haswell may Intel Xeon E5-2687W v3 testing with a MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS) and NVIDIA GeForce GTX 770 on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2105102-IB-HASWELLMA37&grs .
haswell may Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Compiler File-System Screen Resolution 1 2 3 Intel Xeon E5-2687W v3 @ 3.50GHz (10 Cores / 20 Threads) MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS) Intel Xeon E7 v3/Xeon 32GB 80GB INTEL SSDSCKGW08 NVIDIA GeForce GTX 770 Realtek ALC892 LG Ultra HD Intel I218-V Ubuntu 20.04 5.9.0-050900rc7daily20200928-generic (x86_64) 20200927 GNOME Shell 3.36.4 X Server 1.20.9 GCC 9.3.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise 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++,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 Processor Details - Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x44 Python Details - Python 3.8.5 Security Details - itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
haswell may svt-vp9: VMAF Optimized - Bosphorus 1080p srslte: OFDM_Test viennacl: CPU BLAS - dAXPY vosk: toktx: Zstd Compression 9 helsing: 14 digit qmcpack: simple-H2O embree: Pathtracer - Crown aom-av1: Speed 0 Two-Pass - Bosphorus 1080p stockfish: Total Time svt-av1: Preset 4 - Bosphorus 1080p luaradio: Five Back to Back FIR Filters viennacl: CPU BLAS - dGEMM-TN chia-vdf: Square Assembly Optimized avifenc: 10, Lossless aom-av1: Speed 9 Realtime - Bosphorus 1080p basis: UASTC Level 0 onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU sysbench: RAM / Memory aom-av1: Speed 6 Realtime - Bosphorus 4K compress-zstd: 3, Long Mode - Compression Speed embree: Pathtracer - Asian Dragon aom-av1: Speed 6 Two-Pass - Bosphorus 4K compress-zstd: 3 - Compression Speed mnn: MobileNetV2_224 mrbayes: Primate Phylogeny Analysis astcenc: Medium avifenc: 2 build-linux-kernel: Time To Compile avifenc: 0 xmrig: Wownero - 1M compress-zstd: 19, Long Mode - Compression Speed avifenc: 6 tjbench: Decompression Throughput liquid-dsp: 16 - 256 - 57 srslte: PHY_DL_Test incompact3d: input.i3d 193 Cells Per Direction aom-av1: Speed 8 Realtime - Bosphorus 1080p mnn: resnet-v2-50 viennacl: CPU BLAS - dGEMM-NN srslte: PHY_DL_Test toktx: UASTC 3 + Zstd Compression 19 draco: Lion compress-zstd: 19, Long Mode - Decompression Speed build-llvm: Unix Makefiles compress-zstd: 8 - Compression Speed gnuradio: Five Back to Back FIR Filters build-erlang: Time To Compile incompact3d: input.i3d 129 Cells Per Direction aom-av1: Speed 6 Realtime - Bosphorus 1080p draco: Church Facade svt-av1: Preset 8 - Bosphorus 1080p viennacl: CPU BLAS - dGEMM-NT blender: Pabellon Barcelona - CPU-Only compress-zstd: 8, Long Mode - Compression Speed compress-zstd: 8, Long Mode - Decompression Speed dav1d: Summer Nature 1080p liquid-dsp: 2 - 256 - 57 compress-zstd: 19 - Compression Speed liquid-dsp: 4 - 256 - 57 viennacl: CPU BLAS - dGEMV-N xmrig: Monero - 1M basis: ETC1S luaradio: FM Deemphasis Filter dav1d: Chimera 1080p avifenc: 10 luaradio: Hilbert Transform gnuradio: FIR Filter aom-av1: Speed 4 Two-Pass - Bosphorus 1080p compress-zstd: 3, Long Mode - Decompression Speed viennacl: CPU BLAS - dGEMM-TT gnuradio: Signal Source (Cosine) build-mesa: Time To Compile blender: Classroom - CPU-Only mnn: inception-v3 toktx: UASTC 3 onednn: IP Shapes 3D - f32 - CPU libgav1: Chimera 1080p botan: AES-256 embree: Pathtracer ISPC - Crown dav1d: Summer Nature 4K viennacl: CPU BLAS - dGEMV-T toktx: Zstd Compression 19 onednn: Deconvolution Batch shapes_1d - f32 - CPU avifenc: 6, Lossless libgav1: Summer Nature 1080p mnn: SqueezeNetV1.0 onednn: IP Shapes 1D - u8s8f32 - CPU embree: Pathtracer ISPC - Asian Dragon Obj svt-hevc: 7 - Bosphorus 1080p onednn: Recurrent Neural Network Inference - u8s8f32 - CPU libgav1: Summer Nature 4K svt-vp9: PSNR/SSIM Optimized - Bosphorus 1080p aom-av1: Speed 9 Realtime - Bosphorus 4K toybrot: C++ Tasks blender: Barbershop - CPU-Only onednn: IP Shapes 1D - f32 - CPU svt-av1: Preset 4 - Bosphorus 4K svt-av1: Preset 8 - Bosphorus 4K mnn: mobilenet-v1-1.0 aom-av1: Speed 4 Two-Pass - Bosphorus 4K pjsip: OPTIONS, Stateful gnuradio: IIR Filter gnuradio: FM Deemphasis Filter onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU blender: Fishy Cat - CPU-Only onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU viennacl: CPU BLAS - sCOPY pjsip: OPTIONS, Stateless embree: Pathtracer - Asian Dragon Obj libgav1: Chimera 1080p 10-bit onednn: IP Shapes 3D - u8s8f32 - CPU build-nodejs: Time To Compile luaradio: Complex Phase botan: CAST-256 gmpbench: Total Time botan: ChaCha20Poly1305 - Decrypt compress-zstd: 19 - Decompression Speed liquid-dsp: 8 - 256 - 57 aom-av1: Speed 8 Realtime - Bosphorus 4K botan: CAST-256 - Decrypt blender: BMW27 - CPU-Only onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU botan: ChaCha20Poly1305 viennacl: CPU BLAS - sAXPY embree: Pathtracer ISPC - Asian Dragon botan: AES-256 - Decrypt viennacl: CPU BLAS - sDOT basis: UASTC Level 2 onednn: Recurrent Neural Network Training - u8s8f32 - CPU compress-zstd: 8 - Decompression Speed hmmer: Pfam Database Search svt-hevc: 1 - Bosphorus 1080p toybrot: C++ Threads build-llvm: Ninja onednn: Deconvolution Batch shapes_3d - f32 - CPU aom-av1: Speed 6 Two-Pass - Bosphorus 1080p botan: Twofish - Decrypt basis: UASTC Level 3 gnuradio: Hilbert Transform toybrot: OpenMP onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU botan: Twofish onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU astcenc: Exhaustive onednn: Recurrent Neural Network Inference - f32 - CPU botan: KASUMI svt-hevc: 10 - Bosphorus 1080p botan: KASUMI - Decrypt onednn: Recurrent Neural Network Training - f32 - CPU astcenc: Thorough svt-vp9: Visual Quality Optimized - Bosphorus 1080p helsing: 12 digit botan: Blowfish - Decrypt onednn: Convolution Batch Shapes Auto - f32 - CPU liquid-dsp: 20 - 256 - 57 botan: Blowfish toktx: UASTC 4 + Zstd Compression 19 dav1d: Chimera 1080p 10-bit pjsip: INVITE securemark: SecureMark-TLS onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU liquid-dsp: 1 - 256 - 57 sysbench: CPU viennacl: CPU BLAS - dCOPY aom-av1: Speed 0 Two-Pass - Bosphorus 4K chia-vdf: Square Plain C++ viennacl: CPU BLAS - dDOT 1 2 3 131.62 72300000 36.2 26.641 4.174 599.723 52.540 9.8735 0.26 21549515 3.065 678.6 29.4 119633 9.080 72.30 11.079 3.48130 16674.21 8.98 609.1 11.4729 4.55 1958.4 5.116 187.452 7.9701 53.703 121.584 101.116 4762.3 29.6 18.928 158.053214 441220000 173.6 127.615463 60.92 53.833 27.6 70.3 23.681 7235 2477.9 964.590 355.1 497.7 168.766 31.8818099 15.19 10529 37.198 26.9 652.69 446.9 2860.7 394.33 93297000 32.1 168710000 45.7 3039.3 34.099 328.2 472.73 5.375 77.9 474.6 4.09 2799.0 27.9 1855.6 75.847 589.06 57.177 17.165 7.46460 89.69 3116.537 10.9213 145.15 49.2 23.367 6.55061 81.115 126.28 8.740 2.78165 11.8105 82.79 2227.68 43.76 137.41 37.40 66632 775.57 4.12581 0.9 9.987 6.133 2.37 5680 411.9 517.1 3.35073 267.61 3.55190 27.5 36173 10.5853 31.64 2.44319 595.295 417.4 109.393 3657.3 572.534 2469.6 343806667 27.51 109.251 196.01 5.66569 578.937 41.3 13.1137 3103.88 43.7 43.769 4049.70 2688.1 187.834 5.72 66253 925.475 8.39311 12.39 286.169 80.198 264.7 66314 12.7023 288.183 2213.82 141.7632 2215.41 75.999 173.18 71.946 4053.66 18.8220 108.08 6.450 345.055 13.7665 452546667 346.405 444.730 88.68 3321 191107 4051.75 49054667 15805.17 0.09 108100 38.8 112.13 63100000 41.4 29.277 4.102 601.194 52.487 9.5853 0.26 20756949 3.018 662.4 29.2 117600 9.289 73.91 10.867 3.4953 16808.94 9.17 602.4 11.2554 4.63 1970.8 5.178 184.004 8.121 52.725 122.873 102.93 4805.3 29.1 19.232 155.494139 444300000 175.3 127.763268 61.77 53.942 28 70.8 23.467 7316 2476.3 964.077 353.7 493.3 166.64 31.7161522 15.37 10468 36.775 26.6 659.72 442.5 2843.5 395.79 93119000 32.4 167210000 45.6 3012.8 34.278 328.2 475.88 5.41 78.3 472.1 4.06 2819.5 27.7 1844.4 75.824 586.2 56.812 17.263 7.50073 89.52 3125.518 10.9347 144.78 48.9 23.264 6.56384 80.642 125.69 8.692 2.76679 11.7505 82.63 2217.15 43.76 137.02 37.23 66341 778.21 4.13077 0.904 9.943 6.106 2.36 5666 413.5 517.9 3.36355 268.21 3.54188 27.5 36043 10.5473 31.66 2.44752 593.248 417.3 109.742 3645.3 573.601 2477.7 343480000 27.45 109.567 195.56 5.65039 577.483 41.2 13.1354 3100.643 43.6 43.677 4053.76 2693.4 187.917 5.73 66336 925.736 8.39791 12.41 286.613 80.076 264.3 66391 12.6855 288.542 2213.73 141.9129 2217.72 75.92 173.01 71.885 4050.46 18.8103 108.03 6.446 344.851 13.7675 452770000 346.287 444.538 88.67 3320 191158 4052.37 49056000 15806.12 27.5 0.09 108100 43.8 113.45 72700000 41.4 29.207 4.435 627.965 50.184 10.0219 0.25 20973440 2.968 682.5 29.4 116500 9.065 72.17 11.115 3.55843 16459.64 9.11 614.4 11.2491 4.64 1996.5 5.081 187.497 8.1096 53.514 123.825 101.731 4844.9 29.1 18.908 156.529497 437390000 172.6 129.590805 60.85 54.614 27.7 69.8 23.801 7335 2510.4 951.556 350.4 499.6 167.352 31.4801865 15.37 10590 36.846 26.7 656.27 443.4 2832.8 392.01 94009000 32.3 168540000 45.4 3032.8 34.387 330.9 476.55 5.368 77.7 471 4.09 2816.2 27.7 1843 76.334 585.15 56.799 17.279 7.5126 90.08 3106.226 10.989 144.26 48.9 23.404 6.52527 80.774 126.39 8.701 2.76696 11.7876 82.37 2216.7 43.55 137.66 37.3 66331 779.06 4.11242 0.902 9.95 6.128 2.36 5689 413.1 519.1 3.35822 268.63 3.53866 27.4 36169 10.5545 31.55 2.43904 593.916 418.7 109.754 3653.9 574.418 2471.8 342790000 27.43 109.563 195.45 5.66558 578.173 41.3 13.1041 3096.616 43.7 43.689 4058.11 2693.4 187.558 5.73 66224 927.025 8.40666 12.41 286.502 80.104 264.6 66299 12.6892 288.535 2216.1 141.7768 2216.51 75.936 173.06 71.948 4051.36 18.8248 108 6.447 344.919 13.7742 452640000 346.248 444.664 88.65 3320 191151 4051.76 49052000 15805.7 27.5 0.09 108100 43.8 OpenBenchmarking.org
SVT-VP9 Tuning: VMAF Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p 1 2 3 30 60 90 120 150 SE +/- 1.64, N = 12 131.62 112.13 113.45 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
srsLTE Test: OFDM_Test OpenBenchmarking.org Samples / Second, More Is Better srsLTE 20.10.1 Test: OFDM_Test 1 2 3 16M 32M 48M 64M 80M 72300000 63100000 72700000 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
ViennaCL Test: CPU BLAS - dAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dAXPY 1 2 3 9 18 27 36 45 36.2 41.4 41.4 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
VOSK Speech Recognition Toolkit OpenBenchmarking.org Seconds, Fewer Is Better VOSK Speech Recognition Toolkit 0.3.21 1 2 3 7 14 21 28 35 SE +/- 0.22, N = 13 26.64 29.28 29.21
KTX-Software toktx Settings: Zstd Compression 9 OpenBenchmarking.org Seconds, Fewer Is Better KTX-Software toktx 4.0 Settings: Zstd Compression 9 1 2 3 0.9979 1.9958 2.9937 3.9916 4.9895 SE +/- 0.043, N = 8 4.174 4.102 4.435
Helsing Digit Range: 14 digit OpenBenchmarking.org Seconds, Fewer Is Better Helsing 1.0-beta Digit Range: 14 digit 1 2 3 140 280 420 560 700 SE +/- 0.57, N = 3 599.72 601.19 627.97 1. (CC) gcc options: -O2 -pthread
QMCPACK Input: simple-H2O OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.11 Input: simple-H2O 1 2 3 12 24 36 48 60 SE +/- 0.70, N = 5 52.54 52.49 50.18 1. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Crown 1 2 3 3 6 9 12 15 SE +/- 0.0077, N = 3 9.8735 9.5853 10.0219 MIN: 9.81 / MAX: 10.02 MIN: 9.53 / MAX: 9.7 MIN: 9.96 / MAX: 10.15
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p 1 2 3 0.0585 0.117 0.1755 0.234 0.2925 SE +/- 0.00, N = 3 0.26 0.26 0.25 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 13 Total Time 1 2 3 5M 10M 15M 20M 25M SE +/- 245397.16, N = 3 21549515 20756949 20973440 1. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p 1 2 3 0.6896 1.3792 2.0688 2.7584 3.448 3.065 3.018 2.968 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
LuaRadio Test: Five Back to Back FIR Filters OpenBenchmarking.org MiB/s, More Is Better LuaRadio 0.9.1 Test: Five Back to Back FIR Filters 1 2 3 150 300 450 600 750 SE +/- 6.78, N = 3 678.6 662.4 682.5
ViennaCL Test: CPU BLAS - dGEMM-TN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TN 1 2 3 7 14 21 28 35 SE +/- 0.00, N = 3 29.4 30.0 29.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Chia Blockchain VDF Test: Square Assembly Optimized OpenBenchmarking.org IPS, More Is Better Chia Blockchain VDF 1.0.1 Test: Square Assembly Optimized 1 2 3 30K 60K 90K 120K 150K SE +/- 384.42, N = 3 119633 117600 116500 1. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.9.0 Encoder Speed: 10, Lossless 1 2 3 3 6 9 12 15 SE +/- 0.010, N = 3 9.080 9.289 9.065 1. (CXX) g++ options: -O3 -fPIC -lm
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p 1 2 3 16 32 48 64 80 SE +/- 0.31, N = 3 72.30 73.91 72.17 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Basis Universal Settings: UASTC Level 0 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.13 Settings: UASTC Level 0 1 2 3 3 6 9 12 15 SE +/- 0.05, N = 3 11.08 10.87 11.12 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.8006 1.6012 2.4018 3.2024 4.003 SE +/- 0.00516, N = 3 3.48130 3.49530 3.55843 MIN: 3.4 MIN: 3.43 MIN: 3.47 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Sysbench Test: RAM / Memory OpenBenchmarking.org MiB/sec, More Is Better Sysbench 1.0.20 Test: RAM / Memory 1 2 3 4K 8K 12K 16K 20K SE +/- 67.40, N = 3 16674.21 16808.94 16459.64 1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K 1 2 3 3 6 9 12 15 SE +/- 0.11, N = 3 8.98 9.17 9.11 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Zstd Compression Compression Level: 3, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 3, Long Mode - Compression Speed 1 2 3 130 260 390 520 650 SE +/- 1.16, N = 3 609.1 602.4 614.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon 1 2 3 3 6 9 12 15 SE +/- 0.05, N = 3 11.47 11.26 11.25 MIN: 11.35 / MAX: 11.67 MIN: 11.22 / MAX: 11.35 MIN: 11.21 / MAX: 11.38
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K 1 2 3 1.044 2.088 3.132 4.176 5.22 SE +/- 0.01, N = 3 4.55 4.63 4.64 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Zstd Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 3 - Compression Speed 1 2 3 400 800 1200 1600 2000 SE +/- 16.60, N = 3 1958.4 1970.8 1996.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: MobileNetV2_224 1 2 3 1.1651 2.3302 3.4953 4.6604 5.8255 SE +/- 0.026, N = 3 5.116 5.178 5.081 MIN: 5.04 / MAX: 6.71 MIN: 5.08 / MAX: 5.99 MIN: 5.04 / MAX: 5.98 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
Timed MrBayes Analysis Primate Phylogeny Analysis OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis 1 2 3 40 80 120 160 200 SE +/- 2.70, N = 3 187.45 184.00 187.50 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -mabm -O3 -std=c99 -pedantic -lm
ASTC Encoder Preset: Medium OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.4 Preset: Medium 1 2 3 2 4 6 8 10 SE +/- 0.0531, N = 3 7.9701 8.1210 8.1096 1. (CXX) g++ options: -O3 -flto -pthread
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.9.0 Encoder Speed: 2 1 2 3 12 24 36 48 60 SE +/- 0.36, N = 3 53.70 52.73 53.51 1. (CXX) g++ options: -O3 -fPIC -lm
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.10.20 Time To Compile 1 2 3 30 60 90 120 150 SE +/- 1.30, N = 3 121.58 122.87 123.83
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.9.0 Encoder Speed: 0 1 2 3 20 40 60 80 100 SE +/- 0.79, N = 3 101.12 102.93 101.73 1. (CXX) g++ options: -O3 -fPIC -lm
Xmrig Variant: Wownero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Wownero - Hash Count: 1M 1 2 3 1000 2000 3000 4000 5000 SE +/- 28.81, N = 3 4762.3 4805.3 4844.9 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Zstd Compression Compression Level: 19, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 19, Long Mode - Compression Speed 1 2 3 7 14 21 28 35 SE +/- 0.46, N = 3 29.6 29.1 29.1 1. (CC) gcc options: -O3 -pthread -lz -llzma
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.9.0 Encoder Speed: 6 1 2 3 5 10 15 20 25 SE +/- 0.06, N = 3 18.93 19.23 18.91 1. (CXX) g++ options: -O3 -fPIC -lm
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.1.0 Test: Decompression Throughput 1 2 3 30 60 90 120 150 SE +/- 0.22, N = 3 158.05 155.49 156.53 1. (CC) gcc options: -O3 -rdynamic
Liquid-DSP Threads: 16 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 16 - Buffer Length: 256 - Filter Length: 57 1 2 3 100M 200M 300M 400M 500M SE +/- 3448975.69, N = 3 441220000 444300000 437390000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
srsLTE Test: PHY_DL_Test OpenBenchmarking.org eNb Mb/s, More Is Better srsLTE 20.10.1 Test: PHY_DL_Test 1 2 3 40 80 120 160 200 SE +/- 0.25, N = 3 173.6 175.3 172.6 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
Xcompact3d Incompact3d Input: input.i3d 193 Cells Per Direction OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction 1 2 3 30 60 90 120 150 SE +/- 1.48, N = 3 127.62 127.76 129.59 1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p 1 2 3 14 28 42 56 70 SE +/- 0.40, N = 3 60.92 61.77 60.85 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: resnet-v2-50 1 2 3 12 24 36 48 60 SE +/- 0.09, N = 3 53.83 53.94 54.61 MIN: 53.59 / MAX: 57.32 MIN: 53.82 / MAX: 54.84 MIN: 54.3 / MAX: 55.77 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
ViennaCL Test: CPU BLAS - dGEMM-NN OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NN 1 2 3 7 14 21 28 35 SE +/- 0.09, N = 3 27.6 28.0 27.7 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
srsLTE Test: PHY_DL_Test OpenBenchmarking.org UE Mb/s, More Is Better srsLTE 20.10.1 Test: PHY_DL_Test 1 2 3 16 32 48 64 80 SE +/- 0.09, N = 3 70.3 70.8 69.8 1. (CXX) g++ options: -std=c++11 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -rdynamic -lpthread -lmbedcrypto -lconfig++ -lsctp -lbladeRF -lm -lfftw3f
KTX-Software toktx Settings: UASTC 3 + Zstd Compression 19 OpenBenchmarking.org Seconds, Fewer Is Better KTX-Software toktx 4.0 Settings: UASTC 3 + Zstd Compression 19 1 2 3 6 12 18 24 30 SE +/- 0.05, N = 3 23.68 23.47 23.80
Google Draco Model: Lion OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.4.1 Model: Lion 1 2 3 1600 3200 4800 6400 8000 SE +/- 67.87, N = 3 7235 7316 7335 1. (CXX) g++ options: -O3
Zstd Compression Compression Level: 19, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 19, Long Mode - Decompression Speed 1 2 3 500 1000 1500 2000 2500 SE +/- 17.64, N = 3 2477.9 2476.3 2510.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Timed LLVM Compilation Build System: Unix Makefiles OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 12.0 Build System: Unix Makefiles 1 2 3 200 400 600 800 1000 SE +/- 10.37, N = 3 964.59 964.08 951.56
Zstd Compression Compression Level: 8 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 8 - Compression Speed 1 2 3 80 160 240 320 400 SE +/- 0.92, N = 3 355.1 353.7 350.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
GNU Radio Test: Five Back to Back FIR Filters OpenBenchmarking.org MiB/s, More Is Better GNU Radio Test: Five Back to Back FIR Filters 1 2 3 110 220 330 440 550 SE +/- 1.44, N = 3 497.7 493.3 499.6 1. 3.8.1.0
Timed Erlang/OTP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Erlang/OTP Compilation 23.2 Time To Compile 1 2 3 40 80 120 160 200 SE +/- 0.95, N = 3 168.77 166.64 167.35
Xcompact3d Incompact3d Input: input.i3d 129 Cells Per Direction OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 129 Cells Per Direction 1 2 3 7 14 21 28 35 SE +/- 0.24, N = 3 31.88 31.72 31.48 1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p 1 2 3 4 8 12 16 20 SE +/- 0.08, N = 3 15.19 15.37 15.37 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Google Draco Model: Church Facade OpenBenchmarking.org ms, Fewer Is Better Google Draco 1.4.1 Model: Church Facade 1 2 3 2K 4K 6K 8K 10K SE +/- 29.72, N = 3 10529 10468 10590 1. (CXX) g++ options: -O3
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p 1 2 3 9 18 27 36 45 37.20 36.78 36.85 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
ViennaCL Test: CPU BLAS - dGEMM-NT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NT 1 2 3 6 12 18 24 30 SE +/- 0.20, N = 3 26.9 26.6 26.7 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Pabellon Barcelona - Compute: CPU-Only 1 2 3 140 280 420 560 700 SE +/- 0.85, N = 3 652.69 659.72 656.27
Zstd Compression Compression Level: 8, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 8, Long Mode - Compression Speed 1 2 3 100 200 300 400 500 SE +/- 1.64, N = 3 446.9 442.5 443.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Zstd Compression Compression Level: 8, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 8, Long Mode - Decompression Speed 1 2 3 600 1200 1800 2400 3000 SE +/- 2.20, N = 3 2860.7 2843.5 2832.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.8.2 Video Input: Summer Nature 1080p 1 2 3 90 180 270 360 450 SE +/- 1.59, N = 3 394.33 395.79 392.01 MIN: 310.75 / MAX: 435.18 MIN: 309.28 / MAX: 431.68 MIN: 304.75 / MAX: 426.38 1. (CC) gcc options: -pthread -lm
Liquid-DSP Threads: 2 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 2 - Buffer Length: 256 - Filter Length: 57 1 2 3 20M 40M 60M 80M 100M SE +/- 357366.76, N = 3 93297000 93119000 94009000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Zstd Compression Compression Level: 19 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 19 - Compression Speed 1 2 3 8 16 24 32 40 SE +/- 0.17, N = 3 32.1 32.4 32.3 1. (CC) gcc options: -O3 -pthread -lz -llzma
Liquid-DSP Threads: 4 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 4 - Buffer Length: 256 - Filter Length: 57 1 2 3 40M 80M 120M 160M 200M SE +/- 5773.50, N = 3 168710000 167210000 168540000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
ViennaCL Test: CPU BLAS - dGEMV-N OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-N 1 2 3 10 20 30 40 50 SE +/- 0.03, N = 3 45.7 45.6 45.3 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Xmrig Variant: Monero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Monero - Hash Count: 1M 1 2 3 700 1400 2100 2800 3500 SE +/- 13.28, N = 3 3039.3 3012.8 3032.8 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Basis Universal Settings: ETC1S OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.13 Settings: ETC1S 1 2 3 8 16 24 32 40 SE +/- 0.20, N = 3 34.10 34.28 34.39 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
LuaRadio Test: FM Deemphasis Filter OpenBenchmarking.org MiB/s, More Is Better LuaRadio 0.9.1 Test: FM Deemphasis Filter 1 2 3 70 140 210 280 350 SE +/- 1.17, N = 3 328.2 328.2 330.9
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.8.2 Video Input: Chimera 1080p 1 2 3 100 200 300 400 500 SE +/- 2.47, N = 3 472.73 475.88 476.55 MIN: 350.42 / MAX: 614.21 MIN: 357.72 / MAX: 609.66 MIN: 358.56 / MAX: 609.47 1. (CC) gcc options: -pthread -lm
libavif avifenc Encoder Speed: 10 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.9.0 Encoder Speed: 10 1 2 3 1.2173 2.4346 3.6519 4.8692 6.0865 SE +/- 0.015, N = 3 5.375 5.410 5.368 1. (CXX) g++ options: -O3 -fPIC -lm
LuaRadio Test: Hilbert Transform OpenBenchmarking.org MiB/s, More Is Better LuaRadio 0.9.1 Test: Hilbert Transform 1 2 3 20 40 60 80 100 SE +/- 0.22, N = 3 77.9 78.3 77.7
GNU Radio Test: FIR Filter OpenBenchmarking.org MiB/s, More Is Better GNU Radio Test: FIR Filter 1 2 3 100 200 300 400 500 SE +/- 0.32, N = 3 474.6 472.1 471.0 1. 3.8.1.0
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p 1 2 3 0.9203 1.8406 2.7609 3.6812 4.6015 SE +/- 0.01, N = 3 4.09 4.06 4.09 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Zstd Compression Compression Level: 3, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 3, Long Mode - Decompression Speed 1 2 3 600 1200 1800 2400 3000 SE +/- 9.64, N = 3 2799.0 2819.5 2816.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
ViennaCL Test: CPU BLAS - dGEMM-TT OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TT 1 2 3 7 14 21 28 35 SE +/- 0.13, N = 3 27.9 27.7 27.8 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
GNU Radio Test: Signal Source (Cosine) OpenBenchmarking.org MiB/s, More Is Better GNU Radio Test: Signal Source (Cosine) 1 2 3 400 800 1200 1600 2000 SE +/- 12.37, N = 3 1855.6 1844.4 1843.0 1. 3.8.1.0
Timed Mesa Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Mesa Compilation 21.0 Time To Compile 1 2 3 20 40 60 80 100 SE +/- 0.19, N = 3 75.85 75.82 76.33
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Classroom - Compute: CPU-Only 1 2 3 130 260 390 520 650 SE +/- 2.36, N = 3 589.06 586.20 585.15
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: inception-v3 1 2 3 13 26 39 52 65 SE +/- 0.12, N = 3 57.18 56.81 56.80 MIN: 56.86 / MAX: 126.61 MIN: 56.64 / MAX: 59.14 MIN: 56.68 / MAX: 60.57 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
KTX-Software toktx Settings: UASTC 3 OpenBenchmarking.org Seconds, Fewer Is Better KTX-Software toktx 4.0 Settings: UASTC 3 1 2 3 4 8 12 16 20 SE +/- 0.13, N = 3 17.17 17.26 17.28
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU 1 2 3 2 4 6 8 10 SE +/- 0.00495, N = 3 7.46460 7.50073 7.51260 MIN: 7.41 MIN: 7.46 MIN: 7.47 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
libgav1 Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Chimera 1080p 1 2 3 20 40 60 80 100 SE +/- 0.07, N = 3 89.69 89.52 90.08 1. (CXX) g++ options: -O3 -lpthread -lrt
Botan Test: AES-256 OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: AES-256 1 2 3 700 1400 2100 2800 3500 SE +/- 7.97, N = 3 3116.54 3125.52 3106.23 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Crown 1 2 3 3 6 9 12 15 SE +/- 0.01, N = 3 10.92 10.93 10.99 MIN: 10.84 / MAX: 11.11 MIN: 10.86 / MAX: 11.08 MIN: 10.92 / MAX: 11.15
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.8.2 Video Input: Summer Nature 4K 1 2 3 30 60 90 120 150 SE +/- 0.38, N = 3 145.15 144.78 144.26 MIN: 134.65 / MAX: 164.89 MIN: 133.45 / MAX: 163.18 MIN: 129.68 / MAX: 162.64 1. (CC) gcc options: -pthread -lm
ViennaCL Test: CPU BLAS - dGEMV-T OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-T 1 2 3 11 22 33 44 55 SE +/- 0.03, N = 3 49.2 49.1 49.1 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
KTX-Software toktx Settings: Zstd Compression 19 OpenBenchmarking.org Seconds, Fewer Is Better KTX-Software toktx 4.0 Settings: Zstd Compression 19 1 2 3 6 12 18 24 30 SE +/- 0.07, N = 3 23.37 23.26 23.40
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU 1 2 3 2 4 6 8 10 SE +/- 0.00898, N = 3 6.55061 6.56384 6.52527 MIN: 6.33 MIN: 6.37 MIN: 6.33 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.9.0 Encoder Speed: 6, Lossless 1 2 3 20 40 60 80 100 SE +/- 0.35, N = 3 81.12 80.64 80.77 1. (CXX) g++ options: -O3 -fPIC -lm
libgav1 Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Summer Nature 1080p 1 2 3 30 60 90 120 150 SE +/- 0.29, N = 3 126.28 125.69 126.39 1. (CXX) g++ options: -O3 -lpthread -lrt
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: SqueezeNetV1.0 1 2 3 2 4 6 8 10 SE +/- 0.012, N = 3 8.740 8.692 8.701 MIN: 8.62 / MAX: 9.61 MIN: 8.58 / MAX: 10.15 MIN: 8.61 / MAX: 9.65 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: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.6259 1.2518 1.8777 2.5036 3.1295 SE +/- 0.00273, N = 3 2.78165 2.76679 2.76696 MIN: 2.75 MIN: 2.75 MIN: 2.74 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon Obj 1 2 3 3 6 9 12 15 SE +/- 0.01, N = 3 11.81 11.75 11.79 MIN: 11.75 / MAX: 11.97 MIN: 11.7 / MAX: 11.89 MIN: 11.74 / MAX: 11.94
SVT-HEVC Tuning: 7 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 1080p 1 2 3 20 40 60 80 100 SE +/- 0.06, N = 3 82.79 82.63 82.37 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU 1 2 3 500 1000 1500 2000 2500 SE +/- 16.51, N = 3 2227.68 2217.15 2216.70 MIN: 2203.44 MIN: 2215.26 MIN: 2214.59 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
libgav1 Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Summer Nature 4K 1 2 3 10 20 30 40 50 SE +/- 0.13, N = 3 43.76 43.76 43.55 1. (CXX) g++ options: -O3 -lpthread -lrt
SVT-VP9 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p 1 2 3 30 60 90 120 150 SE +/- 0.08, N = 3 137.41 137.02 137.66 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K 1 2 3 9 18 27 36 45 SE +/- 0.06, N = 3 37.40 37.23 37.30 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
toyBrot Fractal Generator Implementation: C++ Tasks OpenBenchmarking.org ms, Fewer Is Better toyBrot Fractal Generator 2020-11-18 Implementation: C++ Tasks 1 2 3 14K 28K 42K 56K 70K SE +/- 177.25, N = 3 66632 66341 66331 1. (CXX) g++ options: -O3 -lpthread
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Barbershop - Compute: CPU-Only 1 2 3 200 400 600 800 1000 SE +/- 0.90, N = 3 775.57 778.21 779.06
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU 1 2 3 0.9294 1.8588 2.7882 3.7176 4.647 SE +/- 0.01500, N = 3 4.12581 4.13077 4.11242 MIN: 4.05 MIN: 4.04 MIN: 4.03 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K 1 2 3 0.2034 0.4068 0.6102 0.8136 1.017 0.900 0.904 0.902 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K 1 2 3 3 6 9 12 15 9.987 9.943 9.950 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 1.1.3 Model: mobilenet-v1-1.0 1 2 3 2 4 6 8 10 SE +/- 0.034, N = 3 6.133 6.106 6.128 MIN: 6.04 / MAX: 6.26 MIN: 6.05 / MAX: 6.81 MIN: 6.07 / MAX: 6.87 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
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K 1 2 3 0.5333 1.0666 1.5999 2.1332 2.6665 SE +/- 0.00, N = 3 2.37 2.36 2.36 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
PJSIP Method: OPTIONS, Stateful OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateful 1 2 3 1200 2400 3600 4800 6000 SE +/- 1.86, N = 3 5680 5666 5689 1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
GNU Radio Test: IIR Filter OpenBenchmarking.org MiB/s, More Is Better GNU Radio Test: IIR Filter 1 2 3 90 180 270 360 450 SE +/- 1.05, N = 3 411.9 413.5 413.1 1. 3.8.1.0
GNU Radio Test: FM Deemphasis Filter OpenBenchmarking.org MiB/s, More Is Better GNU Radio Test: FM Deemphasis Filter 1 2 3 110 220 330 440 550 SE +/- 1.66, N = 3 517.1 517.9 519.1 1. 3.8.1.0
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 1 2 3 0.7568 1.5136 2.2704 3.0272 3.784 SE +/- 0.00231, N = 3 3.35073 3.36355 3.35822 MIN: 3.26 MIN: 3.3 MIN: 3.28 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Fishy Cat - Compute: CPU-Only 1 2 3 60 120 180 240 300 SE +/- 0.20, N = 3 267.61 268.21 268.63
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.7992 1.5984 2.3976 3.1968 3.996 SE +/- 0.01251, N = 3 3.55190 3.54188 3.53866 MIN: 3.52 MIN: 3.52 MIN: 3.52 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
ViennaCL Test: CPU BLAS - sCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sCOPY 1 2 3 6 12 18 24 30 SE +/- 0.00, N = 3 27.5 27.4 27.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
PJSIP Method: OPTIONS, Stateless OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateless 1 2 3 8K 16K 24K 32K 40K SE +/- 173.85, N = 3 36173 36043 36169 1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon Obj 1 2 3 3 6 9 12 15 SE +/- 0.02, N = 3 10.59 10.55 10.55 MIN: 10.51 / MAX: 10.72 MIN: 10.51 / MAX: 10.68 MIN: 10.51 / MAX: 10.69
libgav1 Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better libgav1 0.16.3 Video Input: Chimera 1080p 10-bit 1 2 3 7 14 21 28 35 SE +/- 0.03, N = 3 31.64 31.66 31.55 1. (CXX) g++ options: -O3 -lpthread -lrt
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.5507 1.1014 1.6521 2.2028 2.7535 SE +/- 0.00083, N = 3 2.44319 2.44752 2.43904 MIN: 2.41 MIN: 2.42 MIN: 2.41 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Timed Node.js Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 15.11 Time To Compile 1 2 3 130 260 390 520 650 SE +/- 0.41, N = 3 595.30 593.25 593.92
LuaRadio Test: Complex Phase OpenBenchmarking.org MiB/s, More Is Better LuaRadio 0.9.1 Test: Complex Phase 1 2 3 90 180 270 360 450 SE +/- 0.09, N = 3 417.4 417.3 418.7
Botan Test: CAST-256 OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: CAST-256 1 2 3 20 40 60 80 100 SE +/- 0.44, N = 3 109.39 109.74 109.75 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
GNU GMP GMPbench Total Time OpenBenchmarking.org GMPbench Score, More Is Better GNU GMP GMPbench 6.2.1 Total Time 1 2 3 800 1600 2400 3200 4000 3657.3 3645.3 3653.9 1. (CC) gcc options: -O3 -fomit-frame-pointer -lm
Botan Test: ChaCha20Poly1305 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: ChaCha20Poly1305 - Decrypt 1 2 3 120 240 360 480 600 SE +/- 0.35, N = 3 572.53 573.60 574.42 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Zstd Compression Compression Level: 19 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 19 - Decompression Speed 1 2 3 500 1000 1500 2000 2500 SE +/- 22.95, N = 3 2469.6 2477.7 2471.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
Liquid-DSP Threads: 8 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 8 - Buffer Length: 256 - Filter Length: 57 1 2 3 70M 140M 210M 280M 350M SE +/- 99554.56, N = 3 343806667 343480000 342790000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K 1 2 3 6 12 18 24 30 SE +/- 0.03, N = 3 27.51 27.45 27.43 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Botan Test: CAST-256 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: CAST-256 - Decrypt 1 2 3 20 40 60 80 100 SE +/- 0.40, N = 3 109.25 109.57 109.56 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: BMW27 - Compute: CPU-Only 1 2 3 40 80 120 160 200 SE +/- 0.36, N = 3 196.01 195.56 195.45
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU 1 2 3 1.2748 2.5496 3.8244 5.0992 6.374 SE +/- 0.01512, N = 3 5.66569 5.65039 5.66558 MIN: 5.57 MIN: 5.56 MIN: 5.58 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Botan Test: ChaCha20Poly1305 OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: ChaCha20Poly1305 1 2 3 130 260 390 520 650 SE +/- 0.68, N = 3 578.94 577.48 578.17 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
ViennaCL Test: CPU BLAS - sAXPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sAXPY 1 2 3 9 18 27 36 45 SE +/- 0.03, N = 3 41.3 41.2 41.3 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon 1 2 3 3 6 9 12 15 SE +/- 0.05, N = 3 13.11 13.14 13.10 MIN: 13 / MAX: 13.35 MIN: 13.08 / MAX: 13.27 MIN: 13.05 / MAX: 13.24
Botan Test: AES-256 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: AES-256 - Decrypt 1 2 3 700 1400 2100 2800 3500 SE +/- 5.86, N = 3 3103.88 3100.64 3096.62 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
ViennaCL Test: CPU BLAS - sDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sDOT 1 2 3 10 20 30 40 50 SE +/- 0.00, N = 3 43.7 43.7 43.7 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Basis Universal Settings: UASTC Level 2 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.13 Settings: UASTC Level 2 1 2 3 10 20 30 40 50 SE +/- 0.02, N = 3 43.77 43.68 43.69 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU 1 2 3 900 1800 2700 3600 4500 SE +/- 1.22, N = 3 4049.70 4053.76 4058.11 MIN: 4043.71 MIN: 4050.14 MIN: 4053.28 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Zstd Compression Compression Level: 8 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.9 Compression Level: 8 - Decompression Speed 1 2 3 600 1200 1800 2400 3000 SE +/- 5.69, N = 3 2688.1 2693.4 2693.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 3.3.2 Pfam Database Search 1 2 3 40 80 120 160 200 SE +/- 0.26, N = 3 187.83 187.92 187.56 1. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm -lmpi
SVT-HEVC Tuning: 1 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 1 - Input: Bosphorus 1080p 1 2 3 1.2893 2.5786 3.8679 5.1572 6.4465 SE +/- 0.00, N = 3 5.72 5.73 5.73 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
toyBrot Fractal Generator Implementation: C++ Threads OpenBenchmarking.org ms, Fewer Is Better toyBrot Fractal Generator 2020-11-18 Implementation: C++ Threads 1 2 3 14K 28K 42K 56K 70K SE +/- 26.27, N = 3 66253 66336 66224 1. (CXX) g++ options: -O3 -lpthread
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 12.0 Build System: Ninja 1 2 3 200 400 600 800 1000 SE +/- 0.12, N = 3 925.48 925.74 927.03
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU 1 2 3 2 4 6 8 10 SE +/- 0.01285, N = 3 8.39311 8.39791 8.40666 MIN: 8.34 MIN: 8.36 MIN: 8.37 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p 1 2 3 3 6 9 12 15 SE +/- 0.02, N = 3 12.39 12.41 12.41 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Botan Test: Twofish - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Twofish - Decrypt 1 2 3 60 120 180 240 300 SE +/- 0.19, N = 3 286.17 286.61 286.50 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Basis Universal Settings: UASTC Level 3 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.13 Settings: UASTC Level 3 1 2 3 20 40 60 80 100 SE +/- 0.01, N = 3 80.20 80.08 80.10 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
GNU Radio Test: Hilbert Transform OpenBenchmarking.org MiB/s, More Is Better GNU Radio Test: Hilbert Transform 1 2 3 60 120 180 240 300 SE +/- 3.05, N = 3 264.7 264.3 264.6 1. 3.8.1.0
toyBrot Fractal Generator Implementation: OpenMP OpenBenchmarking.org ms, Fewer Is Better toyBrot Fractal Generator 2020-11-18 Implementation: OpenMP 1 2 3 14K 28K 42K 56K 70K SE +/- 20.85, N = 3 66314 66391 66299 1. (CXX) g++ options: -O3 -lpthread
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 1 2 3 3 6 9 12 15 SE +/- 0.00, N = 3 12.70 12.69 12.69 MIN: 12.47 MIN: 12.57 MIN: 12.64 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Botan Test: Twofish OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Twofish 1 2 3 60 120 180 240 300 SE +/- 0.14, N = 3 288.18 288.54 288.54 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU 1 2 3 500 1000 1500 2000 2500 SE +/- 1.12, N = 3 2213.82 2213.73 2216.10 MIN: 2209.54 MIN: 2211.69 MIN: 2214.02 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
ASTC Encoder Preset: Exhaustive OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.4 Preset: Exhaustive 1 2 3 30 60 90 120 150 SE +/- 0.01, N = 3 141.76 141.91 141.78 1. (CXX) g++ options: -O3 -flto -pthread
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 1 2 3 500 1000 1500 2000 2500 SE +/- 1.49, N = 3 2215.41 2217.72 2216.51 MIN: 2210.71 MIN: 2215.61 MIN: 2214.59 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Botan Test: KASUMI OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: KASUMI 1 2 3 20 40 60 80 100 SE +/- 0.04, N = 3 76.00 75.92 75.94 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
SVT-HEVC Tuning: 10 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 1080p 1 2 3 40 80 120 160 200 SE +/- 0.32, N = 3 173.18 173.01 173.06 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
Botan Test: KASUMI - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: KASUMI - Decrypt 1 2 3 16 32 48 64 80 SE +/- 0.03, N = 3 71.95 71.89 71.95 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 1 2 3 900 1800 2700 3600 4500 SE +/- 5.40, N = 3 4053.66 4050.46 4051.36 MIN: 4038.97 MIN: 4046.74 MIN: 4047.46 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
ASTC Encoder Preset: Thorough OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.4 Preset: Thorough 1 2 3 5 10 15 20 25 SE +/- 0.01, N = 3 18.82 18.81 18.82 1. (CXX) g++ options: -O3 -flto -pthread
SVT-VP9 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p 1 2 3 20 40 60 80 100 SE +/- 0.18, N = 3 108.08 108.03 108.00 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Helsing Digit Range: 12 digit OpenBenchmarking.org Seconds, Fewer Is Better Helsing 1.0-beta Digit Range: 12 digit 1 2 3 2 4 6 8 10 SE +/- 0.004, N = 3 6.450 6.446 6.447 1. (CC) gcc options: -O2 -pthread
Botan Test: Blowfish - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Blowfish - Decrypt 1 2 3 70 140 210 280 350 SE +/- 0.00, N = 3 345.06 344.85 344.92 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 1 2 3 4 8 12 16 20 SE +/- 0.00, N = 3 13.77 13.77 13.77 MIN: 13.7 MIN: 13.7 MIN: 13.71 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Liquid-DSP Threads: 20 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 20 - Buffer Length: 256 - Filter Length: 57 1 2 3 100M 200M 300M 400M 500M SE +/- 79652.02, N = 3 452546667 452770000 452640000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Botan Test: Blowfish OpenBenchmarking.org MiB/s, More Is Better Botan 2.17.3 Test: Blowfish 1 2 3 80 160 240 320 400 SE +/- 0.06, N = 3 346.41 346.29 346.25 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
KTX-Software toktx Settings: UASTC 4 + Zstd Compression 19 OpenBenchmarking.org Seconds, Fewer Is Better KTX-Software toktx 4.0 Settings: UASTC 4 + Zstd Compression 19 1 2 3 100 200 300 400 500 SE +/- 0.05, N = 3 444.73 444.54 444.66
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 0.8.2 Video Input: Chimera 1080p 10-bit 1 2 3 20 40 60 80 100 SE +/- 0.01, N = 3 88.68 88.67 88.65 MIN: 56.07 / MAX: 221.22 MIN: 56.08 / MAX: 223.41 MIN: 56.17 / MAX: 216.38 1. (CC) gcc options: -pthread -lm
PJSIP Method: INVITE OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: INVITE 1 2 3 700 1400 2100 2800 3500 SE +/- 3.71, N = 3 3321 3320 3320 1. (CC) gcc options: -lSDL2 -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
SecureMark Benchmark: SecureMark-TLS OpenBenchmarking.org marks, More Is Better SecureMark 1.0.4 Benchmark: SecureMark-TLS 1 2 3 40K 80K 120K 160K 200K SE +/- 56.19, N = 3 191107 191158 191151 1. (CC) gcc options: -pedantic -O3
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU 1 2 3 900 1800 2700 3600 4500 SE +/- 2.36, N = 3 4051.75 4052.37 4051.76 MIN: 4043.32 MIN: 4048.89 MIN: 4048.8 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 1 - Buffer Length: 256 - Filter Length: 57 1 2 3 11M 22M 33M 44M 55M SE +/- 3179.80, N = 3 49054667 49056000 49052000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Sysbench Test: CPU OpenBenchmarking.org Events Per Second, More Is Better Sysbench 1.0.20 Test: CPU 1 2 3 3K 6K 9K 12K 15K SE +/- 0.16, N = 3 15805.17 15806.12 15805.70 1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
ViennaCL Test: CPU BLAS - dCOPY OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dCOPY 2 3 6 12 18 24 30 27.5 27.5 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K 1 2 3 0.0203 0.0406 0.0609 0.0812 0.1015 SE +/- 0.00, N = 12 0.09 0.09 0.09 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Chia Blockchain VDF Test: Square Plain C++ OpenBenchmarking.org IPS, More Is Better Chia Blockchain VDF 1.0.1 Test: Square Plain C++ 1 2 3 20K 40K 60K 80K 100K 108100 108100 108100 1. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
ViennaCL Test: CPU BLAS - dDOT OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dDOT 1 2 3 10 20 30 40 50 SE +/- 4.97, N = 3 38.8 43.8 43.8 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
Phoronix Test Suite v10.8.4