AMD Ryzen 9 3900X New Tests With Kernel Changes AMD Ryzen 9 3900X 12-Core testing with a ASUS TUF GAMING X570-PLUS (WI-FI) (2203 BIOS) and MSI AMD Radeon RX 470/480/570/570X/580/580X/590 8GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2009220-PTS-AMDRYZEN18&sor .
AMD Ryzen 9 3900X New Tests With Kernel Changes Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Vulkan Compiler File-System Screen Resolution Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git AMD Ryzen 9 3900X 12-Core @ 3.80GHz (12 Cores / 24 Threads) ASUS TUF GAMING X570-PLUS (WI-FI) (2203 BIOS) AMD Starship/Matisse 16GB Samsung SSD 970 EVO Plus 250GB MSI AMD Radeon RX 470/480/570/570X/580/580X/590 8GB (1366/2000MHz) AMD Ellesmere HDMI Audio G237HL Realtek RTL8111/8168/8411 + Intel-AC 9260 Ubuntu 20.04 5.4.0-42-generic (x86_64) GNOME Shell 3.36.3 X Server 1.20.8 modesetting 1.20.8 4.6 Mesa 20.2.0-devel (git-64cdc13 2020-07-02 focal-oibaf-ppa) (LLVM 10.0.0) 1.2.131 GCC 9.3.0 ext4 1920x1080 5.8.10-050810-generic (x86_64) 5.9.0-050900rc6daily20200922-generic (x86_64) 20200921 OpenBenchmarking.org Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8701021 Graphics Details - GLAMOR Python Details - Python 3.8.2 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
AMD Ryzen 9 3900X New Tests With Kernel Changes glmark2: 1920 x 1080 lczero: BLAS lczero: Eigen namd: ATPase Simulation - 327,506 Atoms incompact3d: Cylinder mocassin: Dust 2D tau100.0 lammps: 20k Atoms lammps: Rhodopsin Protein compress-zstd: 19 aom-av1: Speed 4 Two-Pass aom-av1: Speed 6 Realtime aom-av1: Speed 6 Two-Pass aom-av1: Speed 8 Realtime build-llvm: Time To Compile system-decompress-gzip: perf-bench: Epoll Wait perf-bench: Futex Hash perf-bench: Memcpy 1MB perf-bench: Memset 1MB perf-bench: Sched Pipe perf-bench: Futex Lock-Pi perf-bench: Syscall Basic gromacs: Water Benchmark gpaw: Carbon Nanotube mnn: SqueezeNetV1.0 mnn: resnet-v2-50 mnn: MobileNetV2_224 mnn: mobilenet-v1-1.0 mnn: inception-v3 ncnn: CPU - squeezenet_int8 ncnn: CPU - mobilenet_v3 ncnn: CPU - squeezenet ncnn: CPU - mnasnet ncnn: CPU - blazeface ncnn: CPU - googlenet_int8 ncnn: CPU - vgg16_int8 ncnn: CPU - resnet18_int8 ncnn: CPU - alexnet ncnn: CPU - resnet50_int8 ncnn: CPU - mobilenetv2_yolov3 kripke: Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 4640 511 530 1.57422 262.275299 212 9.359 9.116 38.8 2.63 19.39 4.12 37.57 444.847 2.882 43847 4820951 15.142763 70.561859 404240 669 22276401 1.121 309.675 7.378 32.546 4.062 6.868 31.428 13.34 4.84 4.83 4.86 1.95 35.66 103.63 19.73 16.64 63.31 16.39 4614547 4645 520 507 1.56619 267.783132 212 9.201 9.025 38.8 2.61 19.43 4.08 37.51 446.209 2.758 45922 4891875 14.236737 70.589682 328023 667 20388216 1.121 310.856 7.403 33.105 3.979 6.865 31.528 13.37 4.83 4.81 4.91 1.96 36.19 104.48 20.07 16.65 63.72 16.49 11727667 4903 504 521 1.57290 269.579844 213 9.283 9.000 38.7 2.60 19.47 4.05 37.56 444.772 2.848 46204 4798809 14.697864 73.410713 312435 668 20858233 1.117 310.991 7.328 33.030 3.977 6.906 31.472 13.26 4.84 4.86 4.91 1.96 35.79 105.09 20.28 16.64 64.38 16.77 16249158 OpenBenchmarking.org
GLmark2 Resolution: 1920 x 1080 OpenBenchmarking.org Score, More Is Better GLmark2 2020.04 Resolution: 1920 x 1080 Linux 5.9 Git Linux 5.8.10 Ubuntu Default 5.4 1100 2200 3300 4400 5500 4903 4645 4640
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: BLAS Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 110 220 330 440 550 SE +/- 6.43, N = 3 SE +/- 7.62, N = 3 SE +/- 6.08, N = 9 520 511 504 1. (CXX) g++ options: -flto -pthread
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.26 Backend: Eigen Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 110 220 330 440 550 SE +/- 8.65, N = 3 SE +/- 5.36, N = 3 SE +/- 7.88, N = 3 530 521 507 1. (CXX) g++ options: -flto -pthread
NAMD ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms Linux 5.8.10 Linux 5.9 Git Ubuntu Default 5.4 0.3542 0.7084 1.0626 1.4168 1.771 SE +/- 0.00018, N = 3 SE +/- 0.00093, N = 3 SE +/- 0.00284, N = 3 1.56619 1.57290 1.57422
Incompact3D Input: Cylinder OpenBenchmarking.org Seconds, Fewer Is Better Incompact3D 2020-09-17 Input: Cylinder Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 60 120 180 240 300 SE +/- 0.51, N = 3 SE +/- 1.03, N = 3 SE +/- 0.77, N = 3 262.28 267.78 269.58 1. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Monte Carlo Simulations of Ionised Nebulae Input: Dust 2D tau100.0 OpenBenchmarking.org Seconds, Fewer Is Better Monte Carlo Simulations of Ionised Nebulae 2019-03-24 Input: Dust 2D tau100.0 Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 50 100 150 200 250 212 212 213 1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
LAMMPS Molecular Dynamics Simulator Model: 20k Atoms OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 24Aug2020 Model: 20k Atoms Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 3 6 9 12 15 SE +/- 0.007, N = 3 SE +/- 0.008, N = 3 SE +/- 0.037, N = 3 9.359 9.283 9.201 1. (CXX) g++ options: -O3 -pthread -lm
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 24Aug2020 Model: Rhodopsin Protein Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 3 6 9 12 15 SE +/- 0.117, N = 14 SE +/- 0.116, N = 14 SE +/- 0.091, N = 15 9.116 9.025 9.000 1. (CXX) g++ options: -O3 -pthread -lm
Zstd Compression Compression Level: 19 OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.4.5 Compression Level: 19 Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 9 18 27 36 45 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.12, N = 3 38.8 38.8 38.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
AOM AV1 Encoder Mode: Speed 4 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 4 Two-Pass Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 0.5918 1.1836 1.7754 2.3672 2.959 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 2.63 2.61 2.60 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 6 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Realtime Linux 5.9 Git Linux 5.8.10 Ubuntu Default 5.4 5 10 15 20 25 SE +/- 0.08, N = 3 SE +/- 0.14, N = 3 SE +/- 0.03, N = 3 19.47 19.43 19.39 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 6 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 6 Two-Pass Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 0.927 1.854 2.781 3.708 4.635 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 4.12 4.08 4.05 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
AOM AV1 Encoder Mode: Speed 8 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2.0 Encoder Mode: Speed 8 Realtime Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 9 18 27 36 45 SE +/- 0.17, N = 3 SE +/- 0.22, N = 3 SE +/- 0.25, N = 3 37.57 37.56 37.51 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 10.0 Time To Compile Linux 5.9 Git Ubuntu Default 5.4 Linux 5.8.10 100 200 300 400 500 SE +/- 4.89, N = 3 SE +/- 3.65, N = 3 SE +/- 2.69, N = 3 444.77 444.85 446.21
System GZIP Decompression OpenBenchmarking.org Seconds, Fewer Is Better System GZIP Decompression Linux 5.8.10 Linux 5.9 Git Ubuntu Default 5.4 0.6485 1.297 1.9455 2.594 3.2425 SE +/- 0.005, N = 3 SE +/- 0.047, N = 3 SE +/- 0.040, N = 3 2.758 2.848 2.882
perf-bench Benchmark: Epoll Wait OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Epoll Wait Linux 5.9 Git Linux 5.8.10 Ubuntu Default 5.4 10K 20K 30K 40K 50K SE +/- 122.75, N = 3 SE +/- 147.89, N = 3 SE +/- 237.60, N = 3 46204 45922 43847 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
perf-bench Benchmark: Futex Hash OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Hash Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 1000K 2000K 3000K 4000K 5000K SE +/- 6026.98, N = 3 SE +/- 3705.18, N = 3 SE +/- 12288.80, N = 3 4891875 4820951 4798809 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
perf-bench Benchmark: Memcpy 1MB OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memcpy 1MB Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 4 8 12 16 20 SE +/- 0.25, N = 15 SE +/- 0.24, N = 3 SE +/- 0.09, N = 3 15.14 14.70 14.24 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
perf-bench Benchmark: Memset 1MB OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memset 1MB Linux 5.9 Git Linux 5.8.10 Ubuntu Default 5.4 16 32 48 64 80 SE +/- 0.51, N = 3 SE +/- 0.85, N = 3 SE +/- 0.55, N = 3 73.41 70.59 70.56 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
perf-bench Benchmark: Sched Pipe OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Sched Pipe Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 90K 180K 270K 360K 450K SE +/- 3596.98, N = 3 SE +/- 1251.89, N = 3 SE +/- 3676.30, N = 3 404240 328023 312435 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
perf-bench Benchmark: Futex Lock-Pi OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Lock-Pi Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 140 280 420 560 700 SE +/- 2.73, N = 3 SE +/- 2.85, N = 3 SE +/- 3.06, N = 3 669 668 667 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
perf-bench Benchmark: Syscall Basic OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Syscall Basic Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 5M 10M 15M 20M 25M SE +/- 25407.07, N = 3 SE +/- 241562.93, N = 3 SE +/- 214159.67, N = 7 22276401 20858233 20388216 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
GROMACS Water Benchmark OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2020.1 Water Benchmark Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 0.2522 0.5044 0.7566 1.0088 1.261 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 SE +/- 0.001, N = 3 1.121 1.121 1.117 1. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
GPAW Input: Carbon Nanotube OpenBenchmarking.org Seconds, Fewer Is Better GPAW 20.1 Input: Carbon Nanotube Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 70 140 210 280 350 SE +/- 0.20, N = 3 SE +/- 0.24, N = 3 SE +/- 0.32, N = 3 309.68 310.86 310.99 1. (CC) gcc options: -pthread -shared -fwrapv -O2 -lxc -lblas -lmpi
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: SqueezeNetV1.0 Linux 5.9 Git Ubuntu Default 5.4 Linux 5.8.10 2 4 6 8 10 SE +/- 0.048, N = 3 SE +/- 0.045, N = 3 SE +/- 0.023, N = 3 7.328 7.378 7.403 MIN: 7.14 / MAX: 18.84 MIN: 7.18 / MAX: 16.39 MIN: 7.25 / MAX: 18.15 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: resnet-v2-50 Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 8 16 24 32 40 SE +/- 0.24, N = 3 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 32.55 33.03 33.11 MIN: 31.64 / MAX: 42.77 MIN: 32.07 / MAX: 43.69 MIN: 32.16 / MAX: 51.02 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: MobileNetV2_224 Linux 5.9 Git Linux 5.8.10 Ubuntu Default 5.4 0.914 1.828 2.742 3.656 4.57 SE +/- 0.028, N = 3 SE +/- 0.012, N = 3 SE +/- 0.005, N = 3 3.977 3.979 4.062 MIN: 3.88 / MAX: 10.11 MIN: 3.89 / MAX: 8.58 MIN: 3.98 / MAX: 6.98 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: mobilenet-v1-1.0 Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 2 4 6 8 10 SE +/- 0.010, N = 3 SE +/- 0.003, N = 3 SE +/- 0.025, N = 3 6.865 6.868 6.906 MIN: 6.77 / MAX: 18.04 MIN: 6.74 / MAX: 7.88 MIN: 6.77 / MAX: 16.25 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: inception-v3 Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 7 14 21 28 35 SE +/- 0.12, N = 3 SE +/- 0.28, N = 3 SE +/- 0.18, N = 3 31.43 31.47 31.53 MIN: 30.78 / MAX: 43.53 MIN: 30.41 / MAX: 87.19 MIN: 30.68 / MAX: 85.71 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
NCNN Target: CPU - Model: squeezenet_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: squeezenet_int8 Linux 5.9 Git Ubuntu Default 5.4 Linux 5.8.10 3 6 9 12 15 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 13.26 13.34 13.37 MIN: 13.05 / MAX: 17.25 MIN: 13.15 / MAX: 14.14 MIN: 13.07 / MAX: 48.91 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenet_v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenet_v3 Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 1.089 2.178 3.267 4.356 5.445 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 4.83 4.84 4.84 MIN: 4.76 / MAX: 5.01 MIN: 4.75 / MAX: 5.5 MIN: 4.72 / MAX: 6.91 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: squeezenet Linux 5.8.10 Ubuntu Default 5.4 Linux 5.9 Git 1.0935 2.187 3.2805 4.374 5.4675 SE +/- 0.05, N = 3 SE +/- 0.00, N = 2 SE +/- 0.03, N = 2 4.81 4.83 4.86 MIN: 4.66 / MAX: 9.77 MIN: 4.79 / MAX: 4.88 MIN: 4.79 / MAX: 5.43 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mnasnet Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 1.1048 2.2096 3.3144 4.4192 5.524 SE +/- 0.02, N = 2 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 4.86 4.91 4.91 MIN: 4.8 / MAX: 4.93 MIN: 4.82 / MAX: 5.44 MIN: 4.78 / MAX: 5.79 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: blazeface Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 0.441 0.882 1.323 1.764 2.205 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 1.95 1.96 1.96 MIN: 1.91 / MAX: 2.04 MIN: 1.92 / MAX: 2.05 MIN: 1.91 / MAX: 2.05 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: googlenet_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: googlenet_int8 Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 8 16 24 32 40 SE +/- 0.10, N = 3 SE +/- 0.29, N = 3 SE +/- 0.26, N = 3 35.66 35.79 36.19 MIN: 34.99 / MAX: 37.26 MIN: 34.65 / MAX: 41.7 MIN: 34.82 / MAX: 40.62 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vgg16_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: vgg16_int8 Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 20 40 60 80 100 SE +/- 0.56, N = 3 SE +/- 0.29, N = 3 SE +/- 0.25, N = 3 103.63 104.48 105.09 MIN: 101.71 / MAX: 114.82 MIN: 101.26 / MAX: 155.23 MIN: 101.92 / MAX: 191.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet18_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet18_int8 Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.16, N = 3 SE +/- 0.43, N = 3 19.73 20.07 20.28 MIN: 18.45 / MAX: 29.88 MIN: 18.74 / MAX: 25.55 MIN: 18.22 / MAX: 95.37 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: alexnet Ubuntu Default 5.4 Linux 5.9 Git Linux 5.8.10 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 16.64 16.64 16.65 MIN: 16.55 / MAX: 17.18 MIN: 16.49 / MAX: 20.92 MIN: 16.49 / MAX: 21.18 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet50_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet50_int8 Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 14 28 42 56 70 SE +/- 0.39, N = 3 SE +/- 0.58, N = 3 SE +/- 0.21, N = 3 63.31 63.72 64.38 MIN: 62.23 / MAX: 64.82 MIN: 62.27 / MAX: 129.81 MIN: 63.68 / MAX: 142.28 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenetv2_yolov3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenetv2_yolov3 Ubuntu Default 5.4 Linux 5.8.10 Linux 5.9 Git 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.19, N = 3 SE +/- 0.27, N = 3 16.39 16.49 16.77 MIN: 16.04 / MAX: 17.01 MIN: 15.89 / MAX: 20.78 MIN: 15.99 / MAX: 98.19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Kripke OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 Linux 5.9 Git Linux 5.8.10 Ubuntu Default 5.4 3M 6M 9M 12M 15M SE +/- 222150.06, N = 4 SE +/- 518659.22, N = 9 SE +/- 152326.40, N = 9 16249158 11727667 4614547 1. (CXX) g++ options: -O3 -fopenmp
Phoronix Test Suite v10.8.4