3300X NN + More AMD Ryzen 7 1700 Eight-Core testing with a MSI B350 TOMAHAWK (MS-7A34) v1.0 (1.H0 BIOS) and AMD Radeon RX 470/480/570/570X/580/580X/590 8GB on Ubuntu 19.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2009269-FI-3300XNNMO11&grr .
3300X NN + More Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Vulkan Compiler File-System Screen Resolution 1 2 3 4 AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (8 Cores / 16 Threads) MSI B350 TOMAHAWK (MS-7A34) v1.0 (1.H0 BIOS) AMD 17h 16GB 120GB Samsung SSD 840 AMD Radeon RX 470/480/570/570X/580/580X/590 8GB (1266/2000MHz) AMD Ellesmere HDMI Audio VA2431 Realtek RTL8111/8168/8411 Ubuntu 19.10 5.9.0-050900rc2-generic (x86_64) 20200823 GNOME Shell 3.34.1 X Server 1.20.5 amdgpu 19.0.1 4.5 Mesa 19.2.8 (LLVM 9.0.0) 1.1.107 GCC 9.2.1 20191008 ext4 1920x1080 OpenBenchmarking.org Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-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: 0x8001137 Graphics Details - GLAMOR Python Details - Python 2.7.17rc1 + Python 3.7.5 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + srbds: Not affected + tsx_async_abort: Not affected
3300X NN + More lammps: 20k Atoms kripke: influxdb: 4 - 10000 - 2,5000,1 - 10000 incompact3d: Cylinder gpaw: Carbon Nanotube realsr-ncnn: 4x - Yes influxdb: 64 - 10000 - 2,5000,1 - 10000 influxdb: 1024 - 10000 - 2,5000,1 - 10000 mocassin: Dust 2D tau100.0 mnn: inception-v3 mnn: mobilenet-v1-1.0 mnn: MobileNetV2_224 mnn: resnet-v2-50 mnn: SqueezeNetV1.0 espeak: Text-To-Speech Synthesis ncnn: CPU - yolov4-tiny ncnn: CPU - resnet50 ncnn: CPU - alexnet ncnn: CPU - resnet18 ncnn: CPU - vgg16 ncnn: CPU - googlenet ncnn: CPU - blazeface ncnn: CPU - efficientnet-b0 ncnn: CPU - mnasnet ncnn: CPU - shufflenet-v2 ncnn: CPU-v3-v3 - mobilenet-v3 ncnn: CPU-v2-v2 - mobilenet-v2 ncnn: CPU - mobilenet ncnn: CPU - squeezenet glmark2: 1280 x 1024 glmark2: 1920 x 1080 opencv: DNN - Deep Neural Network ncnn: Vulkan GPU - yolov4-tiny ncnn: Vulkan GPU - resnet50 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - vgg16 ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU - efficientnet-b0 ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU - mobilenet ncnn: Vulkan GPU - squeezenet realsr-ncnn: 4x - No dcraw: RAW To PPM Image Conversion libraw: Post-Processing Benchmark webp: Quality 100, Lossless, Highest Compression mpv: Big Buck Bunny Sunflower 4K - Software Only osbench: Create Processes webp: Quality 100, Lossless tnn: CPU - MobileNet v2 tnn: CPU - SqueezeNet v1.1 mpv: Big Buck Bunny Sunflower 1080p - Software Only lammps: Rhodopsin Protein webp: Quality 100, Highest Compression osbench: Create Files osbench: Memory Allocations osbench: Launch Programs osbench: Create Threads webp: Quality 100 webp: Default 1 2 3 4 4.571 2274229 571253.4 532.993001 518.236 443.805 897546.8 963918.4 283 75.264 13.386 6.843 70.600 12.499 41.484 50.54 56.22 29.08 24.09 111.92 31.08 2.86 14.12 9.91 7.43 9.34 10.77 37.46 30.75 5886 4443 21458 28.52 28.59 12.02 11.55 66.60 20.77 1.78 19.33 8.99 5.68 9.76 8.12 17.86 18.86 58.452 53.807 24.24 49.640 558.92 35.312017 23.735 333.936 304.720 1496.59 4.377 9.456 19.150593 91.620366 57.036877 16.569297 2.947 1.976 4.557 528243.2 532.589945 515.265 443.827 848484.3 899150.3 283 73.795 13.428 6.873 69.864 12.276 43.635 50.46 52.76 29.58 23.57 111.22 29.05 2.74 14.12 9.69 7.38 9.67 10.77 38.03 30.87 5896 4438 20024 28.44 28.45 11.99 11.61 66.55 20.76 1.73 19.18 8.96 5.65 9.76 8.13 17.84 18.83 58.413 53.810 24.29 50.313 559.45 34.882704 23.666 333.412 306.242 1497.78 4.381 9.447 19.078225 91.819684 56.003730 16.277631 2.969 1.975 4.510 507255.7 537.422607 517.886 443.843 764198.1 283 73.968 13.365 6.940 69.864 12.452 38.142 50.44 56.21 29.73 24.98 110.78 29.02 3.29 13.75 9.93 7.37 9.57 11.09 37.54 30.57 5875 4435 21553 28.61 28.46 12.04 11.57 66.55 20.76 1.74 19.36 8.96 5.67 9.78 8.10 17.87 18.86 58.385 54.359 24.19 50.630 558.11 35.650253 23.440 332.492 306.045 1490.85 4.490 9.445 18.783694 91.783682 60.716470 16.246637 2.943 1.967 OpenBenchmarking.org
LAMMPS Molecular Dynamics Simulator Model: 20k Atoms OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 24Aug2020 Model: 20k Atoms 1 2 4 1.0285 2.057 3.0855 4.114 5.1425 SE +/- 0.024, N = 3 SE +/- 0.015, N = 3 SE +/- 0.049, N = 9 4.571 4.557 4.510 1. (CXX) g++ options: -O3 -pthread -lm
Kripke OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.4 1 500K 1000K 1500K 2000K 2500K SE +/- 40613.50, N = 2 2274229 1. (CXX) g++ options: -O3 -fopenmp
InfluxDB Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 1 2 4 120K 240K 360K 480K 600K SE +/- 18598.16, N = 12 SE +/- 21790.68, N = 12 SE +/- 18076.30, N = 12 571253.4 528243.2 507255.7
Incompact3D Input: Cylinder OpenBenchmarking.org Seconds, Fewer Is Better Incompact3D 2020-09-17 Input: Cylinder 1 2 4 120 240 360 480 600 SE +/- 2.71, N = 3 SE +/- 3.03, N = 3 SE +/- 4.14, N = 3 532.99 532.59 537.42 1. (F9X) gfortran options: -cpp -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
GPAW Input: Carbon Nanotube OpenBenchmarking.org Seconds, Fewer Is Better GPAW 20.1 Input: Carbon Nanotube 1 2 4 110 220 330 440 550 SE +/- 1.11, N = 3 SE +/- 1.72, N = 3 SE +/- 0.91, N = 3 518.24 515.27 517.89 1. (CC) gcc options: -pthread -shared -lxc -lblas -lmpi
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes 1 2 4 100 200 300 400 500 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 443.81 443.83 443.84
InfluxDB Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 1 2 4 200K 400K 600K 800K 1000K SE +/- 7142.37, N = 3 SE +/- 14288.50, N = 12 SE +/- 13119.06, N = 12 897546.8 848484.3 764198.1
InfluxDB Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 1 2 200K 400K 600K 800K 1000K SE +/- 4179.09, N = 3 SE +/- 39090.53, N = 13 963918.4 899150.3
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 1 2 4 60 120 180 240 300 SE +/- 0.88, N = 3 283 283 283 1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O3 -O2 -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: inception-v3 1 2 4 20 40 60 80 100 SE +/- 1.22, N = 3 SE +/- 0.11, N = 3 SE +/- 0.47, N = 3 75.26 73.80 73.97 MIN: 70.86 / MAX: 132.02 MIN: 70.29 / MAX: 128.12 MIN: 70.72 / MAX: 107.56 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 1 2 4 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 13.39 13.43 13.37 MIN: 12.62 / MAX: 27.98 MIN: 13.04 / MAX: 29.42 MIN: 10.78 / MAX: 28.87 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: MobileNetV2_224 1 2 4 2 4 6 8 10 SE +/- 0.017, N = 3 SE +/- 0.064, N = 3 SE +/- 0.030, N = 3 6.843 6.873 6.940 MIN: 6.34 / MAX: 22.76 MIN: 6.37 / MAX: 8.93 MIN: 6.37 / MAX: 10.3 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 1 2 4 16 32 48 64 80 SE +/- 0.45, N = 3 SE +/- 0.16, N = 3 SE +/- 0.35, N = 3 70.60 69.86 69.86 MIN: 66.38 / MAX: 106.89 MIN: 67.17 / MAX: 89.11 MIN: 65.78 / MAX: 106.91 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: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2020-09-17 Model: SqueezeNetV1.0 1 2 4 3 6 9 12 15 SE +/- 0.09, N = 3 SE +/- 0.13, N = 3 SE +/- 0.05, N = 3 12.50 12.28 12.45 MIN: 11.62 / MAX: 26.02 MIN: 11.47 / MAX: 30.07 MIN: 11.56 / MAX: 29.12 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
eSpeak-NG Speech Engine Text-To-Speech Synthesis OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis 1 2 4 10 20 30 40 50 SE +/- 0.90, N = 16 SE +/- 1.05, N = 20 SE +/- 0.51, N = 5 41.48 43.64 38.14 1. (CC) gcc options: -O2 -std=c99
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: yolov4-tiny 1 2 4 11 22 33 44 55 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 50.54 50.46 50.44 MIN: 42.79 / MAX: 71.94 MIN: 42.79 / MAX: 79.82 MIN: 43.22 / MAX: 70.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet50 1 2 4 13 26 39 52 65 SE +/- 0.17, N = 3 SE +/- 1.78, N = 3 SE +/- 0.13, N = 3 56.22 52.76 56.21 MIN: 47.73 / MAX: 101.17 MIN: 47.28 / MAX: 120.42 MIN: 47.44 / MAX: 100.48 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 1 2 4 7 14 21 28 35 SE +/- 0.19, N = 3 SE +/- 0.80, N = 3 SE +/- 0.43, N = 3 29.08 29.58 29.73 MIN: 18.15 / MAX: 57.79 MIN: 18.17 / MAX: 59.76 MIN: 18.19 / MAX: 94.45 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet18 1 2 4 6 12 18 24 30 SE +/- 0.60, N = 3 SE +/- 0.91, N = 3 SE +/- 0.66, N = 3 24.09 23.57 24.98 MIN: 19.49 / MAX: 77.09 MIN: 19.44 / MAX: 72.07 MIN: 21.61 / MAX: 75.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: vgg16 1 2 4 30 60 90 120 150 SE +/- 0.91, N = 3 SE +/- 0.30, N = 3 SE +/- 0.98, N = 3 111.92 111.22 110.78 MIN: 87.88 / MAX: 165.26 MIN: 86.34 / MAX: 166.75 MIN: 84.59 / MAX: 174.23 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: googlenet 1 2 4 7 14 21 28 35 SE +/- 0.32, N = 3 SE +/- 1.68, N = 3 SE +/- 1.45, N = 3 31.08 29.05 29.02 MIN: 27.19 / MAX: 97.75 MIN: 23.64 / MAX: 72.29 MIN: 23.86 / MAX: 69.21 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 1 2 4 0.7403 1.4806 2.2209 2.9612 3.7015 SE +/- 0.14, N = 3 SE +/- 0.05, N = 3 SE +/- 0.10, N = 3 2.86 2.74 3.29 MIN: 2.54 / MAX: 18.63 MIN: 2.51 / MAX: 9.36 MIN: 2.55 / MAX: 92 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: efficientnet-b0 1 2 4 4 8 12 16 20 SE +/- 0.38, N = 3 SE +/- 0.45, N = 3 SE +/- 0.14, N = 3 14.12 14.12 13.75 MIN: 12.85 / MAX: 66.8 MIN: 12.97 / MAX: 29.34 MIN: 12.89 / MAX: 86.55 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 1 2 4 3 6 9 12 15 SE +/- 0.18, N = 3 SE +/- 0.26, N = 3 SE +/- 0.24, N = 3 9.91 9.69 9.93 MIN: 9.06 / MAX: 23.96 MIN: 8.77 / MAX: 11.64 MIN: 8.86 / MAX: 45.83 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: shufflenet-v2 1 2 4 2 4 6 8 10 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 7.43 7.38 7.37 MIN: 6.94 / MAX: 8.1 MIN: 6.85 / MAX: 8.04 MIN: 6.87 / MAX: 17.87 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU-v3-v3 - Model: mobilenet-v3 1 2 4 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.12, N = 3 SE +/- 0.37, N = 3 9.34 9.67 9.57 MIN: 8.58 / MAX: 23.21 MIN: 8.5 / MAX: 32.82 MIN: 8.51 / MAX: 64.81 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU-v2-v2 - Model: mobilenet-v2 1 2 4 3 6 9 12 15 SE +/- 0.10, N = 3 SE +/- 0.54, N = 3 SE +/- 0.34, N = 3 10.77 10.77 11.09 MIN: 9.68 / MAX: 26.18 MIN: 9.36 / MAX: 64.52 MIN: 9.74 / MAX: 30.81 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenet 1 2 4 9 18 27 36 45 SE +/- 0.21, N = 3 SE +/- 0.47, N = 3 SE +/- 0.13, N = 3 37.46 38.03 37.54 MIN: 33.76 / MAX: 70.94 MIN: 34.21 / MAX: 102.74 MIN: 34.06 / MAX: 82.61 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 1 2 4 7 14 21 28 35 SE +/- 0.13, N = 3 SE +/- 0.26, N = 3 SE +/- 0.30, N = 3 30.75 30.87 30.57 MIN: 27.03 / MAX: 82.21 MIN: 27.17 / MAX: 74.24 MIN: 26.49 / MAX: 75.98 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
GLmark2 Resolution: 1280 x 1024 OpenBenchmarking.org Score, More Is Better GLmark2 2020.04 Resolution: 1280 x 1024 1 2 4 1300 2600 3900 5200 6500 5886 5896 5875
GLmark2 Resolution: 1920 x 1080 OpenBenchmarking.org Score, More Is Better GLmark2 2020.04 Resolution: 1920 x 1080 1 2 4 1000 2000 3000 4000 5000 4443 4438 4435
OpenCV Test: DNN - Deep Neural Network OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.4 Test: DNN - Deep Neural Network 1 2 4 5K 10K 15K 20K 25K SE +/- 431.59, N = 15 SE +/- 400.07, N = 15 SE +/- 324.81, N = 3 21458 20024 21553 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: yolov4-tiny 1 2 4 7 14 21 28 35 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 28.52 28.44 28.61 MIN: 24.35 / MAX: 33.26 MIN: 24.22 / MAX: 32.89 MIN: 24.14 / MAX: 32.68 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: resnet50 1 2 4 7 14 21 28 35 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 28.59 28.45 28.46 MIN: 26.36 / MAX: 37.83 MIN: 26.33 / MAX: 35.78 MIN: 26.08 / MAX: 35.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: alexnet 1 2 4 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 12.02 11.99 12.04 MIN: 11.21 / MAX: 15.88 MIN: 11.31 / MAX: 15.43 MIN: 11.3 / MAX: 16.33 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: resnet18 1 2 4 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 11.55 11.61 11.57 MIN: 10.44 / MAX: 17.06 MIN: 10.49 / MAX: 17.22 MIN: 10.49 / MAX: 16.83 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: vgg16 1 2 4 15 30 45 60 75 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 66.60 66.55 66.55 MIN: 65.77 / MAX: 69.74 MIN: 65.83 / MAX: 69.88 MIN: 65.79 / MAX: 69.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: googlenet 1 2 4 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 20.77 20.76 20.76 MIN: 19.45 / MAX: 25.92 MIN: 19.53 / MAX: 25.18 MIN: 19.47 / MAX: 25.41 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: blazeface 1 2 4 0.4005 0.801 1.2015 1.602 2.0025 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 1.78 1.73 1.74 MIN: 1.52 / MAX: 4.36 MIN: 1.51 / MAX: 3.65 MIN: 1.54 / MAX: 3.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: efficientnet-b0 1 2 4 5 10 15 20 25 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 19.33 19.18 19.36 MIN: 17.83 / MAX: 25.94 MIN: 17.58 / MAX: 25.38 MIN: 17.63 / MAX: 26.89 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: mnasnet 1 2 4 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 8.99 8.96 8.96 MIN: 8.26 / MAX: 12.37 MIN: 8.14 / MAX: 12.22 MIN: 8.26 / MAX: 11.88 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: shufflenet-v2 1 2 4 1.278 2.556 3.834 5.112 6.39 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 5.68 5.65 5.67 MIN: 4.95 / MAX: 7.6 MIN: 4.79 / MAX: 7.91 MIN: 4.91 / MAX: 7.76 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 1 2 4 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 9.76 9.76 9.78 MIN: 8.77 / MAX: 14 MIN: 8.8 / MAX: 13.68 MIN: 8.84 / MAX: 14.34 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 1 2 4 2 4 6 8 10 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 8.12 8.13 8.10 MIN: 7.45 / MAX: 11.88 MIN: 7.57 / MAX: 11.54 MIN: 7.51 / MAX: 10.82 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: mobilenet 1 2 4 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 SE +/- 0.17, N = 3 17.86 17.84 17.87 MIN: 14.9 / MAX: 24.7 MIN: 14.88 / MAX: 23.2 MIN: 14.98 / MAX: 22.57 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: squeezenet 1 2 4 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 18.86 18.83 18.86 MIN: 17.74 / MAX: 23.12 MIN: 17.59 / MAX: 21.01 MIN: 17.79 / MAX: 22.35 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No 1 2 4 13 26 39 52 65 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 58.45 58.41 58.39
dcraw RAW To PPM Image Conversion OpenBenchmarking.org Seconds, Fewer Is Better dcraw RAW To PPM Image Conversion 1 2 4 12 24 36 48 60 SE +/- 0.02, N = 3 SE +/- 0.29, N = 3 SE +/- 0.12, N = 3 53.81 53.81 54.36 1. (CC) gcc options: -lm
LibRaw Post-Processing Benchmark OpenBenchmarking.org Mpix/sec, More Is Better LibRaw 0.20 Post-Processing Benchmark 1 2 4 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 24.24 24.29 24.19 1. (CXX) g++ options: -O2 -fopenmp -ljpeg -lz -lm
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 1 2 4 11 22 33 44 55 SE +/- 0.30, N = 3 SE +/- 0.17, N = 3 SE +/- 0.35, N = 3 49.64 50.31 50.63 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
MPV Video Input: Big Buck Bunny Sunflower 4K - Decode: Software Only OpenBenchmarking.org FPS, More Is Better MPV Video Input: Big Buck Bunny Sunflower 4K - Decode: Software Only 1 2 4 120 240 360 480 600 SE +/- 0.53, N = 3 SE +/- 0.83, N = 3 SE +/- 0.60, N = 3 558.92 559.45 558.11 MIN: 353.51 / MAX: 741.98 MIN: 355.56 / MAX: 751.21 MIN: 354.93 / MAX: 741.6 1. mpv 0.29.1
OSBench Test: Create Processes OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes 1 2 4 8 16 24 32 40 SE +/- 0.47, N = 15 SE +/- 0.39, N = 15 SE +/- 0.54, N = 15 35.31 34.88 35.65 1. (CC) gcc options: -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 1 2 4 6 12 18 24 30 SE +/- 0.19, N = 3 SE +/- 0.18, N = 3 SE +/- 0.05, N = 3 23.74 23.67 23.44 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 1 2 4 70 140 210 280 350 SE +/- 0.97, N = 3 SE +/- 1.19, N = 3 SE +/- 0.92, N = 3 333.94 333.41 332.49 MIN: 322.28 / MAX: 361.07 MIN: 320.39 / MAX: 360.51 MIN: 320.63 / MAX: 353.14 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -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 1 2 4 70 140 210 280 350 SE +/- 0.98, N = 3 SE +/- 1.73, N = 3 SE +/- 0.40, N = 3 304.72 306.24 306.05 MIN: 294.26 / MAX: 317.02 MIN: 294.08 / MAX: 321.12 MIN: 293.97 / MAX: 314.93 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
MPV Video Input: Big Buck Bunny Sunflower 1080p - Decode: Software Only OpenBenchmarking.org FPS, More Is Better MPV Video Input: Big Buck Bunny Sunflower 1080p - Decode: Software Only 1 2 4 300 600 900 1200 1500 SE +/- 4.23, N = 3 SE +/- 3.04, N = 3 SE +/- 0.82, N = 3 1496.59 1497.78 1490.85 MIN: 890.33 / MAX: 2242.83 MIN: 916.25 / MAX: 2144.91 MIN: 901.12 / MAX: 2234.81 1. mpv 0.29.1
LAMMPS Molecular Dynamics Simulator Model: Rhodopsin Protein OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 24Aug2020 Model: Rhodopsin Protein 1 2 4 1.0103 2.0206 3.0309 4.0412 5.0515 SE +/- 0.048, N = 3 SE +/- 0.068, N = 15 SE +/- 0.036, N = 3 4.377 4.381 4.490 1. (CXX) g++ options: -O3 -pthread -lm
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 1 2 4 3 6 9 12 15 SE +/- 0.030, N = 3 SE +/- 0.026, N = 3 SE +/- 0.016, N = 3 9.456 9.447 9.445 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
OSBench Test: Create Files OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files 1 2 4 5 10 15 20 25 SE +/- 0.22, N = 3 SE +/- 0.15, N = 3 SE +/- 0.06, N = 3 19.15 19.08 18.78 1. (CC) gcc options: -lm
OSBench Test: Memory Allocations OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations 1 2 4 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.21, N = 3 SE +/- 0.38, N = 3 91.62 91.82 91.78 1. (CC) gcc options: -lm
OSBench Test: Launch Programs OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs 1 2 4 14 28 42 56 70 SE +/- 0.36, N = 3 SE +/- 0.70, N = 3 SE +/- 0.74, N = 3 57.04 56.00 60.72 1. (CC) gcc options: -lm
OSBench Test: Create Threads OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads 1 2 4 4 8 12 16 20 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 SE +/- 0.17, N = 3 16.57 16.28 16.25 1. (CC) gcc options: -lm
WebP Image Encode Encode Settings: Quality 100 OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Quality 100 1 2 4 0.668 1.336 2.004 2.672 3.34 SE +/- 0.016, N = 3 SE +/- 0.004, N = 3 SE +/- 0.012, N = 3 2.947 2.969 2.943 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
WebP Image Encode Encode Settings: Default OpenBenchmarking.org Encode Time - Seconds, Fewer Is Better WebP Image Encode 1.1 Encode Settings: Default 1 2 4 0.4446 0.8892 1.3338 1.7784 2.223 SE +/- 0.006, N = 3 SE +/- 0.014, N = 3 SE +/- 0.010, N = 3 1.976 1.975 1.967 1. (CC) gcc options: -fvisibility=hidden -O2 -pthread -lm -ljpeg -lpng16 -ltiff
Phoronix Test Suite v10.8.4