 56 70 SE +/- 0.27, N = 3 SE +/- 0.57, N = 3 62.32 40.63 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Zstd Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Threadripper 3970X Ryzen 9 3950X 4 8 12 16 20 SE +/- 0.021, N = 3 SE +/- 0.006, N = 3 8.463 15.819 1. (CC) gcc options: -O3 -pthread -lz -llzma
GIMP Test: auto-levels OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.14 Test: auto-levels Ryzen 9 3950X Threadripper 3970X 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 10.78 13.01
Rodinia Test: OpenMP LavaMD OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD Threadripper 3970X Ryzen 9 3950X 4 8 12 16 20 SE +/- 0.010, N = 3 SE +/- 0.088, N = 3 8.063 15.213 1. (CXX) g++ options: -O2 -lOpenCL
Tungsten Renderer Scene: Hair OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Hair Threadripper 3970X Ryzen 9 3950X 4 8 12 16 20 SE +/- 0.03721, N = 3 SE +/- 0.05265, N = 3 8.13851 13.62950 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -ljpeg -lpthread -ldl
GIMP Test: rotate OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.14 Test: rotate Ryzen 9 3950X Threadripper 3970X 3 6 9 12 15 SE +/- 0.007, N = 3 SE +/- 0.017, N = 3 9.619 10.774
CloverLeaf Lagrangian-Eulerian Hydrodynamics OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics Threadripper 3970X Ryzen 9 3950X 1.098 2.196 3.294 4.392 5.49 SE +/- 0.00, N = 3 SE +/- 0.09, N = 12 0.57 4.88 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
Rodinia Test: OpenMP CFD Solver OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver Threadripper 3970X Ryzen 9 3950X 3 6 9 12 15 SE +/- 0.021, N = 3 SE +/- 0.076, N = 3 7.634 12.142 1. (CXX) g++ options: -O2 -lOpenCL
Darktable Test: Boat - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.3 Test: Boat - Acceleration: CPU-only Threadripper 3970X Ryzen 9 3950X 3 6 9 12 15 SE +/- 0.018, N = 3 SE +/- 0.138, N = 3 4.601 13.446
x265 H.265 1080p Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x265 3.1.2 H.265 1080p Video Encoding Ryzen 9 3950X Threadripper 3970X 20 40 60 80 100 SE +/- 0.39, N = 3 SE +/- 0.21, N = 3 74.32 65.68 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 9Jan2020 Model: Rhodopsin Protein Threadripper 3970X Ryzen 9 3950X 5 10 15 20 25 SE +/- 0.45, N = 15 SE +/- 0.27, N = 12 18.92 10.96 1. (CXX) g++ options: -O3 -rdynamic -ljpeg -lpng -lz -lfftw3 -lm
GEGL Operation: Crop OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Crop Threadripper 3970X Ryzen 9 3950X 2 4 6 8 10 SE +/- 0.031, N = 3 SE +/- 0.100, N = 3 7.156 7.465
SVT-AV1 Encoder Mode: Enc Mode 8 - Input: 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p Threadripper 3970X Ryzen 9 3950X 16 32 48 64 80 SE +/- 0.29, N = 3 SE +/- 0.28, N = 3 71.14 45.25 1. (CXX) g++ options: -fPIE -fPIC -pie
GIMP Test: resize OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.14 Test: resize Ryzen 9 3950X Threadripper 3970X 2 4 6 8 10 SE +/- 0.083, N = 4 SE +/- 0.049, N = 3 5.798 6.595
rays1bench Large Scene OpenBenchmarking.org mrays/s, More Is Better rays1bench 2020-01-09 Large Scene Threadripper 3970X Ryzen 9 3950X 50 100 150 200 250 SE +/- 0.35, N = 3 SE +/- 0.08, N = 3 228.00 109.46
Intel Open Image Denoise Scene: Memorial OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.0.0 Scene: Memorial Threadripper 3970X Ryzen 9 3950X 6 12 18 24 30 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 24.03 11.57
GEGL Operation: Scale OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Scale Ryzen 9 3950X Threadripper 3970X 1.1459 2.2918 3.4377 4.5836 5.7295 SE +/- 0.010, N = 3 SE +/- 0.013, N = 3 4.759 5.093
Darktable Test: Masskrug - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.3 Test: Masskrug - Acceleration: CPU-only Threadripper 3970X Ryzen 9 3950X 1.089 2.178 3.267 4.356 5.445 SE +/- 0.005, N = 3 SE +/- 0.034, N = 3 2.828 4.840
Tungsten Renderer Scene: Volumetric Caustic OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Volumetric Caustic Threadripper 3970X Ryzen 9 3950X 1.2514 2.5028 3.7542 5.0056 6.257 SE +/- 0.00591, N = 3 SE +/- 0.00496, N = 3 3.51755 5.56177 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -ljpeg -lpthread -ldl
Smallpt Global Illumination Renderer; 128 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples Threadripper 3970X Ryzen 9 3950X 1.1918 2.3836 3.5754 4.7672 5.959 SE +/- 0.019, N = 3 SE +/- 0.005, N = 3 2.804 5.297 1. (CXX) g++ options: -fopenmp -O3
Tungsten Renderer Scene: Non-Exponential OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Non-Exponential Threadripper 3970X Ryzen 9 3950X 1.1846 2.3692 3.5538 4.7384 5.923 SE +/- 0.00244, N = 3 SE +/- 0.01428, N = 3 2.60739 5.26490 1. (CXX) g++ options: -std=c++0x -march=znver1 -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -ljpeg -lpthread -ldl
Darktable Test: Server Room - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.3 Test: Server Room - Acceleration: CPU-only Threadripper 3970X Ryzen 9 3950X 0.8946 1.7892 2.6838 3.5784 4.473 SE +/- 0.001, N = 3 SE +/- 0.033, N = 3 1.733 3.976
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-09-25 H.264 Video Encoding Threadripper 3970X Ryzen 9 3950X 50 100 150 200 250 SE +/- 1.11, N = 3 SE +/- 2.98, N = 3 210.42 182.51 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
MKL-DNN DNNL Harness: Deconvolution Batch deconv_3d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN DNNL 1.1 Harness: Deconvolution Batch deconv_3d - Data Type: f32 Threadripper 3970X Ryzen 9 3950X 1.1203 2.2406 3.3609 4.4812 5.6015 SE +/- 0.00752, N = 3 SE +/- 0.07757, N = 3 2.69438 4.97893 MIN: 2.63 MIN: 4.16 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl
SVT-VP9 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Threadripper 3970X Ryzen 9 3950X 80 160 240 320 400 SE +/- 0.85, N = 3 SE +/- 2.36, N = 3 368.40 201.76 1. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance Ryzen 9 3950X Threadripper 3970X 300K 600K 900K 1200K 1500K SE +/- 1300.71, N = 5 SE +/- 629.00, N = 5 1383174 1317021 1. (CC) gcc options: -O3 -march=native
Darktable Test: Server Rack - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.3 Test: Server Rack - Acceleration: CPU-only Threadripper 3970X Ryzen 9 3950X 0.0369 0.0738 0.1107 0.1476 0.1845 SE +/- 0.000, N = 14 SE +/- 0.001, N = 3 0.068 0.164
OSPray Demo: Magnetic Reconnection - Renderer: Path Tracer OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: Path Tracer Threadripper 3970X Ryzen 9 3950X 110 220 330 440 550 500 250 MIN: 250 MIN: 200 / MAX: 333.33
Parboil Test: OpenMP CUTCP OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP Threadripper 3970X Ryzen 9 3950X 0.311 0.622 0.933 1.244 1.555 SE +/- 0.011716, N = 3 SE +/- 0.009060, N = 3 0.982801 1.382408 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
ctx_clock Context Switch Time OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time Ryzen 9 3950X Threadripper 3970X 30 60 90 120 150 140 148
Phoronix Test Suite v10.8.4