AMD Ryzen Threadripper 3990X benchmarking by Michael Larabel for a future article.
2020 Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Pop 20.04, Kernel: 5.4.0-7642-generic (x86_64), Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: amdgpu 19.1.0, OpenGL: 4.6 Mesa 20.0.8 (LLVM 10.0.0), Vulkan: 1.2.128, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: snd_usb_audio.ignore_ctl_error=1Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,noatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Graphics Notes: GLAMORJava Notes: OpenJDK Runtime Environment (build 11.0.8+10-post-Ubuntu-0ubuntu120.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
February 2023 Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB + 257GB Flash DriveGraphics: AMD Radeon RX 5700 8GB (1750/875MHz)Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 23.04, Kernel: 6.2.0-060200rc7daily20230206-generic (x86_64), Desktop: GNOME Shell 43.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.5 (LLVM 15.0.6 DRM 3.49), Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-AKimc9/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-AKimc9/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301055Graphics Notes: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D1820201-101Java Notes: OpenJDK Runtime Environment (build 11.0.18+10-post-Ubuntu-0ubuntu1)Python Notes: Python 3.11.1Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
System76 Thelio Major 3990X 2023 Benchmarks Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Vulkan Compiler File-System Screen Resolution 2020 February 2023 AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads) Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS) AMD Starship/Matisse 128GB Samsung SSD 970 EVO Plus 500GB AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz) AMD Navi 10 HDMI Audio DELL P2415Q Intel I211 + Intel Wi-Fi 6 AX200 Pop 20.04 5.4.0-7642-generic (x86_64) GNOME Shell 3.36.4 X Server 1.20.8 amdgpu 19.1.0 4.6 Mesa 20.0.8 (LLVM 10.0.0) 1.2.128 GCC 9.3.0 ext4 3840x2160 Samsung SSD 970 EVO Plus 500GB + 257GB Flash Drive AMD Radeon RX 5700 8GB (1750/875MHz) Ubuntu 23.04 6.2.0-060200rc7daily20230206-generic (x86_64) GNOME Shell 43.2 X Server + Wayland 4.6 Mesa 22.2.5 (LLVM 15.0.6 DRM 3.49) GCC 12.2.0 OpenBenchmarking.org Kernel Details - 2020: snd_usb_audio.ignore_ctl_error=1 - February 2023: Transparent Huge Pages: madvise Compiler Details - 2020: --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 - February 2023: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-AKimc9/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-AKimc9/gcc-12-12.2.0/debian/tmp-gcn/usr --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 Disk Details - 2020: NONE / errors=remount-ro,noatime,rw - February 2023: NONE / errors=remount-ro,relatime,rw / Block Size: 4096 Processor Details - 2020: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025 - February 2023: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301055 Graphics Details - 2020: GLAMOR - February 2023: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D1820201-101 Java Details - 2020: OpenJDK Runtime Environment (build 11.0.8+10-post-Ubuntu-0ubuntu120.04) - February 2023: OpenJDK Runtime Environment (build 11.0.18+10-post-Ubuntu-0ubuntu1) Python Details - 2020: Python 3.8.5 - February 2023: Python 3.11.1 Security Details - 2020: 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 + srbds: Not affected + tsx_async_abort: Not affected - February 2023: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
2020 vs. February 2023 Comparison Phoronix Test Suite Baseline +111.2% +111.2% +222.4% +222.4% +333.6% +333.6% 444.7% 178.2% 172.3% 120.2% 110.2% 108.1% 100% 99.1% 91.7% 90.7% 79.6% 71.3% 68.9% 67.1% 66.2% 61.2% 59.8% 58.5% 56.3% 55.7% 54.1% 50.8% 49.9% 47% 41.9% 40.7% 36.4% 34.7% 33.2% 32.9% 24% 19.6% 18.4% 17.3% 16.6% 15.3% 15.2% 14.5% 14.2% 13.5% 11.9% 11.8% 11.3% 10.5% 9.4% 9.1% 8.6% 8.3% 8.2% 7.6% 7.6% 6.9% 6.6% 6.5% 6.5% 6.4% 6.1% 6% 4.9% 4.6% 4.5% 4.4% 4.1% 4% 3.9% 3.9% 3.8% 3.7% 3.6% 3.4% 3.3% 2.6% 2.5% 2.5% 2.4% 2.3% 2.2% 2.1% Vulkan GPU - alexnet Vulkan GPU - vgg16 Vulkan GPU-v3-v3 - mobilenet-v3 2.F.P.1.T 4x - Yes Vulkan GPU-v2-v2 - mobilenet-v2 Vulkan GPU - shufflenet-v2 Vulkan GPU - mnasnet Vulkan GPU - googlenet Vulkan GPU - resnet50 Vulkan GPU - resnet18 go Dust 2D tau100.0 Vulkan GPU - yolov4-tiny float chaos pickle_pure_python raytrace Vulkan GPU - efficientnet-b0 django_template 2x - 3 - Yes Vulkan GPU - squeezenet crypto_pyaes Vulkan GPU - mobilenet P.I.O.A.3.V.1.T Image Rendering 4x - No nbody Time To Compile 34.5% fatigue2 1920 x 1080 Bosphorus 4K 24.6% regex_compile Bosphorus 1080p 23.2% V.Q.O - Bosphorus 1080p 20% air 3840 x 2160 - High 1.8.b.Y.T.H.V.E 18.3% 3840 x 2160 - Ultra P.S.O - Bosphorus 1080p 16.7% json_loads Color Enhance Dhrystone 2 Time To Compile 15% Time To Compile 14.6% Wavelet Blur 8 A.R.T 3840 x 2160 - Ultimate T.F.A.T.T 3840 x 2160 Antialias Classroom - CPU-Only 10.4% 2to3 Time To Compile Barbershop - CPU-Only 9% vklBenchmark 9% 3840 x 2160 - Low Seek Rand F.F.T Wavelet Contour - 1920 x 1080 Wavelet Contour - 1920 x 1080 Reflect Tile Glass Renderer2 - 3840 x 2160 S.M.M M.M.B.S.T - f32 - CPU 6.5% Memorial Fishy Cat - CPU-Only 6.3% D.L.M.F Text Rendering R.N.N.I - f32 - CPU 5.9% CoreMark Size 666 - I.P.S 5.8% T.B.T 5.7% IP Batch 1D - f32 - CPU 5.2% O.B 5.1% San Miguel - SciVis Crop Composite SqueezeNet XFrog Forest - Path Tracer 4.1% CPU - yolov4-tiny V.G.R P.P.A.S.T Rand Read Cartoon 3.8% 20k Atoms 3.8% tfft2 San Miguel - Path Tracer 3.6% WAV To FLAC NASA Streamlines - Path Tracer 3.4% BMW27 - CPU-Only 3.4% N.2.3.C.F.R Q.1.H.C 3.4% Rhodopsin Protein IP Batch All - f32 - CPU 3.2% R.N.N.T - f32 - CPU 3% Total Time 2.9% XFrog Forest - SciVis 2.8% Rand Fill pathlib Rand Fill CPU - SqueezeNet v1.1 protein Emily D.B.d - f32 - CPU NCNN NCNN NCNN G'MIC RealSR-NCNN NCNN NCNN NCNN NCNN NCNN NCNN PyPerformance Monte Carlo Simulations of Ionised Nebulae NCNN PyPerformance PyPerformance PyPerformance PyPerformance NCNN PyPerformance Waifu2x-NCNN Vulkan NCNN PyPerformance NCNN G'MIC Java 2D Microbenchmark RealSR-NCNN PyPerformance Timed PHP Compilation Polyhedron Fortran Benchmarks Tesseract x265 PyPerformance x265 SVT-VP9 Polyhedron Fortran Benchmarks Xonotic SVT-HEVC Xonotic SVT-VP9 PyPerformance GEGL BYTE Unix Benchmark Timed Apache Compilation Timed MPlayer Compilation GEGL SQLite Java 2D Microbenchmark Xonotic PyBench Tesseract GEGL Blender PyPerformance Timed FFmpeg Compilation Blender OpenVKL Xonotic LevelDB LuaJIT ParaView ParaView GEGL GEGL ET: Legacy LuaJIT oneDNN Intel Open Image Denoise Blender LuaJIT Java 2D Microbenchmark oneDNN Coremark RawTherapee oneDNN Optcarrot OSPray GEGL LuaJIT TensorFlow Lite OSPray RNNoise NCNN Java 2D Microbenchmark Hugin LevelDB GEGL LAMMPS Molecular Dynamics Simulator Polyhedron Fortran Benchmarks OSPray FLAC Audio Encoding OSPray Blender FFTE WebP Image Encode LAMMPS Molecular Dynamics Simulator oneDNN oneDNN Stockfish OSPray LevelDB PyPerformance LevelDB TNN Polyhedron Fortran Benchmarks Appleseed oneDNN 2020 February 2023
System76 Thelio Major 3990X 2023 Benchmarks ospray: San Miguel - SciVis ospray: XFrog Forest - SciVis ospray: San Miguel - Path Tracer ospray: NASA Streamlines - SciVis ospray: XFrog Forest - Path Tracer ospray: Magnetic Reconnection - SciVis ospray: NASA Streamlines - Path Tracer optcarrot: Optimized Benchmark paraview: Wavelet Contour - 1920 x 1080 etlegacy: Renderer2 - 3840 x 2160 tesseract: 1920 x 1080 tesseract: 3840 x 2160 xonotic: 3840 x 2160 - Low xonotic: 3840 x 2160 - High xonotic: 3840 x 2160 - Ultra xonotic: 3840 x 2160 - Ultimate svt-hevc: 1080p 8-bit YUV To HEVC Video Encode svt-vp9: PSNR/SSIM Optimized - Bosphorus 1080p svt-vp9: Visual Quality Optimized - Bosphorus 1080p x265: Bosphorus 4K x265: Bosphorus 1080p oidn: Memorial openvkl: vklBenchmark coremark: CoreMark Size 666 - Iterations Per Second byte: Dhrystone 2 compress-zstd: 3 compress-zstd: 19 leveldb: Fill Sync leveldb: Overwrite leveldb: Rand Fill leveldb: Seq Fill ffte: N=256, 3D Complex FFT Routine luajit: Composite luajit: Monte Carlo luajit: Fast Fourier Transform luajit: Sparse Matrix Multiply luajit: Dense LU Matrix Factorization luajit: Jacobi Successive Over-Relaxation paraview: Wavelet Contour - 1920 x 1080 libraw: Post-Processing Benchmark rays1bench: Large Scene stockfish: Total Time lammps: 20k Atoms lammps: Rhodopsin Protein hint: FLOAT openssl: RSA 4096-bit Performance kripke: j2dbench: Text Rendering j2dbench: Image Rendering j2dbench: All Rendering Tests j2dbench: Vector Graphics Rendering namd: ATPase Simulation - 327,506 Atoms webp: Quality 100, Lossless webp: Quality 100, Highest Compression webp: Quality 100, Lossless, Highest Compression leveldb: Fill Sync leveldb: Overwrite leveldb: Rand Fill leveldb: Rand Read leveldb: Seek Rand leveldb: Rand Delete leveldb: Seq Fill tensorflow-lite: SqueezeNet pybench: Total For Average Test Times pyperformance: go pyperformance: 2to3 pyperformance: chaos pyperformance: float pyperformance: nbody pyperformance: pathlib pyperformance: raytrace pyperformance: json_loads pyperformance: crypto_pyaes pyperformance: regex_compile pyperformance: python_startup pyperformance: django_template pyperformance: pickle_pure_python onednn: IP Batch 1D - f32 - CPU onednn: IP Batch All - f32 - CPU onednn: Deconvolution Batch deconv_1d - f32 - CPU onednn: Deconvolution Batch deconv_3d - f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU ncnn: CPU - yolov4-tiny ncnn: Vulkan GPU - squeezenet ncnn: Vulkan GPU - mobilenet ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - efficientnet-b0 ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - vgg16 ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - yolov4-tiny ncnn: Vulkan GPU - resnet50 tnn: CPU - SqueezeNet v1.1 sqlite: 8 realsr-ncnn: 4x - No realsr-ncnn: 4x - Yes waifu2x-ncnn: 2x - 3 - Yes polyhedron: air polyhedron: tfft2 polyhedron: induct2 polyhedron: protein polyhedron: fatigue2 hmmer: Pfam Database Search mocassin: Dust 2D tau100.0 build-apache: Time To Compile build-ffmpeg: Time To Compile build-mplayer: Time To Compile build-php: Time To Compile encode-flac: WAV To FLAC rnnoise: couchdb: 100 - 1000 - 24 gegl: Crop gegl: Cartoon gegl: Reflect gegl: Antialias gegl: Tile Glass gegl: Wavelet Blur gegl: Color Enhance gmic: 2D Function Plotting, 1000 Times gmic: Plotting Isosurface Of A 3D Volume, 1000 Times gmic: 3D Elevated Function In Rand Colors, 100 Times hugin: Panorama Photo Assistant + Stitching Time rawtherapee: Total Benchmark Time blender: BMW27 - CPU-Only blender: Classroom - CPU-Only blender: Fishy Cat - CPU-Only blender: Barbershop - CPU-Only appleseed: Emily git: Time To Complete Common Git Commands 2020 February 2023 67.46 13.76 5.70 90.91 7.32 43.48 20.41 131.33 301.88 241.5 369.0690 311.5822 506.6888296 383.4692564 335.6182130 268.3078375 208.29 394.07 326.72 26.25 58.19 27.01 461.15 2256436.977296 43076608.7 7285.8 81.7 2.4 22.8 22.7 23.9 129515.00422893 1541.47 506.84 254.69 1221.29 3668.79 2055.73 3145.937 41.83 311.30 146463486 26.908 23.845 373293210.02252 15077.3 44881177 21836.339072177 8553596.5131508 7815320.9604010 7844283.9750621 0.42677 16.295 7.138 33.500 5953.960 621.251 623.726 109.997 171.029 597.960 591.612 50791.3 941 233 302 104 108 108 16.3 439 23.9 103 160 12.8 47.5 441 1.26441 23.7781 1.55097 2.07153 328.325 115.068 0.508544 36.30 5.82 9.98 4.60 8.47 3.48 4.66 11.55 8.51 84.36 3.34 29.85 15.04 9.65 252.180 114.997 16.512 114.444 9.101 1.71 49.08 21.92 12.72 48.61 165.035 223 21.267 19.736 10.613 35.452 7.971 18.809 119.187 8.843 94.521 31.591 41.076 32.965 64.725 58.097 222.556 19.379 84.886 46.739 50.510 32.43 84.95 46.34 131.12 133.244761 48.365 70.75 13.39 5.50 90.91 7.03 43.48 19.74 124.95 324.76 257.3 490.6299 346.8381 550.0451355 453.8810679 393.5244615 300.1137937 176.07 337.63 272.24 21.06 47.23 28.74 423 2132341.733949 49644913.1 7196.0 82.3 2.4 23.2 23.3 23.9 133898.12115368 1610.41 513.88 275.53 1301.16 3894.04 2067.42 3384.404 41.90 309.57 142312246 25.935 24.629 367154749.55730 14942.8 45462287 23141.430420648 12036752.943046 8868529.0188221 8150041.0127656 0.42898 16.120 7.378 33.289 5975.095 611.019 608.572 105.932 157.938 595.580 592.276 48673.6 842 136 276 64.5 65.0 80.2 15.9 277 20.5 68.7 129 12.6 30.5 276 1.32984 24.5278 1.55913 2.02930 338.110 121.800 0.541529 34.92 3.86 6.79 2.21 3.11 1.74 2.34 7.39 4.44 30.32 1.86 5.48 9.00 5.06 246.191 100.692 12.109 54.444 5.904 1.43 47.35 21.85 12.44 36.5 167.563 132 24.458 18.097 12.162 47.695 7.692 18.069 120.379 8.453 98.075 29.545 37.163 30.932 56.548 50.397 101.076 13.654 86.383 44.999 53.400 33.53 93.81 49.26 142.95 130.329085 47.817 OpenBenchmarking.org
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.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: SciVis February 2023 2020 16 32 48 64 80 SE +/- 0.46, N = 14 SE +/- 0.79, N = 6 70.75 67.46 MIN: 62.5 / MAX: 76.92 MIN: 62.5 / MAX: 76.92
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: SciVis 2020 February 2023 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 13.76 13.39 MIN: 13.51 / MAX: 14.08 MIN: 12.99 / MAX: 13.7
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: San Miguel - Renderer: Path Tracer 2020 February 2023 1.2825 2.565 3.8475 5.13 6.4125 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.70 5.50 MIN: 5.65 / MAX: 5.78 MIN: 5.46 / MAX: 5.68
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: SciVis February 2023 2020 20 40 60 80 100 SE +/- 0.00, N = 3 SE +/- 0.00, N = 13 90.91 90.91 MIN: 83.33 / MAX: 100 MIN: 83.33 / MAX: 100
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: XFrog Forest - Renderer: Path Tracer 2020 February 2023 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 7.32 7.03 MIN: 7.04 / MAX: 7.52 MIN: 6.9 / MAX: 7.25
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: Magnetic Reconnection - Renderer: SciVis February 2023 2020 10 20 30 40 50 SE +/- 0.00, N = 3 SE +/- 0.00, N = 12 43.48 43.48 MIN: 41.67 / MAX: 45.45 MIN: 41.67 / MAX: 45.45
OpenBenchmarking.org FPS, More Is Better OSPray 1.8.5 Demo: NASA Streamlines - Renderer: Path Tracer 2020 February 2023 5 10 15 20 25 SE +/- 0.00, N = 12 SE +/- 0.13, N = 3 20.41 19.74 MIN: 20 / MAX: 21.28 MIN: 19.23 / MAX: 20.41
Optcarrot Optcarrot is an NES emulator benchmark for the Ruby language. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better Optcarrot Optimized Benchmark 2020 February 2023 30 60 90 120 150 SE +/- 0.25, N = 4 SE +/- 1.09, N = 4 131.33 124.95 1. 2020: ruby 2.7.0p0 (2019-12-25 revision 647ee6f091) [x86_64-linux-gnu] 2. February 2023: ruby 3.0.4p208 (2022-04-12 revision 3fa771dded) [x86_64-linux-gnu]
ET: Legacy ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better ET: Legacy 2.75 Renderer: Renderer2 - Resolution: 3840 x 2160 February 2023 2020 60 120 180 240 300 SE +/- 2.02, N = 3 SE +/- 3.22, N = 3 257.3 241.5
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: Low February 2023 2020 120 240 360 480 600 SE +/- 3.70, N = 3 SE +/- 1.68, N = 3 550.05 506.69 MIN: 252 / MAX: 1144 MIN: 316 / MAX: 1048
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: High February 2023 2020 100 200 300 400 500 SE +/- 4.55, N = 3 SE +/- 3.24, N = 3 453.88 383.47 MIN: 209 / MAX: 809 MIN: 235 / MAX: 737
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: Ultra February 2023 2020 90 180 270 360 450 SE +/- 4.49, N = 3 SE +/- 0.35, N = 3 393.52 335.62 MIN: 185 / MAX: 682 MIN: 192 / MAX: 628
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: Ultimate February 2023 2020 70 140 210 280 350 SE +/- 1.55, N = 3 SE +/- 1.88, N = 3 300.11 268.31 MIN: 55 / MAX: 559 MIN: 52 / MAX: 532
SVT-HEVC This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.4.1 1080p 8-bit YUV To HEVC Video Encode 2020 February 2023 50 100 150 200 250 SE +/- 1.79, N = 3 SE +/- 2.46, N = 3 208.29 176.07 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -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.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p 2020 February 2023 90 180 270 360 450 SE +/- 2.19, N = 3 SE +/- 3.73, N = 5 394.07 337.63 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p 2020 February 2023 70 140 210 280 350 SE +/- 2.60, N = 3 SE +/- 0.81, N = 3 326.72 272.24 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
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.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 4K 2020 February 2023 6 12 18 24 30 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 26.25 21.06 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 1080p 2020 February 2023 13 26 39 52 65 SE +/- 0.21, N = 3 SE +/- 0.14, N = 3 58.19 47.23 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
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.org Items / Sec, More Is Better OpenVKL 0.9 Benchmark: vklBenchmark 2020 February 2023 100 200 300 400 500 SE +/- 1.89, N = 3 SE +/- 2.19, N = 3 461.15 423.00 MIN: 1 / MAX: 1298 MIN: 1 / MAX: 1246
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Overwrite February 2023 2020 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 23.2 22.8 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Random Fill February 2023 2020 6 12 18 24 30 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 23.3 22.7 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Sequential Fill February 2023 2020 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.27, N = 3 23.9 23.9 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
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.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine February 2023 2020 30K 60K 90K 120K 150K SE +/- 232.46, N = 3 SE +/- 641.63, N = 3 133898.12 129515.00 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Monte Carlo February 2023 2020 110 220 330 440 550 SE +/- 0.81, N = 3 SE +/- 0.10, N = 3 513.88 506.84 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Fast Fourier Transform February 2023 2020 60 120 180 240 300 SE +/- 0.61, N = 3 SE +/- 23.04, N = 3 275.53 254.69 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Sparse Matrix Multiply February 2023 2020 300 600 900 1200 1500 SE +/- 1.63, N = 3 SE +/- 3.63, N = 3 1301.16 1221.29 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Dense LU Matrix Factorization February 2023 2020 800 1600 2400 3200 4000 SE +/- 7.37, N = 3 SE +/- 18.22, N = 3 3894.04 3668.79 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation February 2023 2020 400 800 1200 1600 2000 SE +/- 4.22, N = 3 SE +/- 3.21, N = 3 2067.42 2055.73 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
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.org Nodes Per Second, More Is Better Stockfish 9 Total Time 2020 February 2023 30M 60M 90M 120M 150M SE +/- 263429.87, N = 3 SE +/- 803242.53, N = 3 146463486 142312246 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance 2020 February 2023 3K 6K 9K 12K 15K SE +/- 116.95, N = 3 SE +/- 81.34, N = 3 15077.3 14942.8 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Kripke Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 February 2023 2020 10M 20M 30M 40M 50M SE +/- 33206.63, N = 3 SE +/- 167341.17, N = 3 45462287 44881177 1. (CXX) g++ options: -O3 -fopenmp
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.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms 2020 February 2023 0.0965 0.193 0.2895 0.386 0.4825 SE +/- 0.00043, N = 3 SE +/- 0.00183, N = 3 0.42677 0.42898
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.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless February 2023 2020 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 16.12 16.30 -pthread 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression 2020 February 2023 2 4 6 8 10 SE +/- 0.015, N = 3 SE +/- 0.005, N = 3 7.138 7.378 -pthread 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff
OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression February 2023 2020 8 16 24 32 40 SE +/- 0.14, N = 3 SE +/- 0.22, N = 3 33.29 33.50 -pthread 1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16 -ltiff
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Fill Sync 2020 February 2023 1300 2600 3900 5200 6500 SE +/- 15.69, N = 3 SE +/- 19.28, N = 3 5953.96 5975.10 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Overwrite February 2023 2020 130 260 390 520 650 SE +/- 0.82, N = 3 SE +/- 0.23, N = 3 611.02 621.25 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Fill February 2023 2020 130 260 390 520 650 SE +/- 1.42, N = 3 SE +/- 1.02, N = 3 608.57 623.73 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Read February 2023 2020 20 40 60 80 100 SE +/- 1.20, N = 3 SE +/- 0.20, N = 3 105.93 110.00 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Seek Random February 2023 2020 40 80 120 160 200 SE +/- 0.23, N = 3 SE +/- 0.18, N = 3 157.94 171.03 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Delete February 2023 2020 130 260 390 520 650 SE +/- 0.94, N = 3 SE +/- 0.60, N = 3 595.58 597.96 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Sequential Fill 2020 February 2023 130 260 390 520 650 SE +/- 6.42, N = 3 SE +/- 1.08, N = 3 591.61 592.28 -lpthread 1. (CXX) g++ options: -O3 -lsnappy
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times February 2023 2020 200 400 600 800 1000 SE +/- 4.67, N = 3 SE +/- 2.40, N = 3 842 941
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 oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: f32 - Engine: CPU 2020 February 2023 0.2992 0.5984 0.8976 1.1968 1.496 SE +/- 0.01741, N = 4 SE +/- 0.04058, N = 15 1.26441 1.32984 -lpthread - MIN: 1.17 MIN: 1.15 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: f32 - Engine: CPU 2020 February 2023 6 12 18 24 30 SE +/- 0.30, N = 3 SE +/- 0.27, N = 4 23.78 24.53 -lpthread - MIN: 22.66 MIN: 23.59 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU 2020 February 2023 0.3508 0.7016 1.0524 1.4032 1.754 SE +/- 0.01640, N = 8 SE +/- 0.01350, N = 3 1.55097 1.55913 -lpthread - MIN: 1.45 MIN: 1.49 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU February 2023 2020 0.4661 0.9322 1.3983 1.8644 2.3305 SE +/- 0.00541, N = 3 SE +/- 0.01201, N = 3 2.02930 2.07153 MIN: 1.95 -lpthread - MIN: 1.93 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 2020 February 2023 70 140 210 280 350 SE +/- 0.49, N = 3 SE +/- 2.43, N = 15 328.33 338.11 -lpthread - MIN: 324.39 MIN: 323.96 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 2020 February 2023 30 60 90 120 150 SE +/- 0.34, N = 3 SE +/- 0.99, N = 3 115.07 121.80 -lpthread - MIN: 112.93 MIN: 118.31 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 2020 February 2023 0.1218 0.2436 0.3654 0.4872 0.609 SE +/- 0.006670, N = 4 SE +/- 0.005684, N = 15 0.508544 0.541529 -lpthread - MIN: 0.46 MIN: 0.46 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -ldl
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: yolov4-tiny February 2023 2020 8 16 24 32 40 SE +/- 0.15, N = 3 SE +/- 0.24, N = 3 34.92 36.30 MIN: 34.15 / MAX: 38.74 MIN: 35.04 / MAX: 56.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: squeezenet February 2023 2020 1.3095 2.619 3.9285 5.238 6.5475 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 3.86 5.82 MIN: 3.7 / MAX: 4.78 MIN: 5.53 / MAX: 19.72 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: mobilenet February 2023 2020 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.16, N = 3 6.79 9.98 MIN: 6.71 / MAX: 7.44 MIN: 8.51 / MAX: 38.16 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 February 2023 2020 1.035 2.07 3.105 4.14 5.175 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 2.21 4.60 MIN: 2.16 / MAX: 4.69 MIN: 4.27 / MAX: 9.6 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 February 2023 2020 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.24, N = 3 3.11 8.47 MIN: 3.06 / MAX: 6.37 MIN: 7.45 / MAX: 27.99 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: shufflenet-v2 February 2023 2020 0.783 1.566 2.349 3.132 3.915 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.74 3.48 MIN: 1.72 / MAX: 2.67 MIN: 3.38 / MAX: 4.25 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: mnasnet February 2023 2020 1.0485 2.097 3.1455 4.194 5.2425 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 2.34 4.66 MIN: 2.3 / MAX: 2.63 MIN: 4.41 / MAX: 5.4 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: efficientnet-b0 February 2023 2020 3 6 9 12 15 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 7.39 11.55 MIN: 6.9 / MAX: 28.13 MIN: 10.46 / MAX: 34.72 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: googlenet February 2023 2020 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 4.44 8.51 MIN: 4.41 / MAX: 4.84 MIN: 7.13 / MAX: 29.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: vgg16 February 2023 2020 20 40 60 80 100 SE +/- 3.74, N = 3 SE +/- 0.04, N = 3 30.32 84.36 MIN: 13.86 / MAX: 49.77 MIN: 67.87 / MAX: 109.54 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: resnet18 February 2023 2020 0.7515 1.503 2.2545 3.006 3.7575 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 1.86 3.34 MIN: 1.79 / MAX: 10.57 MIN: 3.22 / MAX: 5.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: alexnet February 2023 2020 7 14 21 28 35 SE +/- 0.03, N = 3 SE +/- 0.24, N = 3 5.48 29.85 MIN: 5.13 / MAX: 10.61 MIN: 25.42 / MAX: 55.99 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: yolov4-tiny February 2023 2020 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 9.00 15.04 MIN: 8.84 / MAX: 12.27 MIN: 12.6 / MAX: 42.1 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: resnet50 February 2023 2020 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.09, N = 2 5.06 9.65 MIN: 4.98 / MAX: 14.49 MIN: 9.05 / MAX: 31.86 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
TNN TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 February 2023 2020 60 120 180 240 300 SE +/- 0.90, N = 3 SE +/- 0.59, N = 3 246.19 252.18 MIN: 244.08 / MAX: 249.94 MIN: 249.83 / MAX: 254.13 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
SQLite This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.30.1 Threads / Copies: 8 February 2023 2020 30 60 90 120 150 SE +/- 0.31, N = 3 SE +/- 0.34, N = 3 100.69 115.00 -lreadline -ltermcap -ldl -lpthread 1. (CC) gcc options: -O2 -lz -lm
RealSR-NCNN RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No February 2023 2020 4 8 12 16 20 SE +/- 0.23, N = 14 SE +/- 0.06, N = 3 12.11 16.51
Waifu2x-NCNN Vulkan Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: Yes February 2023 2020 3 6 9 12 15 SE +/- 0.343, N = 15 SE +/- 0.010, N = 3 5.904 9.101
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.org Seconds, Fewer Is Better Monte Carlo Simulations of Ionised Nebulae 2019-03-24 Input: Dust 2D tau100.0 February 2023 2020 50 100 150 200 250 SE +/- 1.45, N = 3 SE +/- 0.88, N = 3 132 223 -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lz -pthread 1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -lmpi_usempif08 -lmpi_mpifh -lmpi
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.org Seconds, Fewer Is Better RNNoise 2020-06-28 February 2023 2020 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 18.07 18.81 1. (CC) gcc options: -O2 -pedantic -fvisibility=hidden
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.org Seconds, Fewer Is Better Apache CouchDB 3.1.1 Bulk Size: 100 - Inserts: 1000 - Rounds: 24 2020 February 2023 30 60 90 120 150 SE +/- 0.94, N = 3 SE +/- 0.20, N = 3 119.19 120.38 1. (CXX) g++ options: -std=c++14 -lmozjs-68 -lm -lerl_interface -lei -fPIC -MMD
GEGL GEGL is the Generic Graphics Library and is the library/framework used by GIMP and other applications like GNOME Photos. This test profile times how long it takes to complete various GEGL operations on a static set of sample JPEG images. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Crop February 2023 2020 2 4 6 8 10 SE +/- 0.047, N = 3 SE +/- 0.043, N = 3 8.453 8.843
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Cartoon 2020 February 2023 20 40 60 80 100 SE +/- 0.33, N = 3 SE +/- 0.23, N = 3 94.52 98.08
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Reflect February 2023 2020 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 29.55 31.59
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Antialias February 2023 2020 9 18 27 36 45 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 37.16 41.08
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Tile Glass February 2023 2020 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 30.93 32.97
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Wavelet Blur February 2023 2020 14 28 42 56 70 SE +/- 0.15, N = 3 SE +/- 0.24, N = 3 56.55 64.73
OpenBenchmarking.org Seconds, Fewer Is Better GEGL Operation: Color Enhance February 2023 2020 13 26 39 52 65 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 50.40 58.10
OpenBenchmarking.org Seconds, Fewer Is Better G'MIC Test: Plotting Isosurface Of A 3D Volume, 1000 Times February 2023 2020 5 10 15 20 25 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 13.65 19.38 1. Version 2.4.5, Copyright (c) 2008-2019, David Tschumperle.
OpenBenchmarking.org Seconds, Fewer Is Better G'MIC Test: 3D Elevated Function In Random Colors, 100 Times 2020 February 2023 20 40 60 80 100 SE +/- 0.33, N = 3 SE +/- 0.34, N = 3 84.89 86.38 1. Version 2.4.5, Copyright (c) 2008-2019, David Tschumperle.
Hugin Hugin is an open-source, cross-platform panorama photo stitcher software package. This test profile times how long it takes to run the assistant and panorama photo stitching on a set of images. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Hugin Panorama Photo Assistant + Stitching Time February 2023 2020 11 22 33 44 55 SE +/- 0.12, N = 3 SE +/- 0.31, N = 3 45.00 46.74
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Classroom - Compute: CPU-Only 2020 February 2023 20 40 60 80 100 SE +/- 0.44, N = 3 SE +/- 0.31, N = 3 84.95 93.81
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Fishy Cat - Compute: CPU-Only 2020 February 2023 11 22 33 44 55 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 46.34 49.26
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.90 Blend File: Barbershop - Compute: CPU-Only 2020 February 2023 30 60 90 120 150 SE +/- 0.27, N = 3 SE +/- 0.41, N = 3 131.12 142.95
Git This test measures the time needed to carry out some sample Git operations on an example, static repository that happens to be a copy of the GNOME GTK tool-kit repository. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands February 2023 2020 11 22 33 44 55 SE +/- 0.16, N = 3 SE +/- 0.10, N = 3 47.82 48.37 1. February 2023: git version 2.38.1 2. 2020: git version 2.25.1
Geometric Mean Of All Test Results OpenBenchmarking.org Geometric Mean, More Is Better Geometric Mean Of All Test Results Result Composite - System76 Thelio Major 3990X 2023 Benchmarks February 2023 2020 7 14 21 28 35 29.21 25.51
Hide Result
2020 Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Pop 20.04, Kernel: 5.4.0-7642-generic (x86_64), Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: amdgpu 19.1.0, OpenGL: 4.6 Mesa 20.0.8 (LLVM 10.0.0), Vulkan: 1.2.128, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: snd_usb_audio.ignore_ctl_error=1Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,noatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Graphics Notes: GLAMORJava Notes: OpenJDK Runtime Environment (build 11.0.8+10-post-Ubuntu-0ubuntu120.04)Python Notes: Python 3.8.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 23 October 2020 12:50 by user phoronix.
February 2023 Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB + 257GB Flash Drive, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 23.04, Kernel: 6.2.0-060200rc7daily20230206-generic (x86_64), Desktop: GNOME Shell 43.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.5 (LLVM 15.0.6 DRM 3.49), Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-AKimc9/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-AKimc9/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301055Graphics Notes: BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D1820201-101Java Notes: OpenJDK Runtime Environment (build 11.0.18+10-post-Ubuntu-0ubuntu1)Python Notes: Python 3.11.1Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 6 February 2023 18:19 by user phoronix.