EPYC 7502 AOCC 2.3 Compiler Comparison AMD EPYC 7502 testing of various benchmarks under AMD AOCC 2.3, GCC 10.2, LLVM Clang 11. CFLAGS/CXXFLAGS of "-O3 -march=znver2" throughout. Benchmarks by Michael Larabel for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2012080-HA-EPYC7502A97&sgm=1&swl=1&grw&sro&rro .
EPYC 7502 AOCC 2.3 Compiler Comparison Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver Compiler File-System Screen Resolution GCC 10.2 LLVM Clang 11 AMD AOCC 2.3 AMD EPYC 7502 32-Core @ 2.50GHz (32 Cores / 64 Threads) ASRockRack EPYCD8 (P2.10 BIOS) AMD Starship/Matisse 126GB 280GB INTEL SSDPED1D280GA ASPEED AMD Starship/Matisse VE228 2 x Intel I350 Ubuntu 20.10 5.8.0-31-generic (x86_64) GNOME Shell 3.38.1 X Server 1.20.9 modesetting 1.20.9 GCC 10.2.0 ext4 1920x1080 Clang 11.0.0-2Target: Clang 11.0.0 OpenBenchmarking.org Environment Details - CXXFLAGS="-O3 -march=znver2" CFLAGS="-O3 -march=znver2" Compiler Details - GCC 10.2: --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 - AMD AOCC 2.3: Optimized build with assertions; Default target: x86_64-unknown-linux-gnu; Host CPU: znver2 Processor Details - Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x830101c 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 IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
EPYC 7502 AOCC 2.3 Compiler Comparison tscp: AI Chess Performance cryptopp: Unkeyed Algorithms scimark2: Composite compress-lz4: 1 - Compression Speed compress-lz4: 3 - Compression Speed compress-lz4: 9 - Compression Speed hint: FLOAT basis: UASTC Level 2 basis: UASTC Level 3 basis: UASTC Level 2 + RDO Post-Processing encode-mp3: WAV To MP3 tjbench: Decompression Throughput astcenc: Medium astcenc: Thorough astcenc: Exhaustive libraw: Post-Processing Benchmark webp: Quality 100, Lossless webp: Quality 100, Highest Compression webp: Quality 100, Lossless, Highest Compression daphne: OpenMP - NDT Mapping daphne: OpenMP - Points2Image daphne: OpenMP - Euclidean Cluster mrbayes: Primate Phylogeny Analysis rnnoise: tnn: CPU - MobileNet v2 tnn: CPU - SqueezeNet v1.1 ncnn: CPU - squeezenet ncnn: CPU - mobilenet ncnn: CPU-v2-v2 - mobilenet-v2 ncnn: CPU-v3-v3 - mobilenet-v3 ncnn: CPU - shufflenet-v2 ncnn: CPU - mnasnet ncnn: CPU - efficientnet-b0 ncnn: CPU - blazeface ncnn: CPU - googlenet ncnn: CPU - vgg16 ncnn: CPU - resnet18 ncnn: CPU - alexnet ncnn: CPU - resnet50 ncnn: CPU - yolov4-tiny onednn: IP Batch 1D - f32 - CPU onednn: IP Batch 1D - u8s8f32 - CPU onednn: IP Batch All - u8s8f32 - CPU onednn: Deconvolution Batch deconv_1d - f32 - CPU onednn: Deconvolution Batch deconv_3d - f32 - CPU onednn: Deconvolution Batch deconv_1d - u8s8f32 - CPU onednn: Deconvolution Batch deconv_3d - u8s8f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU stockfish: Total Time compress-zstd: 3 compress-zstd: 19 aobench: 2048 x 2048 - Total Time vpxenc: Speed 0 vpxenc: Speed 5 graphics-magick: Swirl graphics-magick: Rotate graphics-magick: Sharpen graphics-magick: Enhanced graphics-magick: Resizing svt-vp9: VMAF Optimized - Bosphorus 1080p svt-vp9: PSNR/SSIM Optimized - Bosphorus 1080p svt-vp9: Visual Quality Optimized - Bosphorus 1080p x264: H.264 Video Encoding dav1d: Chimera 1080p dav1d: Summer Nature 4K dav1d: Summer Nature 1080p dav1d: Chimera 1080p 10-bit svt-av1: Enc Mode 0 - 1080p svt-av1: Enc Mode 4 - 1080p svt-av1: Enc Mode 8 - 1080p x265: Bosphorus 4K x265: Bosphorus 1080p c-ray: Total Time - 4K, 16 Rays Per Pixel nginx: Static Web Page Serving openssl: RSA 4096-bit Performance redis: LPUSH redis: GET redis: SET sqlite-speedtest: Timed Time - Size 1,000 pgbench: 1 - 1 - Read Only pgbench: 1 - 1 - Read Only - Average Latency pgbench: 1 - 1 - Read Write pgbench: 1 - 1 - Read Write - Average Latency pgbench: 1 - 50 - Read Only pgbench: 1 - 50 - Read Only - Average Latency pgbench: 1 - 50 - Read Write pgbench: 1 - 50 - Read Write - Average Latency GCC 10.2 LLVM Clang 11 AMD AOCC 2.3 1007283 305.900410 2759.35 9459.69 45.57 44.72 291925951.64210 16.507 25.522 755.517 8.798 171.888797 7.00 9.61 73.46 52.54 20.799 8.861 44.383 874.54 18452.623694826 890.71 94.284 21.723 324.172 305.134 17.29 19.46 9.37 8.88 10.59 8.95 11.32 3.99 20.85 30.89 13.12 9.46 23.70 29.47 1.67503 1.17556 13.8793 1.95695 3.68073 2.17957 1.94380 254.610 79.2495 0.532944 58347386 7849.6 109.9 35.783 6.27 19.03 1295 535 434 590 2092 354.98 348.24 279.73 149.35 564.96 269.72 567.42 143.02 0.103 6.775 55.713 22.83 49.05 18.922 30658.67 7395.4 1212068.06 1809976.83 1369757.13 75.140 28919 0.035 3739 0.268 521332 0.096 3453 14.488 1143642 314.389914 2673.79 9838.54 48.40 44.30 292874904.53733 16.485 25.321 837.541 10.078 174.991039 6.05 8.32 67.17 36.98 20.719 7.655 42.377 945.32 11946.566719313 674.01 93.841 21.540 392.204 311.682 15.35 17.85 6.88 6.87 7.35 6.26 8.77 2.84 18.88 36.60 13.19 10.83 22.23 30.70 1.72247 1.03137 13.4649 1.69390 3.72354 2.05539 1.98927 172.513 64.9912 0.575864 62434784 7866.2 111.2 41.655 5.96 18.23 1323 527 384 627 1827 363.72 365.57 286.24 146.43 572.19 275.41 584.65 92.56 0.145 8.591 70.229 22.44 50.02 30.646 30676.68 5412.5 1304842.23 2122749.43 1483322.31 77.642 28921 0.035 3772 0.265 530684 0.094 3443 14.528 1138442 312.637840 2779.62 9780.39 48.52 45.76 294314450.61706 16.315 25.210 833.536 11.026 176.873794 5.81 8.16 66.04 38.47 20.904 7.546 42.704 923.68 13720.155459109 678.51 90.759 20.756 365.390 304.178 14.38 17.02 5.62 5.20 6.25 5.01 6.94 2.19 16.51 32.03 11.99 8.94 19.70 29.38 1.40879 1.00926 11.9233 1.66772 3.66096 1.99141 1.87485 147.523 30.7182 0.464235 62375605 7937.8 114.6 37.758 6.65 20.12 1368 525 316 526 1653 366.61 368.17 291.22 151.92 575.22 274.20 588.62 122.39 0.146 8.657 70.698 23.69 50.60 33.275 31368.86 5413.5 1380024.29 1874175.91 1446872.35 78.067 29645 0.034 3865 0.259 541314 0.092 3413 14.655 OpenBenchmarking.org
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 200K 400K 600K 800K 1000K SE +/- 582.00, N = 5 SE +/- 1467.20, N = 5 SE +/- 471.20, N = 5 1143642 1007283 1138442 1. (CC) gcc options: -O3 -march=znver2 -march=native
Crypto++ Test: Unkeyed Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 70 140 210 280 350 SE +/- 0.20, N = 3 SE +/- 0.09, N = 3 SE +/- 0.16, N = 3 314.39 305.90 312.64 1. (CXX) g++ options: -O3 -march=znver2 -fPIC -pthread -pipe
SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 600 1200 1800 2400 3000 SE +/- 7.04, N = 3 SE +/- 14.97, N = 3 SE +/- 43.99, N = 3 2673.79 2759.35 2779.62 1. (CC) gcc options: -O3 -march=znver2 -lm
LZ4 Compression Compression Level: 1 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2K 4K 6K 8K 10K SE +/- 55.02, N = 3 SE +/- 56.41, N = 3 SE +/- 45.16, N = 3 9838.54 9459.69 9780.39 1. (CC) gcc options: -O3
LZ4 Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 11 22 33 44 55 SE +/- 0.04, N = 3 SE +/- 0.66, N = 3 SE +/- 0.00, N = 3 48.40 45.57 48.52 1. (CC) gcc options: -O3
LZ4 Compression Compression Level: 9 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 10 20 30 40 50 SE +/- 0.02, N = 3 SE +/- 0.60, N = 3 SE +/- 0.03, N = 3 44.30 44.72 45.76 1. (CC) gcc options: -O3
Hierarchical INTegration Test: FLOAT OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 60M 120M 180M 240M 300M SE +/- 170353.62, N = 3 SE +/- 15707.07, N = 3 SE +/- 30193.74, N = 3 292874904.54 291925951.64 294314450.62 1. (CC) gcc options: -O3 -march=znver2 -march=native -lm
Basis Universal Settings: UASTC Level 2 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 16.49 16.51 16.32 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Basis Universal Settings: UASTC Level 3 OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 3 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 25.32 25.52 25.21 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
Basis Universal Settings: UASTC Level 2 + RDO Post-Processing OpenBenchmarking.org Seconds, Fewer Is Better Basis Universal 1.12 Settings: UASTC Level 2 + RDO Post-Processing LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 200 400 600 800 1000 SE +/- 0.24, N = 3 SE +/- 0.14, N = 3 SE +/- 0.04, N = 3 837.54 755.52 833.54 1. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.003, N = 3 SE +/- 0.004, N = 3 SE +/- 0.004, N = 3 10.078 8.798 11.026 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr 1. (CC) gcc options: -O3 -pipe -march=znver2 -lncurses -lm
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 40 80 120 160 200 SE +/- 0.21, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 174.99 171.89 176.87 1. (CC) gcc options: -O3 -march=znver2 -rdynamic
ASTC Encoder Preset: Medium OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Medium LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 6.05 7.00 5.81 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
ASTC Encoder Preset: Thorough OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Thorough LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 8.32 9.61 8.16 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
ASTC Encoder Preset: Exhaustive OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 2.0 Preset: Exhaustive LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 16 32 48 64 80 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 67.17 73.46 66.04 1. (CXX) g++ options: -std=c++14 -fvisibility=hidden -O3 -flto -mfpmath=sse -mavx2 -mpopcnt -lpthread
LibRaw Post-Processing Benchmark OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 12 24 36 48 60 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 SE +/- 0.10, N = 3 36.98 52.54 38.47 1. (CXX) g++ options: -O3 -march=znver2 -fopenmp -ljpeg -lz -lm
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 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 20.72 20.80 20.90 1. (CC) gcc options: -fvisibility=hidden -O3 -march=znver2 -pthread -lm -ljpeg
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 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2 4 6 8 10 SE +/- 0.008, N = 3 SE +/- 0.003, N = 3 SE +/- 0.008, N = 3 7.655 8.861 7.546 1. (CC) gcc options: -fvisibility=hidden -O3 -march=znver2 -pthread -lm -ljpeg
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 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 10 20 30 40 50 SE +/- 0.12, N = 3 SE +/- 0.12, N = 3 SE +/- 0.01, N = 3 42.38 44.38 42.70 1. (CC) gcc options: -fvisibility=hidden -O3 -march=znver2 -pthread -lm -ljpeg
Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: NDT Mapping LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 200 400 600 800 1000 SE +/- 2.29, N = 3 SE +/- 2.97, N = 3 SE +/- 1.07, N = 3 945.32 874.54 923.68 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Points2Image LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 4K 8K 12K 16K 20K SE +/- 96.60, N = 3 SE +/- 140.90, N = 3 SE +/- 163.63, N = 6 11946.57 18452.62 13720.16 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster OpenBenchmarking.org Test Cases Per Minute, More Is Better Darmstadt Automotive Parallel Heterogeneous Suite Backend: OpenMP - Kernel: Euclidean Cluster LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 200 400 600 800 1000 SE +/- 2.53, N = 3 SE +/- 0.77, N = 3 SE +/- 0.47, N = 3 674.01 890.71 678.51 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp
Timed MrBayes Analysis Primate Phylogeny Analysis OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 20 40 60 80 100 SE +/- 0.04, N = 3 SE +/- 0.17, N = 3 SE +/- 0.04, N = 3 93.84 94.28 90.76 -mabm 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -O3 -std=c99 -pedantic -march=znver2 -lm
RNNoise OpenBenchmarking.org Seconds, Fewer Is Better RNNoise 2020-06-28 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 21.54 21.72 20.76 1. (CC) gcc options: -O3 -march=znver2 -pedantic -fvisibility=hidden
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 90 180 270 360 450 SE +/- 0.45, N = 3 SE +/- 0.62, N = 3 SE +/- 0.24, N = 3 392.20 324.17 365.39 -fopenmp=libomp - MIN: 390.71 / MAX: 393.87 -fopenmp - MIN: 311.9 / MAX: 354.48 -fopenmp=libomp - MIN: 364.57 / MAX: 366.34 1. (CXX) g++ options: -O3 -march=znver2 -pthread -fvisibility=hidden -rdynamic -ldl
TNN Target: CPU - Model: SqueezeNet v1.1 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: SqueezeNet v1.1 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 70 140 210 280 350 SE +/- 1.92, N = 3 SE +/- 0.20, N = 3 SE +/- 0.66, N = 3 311.68 305.13 304.18 -fopenmp=libomp - MIN: 306.99 / MAX: 314.32 -fopenmp - MIN: 304.36 / MAX: 306.06 -fopenmp=libomp - MIN: 302.78 / MAX: 315.91 1. (CXX) g++ options: -O3 -march=znver2 -pthread -fvisibility=hidden -rdynamic -ldl
NCNN Target: CPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: squeezenet LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 4 8 12 16 20 SE +/- 0.14, N = 15 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 15.35 17.29 14.38 -lomp - MIN: 14.29 / MAX: 18.88 -lgomp - MIN: 16.89 / MAX: 19.32 -lomp - MIN: 13.96 / MAX: 16.87 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenet LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 5 10 15 20 25 SE +/- 0.13, N = 15 SE +/- 0.12, N = 3 SE +/- 0.34, N = 3 17.85 19.46 17.02 -lomp - MIN: 16.89 / MAX: 21.04 -lgomp - MIN: 18.87 / MAX: 31.76 -lomp - MIN: 16.39 / MAX: 20.11 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU-v2-v2 - Model: mobilenet-v2 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.06, N = 15 SE +/- 0.24, N = 3 SE +/- 0.08, N = 3 6.88 9.37 5.62 -lomp - MIN: 6.35 / MAX: 16.49 -lgomp - MIN: 8.62 / MAX: 11 -lomp - MIN: 5.35 / MAX: 7.49 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU-v3-v3 - Model: mobilenet-v3 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2 4 6 8 10 SE +/- 0.06, N = 15 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 6.87 8.88 5.20 -lomp - MIN: 6.16 / MAX: 8.64 -lgomp - MIN: 8.58 / MAX: 10.83 -lomp - MIN: 5.05 / MAX: 7.61 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: shufflenet-v2 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.02, N = 15 SE +/- 0.68, N = 3 SE +/- 0.12, N = 3 7.35 10.59 6.25 -lomp - MIN: 7.09 / MAX: 11.01 -lgomp - MIN: 9.42 / MAX: 13.72 -lomp - MIN: 5.96 / MAX: 6.53 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mnasnet LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.09, N = 15 SE +/- 0.29, N = 3 SE +/- 0.03, N = 3 6.26 8.95 5.01 -lomp - MIN: 5.63 / MAX: 8.5 -lgomp - MIN: 8.26 / MAX: 11.03 -lomp - MIN: 4.87 / MAX: 5.39 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: efficientnet-b0 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.09, N = 15 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 8.77 11.32 6.94 -lomp - MIN: 7.99 / MAX: 13.6 -lgomp - MIN: 11.03 / MAX: 13.16 -lomp - MIN: 6.72 / MAX: 9 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: blazeface LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.8978 1.7956 2.6934 3.5912 4.489 SE +/- 0.02, N = 15 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 2.84 3.99 2.19 -lomp - MIN: 2.67 / MAX: 4.67 -lgomp - MIN: 3.77 / MAX: 4.73 -lomp - MIN: 2.1 / MAX: 2.43 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: googlenet LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 5 10 15 20 25 SE +/- 0.27, N = 15 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 18.88 20.85 16.51 -lomp - MIN: 16.91 / MAX: 23.31 -lgomp - MIN: 19.8 / MAX: 22.84 -lomp - MIN: 16.26 / MAX: 18.75 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: vgg16 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 8 16 24 32 40 SE +/- 0.42, N = 15 SE +/- 0.30, N = 3 SE +/- 0.31, N = 3 36.60 30.89 32.03 -lomp - MIN: 32.93 / MAX: 48.08 -lgomp - MIN: 30.04 / MAX: 50.21 -lomp - MIN: 30.86 / MAX: 35.2 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet18 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.12, N = 15 SE +/- 0.09, N = 3 SE +/- 0.15, N = 3 13.19 13.12 11.99 -lomp - MIN: 12.17 / MAX: 23.53 -lgomp - MIN: 12.79 / MAX: 15.11 -lomp - MIN: 11.71 / MAX: 14.27 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: alexnet LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 3 6 9 12 15 SE +/- 0.18, N = 15 SE +/- 0.14, N = 3 SE +/- 0.01, N = 3 10.83 9.46 8.94 -lomp - MIN: 9.15 / MAX: 60.4 -lgomp - MIN: 9.17 / MAX: 11.49 -lomp - MIN: 8.81 / MAX: 13.49 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet50 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 6 12 18 24 30 SE +/- 0.20, N = 15 SE +/- 0.16, N = 3 SE +/- 0.18, N = 3 22.23 23.70 19.70 -lomp - MIN: 20.63 / MAX: 31.79 -lgomp - MIN: 23.21 / MAX: 25.68 -lomp - MIN: 19.16 / MAX: 22.41 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: yolov4-tiny LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 7 14 21 28 35 SE +/- 0.19, N = 15 SE +/- 0.12, N = 3 SE +/- 0.15, N = 3 30.70 29.47 29.38 -lomp - MIN: 29.08 / MAX: 40.6 -lgomp - MIN: 28.89 / MAX: 31.49 -lomp - MIN: 28.77 / MAX: 31.68 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread
oneDNN Harness: IP Batch 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.3876 0.7752 1.1628 1.5504 1.938 SE +/- 0.00256, N = 3 SE +/- 0.00519, N = 3 SE +/- 0.00432, N = 3 1.72247 1.67503 1.40879 -fopenmp=libomp - MIN: 1.62 -fopenmp - MIN: 1.59 -fopenmp=libomp - MIN: 1.36 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch 1D - Data Type: u8s8f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.2645 0.529 0.7935 1.058 1.3225 SE +/- 0.00137, N = 3 SE +/- 0.00350, N = 3 SE +/- 0.00275, N = 3 1.03137 1.17556 1.00926 -fopenmp=libomp - MIN: 0.97 -fopenmp - MIN: 1.13 -fopenmp=libomp - MIN: 0.95 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: IP Batch All - Data Type: u8s8f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 13.46 13.88 11.92 -fopenmp=libomp - MIN: 13.14 -fopenmp - MIN: 13.35 -fopenmp=libomp - MIN: 11.65 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.4403 0.8806 1.3209 1.7612 2.2015 SE +/- 0.01023, N = 3 SE +/- 0.01336, N = 3 SE +/- 0.00186, N = 3 1.69390 1.95695 1.66772 -fopenmp=libomp - MIN: 1.63 -fopenmp - MIN: 1.87 -fopenmp=libomp - MIN: 1.61 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.8378 1.6756 2.5134 3.3512 4.189 SE +/- 0.00775, N = 3 SE +/- 0.01624, N = 3 SE +/- 0.01101, N = 3 3.72354 3.68073 3.66096 -fopenmp=libomp - MIN: 3.57 -fopenmp - MIN: 3.53 -fopenmp=libomp - MIN: 3.49 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.4904 0.9808 1.4712 1.9616 2.452 SE +/- 0.00233, N = 3 SE +/- 0.00145, N = 3 SE +/- 0.00112, N = 3 2.05539 2.17957 1.99141 -fopenmp=libomp - MIN: 1.96 -fopenmp - MIN: 2.05 -fopenmp=libomp - MIN: 1.92 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.4476 0.8952 1.3428 1.7904 2.238 SE +/- 0.00691, N = 3 SE +/- 0.00392, N = 3 SE +/- 0.00189, N = 3 1.98927 1.94380 1.87485 -fopenmp=libomp - MIN: 1.9 -fopenmp - MIN: 1.82 -fopenmp=libomp - MIN: 1.82 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 60 120 180 240 300 SE +/- 1.10, N = 3 SE +/- 0.73, N = 3 SE +/- 0.06, N = 3 172.51 254.61 147.52 -fopenmp=libomp - MIN: 169.54 -fopenmp - MIN: 251.67 -fopenmp=libomp - MIN: 146.31 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 20 40 60 80 100 SE +/- 2.37, N = 15 SE +/- 0.80, N = 3 SE +/- 0.32, N = 3 64.99 79.25 30.72 -fopenmp=libomp - MIN: 50.95 -fopenmp - MIN: 77.35 -fopenmp=libomp - MIN: 29.35 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 1.5 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.1296 0.2592 0.3888 0.5184 0.648 SE +/- 0.001767, N = 3 SE +/- 0.001691, N = 3 SE +/- 0.000544, N = 3 0.575864 0.532944 0.464235 -fopenmp=libomp - MIN: 0.55 -fopenmp - MIN: 0.51 -fopenmp=libomp - MIN: 0.45 1. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -pie -lpthread -ldl
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 12 Total Time LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 13M 26M 39M 52M 65M SE +/- 877956.02, N = 4 SE +/- 872585.44, N = 3 SE +/- 379890.17, N = 3 62434784 58347386 62375605 -flto=thin -flto -flto=jobserver -flto=thin 1. (CXX) g++ options: -m64 -lpthread -O3 -march=znver2 -fno-exceptions -std=c++17 -pedantic -msse -msse3 -mpopcnt -msse4.1 -mssse3 -msse2
Zstd Compression Compression Level: 3 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 3 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2K 4K 6K 8K 10K SE +/- 30.93, N = 3 SE +/- 6.11, N = 3 SE +/- 3.73, N = 3 7866.2 7849.6 7937.8 1. (CC) gcc options: -O3 -march=znver2 -pthread -lz
Zstd Compression Compression Level: 19 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 30 60 90 120 150 SE +/- 0.30, N = 3 SE +/- 0.23, N = 3 SE +/- 0.26, N = 3 111.2 109.9 114.6 1. (CC) gcc options: -O3 -march=znver2 -pthread -lz
AOBench Size: 2048 x 2048 - Total Time OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 10 20 30 40 50 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 41.66 35.78 37.76 1. (CC) gcc options: -lm -O3 -march=znver2
VP9 libvpx Encoding Speed: Speed 0 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.2 Speed: Speed 0 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2 4 6 8 10 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 5.96 6.27 6.65 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=znver2 -fPIC -U_FORTIFY_SOURCE -std=c++11
VP9 libvpx Encoding Speed: Speed 5 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.2 Speed: Speed 5 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 5 10 15 20 25 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 18.23 19.03 20.12 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=znver2 -fPIC -U_FORTIFY_SOURCE -std=c++11
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Swirl LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 300 600 900 1200 1500 SE +/- 2.40, N = 3 SE +/- 1.00, N = 3 SE +/- 4.26, N = 3 1323 1295 1368 1. (CC) gcc options: -fopenmp -O3 -march=znver2 -pthread -ljpeg -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Rotate LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 120 240 360 480 600 SE +/- 5.24, N = 3 SE +/- 1.00, N = 3 527 535 525 1. (CC) gcc options: -fopenmp -O3 -march=znver2 -pthread -ljpeg -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Sharpen LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 90 180 270 360 450 384 434 316 1. (CC) gcc options: -fopenmp -O3 -march=znver2 -pthread -ljpeg -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Enhanced LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 140 280 420 560 700 627 590 526 1. (CC) gcc options: -fopenmp -O3 -march=znver2 -pthread -ljpeg -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Resizing LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 400 800 1200 1600 2000 SE +/- 10.17, N = 3 SE +/- 17.89, N = 3 SE +/- 22.27, N = 3 1827 2092 1653 1. (CC) gcc options: -fopenmp -O3 -march=znver2 -pthread -ljpeg -lX11 -lz -lm -lpthread
SVT-VP9 Tuning: VMAF Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: VMAF Optimized - Input: Bosphorus 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 80 160 240 320 400 SE +/- 1.70, N = 3 SE +/- 2.15, N = 3 SE +/- 1.39, N = 3 363.72 354.98 366.61 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
SVT-VP9 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 80 160 240 320 400 SE +/- 1.74, N = 3 SE +/- 1.42, N = 3 SE +/- 0.54, N = 3 365.57 348.24 368.17 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
SVT-VP9 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 60 120 180 240 300 SE +/- 1.68, N = 3 SE +/- 1.18, N = 3 SE +/- 0.90, N = 3 286.24 279.73 291.22 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2019-12-17 H.264 Video Encoding LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 30 60 90 120 150 SE +/- 0.59, N = 3 SE +/- 1.24, N = 3 SE +/- 0.67, N = 3 146.43 149.35 151.92 -mstack-alignment=64 -mstack-alignment=64 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -march=znver2 -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.7.0 Video Input: Chimera 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 120 240 360 480 600 SE +/- 0.97, N = 3 SE +/- 1.05, N = 3 SE +/- 0.49, N = 3 572.19 564.96 575.22 MIN: 404.8 / MAX: 726.7 MIN: 399.64 / MAX: 724.27 MIN: 414.12 / MAX: 729.95 1. (CC) gcc options: -O3 -march=znver2 -pthread
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.7.0 Video Input: Summer Nature 4K LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 60 120 180 240 300 SE +/- 1.08, N = 3 SE +/- 0.57, N = 3 SE +/- 1.22, N = 3 275.41 269.72 274.20 MIN: 155.99 / MAX: 295.35 MIN: 160.12 / MAX: 288.9 MIN: 151.98 / MAX: 293.44 1. (CC) gcc options: -O3 -march=znver2 -pthread
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.7.0 Video Input: Summer Nature 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 130 260 390 520 650 SE +/- 0.95, N = 3 SE +/- 0.36, N = 3 SE +/- 2.09, N = 3 584.65 567.42 588.62 MIN: 337.56 / MAX: 641.35 MIN: 337.19 / MAX: 625.71 MIN: 345.64 / MAX: 651.08 1. (CC) gcc options: -O3 -march=znver2 -pthread
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 0.7.0 Video Input: Chimera 1080p 10-bit LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 30 60 90 120 150 SE +/- 0.26, N = 3 SE +/- 0.18, N = 3 SE +/- 0.34, N = 3 92.56 143.02 122.39 MIN: 61.05 / MAX: 158.66 MIN: 98.76 / MAX: 246.17 MIN: 85.78 / MAX: 202.39 1. (CC) gcc options: -O3 -march=znver2 -pthread
SVT-AV1 Encoder Mode: Enc Mode 0 - Input: 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 0 - Input: 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.0329 0.0658 0.0987 0.1316 0.1645 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.145 0.103 0.146 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
SVT-AV1 Encoder Mode: Enc Mode 4 - Input: 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 4 - Input: 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 2 4 6 8 10 SE +/- 0.026, N = 3 SE +/- 0.025, N = 3 SE +/- 0.042, N = 3 8.591 6.775 8.657 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
SVT-AV1 Encoder Mode: Enc Mode 8 - Input: 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 16 32 48 64 80 SE +/- 0.30, N = 3 SE +/- 0.29, N = 3 SE +/- 0.46, N = 3 70.23 55.71 70.70 1. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie
x265 Video Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 4K LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 22.44 22.83 23.69 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread -lrt -ldl -lnuma
x265 Video Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better x265 3.4 Video Input: Bosphorus 1080p LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 11 22 33 44 55 SE +/- 0.09, N = 3 SE +/- 0.13, N = 3 SE +/- 0.14, N = 3 50.02 49.05 50.60 1. (CXX) g++ options: -O3 -march=znver2 -rdynamic -lpthread -lrt -ldl -lnuma
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 8 16 24 32 40 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 30.65 18.92 33.28 1. (CC) gcc options: -lm -lpthread -O3 -march=znver2
NGINX Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 7K 14K 21K 28K 35K SE +/- 159.74, N = 3 SE +/- 381.73, N = 4 SE +/- 254.59, N = 15 30676.68 30658.67 31368.86 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native -march=znver2
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 1600 3200 4800 6400 8000 SE +/- 1.37, N = 3 SE +/- 0.73, N = 3 SE +/- 0.64, N = 3 5412.5 7395.4 5413.5 -Qunused-arguments -Qunused-arguments 1. (CC) gcc options: -pthread -m64 -O3 -march=znver2 -lssl -lcrypto -ldl
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: LPUSH LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 300K 600K 900K 1200K 1500K SE +/- 22719.73, N = 15 SE +/- 21030.89, N = 15 SE +/- 18763.42, N = 3 1304842.23 1212068.06 1380024.29 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=znver2
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: GET LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 500K 1000K 1500K 2000K 2500K SE +/- 49885.99, N = 15 SE +/- 18838.90, N = 3 SE +/- 30693.11, N = 15 2122749.43 1809976.83 1874175.91 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=znver2
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 6.0.9 Test: SET LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 300K 600K 900K 1200K 1500K SE +/- 22282.05, N = 15 SE +/- 24810.19, N = 15 SE +/- 27170.26, N = 15 1483322.31 1369757.13 1446872.35 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=znver2
SQLite Speedtest Timed Time - Size 1,000 OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 20 40 60 80 100 SE +/- 0.18, N = 3 SE +/- 0.13, N = 3 SE +/- 0.09, N = 3 77.64 75.14 78.07 1. (CC) gcc options: -O3 -march=znver2 -ldl -lz -lpthread
PostgreSQL pgbench Scaling Factor: 1 - Clients: 1 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Only LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 6K 12K 18K 24K 30K SE +/- 252.07, N = 3 SE +/- 137.51, N = 3 SE +/- 73.04, N = 3 28921 28919 29645 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average Latency LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.0079 0.0158 0.0237 0.0316 0.0395 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.035 0.035 0.034 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 1 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Write LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 800 1600 2400 3200 4000 SE +/- 3.59, N = 3 SE +/- 46.08, N = 5 SE +/- 18.01, N = 3 3772 3739 3865 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average Latency LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.0603 0.1206 0.1809 0.2412 0.3015 SE +/- 0.000, N = 3 SE +/- 0.003, N = 5 SE +/- 0.001, N = 3 0.265 0.268 0.259 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 50 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Only LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 120K 240K 360K 480K 600K SE +/- 4278.21, N = 3 SE +/- 4440.25, N = 3 SE +/- 798.75, N = 3 530684 521332 541314 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 0.0216 0.0432 0.0648 0.0864 0.108 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 0.094 0.096 0.092 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 50 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Write LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 700 1400 2100 2800 3500 SE +/- 4.43, N = 3 SE +/- 5.86, N = 3 SE +/- 5.71, N = 3 3443 3453 3413 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
PostgreSQL pgbench Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 14.53 14.49 14.66 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=znver2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Geometric Mean Of All Test Results Result Composite - EPYC 7502 AOCC 2.3 Compiler Comparison OpenBenchmarking.org Geometric Mean, More Is Better Geometric Mean Of All Test Results Result Composite - EPYC 7502 AOCC 2.3 Compiler Comparison LLVM Clang 11 GCC 10.2 AMD AOCC 2.3 30 60 90 120 150 115.39 113.07 121.37
Number Of First Place Finishes Wins - 89 Tests LLVM Clang 11 11 [12.4%] GCC 10.2 17 [19.1%] AMD AOCC 2.3 61 [68.5%] Number Of First Place Finishes Wins - 89 Tests OpenBenchmarking.org
Number Of Last Place Finishes Losses - 89 Tests LLVM Clang 11 23 [25.8%] GCC 10.2 56 [62.9%] AMD AOCC 2.3 10 [11.2%] Number Of Last Place Finishes Losses - 89 Tests OpenBenchmarking.org
Phoronix Test Suite v10.8.5