3900xt-november AMD Ryzen 9 3900XT 12-Core testing with a MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS) and AMD Radeon RX 56/64 8GB on Ubuntu 22.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2211180-SYST-3900XTN38&sro&grs .
3900xt-november Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution a aa b AMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads) MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.B3 BIOS) AMD Starship/Matisse 16GB 500GB Seagate FireCuda 520 SSD ZP500GM30002 AMD Radeon RX 56/64 8GB (1630/945MHz) AMD Vega 10 HDMI Audio ASUS MG28U Realtek Device 2600 + Realtek Killer E3000 2.5GbE + Intel Wi-Fi 6 AX200 Ubuntu 22.04 5.15.0-47-generic (x86_64) GNOME Shell 42.2 X Server + Wayland 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.42) 1.3.204 GCC 11.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701021 Graphics Details - BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D0500100-102 Python Details - Python 3.10.4 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
3900xt-november stress-ng: IO_uring spacy: en_core_web_trf stress-ng: Glibc C String Functions cpuminer-opt: Garlicoin stress-ng: Context Switching stress-ng: CPU Stress stress-ng: Glibc Qsort Data Sorting smhasher: t1ha2_atonce stress-ng: CPU Cache quadray: 1 - 4K stress-ng: Futex stress-ng: Socket Activity smhasher: FarmHash32 x86_64 AVX stress-ng: SENDFILE smhasher: t1ha0_aes_avx2 x86_64 smhasher: FarmHash128 nginx: 100 stress-ng: Memory Copying stress-ng: Matrix Math aom-av1: Speed 6 Realtime - Bosphorus 1080p stress-ng: Vector Math smhasher: fasthash32 smhasher: MeowHash x86_64 AES-NI stress-ng: System V Message Passing jpegxl: JPEG - 80 stress-ng: MEMFD encodec: 1.5 kbps nekrs: TurboPipe Periodic jpegxl-decode: 1 libplacebo: polar_nocompute openradioss: INIVOL and Fluid Structure Interaction Drop Container xmrig: Monero - 1M cpuminer-opt: Skeincoin encodec: 6 kbps encodec: 3 kbps quadray: 1 - 1080p quadray: 3 - 4K smhasher: SHA3-256 cpuminer-opt: LBC, LBRY Credits smhasher: wyhash jpegxl: JPEG - 90 deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream spacy: en_core_web_lg avifenc: 2 nginx: 500 aom-av1: Speed 10 Realtime - Bosphorus 1080p nginx: 1000 onednn: Recurrent Neural Network Training - f32 - CPU nginx: 200 cpuminer-opt: Blake-2 S xmrig: Wownero - 1M tensorflow: CPU - 512 - GoogLeNet libplacebo: av1_grain_lap stress-ng: Forking libplacebo: hdr_lut aom-av1: Speed 6 Two-Pass - Bosphorus 1080p aom-av1: Speed 10 Realtime - Bosphorus 4K quadray: 2 - 4K aom-av1: Speed 9 Realtime - Bosphorus 1080p jpegxl-decode: All aom-av1: Speed 8 Realtime - Bosphorus 4K deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream cpuminer-opt: Myriad-Groestl onednn: IP Shapes 1D - f32 - CPU smhasher: Spooky32 stress-ng: Malloc aom-av1: Speed 8 Realtime - Bosphorus 1080p deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream quadray: 3 - 1080p avifenc: 0 libplacebo: deband_heavy deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream y-cruncher: 1B avifenc: 6 cpuminer-opt: Ringcoin deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream quadray: 2 - 1080p openfoam: drivaerFastback, Small Mesh Size - Execution Time onednn: Recurrent Neural Network Inference - f32 - CPU deepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Stream stress-ng: Mutex deepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Stream encodec: 24 kbps ffmpeg: libx264 - Platform ffmpeg: libx264 - Platform ffmpeg: libx265 - Platform ffmpeg: libx265 - Platform quadray: 5 - 1080p aom-av1: Speed 4 Two-Pass - Bosphorus 1080p openradioss: Rubber O-Ring Seal Installation openradioss: Bumper Beam onednn: Deconvolution Batch shapes_3d - f32 - CPU stress-ng: Atomic encode-flac: WAV To FLAC stress-ng: MMAP minibude: OpenMP - BM1 minibude: OpenMP - BM1 openfoam: drivaerFastback, Small Mesh Size - Mesh Time tensorflow: CPU - 64 - ResNet-50 avifenc: 6, Lossless openradioss: Bird Strike on Windshield jpegxl: PNG - 90 ffmpeg: libx265 - Live jpegxl: PNG - 80 ffmpeg: libx265 - Video On Demand aom-av1: Speed 4 Two-Pass - Bosphorus 4K ffmpeg: libx265 - Live ffmpeg: libx265 - Video On Demand cpuminer-opt: x25x cpuminer-opt: Deepcoin tensorflow: CPU - 16 - ResNet-50 stress-ng: NUMA avifenc: 10, Lossless cpuminer-opt: Magi aom-av1: Speed 6 Realtime - Bosphorus 4K libplacebo: hdr_peakdetect tensorflow: CPU - 512 - AlexNet aom-av1: Speed 9 Realtime - Bosphorus 4K deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream stress-ng: Crypto tensorflow: CPU - 32 - GoogLeNet tensorflow: CPU - 32 - ResNet-50 openradioss: Cell Phone Drop Test deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream ffmpeg: libx265 - Upload ffmpeg: libx265 - Upload aom-av1: Speed 6 Two-Pass - Bosphorus 4K cpuminer-opt: scrypt tensorflow: CPU - 16 - AlexNet onednn: IP Shapes 3D - f32 - CPU ffmpeg: libx264 - Upload deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream ffmpeg: libx264 - Upload tensorflow: CPU - 32 - AlexNet tensorflow: CPU - 16 - GoogLeNet deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream ffmpeg: libx264 - Video On Demand tensorflow: CPU - 64 - GoogLeNet deepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Stream ffmpeg: libx264 - Video On Demand ffmpeg: libx264 - Live deepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Stream cpuminer-opt: Quad SHA-256, Pyrite ffmpeg: libx264 - Live deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream tensorflow: CPU - 256 - AlexNet deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream onednn: Convolution Batch Shapes Auto - f32 - CPU stress-ng: Semaphores minibude: OpenMP - BM2 minibude: OpenMP - BM2 tensorflow: CPU - 256 - GoogLeNet y-cruncher: 500M tensorflow: CPU - 64 - AlexNet cpuminer-opt: Triple SHA-256, Onecoin deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream aom-av1: Speed 0 Two-Pass - Bosphorus 1080p aom-av1: Speed 0 Two-Pass - Bosphorus 4K quadray: 5 - 4K jpegxl: JPEG - 100 jpegxl: PNG - 100 onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU onednn: Deconvolution Batch shapes_1d - f32 - CPU smhasher: MeowHash x86_64 AES-NI smhasher: t1ha0_aes_avx2 x86_64 smhasher: FarmHash32 x86_64 AVX smhasher: t1ha2_atonce smhasher: FarmHash128 smhasher: fasthash32 smhasher: Spooky32 smhasher: SHA3-256 smhasher: wyhash a aa b 5126.99 832 2048965.25 1939.65 4076744.52 32900.39 180.77 16686.57 158.89 10.21 2600010.05 8595.2 29093.03 215308.14 66796.46 15884.24 87222.68 3599.42 61078.06 45.2 90999.65 6658.92 37183.28 7925418.15 8.68 747.31 45.14 31559900000 48.69 977.72 633.97 7982.6 106880 47.326 47.231 39.48 2.46 149.08 31260 23915.75 8.62 48.4869 20.6208 12196 63.121 75259.81 109.52 65135.63 4075.57 83513.14 524680 10466.6 28.79 2073.69 40818.92 2797.23 41.57 47.55 2.89 108.47 146.63 34.46 617.6457 21130 4.69026 14909.02 13779284.62 85.22 9.7133 9.71 130.357 476.59 611.8404 9.7748 41.524 6.049 2705.27 180.6052 33.2144 11.12 363.21657 2503.12 113.4603 6319425.86 52.8476 54.416 155.11 48.84 31.31 241.928872179 2.64 15.08 131 127.42 5.27435 575564.96 17.255 294.05 405.816 16.233 41.823568 10.09 10.149 285.69 9.02 66.99 9.08 31.45 7.41 75.38 240.839641179 631.15 11710 10.39 260.68 5.48 603.58 23.1 2610.2 110.36 47.42 73.7085 81.374 22523.79 30.46 10.29 104.51 40.9537 146.4793 16.8159 59.4485 164.797431657 15.32 12.67 226.42 55.63 12.0602 12.64 124.7919 8.0129 199.730241377 76.22 31.89 33.3853 29.9481 48.83 30.55 23.5574 155.137732285 24.66 42.4309 102170 204.75 125.3455 7.9776 112.2815 109.23 12.7499 78.3889 22.5449 2463565.68 409.855 16.394 28.64 19.547 93.8 204970 53.4102 0.73 0.25 0.66 0.68 0.69 1.8817 8.2274 57.74 34.811 40.069 32.254 64.103 36.902 49.056 2598.05 25.309 8765.88 833 2035501.44 1960.71 4103283.77 31954.78 192.23 15706.92 152.46 9.63 2717409.38 8500.17 28582.22 215163.99 69896.41 16614.28 86726.1 3592.78 61203.21 46.98 90998.92 6917.32 37904.32 7932169.44 8.98 773.07 46.082 32575200000 47.21 948.28 650.25 7945.8 108620 48.564 47.659 38.5 2.43 148.14 31820 24070.66 8.76 49.228 20.3104 12217 64.019 76617.7 111.49 65507.92 4145.83 84210.42 528060 10640 28.71 2106.73 40572.79 2806.16 42.17 47.95 2.9 108.91 148.61 34.91 617.0345 21260 4.74859 15026.6 13924963.43 85.9 9.723 9.81 131.57 479.65 617.7405 9.7085 41.841 6.076 2709.03 181.4556 33.0595 11.22 361.47957 2484.9 113.6428 6371966.37 52.7812 54.426 154.57 49.01 31.49 240.520175955 2.62 15.01 130.82 126.51 5.25072 575708.53 17.326 292.1 406.706 16.268 41.852307 10.06 10.094 286.84 9.03 66.89 9.13 31.62 7.45 75.50 239.59 631.74 11720 10.44 261.26 5.505 603.33 23.2 2598.97 110.49 47.62 74.0117 81.044 22525.14 30.58 10.29 104.11 41.0572 146.1108 16.7678 59.619 164.72 15.33 12.68 226.59 55.47 12.0288 12.64 124.7582 8.0151 199.81 76.25 31.83 33.3228 30.0041 48.85 30.51 23.5672 155.058138517 24.63 42.4149 102180 205.03 125.1653 7.989 112.2567 109.38 12.7463 78.4088 22.5504 2465656.19 410.307 16.412 28.67 19.527 93.88 204950 53.4052 0.73 0.25 0.66 0.68 0.69 1.27895 5.47125 56.976 32.875 41.134 34.548 60.336 35.257 48.98 2620.221 25.29 4168.54 550 1531955.88 2112.41 3796353.99 30793.53 184.77 16085.48 149.79 10.14 2563831.67 8961.2 27788.51 205666.12 68310.66 16361.64 83591.65 3456.53 58814.38 45.64 87587.66 6686.25 38610.17 7654922.05 8.84 770.89 46.624 32544333333 48.24 961.48 631.33 8175.8 105667 48.215 48.45 39.26 2.49 151.76 31143 23571.26 8.80 49.4908 20.2026 11978 64.373 76562.05 110.26 66302.86 4147.65 82789.42 519433 10577.3 28.33 2093.70 40197.6 2840.16 42.07 48.23 2.86 107.41 148.59 34.60 609.8345 21397 4.73086 14856.10 13767394.33 84.93 9.8238 9.70 130.149 474.53 618.1363 9.6761 41.426 6.019 2730.30 182.2569 32.914 11.17 359.99903 2481.61 114.4037 6344621.56 52.4162 54.862 155.81 48.62 31.55 240.09 2.63 15.12 130.06 127.37 5.28794 571674.68 17.209 293.29 404.247 16.170 42.077007 10.12 10.152 287.33 9.07 67.26 9.12 31.56 7.41 75.10 240.03 628.47 11660 10.44 260.02 5.497 600.87 23.13 2603.19 110.02 47.60 73.7666 81.3092 22434.69 30.48 10.33 104.27 40.9084 146.6414 16.8258 59.4138 164.26 15.37 12.71 225.89 55.49 12.0313 12.67 125.0469 7.9966 199.35 76.37 31.85 33.3404 29.9886 48.77 30.50 23.596 155.31 24.67 42.3627 102017 204.73 125.2691 7.9824 112.1223 109.30 12.7634 78.3062 22.5210 2466594.73 410.141 16.405 28.65 19.531 93.79 205100 53.4398 0.73 0.25 0.66 0.68 0.69 1.74121 6.22396 55.875 33.612 42.575 33.679 61.812 36.688 49.500 2564.206 25.914 OpenBenchmarking.org
Stress-NG Test: IO_uring OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: IO_uring a aa b 2K 4K 6K 8K 10K 5126.99 8765.88 4168.54 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
spaCy Model: en_core_web_trf OpenBenchmarking.org tokens/sec, More Is Better spaCy 3.4.1 Model: en_core_web_trf a aa b 200 400 600 800 1000 832 833 550
Stress-NG Test: Glibc C String Functions OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Glibc C String Functions a aa b 400K 800K 1200K 1600K 2000K 2048965.25 2035501.44 1531955.88 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Cpuminer-Opt Algorithm: Garlicoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Garlicoin a aa b 500 1000 1500 2000 2500 SE +/- 24.17, N = 15 1939.65 1960.71 2112.41 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Context Switching a aa b 900K 1800K 2700K 3600K 4500K 4076744.52 4103283.77 3796353.99 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: CPU Stress a aa b 7K 14K 21K 28K 35K 32900.39 31954.78 30793.53 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: Glibc Qsort Data Sorting OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Glibc Qsort Data Sorting a aa b 40 80 120 160 200 180.77 192.23 184.77 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
SMHasher Hash: t1ha2_atonce OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: t1ha2_atonce a aa b 4K 8K 12K 16K 20K SE +/- 116.42, N = 15 16686.57 15706.92 16085.48 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Stress-NG Test: CPU Cache OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: CPU Cache a aa b 40 80 120 160 200 158.89 152.46 149.79 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
QuadRay Scene: 1 - Resolution: 4K OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 1 - Resolution: 4K a aa b 3 6 9 12 15 SE +/- 0.04, N = 3 10.21 9.63 10.14 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
Stress-NG Test: Futex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Futex a aa b 600K 1200K 1800K 2400K 3000K 2600010.05 2717409.38 2563831.67 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: Socket Activity OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Socket Activity a aa b 2K 4K 6K 8K 10K 8595.20 8500.17 8961.20 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
SMHasher Hash: FarmHash32 x86_64 AVX OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX a aa b 6K 12K 18K 24K 30K SE +/- 369.00, N = 3 29093.03 28582.22 27788.51 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Stress-NG Test: SENDFILE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: SENDFILE a aa b 50K 100K 150K 200K 250K 215308.14 215163.99 205666.12 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
SMHasher Hash: t1ha0_aes_avx2 x86_64 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 a aa b 15K 30K 45K 60K 75K SE +/- 549.14, N = 15 66796.46 69896.41 68310.66 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: FarmHash128 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: FarmHash128 a aa b 4K 8K 12K 16K 20K SE +/- 128.88, N = 15 15884.24 16614.28 16361.64 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
nginx Connections: 100 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 100 a aa b 20K 40K 60K 80K 100K 87222.68 86726.10 83591.65 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Memory Copying a aa b 800 1600 2400 3200 4000 3599.42 3592.78 3456.53 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Matrix Math a aa b 13K 26K 39K 52K 65K 61078.06 61203.21 58814.38 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p a aa b 11 22 33 44 55 SE +/- 0.20, N = 3 45.20 46.98 45.64 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Vector Math a aa b 20K 40K 60K 80K 100K 90999.65 90998.92 87587.66 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
SMHasher Hash: fasthash32 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: fasthash32 a aa b 1500 3000 4500 6000 7500 SE +/- 77.96, N = 4 6658.92 6917.32 6686.25 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: MeowHash x86_64 AES-NI OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI a aa b 8K 16K 24K 32K 40K SE +/- 285.07, N = 15 37183.28 37904.32 38610.17 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Stress-NG Test: System V Message Passing OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: System V Message Passing a aa b 2M 4M 6M 8M 10M 7925418.15 7932169.44 7654922.05 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
JPEG XL libjxl Input: JPEG - Quality: 80 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: JPEG - Quality: 80 a aa b 3 6 9 12 15 SE +/- 0.03, N = 3 8.68 8.98 8.84 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
Stress-NG Test: MEMFD OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: MEMFD a aa b 170 340 510 680 850 747.31 773.07 770.89 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
EnCodec Target Bandwidth: 1.5 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 1.5 kbps a aa b 11 22 33 44 55 45.14 46.08 46.62
nekRS Input: TurboPipe Periodic OpenBenchmarking.org FLOP/s, More Is Better nekRS 22.0 Input: TurboPipe Periodic a aa b 7000M 14000M 21000M 28000M 35000M SE +/- 20784155.29, N = 3 31559900000 32575200000 32544333333 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -lmpi_cxx -lmpi
JPEG XL Decoding libjxl CPU Threads: 1 OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.7 CPU Threads: 1 a aa b 11 22 33 44 55 SE +/- 0.56, N = 3 48.69 47.21 48.24
Libplacebo Test: polar_nocompute OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: polar_nocompute a aa b 200 400 600 800 1000 SE +/- 3.01, N = 3 977.72 948.28 961.48 1. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
OpenRadioss Model: INIVOL and Fluid Structure Interaction Drop Container OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: INIVOL and Fluid Structure Interaction Drop Container a aa b 140 280 420 560 700 SE +/- 0.82, N = 3 633.97 650.25 631.33
Xmrig Variant: Monero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.18.1 Variant: Monero - Hash Count: 1M a aa b 2K 4K 6K 8K 10K SE +/- 88.16, N = 3 7982.6 7945.8 8175.8 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Cpuminer-Opt Algorithm: Skeincoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Skeincoin a aa b 20K 40K 60K 80K 100K SE +/- 1171.67, N = 3 106880 108620 105667 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
EnCodec Target Bandwidth: 6 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 6 kbps a aa b 11 22 33 44 55 47.33 48.56 48.22
EnCodec Target Bandwidth: 3 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 3 kbps a aa b 11 22 33 44 55 47.23 47.66 48.45
QuadRay Scene: 1 - Resolution: 1080p OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 1 - Resolution: 1080p a aa b 9 18 27 36 45 SE +/- 0.13, N = 3 39.48 38.50 39.26 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
QuadRay Scene: 3 - Resolution: 4K OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 3 - Resolution: 4K a aa b 0.5603 1.1206 1.6809 2.2412 2.8015 SE +/- 0.01, N = 3 2.46 2.43 2.49 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
SMHasher Hash: SHA3-256 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: SHA3-256 a aa b 30 60 90 120 150 SE +/- 1.41, N = 7 149.08 148.14 151.76 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Cpuminer-Opt Algorithm: LBC, LBRY Credits OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: LBC, LBRY Credits a aa b 7K 14K 21K 28K 35K SE +/- 18.56, N = 3 31260 31820 31143 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
SMHasher Hash: wyhash OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: wyhash a aa b 5K 10K 15K 20K 25K SE +/- 338.95, N = 3 23915.75 24070.66 23571.26 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
JPEG XL libjxl Input: JPEG - Quality: 90 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: JPEG - Quality: 90 a aa b 2 4 6 8 10 SE +/- 0.08, N = 3 8.62 8.76 8.80 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream a aa b 11 22 33 44 55 48.49 49.23 49.49
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream a aa b 5 10 15 20 25 20.62 20.31 20.20
spaCy Model: en_core_web_lg OpenBenchmarking.org tokens/sec, More Is Better spaCy 3.4.1 Model: en_core_web_lg a aa b 3K 6K 9K 12K 15K 12196 12217 11978
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 2 a aa b 14 28 42 56 70 SE +/- 0.20, N = 3 63.12 64.02 64.37 1. (CXX) g++ options: -O3 -fPIC -lm
nginx Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 a aa b 16K 32K 48K 64K 80K 75259.81 76617.70 76562.05 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p a aa b 20 40 60 80 100 SE +/- 0.74, N = 3 109.52 111.49 110.26 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 a aa b 14K 28K 42K 56K 70K 65135.63 65507.92 66302.86 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU a aa b 900 1800 2700 3600 4500 SE +/- 9.23, N = 3 4075.57 4145.83 4147.65 MIN: 4066.65 MIN: 4135.73 MIN: 4122.17 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
nginx Connections: 200 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 200 a aa b 20K 40K 60K 80K 100K 83513.14 84210.42 82789.42 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Cpuminer-Opt Algorithm: Blake-2 S OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Blake-2 S a aa b 110K 220K 330K 440K 550K SE +/- 4099.93, N = 10 524680 528060 519433 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Xmrig Variant: Wownero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.18.1 Variant: Wownero - Hash Count: 1M a aa b 2K 4K 6K 8K 10K SE +/- 4.48, N = 3 10466.6 10640.0 10577.3 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
TensorFlow Device: CPU - Batch Size: 512 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 512 - Model: GoogLeNet a aa b 7 14 21 28 35 28.79 28.71 28.33
Libplacebo Test: av1_grain_lap OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: av1_grain_lap a aa b 500 1000 1500 2000 2500 SE +/- 8.76, N = 3 2073.69 2106.73 2093.70 1. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
Stress-NG Test: Forking OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Forking a aa b 9K 18K 27K 36K 45K 40818.92 40572.79 40197.60 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Libplacebo Test: hdr_lut OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: hdr_lut a aa b 600 1200 1800 2400 3000 SE +/- 46.08, N = 3 2797.23 2806.16 2840.16 1. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p a aa b 10 20 30 40 50 SE +/- 0.26, N = 3 41.57 42.17 42.07 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K a aa b 11 22 33 44 55 SE +/- 0.11, N = 3 47.55 47.95 48.23 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
QuadRay Scene: 2 - Resolution: 4K OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 2 - Resolution: 4K a aa b 0.6525 1.305 1.9575 2.61 3.2625 SE +/- 0.01, N = 3 2.89 2.90 2.86 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p a aa b 20 40 60 80 100 SE +/- 0.48, N = 3 108.47 108.91 107.41 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
JPEG XL Decoding libjxl CPU Threads: All OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.7 CPU Threads: All a aa b 30 60 90 120 150 SE +/- 0.27, N = 3 146.63 148.61 148.59
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K a aa b 8 16 24 32 40 SE +/- 0.08, N = 3 34.46 34.91 34.60 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a aa b 130 260 390 520 650 617.65 617.03 609.83
Cpuminer-Opt Algorithm: Myriad-Groestl OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Myriad-Groestl a aa b 5K 10K 15K 20K 25K SE +/- 153.44, N = 3 21130 21260 21397 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU a aa b 1.0684 2.1368 3.2052 4.2736 5.342 SE +/- 0.01435, N = 3 4.69026 4.74859 4.73086 MIN: 4.49 MIN: 4.55 MIN: 4.49 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
SMHasher Hash: Spooky32 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2022-08-22 Hash: Spooky32 a aa b 3K 6K 9K 12K 15K SE +/- 98.93, N = 15 14909.02 15026.60 14856.10 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Stress-NG Test: Malloc OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Malloc a aa b 3M 6M 9M 12M 15M 13779284.62 13924963.43 13767394.33 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p a aa b 20 40 60 80 100 SE +/- 0.06, N = 3 85.22 85.90 84.93 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a aa b 3 6 9 12 15 9.7133 9.7230 9.8238
QuadRay Scene: 3 - Resolution: 1080p OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 3 - Resolution: 1080p a aa b 3 6 9 12 15 SE +/- 0.10, N = 3 9.71 9.81 9.70 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 0 a aa b 30 60 90 120 150 SE +/- 0.52, N = 3 130.36 131.57 130.15 1. (CXX) g++ options: -O3 -fPIC -lm
Libplacebo Test: deband_heavy OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: deband_heavy a aa b 100 200 300 400 500 SE +/- 3.08, N = 3 476.59 479.65 474.53 1. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a aa b 130 260 390 520 650 611.84 617.74 618.14
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a aa b 3 6 9 12 15 9.7748 9.7085 9.6761
Y-Cruncher Pi Digits To Calculate: 1B OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 1B a aa b 10 20 30 40 50 SE +/- 0.11, N = 3 41.52 41.84 41.43
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 6 a aa b 2 4 6 8 10 SE +/- 0.028, N = 3 6.049 6.076 6.019 1. (CXX) g++ options: -O3 -fPIC -lm
Cpuminer-Opt Algorithm: Ringcoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Ringcoin a aa b 600 1200 1800 2400 3000 SE +/- 7.56, N = 3 2705.27 2709.03 2730.30 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream a aa b 40 80 120 160 200 180.61 181.46 182.26
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream a aa b 8 16 24 32 40 33.21 33.06 32.91
QuadRay Scene: 2 - Resolution: 1080p OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 2 - Resolution: 1080p a aa b 3 6 9 12 15 SE +/- 0.06, N = 3 11.12 11.22 11.17 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
OpenFOAM Input: drivaerFastback, Small Mesh Size - Execution Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Execution Time a aa b 80 160 240 320 400 363.22 361.48 360.00 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU a aa b 500 1000 1500 2000 2500 SE +/- 1.05, N = 3 2503.12 2484.90 2481.61 MIN: 2492.43 MIN: 2478.59 MIN: 2470.87 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream a aa b 30 60 90 120 150 113.46 113.64 114.40
Stress-NG Test: Mutex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Mutex a aa b 1.4M 2.8M 4.2M 5.6M 7M 6319425.86 6371966.37 6344621.56 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream a aa b 12 24 36 48 60 52.85 52.78 52.42
EnCodec Target Bandwidth: 24 kbps OpenBenchmarking.org Seconds, Fewer Is Better EnCodec 0.1.1 Target Bandwidth: 24 kbps a aa b 12 24 36 48 60 54.42 54.43 54.86
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Platform a aa b 30 60 90 120 150 SE +/- 0.25, N = 3 155.11 154.57 155.81 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx264 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Platform a aa b 11 22 33 44 55 SE +/- 0.08, N = 3 48.84 49.01 48.62 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Platform a aa b 7 14 21 28 35 SE +/- 0.03, N = 3 31.31 31.49 31.55 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Platform OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Platform a aa b 50 100 150 200 250 SE +/- 0.20, N = 3 241.93 240.52 240.09 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
QuadRay Scene: 5 - Resolution: 1080p OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 5 - Resolution: 1080p a aa b 0.594 1.188 1.782 2.376 2.97 SE +/- 0.00, N = 3 2.64 2.62 2.63 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p a aa b 4 8 12 16 20 SE +/- 0.08, N = 3 15.08 15.01 15.12 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenRadioss Model: Rubber O-Ring Seal Installation OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Rubber O-Ring Seal Installation a aa b 30 60 90 120 150 SE +/- 0.35, N = 3 131.00 130.82 130.06
OpenRadioss Model: Bumper Beam OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Bumper Beam a aa b 30 60 90 120 150 SE +/- 0.16, N = 3 127.42 126.51 127.37
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU a aa b 1.1898 2.3796 3.5694 4.7592 5.949 SE +/- 0.00411, N = 3 5.27435 5.25072 5.28794 MIN: 5.18 MIN: 5.17 MIN: 5.2 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Stress-NG Test: Atomic OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Atomic a aa b 120K 240K 360K 480K 600K 575564.96 575708.53 571674.68 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.4 WAV To FLAC a aa b 4 8 12 16 20 SE +/- 0.04, N = 5 17.26 17.33 17.21 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: MMAP a aa b 60 120 180 240 300 294.05 292.10 293.29 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 a aa b 90 180 270 360 450 SE +/- 0.95, N = 3 405.82 406.71 404.25 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 a aa b 4 8 12 16 20 SE +/- 0.04, N = 3 16.23 16.27 16.17 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenFOAM Input: drivaerFastback, Small Mesh Size - Mesh Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time a aa b 10 20 30 40 50 41.82 41.85 42.08 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: ResNet-50 a aa b 3 6 9 12 15 10.09 10.06 10.12
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 6, Lossless a aa b 3 6 9 12 15 SE +/- 0.04, N = 3 10.15 10.09 10.15 1. (CXX) g++ options: -O3 -fPIC -lm
OpenRadioss Model: Bird Strike on Windshield OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Bird Strike on Windshield a aa b 60 120 180 240 300 SE +/- 0.10, N = 3 285.69 286.84 287.33
JPEG XL libjxl Input: PNG - Quality: 90 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: PNG - Quality: 90 a aa b 3 6 9 12 15 SE +/- 0.02, N = 3 9.02 9.03 9.07 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Live a aa b 15 30 45 60 75 SE +/- 0.65, N = 3 66.99 66.89 67.26 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
JPEG XL libjxl Input: PNG - Quality: 80 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: PNG - Quality: 80 a aa b 3 6 9 12 15 SE +/- 0.02, N = 3 9.08 9.13 9.12 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Video On Demand a aa b 7 14 21 28 35 SE +/- 0.07, N = 3 31.45 31.62 31.56 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K a aa b 2 4 6 8 10 SE +/- 0.01, N = 3 7.41 7.45 7.41 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
FFmpeg Encoder: libx265 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Live a aa b 20 40 60 80 100 SE +/- 0.73, N = 3 75.38 75.50 75.10 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Video On Demand a aa b 50 100 150 200 250 SE +/- 0.53, N = 3 240.84 239.59 240.03 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Cpuminer-Opt Algorithm: x25x OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: x25x a aa b 140 280 420 560 700 SE +/- 0.96, N = 3 631.15 631.74 628.47 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Cpuminer-Opt Algorithm: Deepcoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Deepcoin a aa b 3K 6K 9K 12K 15K SE +/- 5.77, N = 3 11710 11720 11660 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
TensorFlow Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: ResNet-50 a aa b 3 6 9 12 15 SE +/- 0.02, N = 3 10.39 10.44 10.44
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: NUMA a aa b 60 120 180 240 300 260.68 261.26 260.02 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.11 Encoder Speed: 10, Lossless a aa b 1.2386 2.4772 3.7158 4.9544 6.193 SE +/- 0.026, N = 3 5.480 5.505 5.497 1. (CXX) g++ options: -O3 -fPIC -lm
Cpuminer-Opt Algorithm: Magi OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Magi a aa b 130 260 390 520 650 SE +/- 0.50, N = 3 603.58 603.33 600.87 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K a aa b 6 12 18 24 30 SE +/- 0.06, N = 3 23.10 23.20 23.13 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Libplacebo Test: hdr_peakdetect OpenBenchmarking.org FPS, More Is Better Libplacebo 5.229.1 Test: hdr_peakdetect a aa b 600 1200 1800 2400 3000 SE +/- 4.71, N = 3 2610.20 2598.97 2603.19 1. (CXX) g++ options: -lm -pthread -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -ldl -std=c++11 -O2 -fvisibility=hidden -fPIC -MD -MQ -MF
TensorFlow Device: CPU - Batch Size: 512 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 512 - Model: AlexNet a aa b 20 40 60 80 100 SE +/- 0.06, N = 3 110.36 110.49 110.02
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K a aa b 11 22 33 44 55 SE +/- 0.12, N = 3 47.42 47.62 47.60 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a aa b 16 32 48 64 80 73.71 74.01 73.77
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a aa b 20 40 60 80 100 81.37 81.04 81.31
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Crypto a aa b 5K 10K 15K 20K 25K 22523.79 22525.14 22434.69 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
TensorFlow Device: CPU - Batch Size: 32 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: GoogLeNet a aa b 7 14 21 28 35 SE +/- 0.01, N = 3 30.46 30.58 30.48
TensorFlow Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: ResNet-50 a aa b 3 6 9 12 15 SE +/- 0.01, N = 3 10.29 10.29 10.33
OpenRadioss Model: Cell Phone Drop Test OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Cell Phone Drop Test a aa b 20 40 60 80 100 SE +/- 0.05, N = 3 104.51 104.11 104.27
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream a aa b 9 18 27 36 45 40.95 41.06 40.91
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream a aa b 30 60 90 120 150 146.48 146.11 146.64
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a aa b 4 8 12 16 20 16.82 16.77 16.83
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a aa b 13 26 39 52 65 59.45 59.62 59.41
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Upload a aa b 40 80 120 160 200 SE +/- 0.31, N = 3 164.80 164.72 164.26 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx265 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx265 - Scenario: Upload a aa b 4 8 12 16 20 SE +/- 0.03, N = 3 15.32 15.33 15.37 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K a aa b 3 6 9 12 15 SE +/- 0.06, N = 3 12.67 12.68 12.71 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Cpuminer-Opt Algorithm: scrypt OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: scrypt a aa b 50 100 150 200 250 SE +/- 0.18, N = 3 226.42 226.59 225.89 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
TensorFlow Device: CPU - Batch Size: 16 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: AlexNet a aa b 12 24 36 48 60 SE +/- 0.04, N = 3 55.63 55.47 55.49
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU a aa b 3 6 9 12 15 SE +/- 0.01, N = 3 12.06 12.03 12.03 MIN: 11.96 MIN: 11.93 MIN: 11.91 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Upload a aa b 3 6 9 12 15 SE +/- 0.06, N = 3 12.64 12.64 12.67 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a aa b 30 60 90 120 150 124.79 124.76 125.05
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a aa b 2 4 6 8 10 8.0129 8.0151 7.9966
FFmpeg Encoder: libx264 - Scenario: Upload OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Upload a aa b 40 80 120 160 200 SE +/- 0.94, N = 3 199.73 199.81 199.35 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
TensorFlow Device: CPU - Batch Size: 32 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 32 - Model: AlexNet a aa b 20 40 60 80 100 SE +/- 0.07, N = 3 76.22 76.25 76.37
TensorFlow Device: CPU - Batch Size: 16 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 16 - Model: GoogLeNet a aa b 7 14 21 28 35 SE +/- 0.03, N = 3 31.89 31.83 31.85
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream a aa b 8 16 24 32 40 33.39 33.32 33.34
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream a aa b 7 14 21 28 35 29.95 30.00 29.99
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Video On Demand a aa b 11 22 33 44 55 SE +/- 0.09, N = 3 48.83 48.85 48.77 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
TensorFlow Device: CPU - Batch Size: 64 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: GoogLeNet a aa b 7 14 21 28 35 SE +/- 0.01, N = 3 30.55 30.51 30.50
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream a aa b 6 12 18 24 30 23.56 23.57 23.60
FFmpeg Encoder: libx264 - Scenario: Video On Demand OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Video On Demand a aa b 30 60 90 120 150 SE +/- 0.30, N = 3 155.14 155.06 155.31 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Live a aa b 6 12 18 24 30 SE +/- 0.05, N = 3 24.66 24.63 24.67 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream a aa b 10 20 30 40 50 42.43 42.41 42.36
Cpuminer-Opt Algorithm: Quad SHA-256, Pyrite OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Quad SHA-256, Pyrite a aa b 20K 40K 60K 80K 100K SE +/- 58.12, N = 3 102170 102180 102017 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
FFmpeg Encoder: libx264 - Scenario: Live OpenBenchmarking.org FPS, More Is Better FFmpeg 5.1.2 Encoder: libx264 - Scenario: Live a aa b 40 80 120 160 200 SE +/- 0.39, N = 3 204.75 205.03 204.73 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a aa b 30 60 90 120 150 125.35 125.17 125.27
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a aa b 2 4 6 8 10 7.9776 7.9890 7.9824
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a aa b 30 60 90 120 150 112.28 112.26 112.12
TensorFlow Device: CPU - Batch Size: 256 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: AlexNet a aa b 20 40 60 80 100 SE +/- 0.06, N = 3 109.23 109.38 109.30
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream a aa b 3 6 9 12 15 12.75 12.75 12.76
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream a aa b 20 40 60 80 100 78.39 78.41 78.31
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU a aa b 5 10 15 20 25 SE +/- 0.02, N = 3 22.54 22.55 22.52 MIN: 22.24 MIN: 22.11 MIN: 21.81 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Semaphores a aa b 500K 1000K 1500K 2000K 2500K 2463565.68 2465656.19 2466594.73 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 a aa b 90 180 270 360 450 SE +/- 0.03, N = 3 409.86 410.31 410.14 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 a aa b 4 8 12 16 20 SE +/- 0.00, N = 3 16.39 16.41 16.41 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
TensorFlow Device: CPU - Batch Size: 256 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: GoogLeNet a aa b 7 14 21 28 35 28.64 28.67 28.65
Y-Cruncher Pi Digits To Calculate: 500M OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.10.9513 Pi Digits To Calculate: 500M a aa b 5 10 15 20 25 SE +/- 0.02, N = 3 19.55 19.53 19.53
TensorFlow Device: CPU - Batch Size: 64 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.10 Device: CPU - Batch Size: 64 - Model: AlexNet a aa b 20 40 60 80 100 SE +/- 0.06, N = 3 93.80 93.88 93.79
Cpuminer-Opt Algorithm: Triple SHA-256, Onecoin OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.20.3 Algorithm: Triple SHA-256, Onecoin a aa b 40K 80K 120K 160K 200K SE +/- 26.46, N = 3 204970 204950 205100 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a aa b 12 24 36 48 60 53.41 53.41 53.44
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p a aa b 0.1643 0.3286 0.4929 0.6572 0.8215 SE +/- 0.00, N = 3 0.73 0.73 0.73 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.5 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K a aa b 0.0563 0.1126 0.1689 0.2252 0.2815 SE +/- 0.00, N = 3 0.25 0.25 0.25 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
QuadRay Scene: 5 - Resolution: 4K OpenBenchmarking.org FPS, More Is Better QuadRay 2022.05.25 Scene: 5 - Resolution: 4K a aa b 0.1485 0.297 0.4455 0.594 0.7425 SE +/- 0.00, N = 3 0.66 0.66 0.66 1. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread
JPEG XL libjxl Input: JPEG - Quality: 100 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: JPEG - Quality: 100 a aa b 0.153 0.306 0.459 0.612 0.765 SE +/- 0.00, N = 3 0.68 0.68 0.68 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
JPEG XL libjxl Input: PNG - Quality: 100 OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.7 Input: PNG - Quality: 100 a aa b 0.1553 0.3106 0.4659 0.6212 0.7765 SE +/- 0.01, N = 3 0.69 0.69 0.69 1. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU a aa b 0.4234 0.8468 1.2702 1.6936 2.117 SE +/- 0.12895, N = 15 1.88170 1.27895 1.74121 MIN: 1.76 MIN: 1.15 MIN: 1.02 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU a aa b 2 4 6 8 10 SE +/- 0.14562, N = 15 8.22740 5.47125 6.22396 MIN: 7 MIN: 4.57 MIN: 4.61 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
SMHasher Hash: MeowHash x86_64 AES-NI OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: MeowHash x86_64 AES-NI a aa b 13 26 39 52 65 SE +/- 0.39, N = 15 57.74 56.98 55.88 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: t1ha0_aes_avx2 x86_64 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: t1ha0_aes_avx2 x86_64 a aa b 8 16 24 32 40 SE +/- 0.30, N = 15 34.81 32.88 33.61 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: FarmHash32 x86_64 AVX OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: FarmHash32 x86_64 AVX a aa b 10 20 30 40 50 SE +/- 0.70, N = 3 40.07 41.13 42.58 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: t1ha2_atonce OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: t1ha2_atonce a aa b 8 16 24 32 40 SE +/- 0.28, N = 15 32.25 34.55 33.68 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: FarmHash128 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: FarmHash128 a aa b 14 28 42 56 70 SE +/- 0.72, N = 15 64.10 60.34 61.81 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: fasthash32 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: fasthash32 a aa b 8 16 24 32 40 SE +/- 0.48, N = 4 36.90 35.26 36.69 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: Spooky32 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: Spooky32 a aa b 11 22 33 44 55 SE +/- 0.31, N = 15 49.06 48.98 49.50 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: SHA3-256 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: SHA3-256 a aa b 600 1200 1800 2400 3000 SE +/- 19.10, N = 7 2598.05 2620.22 2564.21 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
SMHasher Hash: wyhash OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2022-08-22 Hash: wyhash a aa b 6 12 18 24 30 SE +/- 0.37, N = 3 25.31 25.29 25.91 1. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects
Phoronix Test Suite v10.8.5