AMD Ryzen 3 2200G testing with a ASUS PRIME B350M-E (5220 BIOS) and ASUS AMD Radeon Vega / Mobile 2GB on Ubuntu 20.10 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2101191-HA-RYZEN322022 Ryzen 3 2200G 2021 - Phoronix Test Suite Ryzen 3 2200G 2021 AMD Ryzen 3 2200G testing with a ASUS PRIME B350M-E (5220 BIOS) and ASUS AMD Radeon Vega / Mobile 2GB on Ubuntu 20.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2101191-HA-RYZEN322022&export=txt&sor&grs .
Ryzen 3 2200G 2021 Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Vulkan Compiler File-System Screen Resolution 1 2 3 AMD Ryzen 3 2200G @ 3.50GHz (4 Cores) ASUS PRIME B350M-E (5220 BIOS) AMD Raven/Raven2 6GB Samsung SSD 970 EVO 250GB ASUS AMD Radeon Vega / Mobile 2GB (1100/1600MHz) AMD Raven/Raven2/Fenghuang G237HL Realtek RTL8111/8168/8411 Ubuntu 20.10 5.8.0-38-generic (x86_64) GNOME Shell 3.38.1 X Server 1.20.9 modesetting 1.20.9 4.6 Mesa 20.2.6 (LLVM 11.0.0) 1.2.131 GCC 10.2.0 ext4 1920x1080 OpenBenchmarking.org Compiler Details - --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/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 Processor Details - Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x8101016 Graphics Details - GLAMOR Java Details - OpenJDK Runtime Environment (build 11.0.9.1+1-Ubuntu-0ubuntu1.20.10) Python Details - Python 3.8.6 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: disabled RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Ryzen 3 2200G 2021 redis: LPOP kripke: lczero: BLAS lczero: Eigen redis: GET onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU sunflow: Global Illumination + Image Synthesis node-web-tooling: osbench: Memory Allocations onednn: IP Shapes 1D - u8s8f32 - CPU ncnn: Vulkan GPU - resnet50 onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU tensorflow-lite: Mobilenet Quant onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU mnn: mobilenet-v1-1.0 onednn: Convolution Batch Shapes Auto - f32 - CPU rav1e: 10 ncnn: Vulkan GPU - shufflenet-v2 onednn: Recurrent Neural Network Training - u8s8f32 - CPU darktable: Boat - CPU-only ncnn: CPU-v2-v2 - mobilenet-v2 mnn: SqueezeNetV1.0 onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU sockperf: Latency Ping Pong osbench: Create Processes ncnn: Vulkan GPU - blazeface ncnn: CPU - mnasnet lulesh: ffte: N=256, 3D Complex FFT Routine compress-lz4: 3 - Compression Speed tensorflow-lite: Mobilenet Float onednn: IP Shapes 3D - f32 - CPU libraw: Post-Processing Benchmark gromacs: Water Benchmark dav1d: Chimera 1080p 10-bit rnnoise: onednn: Deconvolution Batch shapes_3d - f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU hugin: Panorama Photo Assistant + Stitching Time ncnn: CPU - regnety_400m ncnn: CPU - shufflenet-v2 ncnn: CPU-v3-v3 - mobilenet-v3 ncnn: CPU - vgg16 ncnn: CPU - efficientnet-b0 ncnn: CPU - resnet50 asmfish: 1024 Hash Memory, 26 Depth basis: UASTC Level 0 stockfish: Total Time onednn: Deconvolution Batch shapes_1d - f32 - CPU darktable: Server Room - CPU-only onednn: Recurrent Neural Network Inference - f32 - CPU keydb: ncnn: Vulkan GPU - mobilenet darktable: Server Rack - CPU-only compress-zstd: 3 aom-av1: Speed 4 Two-Pass influxdb: 4 - 10000 - 2,5000,1 - 10000 darktable: Masskrug - CPU-only compress-zstd: 19 openfoam: Motorbike 30M redis: SADD aom-av1: Speed 6 Realtime tensorflow-lite: SqueezeNet incompact3d: Cylinder ncnn: Vulkan GPU - vgg16 tensorflow-lite: Inception V4 ncnn: CPU - blazeface onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU redis: SET namd: ATPase Simulation - 327,506 Atoms mnn: inception-v3 x265: Bosphorus 1080p ncnn: CPU - googlenet ncnn: CPU - mobilenet osbench: Create Files rav1e: 5 crafty: Elapsed Time dav1d: Summer Nature 1080p lammps: Rhodopsin Protein ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU - alexnet byte: Dhrystone 2 ncnn: Vulkan GPU - regnety_400m embree: Pathtracer - Asian Dragon cp2k: Fayalite-FIST Data ncnn: CPU - alexnet aom-av1: Speed 6 Two-Pass compress-lz4: 1 - Decompression Speed sqlite-speedtest: Timed Time - Size 1,000 onednn: IP Shapes 3D - u8s8f32 - CPU yquake2: OpenGL 3.x - 1920 x 1080 webp: Default redis: LPUSH sockperf: Throughput warsow: 1920 x 1080 indigobench: CPU - Supercar indigobench: CPU - Bedroom webp: Quality 100, Lossless, Highest Compression numpy: phpbench: PHP Benchmark Suite embree: Pathtracer ISPC - Asian Dragon Obj coremark: CoreMark Size 666 - Iterations Per Second embree: Pathtracer - Asian Dragon Obj caffe: AlexNet - CPU - 100 dolfyn: Computational Fluid Dynamics ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - yolov4-tiny osbench: Launch Programs kvazaar: Bosphorus 4K - Medium osbench: Create Threads rav1e: 6 embree: Pathtracer - Crown kvazaar: Bosphorus 1080p - Very Fast onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU encode-wavpack: WAV To WavPack astcenc: Medium ncnn: CPU - squeezenet_ssd astcenc: Fast embree: Pathtracer ISPC - Crown build-eigen: Time To Compile ocrmypdf: Processing 60 Page PDF Document mocassin: Dust 2D tau100.0 onednn: Recurrent Neural Network Inference - u8s8f32 - CPU ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - squeezenet_ssd build-godot: Time To Compile mnn: resnet-v2-50 espeak: Text-To-Speech Synthesis influxdb: 64 - 10000 - 2,5000,1 - 10000 ncnn: CPU - resnet18 mnn: MobileNetV2_224 ncnn: Vulkan GPU - efficientnet-b0 build-ffmpeg: Time To Compile synthmark: VoiceMark_100 tensorflow-lite: NASNet Mobile gimp: auto-levels waifu2x-ncnn: 2x - 3 - No gimp: rotate yquake2: Software CPU - 1920 x 1080 webp: Quality 100 x265: Bosphorus 4K onednn: IP Shapes 1D - f32 - CPU aom-av1: Speed 8 Realtime build2: Time To Compile hmmer: Pfam Database Search tnn: CPU - SqueezeNet v1.1 compress-lz4: 1 - Compression Speed amg: astcenc: Thorough dav1d: Summer Nature 4K astcenc: Exhaustive gimp: resize kvazaar: Bosphorus 4K - Ultra Fast mafft: Multiple Sequence Alignment - LSU RNA rawtherapee: Total Benchmark Time kvazaar: Bosphorus 4K - Very Fast webp: Quality 100, Lossless vkmark: 1920 x 1080 encode-ape: WAV To APE ncnn: Vulkan GPU - googlenet ncnn: CPU - yolov4-tiny encode-opus: WAV To Opus Encode basis: UASTC Level 2 cloverleaf: Lagrangian-Eulerian Hydrodynamics embree: Pathtracer ISPC - Asian Dragon kvazaar: Bosphorus 1080p - Ultra Fast caffe: GoogleNet - CPU - 100 unpack-firefox: firefox-84.0.source.tar.xz dav1d: Chimera 1080p hint: FLOAT glmark2: 1920 x 1080 kvazaar: Bosphorus 1080p - Medium compress-lz4: 9 - Decompression Speed basis: ETC1S tensorflow-lite: Inception ResNet V2 tnn: CPU - MobileNet v2 gimp: unsharp-mask compress-lz4: 3 - Decompression Speed realsr-ncnn: 4x - Yes waifu2x-ncnn: 2x - 3 - Yes realsr-ncnn: 4x - No webp: Quality 100, Highest Compression simdjson: DistinctUserID simdjson: PartialTweets simdjson: LargeRand simdjson: Kostya clomp: Static OMP Speedup ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU compress-lz4: 9 - Compression Speed sockperf: Latency Under Load 1 2 3 2261210.92 4811563 432 448 2064794.83 14.8215 3.206 7.38 81.742684 14.6322 71.68 7913.61 328733 8195.20 7.395 23.0213 2.573 12.63 8193.61 25.206 11.20 9.732 29.7098 6.927 26.190281 3.25 10.50 1180.0971 15392.809951922 42.77 309946 13.0736 19.36 0.333 52.55 22.238 30.8073 8437.12 82.006 19.11 12.65 9.67 117.19 17.07 72.55 7748047 11.898 5718169 22.5413 20.695 7721.13 265074.45 46.40 0.339 2346.0 1.38 706035.5 24.172 14.0 342.98 1735687.33 10.13 467745 810.954712 117.46 6441017 3.31 38.8392 1489969.25 6.75407 63.415 19.49 32.59 46.83 18.245888 0.848 6274275 182.89 2.603 9.59 23.46 35499453.0 18.88 3.3140 1448.589 23.30 2.22 8722.4 81.222 5.81417 814.1 1.648 1216336.46 555055 158.1 1.107 0.494 57.670 242.34 508106 2.8199 102524.442176 2.9903 41877 21.069 29.17 59.00 81.523260 1.49 14.920235 1.082 2.7601 15.60 7.35744 15.082 12.77 59.39 9.75 2.5819 113.520 52.675 342 7746.87 10.38 59.28 501.200 50.216 35.134 721428.3 29.04 5.424 16.98 182.189 596.254 316112 15.924 4.115 14.487 92.9 2.595 4.81 16.5429 27.20 514.482 127.149 287.255 7994.30 213491533 84.33 51.92 696.05 12.855 6.84 15.042 123.684 3.94 24.894 1199 15.955 32.56 59.28 8.936 86.479 191.41 3.2431 27.01 110084 23.847 184.25 301333349.99382 1849 6.51 8565.2 82.060 5691310 279.339 17.332 8554.3 482.609 26.675 63.028 8.872 0.46 0.45 0.35 0.38 2 10.88 36.5596 42.22 53.775 1258380.50 3117717 374 380 1931045.20 15.0825 3.148 7.74 81.999382 14.8700 74.10 7667.94 318685 8438.64 7.526 23.6791 2.566 12.98 8356.19 25.901 10.91 9.613 29.8889 6.751 26.479562 3.31 10.34 1208.3866 15755.585271762 42.34 306270 13.1685 19.66 0.330 52.39 22.693 31.4147 8277.53 83.581 18.87 12.89 9.49 119.38 16.76 73.14 7802828 12.054 5628220 22.3144 21.008 7794.13 267212.95 47.01 0.342 2358.0 1.38 696009.5 24.518 14.2 339.54 1758200.50 10.12 461955 821.055725 118.91 6389943 3.33 38.9728 1486411.63 6.79902 63.997 19.60 32.79 46.49 18.315846 0.844 6255015 183.50 2.586 9.69 23.24 35791748.5 19.06 3.3113 1461.853 23.42 2.22 8646.3 81.423 5.84129 807.2 1.662 1213155.04 559663 159.4 1.098 0.494 57.216 241.36 506055 2.8371 101765.566926 2.9682 41672 21.139 29.08 59.40 81.967513 1.49 14.823278 1.083 2.7659 15.52 7.36061 15.077 12.83 59.71 9.74 2.5670 113.654 52.741 340 7725.63 10.44 59.16 503.977 50.269 35.319 725224.2 29.04 5.419 16.95 183.024 596.615 317213 15.853 4.110 14.470 93.3 2.601 4.83 16.5558 27.18 516.516 127.645 287.063 8015.39 214232233 84.61 52.00 697.50 12.866 6.85 15.000 123.405 3.94 24.957 1199 15.994 32.64 59.42 8.923 86.540 191.45 3.2504 27.07 110320 23.799 184.17 301687480.18842 1851 6.51 8552.1 82.178 5697083 279.626 17.317 8547.8 482.839 26.685 63.023 8.871 0.46 0.45 0.35 0.38 2.0 11.30 35.8461 41.02 52.758 1275489.17 353 377 1930168.38 15.5619 3.302 7.38 85.632006 15.1975 71.69 7750.49 327187 8426.84 7.313 23.6697 2.639 12.75 8419.82 25.451 11.04 9.867 29.1210 6.790 26.857058 3.33 10.25 1208.0185 15437.468578998 41.81 313216 13.3661 19.79 0.326 53.51 22.580 30.8208 8342.73 82.188 18.75 12.70 9.59 117.45 16.89 71.82 7669043 11.863 5648589 22.6619 20.739 7837.72 269044.48 46.32 0.344 2324.2 1.40 700554.6 24.194 14.2 338.27 1734495.83 10.25 467404 820.322815 118.04 6468567 3.29 38.5055 1472539.67 6.83284 63.269 19.71 32.43 46.32 18.441862 0.839 6322069 184.81 2.613 9.60 23.48 35427649.2 19.07 3.3432 1452.469 23.51 2.24 8690.1 81.933 5.79119 807.9 1.657 1223284.42 557665 159.4 1.106 0.498 57.452 243.26 504159 2.8151 102339.408641 2.9782 41573 20.988 29.28 59.28 82.073212 1.50 14.909108 1.089 2.7779 15.62 7.31351 15.174 12.75 59.34 9.80 2.5828 112.964 52.986 341 7701.75 10.43 59.50 502.608 50.489 35.312 723222.1 29.18 5.398 16.90 182.601 593.909 315790 15.894 4.097 14.424 93.3 2.590 4.83 16.6108 27.29 514.799 127.385 286.140 8022.48 214072633 84.49 52.09 695.24 12.826 6.83 15.035 123.339 3.95 24.900 1196 15.995 32.59 59.28 8.915 86.340 191.01 3.2482 27.05 110157 23.806 183.88 301185316.58909 1852 6.50 8562.8 82.076 5689070 279.278 17.328 8547.2 482.551 26.673 63.002 8.869 0.46 0.45 0.35 0.38 2 10.68 33.0587 41.23 50.699 OpenBenchmarking.org
Redis Test: LPOP OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: LPOP 1 3 2 500K 1000K 1500K 2000K 2500K SE +/- 16948.58, N = 3 SE +/- 8952.66, N = 3 SE +/- 11741.76, N = 3 2261210.92 1275489.17 1258380.50 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Kripke OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 1 2 1000K 2000K 3000K 4000K 5000K SE +/- 36406.50, N = 2 SE +/- 35494.54, N = 3 4811563 3117717 1. (CXX) g++ options: -O3 -fopenmp
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: BLAS 1 2 3 90 180 270 360 450 SE +/- 4.54, N = 8 SE +/- 6.01, N = 9 SE +/- 2.52, N = 3 432 374 353 1. (CXX) g++ options: -flto -pthread
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: Eigen 1 2 3 100 200 300 400 500 SE +/- 4.81, N = 3 SE +/- 5.13, N = 9 448 380 377 1. (CXX) g++ options: -flto -pthread
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: GET 1 2 3 400K 800K 1200K 1600K 2000K SE +/- 35016.53, N = 3 SE +/- 23617.43, N = 5 SE +/- 22292.08, N = 3 2064794.83 1931045.20 1930168.38 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU 1 2 3 4 8 12 16 20 SE +/- 0.22, N = 15 SE +/- 0.18, N = 15 SE +/- 0.09, N = 3 14.82 15.08 15.56 MIN: 11.81 MIN: 12.35 MIN: 13.12 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Sunflow Rendering System Global Illumination + Image Synthesis OpenBenchmarking.org Seconds, Fewer Is Better Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis 2 1 3 0.743 1.486 2.229 2.972 3.715 SE +/- 0.032, N = 3 SE +/- 0.041, N = 3 SE +/- 0.028, N = 15 3.148 3.206 3.302 MIN: 2.89 / MAX: 3.84 MIN: 2.88 / MAX: 3.79 MIN: 2.87 / MAX: 4.18
Node.js V8 Web Tooling Benchmark OpenBenchmarking.org runs/s, More Is Better Node.js V8 Web Tooling Benchmark 2 3 1 2 4 6 8 10 SE +/- 0.03, N = 3 SE +/- 0.09, N = 4 SE +/- 0.08, N = 3 7.74 7.38 7.38 1. Nodejs
v12.18.2
OSBench Test: Memory Allocations OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations 1 2 3 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 1.45, N = 3 81.74 82.00 85.63 1. (CC) gcc options: -lm
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU 1 2 3 4 8 12 16 20 SE +/- 0.11, N = 3 SE +/- 0.12, N = 3 SE +/- 0.12, N = 3 14.63 14.87 15.20 MIN: 13.38 MIN: 13.38 MIN: 13.47 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: resnet50 1 3 2 16 32 48 64 80 SE +/- 0.20, N = 3 SE +/- 0.29, N = 3 SE +/- 0.63, N = 4 71.68 71.69 74.10 MIN: 65.87 / MAX: 90.56 MIN: 66.47 / MAX: 91.18 MIN: 66.26 / MAX: 110.26 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU 2 3 1 2K 4K 6K 8K 10K SE +/- 22.70, N = 3 SE +/- 13.61, N = 3 SE +/- 101.76, N = 3 7667.94 7750.49 7913.61 MIN: 7509.56 MIN: 7562.51 MIN: 7617.25 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
TensorFlow Lite Model: Mobilenet Quant OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Quant 2 3 1 70K 140K 210K 280K 350K SE +/- 1025.34, N = 3 SE +/- 1866.22, N = 3 SE +/- 1183.02, N = 3 318685 327187 328733
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU 1 3 2 2K 4K 6K 8K 10K SE +/- 99.84, N = 5 SE +/- 46.50, N = 3 SE +/- 66.22, N = 15 8195.20 8426.84 8438.64 MIN: 7505 MIN: 8003.45 MIN: 7752.96 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: mobilenet-v1-1.0 3 1 2 2 4 6 8 10 SE +/- 0.030, N = 3 SE +/- 0.021, N = 3 SE +/- 0.065, N = 3 7.313 7.395 7.526 MIN: 6.61 / MAX: 17.32 MIN: 6.57 / MAX: 16.47 MIN: 6.6 / MAX: 20.05 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 1 3 2 6 12 18 24 30 SE +/- 0.34, N = 3 SE +/- 0.14, N = 3 SE +/- 0.20, N = 3 23.02 23.67 23.68 MIN: 19.03 MIN: 20.01 MIN: 20.06 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
rav1e Speed: 10 OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 10 3 1 2 0.5938 1.1876 1.7814 2.3752 2.969 SE +/- 0.007, N = 3 SE +/- 0.015, N = 3 SE +/- 0.008, N = 3 2.639 2.573 2.566
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: shufflenet-v2 1 3 2 3 6 9 12 15 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.16, N = 4 12.63 12.75 12.98 MIN: 10.39 / MAX: 25.47 MIN: 10.36 / MAX: 21.98 MIN: 10.26 / MAX: 24.78 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU 1 2 3 2K 4K 6K 8K 10K SE +/- 128.90, N = 3 SE +/- 144.66, N = 3 SE +/- 28.77, N = 3 8193.61 8356.19 8419.82 MIN: 7671.11 MIN: 7776.27 MIN: 8053.44 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Darktable Test: Boat - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.2.1 Test: Boat - Acceleration: CPU-only 1 3 2 6 12 18 24 30 SE +/- 0.09, N = 3 SE +/- 0.09, N = 3 SE +/- 0.25, N = 13 25.21 25.45 25.90
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v2-v2 - Model: mobilenet-v2 2 3 1 3 6 9 12 15 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.13, N = 3 10.91 11.04 11.20 MIN: 8.91 / MAX: 18.07 MIN: 8.96 / MAX: 21.86 MIN: 8.97 / MAX: 20.5 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: SqueezeNetV1.0 2 1 3 3 6 9 12 15 SE +/- 0.039, N = 3 SE +/- 0.127, N = 3 SE +/- 0.066, N = 3 9.613 9.732 9.867 MIN: 8.69 / MAX: 20.7 MIN: 8.67 / MAX: 18.76 MIN: 8.73 / MAX: 39.42 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU 3 1 2 7 14 21 28 35 SE +/- 0.29, N = 3 SE +/- 0.09, N = 3 SE +/- 0.14, N = 3 29.12 29.71 29.89 MIN: 26.35 MIN: 26.4 MIN: 26.42 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Sockperf Test: Latency Ping Pong OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong 2 3 1 2 4 6 8 10 SE +/- 0.049, N = 5 SE +/- 0.074, N = 5 SE +/- 0.065, N = 5 6.751 6.790 6.927 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
OSBench Test: Create Processes OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes 1 2 3 6 12 18 24 30 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.21, N = 3 26.19 26.48 26.86 1. (CC) gcc options: -lm
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: blazeface 1 2 3 0.7493 1.4986 2.2479 2.9972 3.7465 SE +/- 0.01, N = 3 SE +/- 0.02, N = 4 SE +/- 0.03, N = 3 3.25 3.31 3.33 MIN: 2.61 / MAX: 14.28 MIN: 2.6 / MAX: 4.93 MIN: 2.62 / MAX: 5.7 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mnasnet 3 2 1 3 6 9 12 15 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 10.25 10.34 10.50 MIN: 8.43 / MAX: 24.69 MIN: 8.42 / MAX: 16.24 MIN: 8.45 / MAX: 18.4 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
LULESH OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 2 3 1 300 600 900 1200 1500 SE +/- 0.69, N = 3 SE +/- 2.03, N = 3 SE +/- 0.53, N = 3 1208.39 1208.02 1180.10 1. (CXX) g++ options: -O3 -fopenmp -lm -pthread -lmpi_cxx -lmpi
FFTE N=256, 3D Complex FFT Routine OpenBenchmarking.org MFLOPS, More Is Better FFTE 7.0 N=256, 3D Complex FFT Routine 2 3 1 3K 6K 9K 12K 15K SE +/- 110.35, N = 3 SE +/- 161.49, N = 3 SE +/- 120.97, N = 3 15755.59 15437.47 15392.81 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
LZ4 Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed 1 2 3 10 20 30 40 50 SE +/- 0.58, N = 3 SE +/- 0.43, N = 15 SE +/- 0.65, N = 15 42.77 42.34 41.81 1. (CC) gcc options: -O3
TensorFlow Lite Model: Mobilenet Float OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Mobilenet Float 2 1 3 70K 140K 210K 280K 350K SE +/- 1840.34, N = 3 SE +/- 172.14, N = 3 SE +/- 2441.23, N = 3 306270 309946 313216
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU 1 2 3 3 6 9 12 15 SE +/- 0.18, N = 15 SE +/- 0.17, N = 15 SE +/- 0.06, N = 3 13.07 13.17 13.37 MIN: 10.63 MIN: 10.78 MIN: 12.28 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
LibRaw Post-Processing Benchmark OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark 3 2 1 5 10 15 20 25 SE +/- 0.12, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 19.79 19.66 19.36 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
GROMACS Water Benchmark OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.3 Water Benchmark 1 2 3 0.0749 0.1498 0.2247 0.2996 0.3745 SE +/- 0.002, N = 3 SE +/- 0.002, N = 3 SE +/- 0.005, N = 3 0.333 0.330 0.326 1. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p 10-bit 3 1 2 12 24 36 48 60 SE +/- 0.31, N = 3 SE +/- 0.17, N = 3 SE +/- 0.21, N = 3 53.51 52.55 52.39 MIN: 35.6 / MAX: 125.13 MIN: 35.45 / MAX: 124.71 MIN: 35.47 / MAX: 120.48 1. (CC) gcc options: -pthread -ldl -lm
RNNoise OpenBenchmarking.org Seconds, Fewer Is Better RNNoise 2020-06-28 1 3 2 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.35, N = 3 SE +/- 0.23, N = 8 22.24 22.58 22.69 1. (CC) gcc options: -O2 -pedantic -fvisibility=hidden
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU 1 3 2 7 14 21 28 35 SE +/- 0.15, N = 3 SE +/- 0.14, N = 3 SE +/- 0.28, N = 3 30.81 30.82 31.41 MIN: 22.57 MIN: 22.62 MIN: 22.58 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 2 3 1 2K 4K 6K 8K 10K SE +/- 86.65, N = 3 SE +/- 61.07, N = 3 SE +/- 133.51, N = 3 8277.53 8342.73 8437.12 MIN: 7767.68 MIN: 7929.12 MIN: 7874.47 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Hugin Panorama Photo Assistant + Stitching Time OpenBenchmarking.org Seconds, Fewer Is Better Hugin Panorama Photo Assistant + Stitching Time 1 3 2 20 40 60 80 100 SE +/- 0.57, N = 3 SE +/- 0.13, N = 3 SE +/- 0.26, N = 3 82.01 82.19 83.58
NCNN Target: CPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: regnety_400m 3 2 1 5 10 15 20 25 SE +/- 0.18, N = 3 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 18.75 18.87 19.11 MIN: 16.84 / MAX: 32.85 MIN: 16.81 / MAX: 33.23 MIN: 16.69 / MAX: 35.81 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: shufflenet-v2 1 3 2 3 6 9 12 15 SE +/- 0.12, N = 3 SE +/- 0.14, N = 3 SE +/- 0.14, N = 3 12.65 12.70 12.89 MIN: 10.41 / MAX: 23.3 MIN: 10.47 / MAX: 19.71 MIN: 10.42 / MAX: 26.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v3-v3 - Model: mobilenet-v3 2 3 1 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 9.49 9.59 9.67 MIN: 7.78 / MAX: 14.76 MIN: 7.81 / MAX: 16.29 MIN: 7.8 / MAX: 15.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: vgg16 1 3 2 30 60 90 120 150 SE +/- 0.18, N = 3 SE +/- 0.17, N = 3 SE +/- 0.23, N = 3 117.19 117.45 119.38 MIN: 112.25 / MAX: 143.19 MIN: 112.54 / MAX: 135.06 MIN: 113.9 / MAX: 142.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: efficientnet-b0 2 3 1 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.12, N = 3 SE +/- 0.15, N = 3 16.76 16.89 17.07 MIN: 13.96 / MAX: 31.87 MIN: 14.11 / MAX: 30.24 MIN: 14.1 / MAX: 30.1 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet50 3 1 2 16 32 48 64 80 SE +/- 0.19, N = 3 SE +/- 0.82, N = 3 SE +/- 0.39, N = 3 71.82 72.55 73.14 MIN: 65.77 / MAX: 87.35 MIN: 66.75 / MAX: 91.97 MIN: 66.24 / MAX: 103.74 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
asmFish 1024 Hash Memory, 26 Depth OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth 2 1 3 2M 4M 6M 8M 10M SE +/- 50254.79, N = 3 SE +/- 28500.12, N = 3 SE +/- 29445.58, N = 3 7802828 7748047 7669043
Basis Universal Settings: UASTC Level 0 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 0 3 1 2 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 11.86 11.90 12.05 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 12 Total Time 1 3 2 1.2M 2.4M 3.6M 4.8M 6M SE +/- 48644.36, N = 3 SE +/- 39149.01, N = 3 SE +/- 74806.22, N = 3 5718169 5648589 5628220 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2 -flto -flto=jobserver
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU 2 1 3 5 10 15 20 25 SE +/- 0.18, N = 15 SE +/- 0.26, N = 3 SE +/- 0.14, N = 3 22.31 22.54 22.66 MIN: 17.69 MIN: 17.75 MIN: 17.69 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Darktable Test: Server Room - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.2.1 Test: Server Room - Acceleration: CPU-only 1 3 2 5 10 15 20 25 SE +/- 0.21, N = 3 SE +/- 0.19, N = 3 SE +/- 0.12, N = 3 20.70 20.74 21.01
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 1 2 3 2K 4K 6K 8K 10K SE +/- 11.25, N = 3 SE +/- 85.05, N = 3 SE +/- 30.28, N = 3 7721.13 7794.13 7837.72 MIN: 7547.72 MIN: 7534.49 MIN: 7613.07 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
KeyDB OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.0.16 3 2 1 60K 120K 180K 240K 300K SE +/- 1852.38, N = 3 SE +/- 2051.97, N = 3 SE +/- 3138.20, N = 3 269044.48 267212.95 265074.45 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: mobilenet 3 1 2 11 22 33 44 55 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.66, N = 4 46.32 46.40 47.01 MIN: 42.77 / MAX: 61.8 MIN: 42.8 / MAX: 59.97 MIN: 42.66 / MAX: 60.93 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Darktable Test: Server Rack - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.2.1 Test: Server Rack - Acceleration: CPU-only 1 2 3 0.0774 0.1548 0.2322 0.3096 0.387 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 0.339 0.342 0.344
Zstd Compression Compression Level: 3 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 3 2 1 3 500 1000 1500 2000 2500 SE +/- 16.00, N = 3 SE +/- 27.78, N = 3 SE +/- 7.82, N = 3 2358.0 2346.0 2324.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
AOM AV1 Encoder Mode: Speed 4 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 4 Two-Pass 3 2 1 0.315 0.63 0.945 1.26 1.575 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 1.40 1.38 1.38 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
InfluxDB Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 1 3 2 150K 300K 450K 600K 750K SE +/- 8641.46, N = 3 SE +/- 5594.95, N = 3 SE +/- 6558.55, N = 3 706035.5 700554.6 696009.5
Darktable Test: Masskrug - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.2.1 Test: Masskrug - Acceleration: CPU-only 1 3 2 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 24.17 24.19 24.52
Zstd Compression Compression Level: 19 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 3 2 1 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.18, N = 5 14.2 14.2 14.0 1. (CC) gcc options: -O3 -pthread -lz -llzma
OpenFOAM Input: Motorbike 30M OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 8 Input: Motorbike 30M 3 2 1 70 140 210 280 350 SE +/- 2.23, N = 3 SE +/- 0.27, N = 3 SE +/- 1.66, N = 3 338.27 339.54 342.98 1. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -lgenericPatchFields -lOpenFOAM -ldl -lm
Redis Test: SADD OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SADD 2 1 3 400K 800K 1200K 1600K 2000K SE +/- 21162.12, N = 3 SE +/- 11595.14, N = 3 SE +/- 4337.84, N = 3 1758200.50 1735687.33 1734495.83 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
AOM AV1 Encoder Mode: Speed 6 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Realtime 3 1 2 3 6 9 12 15 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 10.25 10.13 10.12 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
TensorFlow Lite Model: SqueezeNet OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: SqueezeNet 2 3 1 100K 200K 300K 400K 500K SE +/- 1459.42, N = 3 SE +/- 564.75, N = 3 SE +/- 158.13, N = 3 461955 467404 467745
Incompact3D Input: Cylinder OpenBenchmarking.org Seconds, Fewer Is Better Incompact3D 2020-09-17 Input: Cylinder 1 3 2 200 400 600 800 1000 SE +/- 3.54, N = 3 SE +/- 2.19, N = 3 SE +/- 10.03, N = 3 810.95 820.32 821.06 1. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lm -lrt -lz
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: vgg16 1 3 2 30 60 90 120 150 SE +/- 0.37, N = 3 SE +/- 0.16, N = 3 SE +/- 0.24, N = 4 117.46 118.04 118.91 MIN: 111.97 / MAX: 149.37 MIN: 112.22 / MAX: 141.41 MIN: 113.22 / MAX: 141.78 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
TensorFlow Lite Model: Inception V4 OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception V4 2 1 3 1.4M 2.8M 4.2M 5.6M 7M SE +/- 7475.10, N = 3 SE +/- 24424.39, N = 3 SE +/- 4440.22, N = 3 6389943 6441017 6468567
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: blazeface 3 1 2 0.7493 1.4986 2.2479 2.9972 3.7465 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 3.29 3.31 3.33 MIN: 2.73 / MAX: 4.72 MIN: 2.64 / MAX: 9.91 MIN: 2.62 / MAX: 5.11 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 3 1 2 9 18 27 36 45 SE +/- 0.51, N = 3 SE +/- 0.18, N = 3 SE +/- 0.23, N = 3 38.51 38.84 38.97 MIN: 35.63 MIN: 35.67 MIN: 35.93 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SET 1 2 3 300K 600K 900K 1200K 1500K SE +/- 5888.95, N = 3 SE +/- 19097.96, N = 3 SE +/- 15253.86, N = 8 1489969.25 1486411.63 1472539.67 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
NAMD ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms 1 2 3 2 4 6 8 10 SE +/- 0.01425, N = 3 SE +/- 0.03865, N = 3 SE +/- 0.08887, N = 5 6.75407 6.79902 6.83284
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: inception-v3 3 1 2 14 28 42 56 70 SE +/- 0.19, N = 3 SE +/- 0.18, N = 3 SE +/- 0.32, N = 3 63.27 63.42 64.00 MIN: 60.45 / MAX: 98.45 MIN: 60.02 / MAX: 120.02 MIN: 60.33 / MAX: 93.06 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
x265 Video Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 1080p 3 2 1 5 10 15 20 25 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.16, N = 3 19.71 19.60 19.49 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: googlenet 3 1 2 8 16 24 32 40 SE +/- 0.18, N = 3 SE +/- 0.18, N = 3 SE +/- 0.05, N = 3 32.43 32.59 32.79 MIN: 28.44 / MAX: 46.34 MIN: 28.72 / MAX: 51.96 MIN: 28.77 / MAX: 48.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mobilenet 3 2 1 11 22 33 44 55 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.46, N = 3 46.32 46.49 46.83 MIN: 43.57 / MAX: 62.16 MIN: 42.72 / MAX: 62.2 MIN: 42.34 / MAX: 64.41 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OSBench Test: Create Files OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files 1 2 3 5 10 15 20 25 SE +/- 0.24, N = 3 SE +/- 0.21, N = 3 SE +/- 0.12, N = 3 18.25 18.32 18.44 1. (CC) gcc options: -lm
rav1e Speed: 5 OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 5 1 2 3 0.1908 0.3816 0.5724 0.7632 0.954 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 0.848 0.844 0.839
Crafty Elapsed Time OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time 3 1 2 1.4M 2.8M 4.2M 5.6M 7M SE +/- 23149.53, N = 3 SE +/- 2855.69, N = 3 SE +/- 20483.82, N = 3 6322069 6274275 6255015 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Summer Nature 1080p 3 2 1 40 80 120 160 200 SE +/- 0.36, N = 3 SE +/- 0.30, N = 3 SE +/- 0.82, N = 3 184.81 183.50 182.89 MIN: 171.93 / MAX: 203.27 MIN: 169.65 / MAX: 201.98 MIN: 167.96 / MAX: 203.4 1. (CC) gcc options: -pthread -ldl -lm
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 29Oct2020 Model: Rhodopsin Protein 3 1 2 0.5879 1.1758 1.7637 2.3516 2.9395 SE +/- 0.031, N = 3 SE +/- 0.015, N = 3 SE +/- 0.030, N = 3 2.613 2.603 2.586 1. (CXX) g++ options: -O3 -pthread -lm
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 1 3 2 3 6 9 12 15 SE +/- 0.13, N = 3 SE +/- 0.09, N = 3 SE +/- 0.14, N = 4 9.59 9.60 9.69 MIN: 7.78 / MAX: 22.63 MIN: 7.73 / MAX: 19.55 MIN: 7.81 / MAX: 18.31 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: alexnet 2 1 3 6 12 18 24 30 SE +/- 0.03, N = 4 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 23.24 23.46 23.48 MIN: 21.11 / MAX: 37.35 MIN: 21.29 / MAX: 37.36 MIN: 21.25 / MAX: 36.41 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
BYTE Unix Benchmark Computational Test: Dhrystone 2 OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 3.6 Computational Test: Dhrystone 2 2 1 3 8M 16M 24M 32M 40M SE +/- 208729.08, N = 3 SE +/- 289585.60, N = 3 SE +/- 503813.27, N = 3 35791748.5 35499453.0 35427649.2
NCNN Target: Vulkan GPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: regnety_400m 1 2 3 5 10 15 20 25 SE +/- 0.09, N = 3 SE +/- 0.15, N = 4 SE +/- 0.01, N = 3 18.88 19.06 19.07 MIN: 16.76 / MAX: 26.52 MIN: 16.61 / MAX: 34.07 MIN: 16.77 / MAX: 34.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer - Model: Asian Dragon 3 1 2 0.7522 1.5044 2.2566 3.0088 3.761 SE +/- 0.0299, N = 3 SE +/- 0.0186, N = 3 SE +/- 0.0143, N = 3 3.3432 3.3140 3.3113 MIN: 3.25 / MAX: 3.45 MIN: 3.25 / MAX: 3.4 MIN: 3.26 / MAX: 3.4
CP2K Molecular Dynamics Fayalite-FIST Data OpenBenchmarking.org Seconds, Fewer Is Better CP2K Molecular Dynamics 8.1 Fayalite-FIST Data 1 3 2 300 600 900 1200 1500 1448.59 1452.47 1461.85
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: alexnet 1 2 3 6 12 18 24 30 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 23.30 23.42 23.51 MIN: 21.24 / MAX: 38.11 MIN: 21.27 / MAX: 37.51 MIN: 21.21 / MAX: 37.48 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 6 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Two-Pass 3 2 1 0.504 1.008 1.512 2.016 2.52 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 2.24 2.22 2.22 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
LZ4 Compression Compression Level: 1 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Decompression Speed 1 3 2 2K 4K 6K 8K 10K SE +/- 6.35, N = 3 SE +/- 57.65, N = 3 SE +/- 8.66, N = 3 8722.4 8690.1 8646.3 1. (CC) gcc options: -O3
SQLite Speedtest Timed Time - Size 1,000 OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 1 2 3 20 40 60 80 100 SE +/- 0.14, N = 3 SE +/- 0.76, N = 3 SE +/- 0.70, N = 3 81.22 81.42 81.93 1. (CC) gcc options: -O2 -ldl -lz -lpthread
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU 3 1 2 1.3143 2.6286 3.9429 5.2572 6.5715 SE +/- 0.01488, N = 3 SE +/- 0.02077, N = 3 SE +/- 0.01362, N = 3 5.79119 5.81417 5.84129 MIN: 5.24 MIN: 5.17 MIN: 5.26 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
yquake2 Renderer: OpenGL 3.x - Resolution: 1920 x 1080 OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: OpenGL 3.x - Resolution: 1920 x 1080 1 3 2 200 400 600 800 1000 SE +/- 4.11, N = 3 SE +/- 4.89, N = 3 SE +/- 3.33, N = 3 814.1 807.9 807.2 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
WebP Image Encode Encode Settings: Default OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Default 1 3 2 0.374 0.748 1.122 1.496 1.87 SE +/- 0.002, N = 3 SE +/- 0.009, N = 3 SE +/- 0.003, N = 3 1.648 1.657 1.662 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: LPUSH 3 1 2 300K 600K 900K 1200K 1500K SE +/- 4396.42, N = 3 SE +/- 16215.78, N = 3 SE +/- 2985.46, N = 3 1223284.42 1216336.46 1213155.04 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Sockperf Test: Throughput OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput 2 3 1 120K 240K 360K 480K 600K SE +/- 3595.18, N = 5 SE +/- 3270.45, N = 5 SE +/- 6800.99, N = 5 559663 557665 555055 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Warsow Resolution: 1920 x 1080 OpenBenchmarking.org Frames Per Second, More Is Better Warsow 2.5 Beta Resolution: 1920 x 1080 3 2 1 40 80 120 160 200 SE +/- 0.10, N = 3 SE +/- 0.12, N = 3 SE +/- 1.30, N = 3 159.4 159.4 158.1
IndigoBench Acceleration: CPU - Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Supercar 1 3 2 0.2491 0.4982 0.7473 0.9964 1.2455 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.009, N = 3 1.107 1.106 1.098
IndigoBench Acceleration: CPU - Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.4 Acceleration: CPU - Scene: Bedroom 3 2 1 0.1121 0.2242 0.3363 0.4484 0.5605 SE +/- 0.002, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.498 0.494 0.494
WebP Image Encode Encode Settings: Quality 100, Lossless, Highest Compression OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless, Highest Compression 2 3 1 13 26 39 52 65 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.26, N = 3 57.22 57.45 57.67 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Numpy Benchmark OpenBenchmarking.org Score, More Is Better Numpy Benchmark 3 1 2 50 100 150 200 250 SE +/- 0.50, N = 3 SE +/- 0.34, N = 3 SE +/- 0.33, N = 3 243.26 242.34 241.36
PHPBench PHP Benchmark Suite OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite 1 2 3 110K 220K 330K 440K 550K SE +/- 423.52, N = 3 SE +/- 1952.23, N = 3 SE +/- 2233.09, N = 3 508106 506055 504159
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer ISPC - Model: Asian Dragon Obj 2 1 3 0.6383 1.2766 1.9149 2.5532 3.1915 SE +/- 0.0157, N = 3 SE +/- 0.0119, N = 3 SE +/- 0.0152, N = 3 2.8371 2.8199 2.8151 MIN: 2.77 / MAX: 2.9 MIN: 2.75 / MAX: 2.92 MIN: 2.75 / MAX: 2.89
Coremark CoreMark Size 666 - Iterations Per Second OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second 1 3 2 20K 40K 60K 80K 100K SE +/- 816.07, N = 3 SE +/- 373.76, N = 3 SE +/- 290.85, N = 3 102524.44 102339.41 101765.57 1. (CC) gcc options: -O2 -lrt" -lrt
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer - Model: Asian Dragon Obj 1 3 2 0.6728 1.3456 2.0184 2.6912 3.364 SE +/- 0.0237, N = 3 SE +/- 0.0217, N = 3 SE +/- 0.0135, N = 3 2.9903 2.9782 2.9682 MIN: 2.9 / MAX: 3.08 MIN: 2.91 / MAX: 3.08 MIN: 2.9 / MAX: 3.07
Caffe Model: AlexNet - Acceleration: CPU - Iterations: 100 OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: AlexNet - Acceleration: CPU - Iterations: 100 3 2 1 9K 18K 27K 36K 45K SE +/- 137.35, N = 3 SE +/- 90.86, N = 3 SE +/- 193.32, N = 3 41573 41672 41877 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
Dolfyn Computational Fluid Dynamics OpenBenchmarking.org Seconds, Fewer Is Better Dolfyn 0.527 Computational Fluid Dynamics 3 1 2 5 10 15 20 25 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 20.99 21.07 21.14
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: resnet18 2 1 3 7 14 21 28 35 SE +/- 0.14, N = 4 SE +/- 0.26, N = 3 SE +/- 0.02, N = 3 29.08 29.17 29.28 MIN: 25.74 / MAX: 44.35 MIN: 26 / MAX: 40.97 MIN: 25.57 / MAX: 39.41 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: yolov4-tiny 1 3 2 13 26 39 52 65 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 SE +/- 0.06, N = 4 59.00 59.28 59.40 MIN: 55.05 / MAX: 74.18 MIN: 54.59 / MAX: 74.87 MIN: 55.12 / MAX: 75.24 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OSBench Test: Launch Programs OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs 1 2 3 20 40 60 80 100 SE +/- 0.27, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 81.52 81.97 82.07 1. (CC) gcc options: -lm
Kvazaar Video Input: Bosphorus 4K - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Medium 3 2 1 0.3375 0.675 1.0125 1.35 1.6875 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.50 1.49 1.49 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OSBench Test: Create Threads OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads 2 3 1 4 8 12 16 20 SE +/- 0.09, N = 3 SE +/- 0.14, N = 3 SE +/- 0.03, N = 3 14.82 14.91 14.92 1. (CC) gcc options: -lm
rav1e Speed: 6 OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.4 Speed: 6 3 2 1 0.245 0.49 0.735 0.98 1.225 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 SE +/- 0.005, N = 3 1.089 1.083 1.082
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer - Model: Crown 3 2 1 0.625 1.25 1.875 2.5 3.125 SE +/- 0.0087, N = 3 SE +/- 0.0043, N = 3 SE +/- 0.0145, N = 3 2.7779 2.7659 2.7601 MIN: 2.75 / MAX: 2.87 MIN: 2.73 / MAX: 2.83 MIN: 2.71 / MAX: 2.86
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 1080p - Video Preset: Very Fast 3 1 2 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 15.62 15.60 15.52 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 3 1 2 2 4 6 8 10 SE +/- 0.00751, N = 3 SE +/- 0.02489, N = 3 SE +/- 0.00711, N = 3 7.31351 7.35744 7.36061 MIN: 6.35 MIN: 6.35 MIN: 6.34 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
WavPack Audio Encoding WAV To WavPack OpenBenchmarking.org Seconds, Fewer Is Better WavPack Audio Encoding 5.3 WAV To WavPack 2 1 3 4 8 12 16 20 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.09, N = 21 15.08 15.08 15.17 1. (CXX) g++ options: -rdynamic
ASTC Encoder Preset: Medium OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Medium 3 1 2 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 12.75 12.77 12.83 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
NCNN Target: CPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: squeezenet_ssd 3 1 2 13 26 39 52 65 SE +/- 0.27, N = 3 SE +/- 0.05, N = 3 SE +/- 0.20, N = 3 59.34 59.39 59.71 MIN: 52.64 / MAX: 71.17 MIN: 53.07 / MAX: 79.98 MIN: 52.9 / MAX: 78.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
ASTC Encoder Preset: Fast OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Fast 2 1 3 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 9.74 9.75 9.80 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer ISPC - Model: Crown 3 1 2 0.5811 1.1622 1.7433 2.3244 2.9055 SE +/- 0.0165, N = 3 SE +/- 0.0040, N = 3 SE +/- 0.0105, N = 3 2.5828 2.5819 2.5670 MIN: 2.51 / MAX: 2.65 MIN: 2.55 / MAX: 2.62 MIN: 2.52 / MAX: 2.63
Timed Eigen Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Eigen Compilation 3.3.9 Time To Compile 3 1 2 30 60 90 120 150 SE +/- 0.21, N = 3 SE +/- 0.54, N = 3 SE +/- 0.19, N = 3 112.96 113.52 113.65
OCRMyPDF Processing 60 Page PDF Document OpenBenchmarking.org Seconds, Fewer Is Better OCRMyPDF 10.3.1+dfsg Processing 60 Page PDF Document 1 2 3 12 24 36 48 60 SE +/- 0.13, N = 3 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 52.68 52.74 52.99
Monte Carlo Simulations of Ionised Nebulae Input: Dust 2D tau100.0 OpenBenchmarking.org Seconds, Fewer Is Better Monte Carlo Simulations of Ionised Nebulae 2019-03-24 Input: Dust 2D tau100.0 2 3 1 70 140 210 280 350 SE +/- 1.76, N = 3 SE +/- 0.67, N = 3 340 341 342 1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -ldl -levent -levent_pthreads -lutil -lrt -lz
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU 3 2 1 1700 3400 5100 6800 8500 SE +/- 16.23, N = 3 SE +/- 27.04, N = 3 SE +/- 38.84, N = 3 7701.75 7725.63 7746.87 MIN: 7494.48 MIN: 7520.66 MIN: 7556.29 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: mnasnet 1 3 2 3 6 9 12 15 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.20, N = 4 10.38 10.43 10.44 MIN: 8.43 / MAX: 17.05 MIN: 8.39 / MAX: 21.7 MIN: 8.36 / MAX: 26.87 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: squeezenet_ssd 2 1 3 13 26 39 52 65 SE +/- 0.14, N = 4 SE +/- 0.08, N = 3 SE +/- 0.16, N = 3 59.16 59.28 59.50 MIN: 51.95 / MAX: 72.82 MIN: 53.06 / MAX: 77.54 MIN: 52.65 / MAX: 71.96 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Timed Godot Game Engine Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 3.2.3 Time To Compile 1 3 2 110 220 330 440 550 SE +/- 0.16, N = 3 SE +/- 0.32, N = 3 SE +/- 0.30, N = 3 501.20 502.61 503.98
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: resnet-v2-50 1 2 3 11 22 33 44 55 SE +/- 0.51, N = 3 SE +/- 0.33, N = 3 SE +/- 0.27, N = 3 50.22 50.27 50.49 MIN: 47.21 / MAX: 83.29 MIN: 47.5 / MAX: 72.85 MIN: 47.85 / MAX: 147.84 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
eSpeak-NG Speech Engine Text-To-Speech Synthesis OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis 1 3 2 8 16 24 32 40 SE +/- 0.11, N = 4 SE +/- 0.11, N = 4 SE +/- 0.12, N = 4 35.13 35.31 35.32 1. (CC) gcc options: -O2 -std=c99
InfluxDB Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 2 3 1 160K 320K 480K 640K 800K SE +/- 1666.30, N = 3 SE +/- 3366.02, N = 3 SE +/- 1902.33, N = 3 725224.2 723222.1 721428.3
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet18 1 2 3 7 14 21 28 35 SE +/- 0.16, N = 3 SE +/- 0.12, N = 3 SE +/- 0.33, N = 3 29.04 29.04 29.18 MIN: 25.89 / MAX: 42.02 MIN: 25.69 / MAX: 36.07 MIN: 25.64 / MAX: 44.22 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.1 Model: MobileNetV2_224 3 2 1 1.2204 2.4408 3.6612 4.8816 6.102 SE +/- 0.037, N = 3 SE +/- 0.038, N = 3 SE +/- 0.019, N = 3 5.398 5.419 5.424 MIN: 4.83 / MAX: 15.64 MIN: 4.88 / MAX: 15.69 MIN: 4.8 / MAX: 14.75 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: efficientnet-b0 3 2 1 4 8 12 16 20 SE +/- 0.14, N = 3 SE +/- 0.26, N = 4 SE +/- 0.07, N = 3 16.90 16.95 16.98 MIN: 13.97 / MAX: 30.89 MIN: 14.04 / MAX: 27.15 MIN: 14.01 / MAX: 31.1 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Timed FFmpeg Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed FFmpeg Compilation 4.2.2 Time To Compile 1 3 2 40 80 120 160 200 SE +/- 0.34, N = 3 SE +/- 0.72, N = 3 SE +/- 0.22, N = 3 182.19 182.60 183.02
Google SynthMark Test: VoiceMark_100 OpenBenchmarking.org Voices, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 2 1 3 130 260 390 520 650 SE +/- 0.89, N = 3 SE +/- 1.56, N = 3 SE +/- 1.20, N = 3 596.62 596.25 593.91 1. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
TensorFlow Lite Model: NASNet Mobile OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: NASNet Mobile 3 1 2 70K 140K 210K 280K 350K SE +/- 504.11, N = 3 SE +/- 995.45, N = 3 SE +/- 384.90, N = 3 315790 316112 317213
GIMP Test: auto-levels OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: auto-levels 2 3 1 4 8 12 16 20 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 15.85 15.89 15.92
Waifu2x-NCNN Vulkan Scale: 2x - Denoise: 3 - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: No 3 2 1 0.9259 1.8518 2.7777 3.7036 4.6295 SE +/- 0.005, N = 3 SE +/- 0.005, N = 3 SE +/- 0.023, N = 3 4.097 4.110 4.115
GIMP Test: rotate OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: rotate 3 2 1 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 14.42 14.47 14.49
yquake2 Renderer: Software CPU - Resolution: 1920 x 1080 OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: Software CPU - Resolution: 1920 x 1080 3 2 1 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.38, N = 3 93.3 93.3 92.9 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
WebP Image Encode Encode Settings: Quality 100 OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 3 1 2 0.5852 1.1704 1.7556 2.3408 2.926 SE +/- 0.004, N = 3 SE +/- 0.009, N = 3 SE +/- 0.007, N = 3 2.590 2.595 2.601 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
x265 Video Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 4K 3 2 1 1.0868 2.1736 3.2604 4.3472 5.434 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 4.83 4.83 4.81 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU 1 2 3 4 8 12 16 20 SE +/- 0.17, N = 3 SE +/- 0.19, N = 3 SE +/- 0.24, N = 3 16.54 16.56 16.61 MIN: 13.54 MIN: 13.67 MIN: 13.5 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
AOM AV1 Encoder Mode: Speed 8 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 8 Realtime 3 1 2 6 12 18 24 30 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 27.29 27.20 27.18 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.13 Time To Compile 1 3 2 110 220 330 440 550 SE +/- 0.43, N = 3 SE +/- 1.15, N = 3 SE +/- 2.15, N = 3 514.48 514.80 516.52
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.1 Pfam Database Search 1 3 2 30 60 90 120 150 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.26, N = 3 127.15 127.39 127.65 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
TNN Target: CPU - Model: SqueezeNet v1.1 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 3 2 1 60 120 180 240 300 SE +/- 0.02, N = 3 SE +/- 0.28, N = 3 SE +/- 0.12, N = 3 286.14 287.06 287.26 MIN: 285.41 / MAX: 286.85 MIN: 286.17 / MAX: 287.91 MIN: 286.22 / MAX: 288.27 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
LZ4 Compression Compression Level: 1 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed 3 2 1 2K 4K 6K 8K 10K SE +/- 92.88, N = 3 SE +/- 41.71, N = 3 SE +/- 51.24, N = 3 8022.48 8015.39 7994.30 1. (CC) gcc options: -O3
Algebraic Multi-Grid Benchmark OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 2 3 1 50M 100M 150M 200M 250M SE +/- 543041.48, N = 3 SE +/- 483314.65, N = 3 SE +/- 326563.65, N = 3 214232233 214072633 213491533 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi
ASTC Encoder Preset: Thorough OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Thorough 1 3 2 20 40 60 80 100 SE +/- 0.21, N = 3 SE +/- 0.04, N = 3 SE +/- 0.21, N = 3 84.33 84.49 84.61 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Summer Nature 4K 3 2 1 12 24 36 48 60 SE +/- 0.44, N = 3 SE +/- 0.37, N = 3 SE +/- 0.35, N = 3 52.09 52.00 51.92 MIN: 48.07 / MAX: 61.57 MIN: 48.01 / MAX: 61.72 MIN: 48.08 / MAX: 61.73 1. (CC) gcc options: -pthread -ldl -lm
ASTC Encoder Preset: Exhaustive OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Exhaustive 3 1 2 150 300 450 600 750 SE +/- 0.20, N = 3 SE +/- 1.18, N = 3 SE +/- 0.48, N = 3 695.24 696.05 697.50 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
GIMP Test: resize OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: resize 3 1 2 3 6 9 12 15 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 12.83 12.86 12.87
Kvazaar Video Input: Bosphorus 4K - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Ultra Fast 2 1 3 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 6.85 6.84 6.83 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Timed MAFFT Alignment Multiple Sequence Alignment - LSU RNA OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.471 Multiple Sequence Alignment - LSU RNA 2 3 1 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.21, N = 3 15.00 15.04 15.04 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
RawTherapee Total Benchmark Time OpenBenchmarking.org Seconds, Fewer Is Better RawTherapee Total Benchmark Time 3 2 1 30 60 90 120 150 SE +/- 0.14, N = 3 SE +/- 0.21, N = 3 SE +/- 0.06, N = 3 123.34 123.41 123.68 1. RawTherapee, version 5.8, command line.
Kvazaar Video Input: Bosphorus 4K - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 4K - Video Preset: Very Fast 3 2 1 0.8888 1.7776 2.6664 3.5552 4.444 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.95 3.94 3.94 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
WebP Image Encode Encode Settings: Quality 100, Lossless OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Lossless 1 3 2 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.21, N = 3 SE +/- 0.10, N = 3 24.89 24.90 24.96 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
VKMark Resolution: 1920 x 1080 OpenBenchmarking.org VKMark Score, More Is Better VKMark 2020-05-21 Resolution: 1920 x 1080 2 1 3 300 600 900 1200 1500 1199 1199 1196 1. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
Monkey Audio Encoding WAV To APE OpenBenchmarking.org Seconds, Fewer Is Better Monkey Audio Encoding 3.99.6 WAV To APE 1 2 3 4 8 12 16 20 SE +/- 0.05, N = 5 SE +/- 0.08, N = 5 SE +/- 0.05, N = 5 15.96 15.99 16.00 1. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: googlenet 1 3 2 8 16 24 32 40 SE +/- 0.20, N = 3 SE +/- 0.13, N = 3 SE +/- 0.12, N = 4 32.56 32.59 32.64 MIN: 28.77 / MAX: 47.43 MIN: 28.42 / MAX: 45.73 MIN: 28.94 / MAX: 42.38 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: yolov4-tiny 1 3 2 13 26 39 52 65 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 59.28 59.28 59.42 MIN: 55.69 / MAX: 71.74 MIN: 55.36 / MAX: 72.94 MIN: 55.13 / MAX: 75.95 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Opus Codec Encoding WAV To Opus Encode OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.3.1 WAV To Opus Encode 3 2 1 2 4 6 8 10 SE +/- 0.021, N = 5 SE +/- 0.030, N = 5 SE +/- 0.033, N = 5 8.915 8.923 8.936 1. (CXX) g++ options: -fvisibility=hidden -logg -lm
Basis Universal Settings: UASTC Level 2 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 3 1 2 20 40 60 80 100 SE +/- 0.16, N = 3 SE +/- 0.24, N = 3 SE +/- 0.18, N = 3 86.34 86.48 86.54 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
CloverLeaf Lagrangian-Eulerian Hydrodynamics OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics 3 1 2 40 80 120 160 200 SE +/- 0.07, N = 3 SE +/- 0.19, N = 3 SE +/- 0.07, N = 3 191.01 191.41 191.45 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer ISPC - Model: Asian Dragon 2 3 1 0.7313 1.4626 2.1939 2.9252 3.6565 SE +/- 0.0130, N = 3 SE +/- 0.0025, N = 3 SE +/- 0.0133, N = 3 3.2504 3.2482 3.2431 MIN: 3.19 / MAX: 3.32 MIN: 3.2 / MAX: 3.32 MIN: 3.18 / MAX: 3.32
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast 2 3 1 6 12 18 24 30 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.16, N = 3 27.07 27.05 27.01 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Caffe Model: GoogleNet - Acceleration: CPU - Iterations: 100 OpenBenchmarking.org Milli-Seconds, Fewer Is Better Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 100 1 3 2 20K 40K 60K 80K 100K SE +/- 106.80, N = 3 SE +/- 223.03, N = 3 SE +/- 38.96, N = 3 110084 110157 110320 1. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas
Unpacking Firefox Extracting: firefox-84.0.source.tar.xz OpenBenchmarking.org Seconds, Fewer Is Better Unpacking Firefox 84.0 Extracting: firefox-84.0.source.tar.xz 2 3 1 6 12 18 24 30 SE +/- 0.03, N = 4 SE +/- 0.08, N = 4 SE +/- 0.03, N = 4 23.80 23.81 23.85
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.8.1 Video Input: Chimera 1080p 1 2 3 40 80 120 160 200 SE +/- 0.04, N = 3 SE +/- 1.76, N = 3 SE +/- 1.89, N = 3 184.25 184.17 183.88 MIN: 129.28 / MAX: 331.16 MIN: 127.79 / MAX: 333.51 MIN: 127.66 / MAX: 340.88 1. (CC) gcc options: -pthread -ldl -lm
Hierarchical INTegration Test: FLOAT OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT 2 1 3 60M 120M 180M 240M 300M SE +/- 109501.99, N = 3 SE +/- 252430.64, N = 3 SE +/- 702868.97, N = 3 301687480.19 301333349.99 301185316.59 1. (CC) gcc options: -O3 -march=native -lm
GLmark2 Resolution: 1920 x 1080 OpenBenchmarking.org Score, More Is Better GLmark2 2020.04 Resolution: 1920 x 1080 3 2 1 400 800 1200 1600 2000 1852 1851 1849
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.0 Video Input: Bosphorus 1080p - Video Preset: Medium 2 1 3 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 6.51 6.51 6.50 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
LZ4 Compression Compression Level: 9 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Decompression Speed 1 3 2 2K 4K 6K 8K 10K SE +/- 6.38, N = 13 SE +/- 3.43, N = 15 SE +/- 5.73, N = 15 8565.2 8562.8 8552.1 1. (CC) gcc options: -O3
Basis Universal Settings: ETC1S OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: ETC1S 1 3 2 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.14, N = 3 82.06 82.08 82.18 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
TensorFlow Lite Model: Inception ResNet V2 OpenBenchmarking.org Microseconds, Fewer Is Better TensorFlow Lite 2020-08-23 Model: Inception ResNet V2 3 1 2 1.2M 2.4M 3.6M 4.8M 6M SE +/- 6688.08, N = 3 SE +/- 3601.25, N = 3 SE +/- 3137.20, N = 3 5689070 5691310 5697083
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 3 1 2 60 120 180 240 300 SE +/- 0.29, N = 3 SE +/- 0.05, N = 3 SE +/- 0.44, N = 3 279.28 279.34 279.63 MIN: 276.39 / MAX: 296.43 MIN: 276.5 / MAX: 293.72 MIN: 276.69 / MAX: 295.47 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
GIMP Test: unsharp-mask OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: unsharp-mask 2 3 1 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 17.32 17.33 17.33
LZ4 Compression Compression Level: 3 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Decompression Speed 1 2 3 2K 4K 6K 8K 10K SE +/- 26.74, N = 3 SE +/- 8.28, N = 15 SE +/- 5.83, N = 15 8554.3 8547.8 8547.2 1. (CC) gcc options: -O3
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes 3 1 2 100 200 300 400 500 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 482.55 482.61 482.84
Waifu2x-NCNN Vulkan Scale: 2x - Denoise: 3 - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: Yes 3 1 2 6 12 18 24 30 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 26.67 26.68 26.69
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No 3 2 1 14 28 42 56 70 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 63.00 63.02 63.03
WebP Image Encode Encode Settings: Quality 100, Highest Compression OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100, Highest Compression 3 2 1 2 4 6 8 10 SE +/- 0.013, N = 3 SE +/- 0.011, N = 3 SE +/- 0.001, N = 3 8.869 8.871 8.872 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
simdjson Throughput Test: DistinctUserID OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: DistinctUserID 3 2 1 0.1035 0.207 0.3105 0.414 0.5175 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.46 0.46 0.46 1. (CXX) g++ options: -O3 -pthread
simdjson Throughput Test: PartialTweets OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: PartialTweets 3 2 1 0.1013 0.2026 0.3039 0.4052 0.5065 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.45 0.45 0.45 1. (CXX) g++ options: -O3 -pthread
simdjson Throughput Test: LargeRandom OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: LargeRandom 3 2 1 0.0788 0.1576 0.2364 0.3152 0.394 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.35 0.35 0.35 1. (CXX) g++ options: -O3 -pthread
simdjson Throughput Test: Kostya OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: Kostya 3 2 1 0.0855 0.171 0.2565 0.342 0.4275 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.38 0.38 0.38 1. (CXX) g++ options: -O3 -pthread
CLOMP Static OMP Speedup OpenBenchmarking.org Speedup, More Is Better CLOMP 1.2 Static OMP Speedup 3 2 1 0.45 0.9 1.35 1.8 2.25 SE +/- 0.03, N = 3 2.0 2.0 2.0 1. (CC) gcc options: -fopenmp -O3 -lm
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 3 1 2 3 6 9 12 15 SE +/- 0.38, N = 3 SE +/- 0.01, N = 3 SE +/- 0.21, N = 4 10.68 10.88 11.30 MIN: 8.89 / MAX: 17.41 MIN: 8.93 / MAX: 27.73 MIN: 8.86 / MAX: 23.26 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU 3 2 1 8 16 24 32 40 SE +/- 1.44, N = 12 SE +/- 1.77, N = 15 SE +/- 1.59, N = 15 33.06 35.85 36.56 MIN: 26.99 MIN: 27.16 MIN: 27.16 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
LZ4 Compression Compression Level: 9 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed 1 3 2 10 20 30 40 50 SE +/- 0.73, N = 13 SE +/- 0.47, N = 15 SE +/- 0.55, N = 15 42.22 41.23 41.02 1. (CC) gcc options: -O3
Sockperf Test: Latency Under Load OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Under Load 3 2 1 12 24 36 48 60 SE +/- 1.93, N = 25 SE +/- 1.88, N = 25 SE +/- 2.25, N = 20 50.70 52.76 53.78 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Phoronix Test Suite v10.8.4