3900XT NN AMD Ryzen 9 3900XT 12-Core testing with a MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.94 BIOS) and AMD Radeon RX 56/64 8GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2009250-PTS-3900XTNN87&grs .
3900XT NN Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution 1 2 3 AMD Ryzen 9 3900XT 12-Core @ 3.80GHz (12 Cores / 24 Threads) MSI MEG X570 GODLIKE (MS-7C34) v1.0 (1.94 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 Device 3000 + Intel Wi-Fi 6 AX200 Ubuntu 20.04 5.9.0-050900rc6daily20200922-generic (x86_64) 20200921 GNOME Shell 3.36.4 X Server 1.20.8 amdgpu 19.1.0 4.6 Mesa 20.3.0-devel (git-31f75aa 2020-08-28 focal-oibaf-ppa) (LLVM 10.0.1) GCC 9.3.0 ext4 3840x2160 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 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
3900XT NN ncnn: Vulkan GPU - vgg16 ncnn: CPU - mnasnet ncnn: Vulkan GPU - googlenet realsr-ncnn: 4x - Yes ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 tnn: CPU - SqueezeNet v1.1 ncnn: CPU - efficientnet-b0 ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: CPU - mobilenet ncnn: Vulkan GPU - mobilenet tnn: CPU - MobileNet v2 ncnn: Vulkan GPU - squeezenet ncnn: CPU - resnet18 ncnn: Vulkan GPU - shufflenet-v2 ncnn: CPU - shufflenet-v2 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - mnasnet ncnn: CPU - squeezenet ncnn: CPU - vgg16 ncnn: Vulkan GPU - efficientnet-b0 ncnn: CPU-v2-v2 - mobilenet-v2 realsr-ncnn: 4x - No ncnn: CPU - alexnet ncnn: CPU - yolov4-tiny ncnn: Vulkan GPU - yolov4-tiny ncnn: CPU - googlenet ncnn: CPU - resnet50 ncnn: CPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU - resnet18 ncnn: CPU - blazeface ncnn: Vulkan GPU - resnet50 ncnn: Vulkan GPU - blazeface 1 2 3 11.95 4.76 5.78 63.327 3.55 230.022 6.54 2.52 16.33 7.77 245.154 4.66 16.20 2.28 4.82 3.92 2.69 15.96 67.41 10.41 5.36 9.845 16.13 28.15 10.70 17.17 27.55 4.71 2.10 1.91 6.02 0.90 11.46 4.80 5.66 64.551 3.50 228.479 6.58 2.49 16.17 7.72 246.961 4.68 16.36 2.28 4.78 3.94 2.67 15.87 67.82 10.47 5.33 9.797 16.18 28.07 10.68 17.18 27.56 4.71 2.1 1.91 6.38 0.87 11.69 4.88 5.71 63.569 3.49 226.902 6.62 2.49 16.36 7.68 244.143 4.71 16.27 2.26 4.80 3.91 2.68 15.97 67.67 10.44 5.35 9.838 16.18 28.08 10.71 17.14 27.50 4.70 2.10 1.91 6.13 0.85 OpenBenchmarking.org
NCNN Target: Vulkan GPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: vgg16 1 2 3 3 6 9 12 15 SE +/- 0.36, N = 3 SE +/- 0.27, N = 3 SE +/- 0.30, N = 3 11.95 11.46 11.69 MIN: 9.6 / MAX: 31.99 MIN: 9.59 / MAX: 31.3 MIN: 9.56 / MAX: 36.67 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 3 1.098 2.196 3.294 4.392 5.49 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.11, N = 3 4.76 4.80 4.88 MIN: 4.68 / MAX: 6.21 MIN: 4.67 / MAX: 6.45 MIN: 4.69 / MAX: 47.76 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 3 1.3005 2.601 3.9015 5.202 6.5025 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 5.78 5.66 5.71 MIN: 5.53 / MAX: 23.31 MIN: 5.53 / MAX: 17.32 MIN: 5.52 / MAX: 13.35 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes 1 2 3 14 28 42 56 70 SE +/- 0.49, N = 3 SE +/- 0.91, N = 3 SE +/- 0.78, N = 3 63.33 64.55 63.57
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 3 0.7988 1.5976 2.3964 3.1952 3.994 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 3.55 3.50 3.49 MIN: 3.47 / MAX: 18.04 MIN: 3.47 / MAX: 4.03 MIN: 3.46 / MAX: 3.99 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
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 3 50 100 150 200 250 SE +/- 3.76, N = 3 SE +/- 0.04, N = 3 SE +/- 0.19, N = 3 230.02 228.48 226.90 MIN: 222.17 / MAX: 240.13 MIN: 225.95 / MAX: 229.21 MIN: 224.24 / MAX: 228.35 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: efficientnet-b0 1 2 3 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 6.54 6.58 6.62 MIN: 6.42 / MAX: 13.95 MIN: 6.45 / MAX: 11.26 MIN: 6.41 / MAX: 8.49 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 3 0.567 1.134 1.701 2.268 2.835 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 2.52 2.49 2.49 MIN: 2.47 / MAX: 3.46 MIN: 2.47 / MAX: 2.85 MIN: 2.46 / MAX: 2.84 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 3 4 8 12 16 20 SE +/- 0.16, N = 3 SE +/- 0.13, N = 3 SE +/- 0.12, N = 3 16.33 16.17 16.36 MIN: 15.8 / MAX: 17.16 MIN: 15.85 / MAX: 16.8 MIN: 16.1 / MAX: 16.93 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 3 2 4 6 8 10 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 7.77 7.72 7.68 MIN: 5.52 / MAX: 11.8 MIN: 5.6 / MAX: 8.83 MIN: 5.58 / MAX: 11.76 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
TNN Target: CPU - Model: MobileNet v2 OpenBenchmarking.org ms, Fewer Is Better TNN 0.2.3 Target: CPU - Model: MobileNet v2 1 2 3 50 100 150 200 250 SE +/- 1.13, N = 3 SE +/- 1.34, N = 3 SE +/- 1.92, N = 3 245.15 246.96 244.14 MIN: 233.67 / MAX: 275.36 MIN: 233.69 / MAX: 262.21 MIN: 232.48 / MAX: 272.94 1. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
NCNN Target: Vulkan GPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: squeezenet 1 2 3 1.0598 2.1196 3.1794 4.2392 5.299 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 4.66 4.68 4.71 MIN: 4.48 / MAX: 6.06 MIN: 4.51 / MAX: 6.25 MIN: 4.51 / MAX: 6.36 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 3 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 16.20 16.36 16.27 MIN: 16.02 / MAX: 21.18 MIN: 16.04 / MAX: 27.68 MIN: 16.02 / MAX: 16.66 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 3 0.513 1.026 1.539 2.052 2.565 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 2.28 2.28 2.26 MIN: 2.25 / MAX: 3.05 MIN: 2.24 / MAX: 6.97 MIN: 2.24 / MAX: 2.82 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 3 1.0845 2.169 3.2535 4.338 5.4225 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 4.82 4.78 4.80 MIN: 4.77 / MAX: 5.98 MIN: 4.61 / MAX: 6.32 MIN: 4.72 / MAX: 5.75 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 3 0.8865 1.773 2.6595 3.546 4.4325 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 3.92 3.94 3.91 MIN: 3.79 / MAX: 4.97 MIN: 3.8 / MAX: 5.12 MIN: 3.8 / MAX: 4.91 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 3 0.6053 1.2106 1.8159 2.4212 3.0265 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 2.69 2.67 2.68 MIN: 2.64 / MAX: 3.72 MIN: 2.64 / MAX: 3.26 MIN: 2.64 / MAX: 3.71 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 3 4 8 12 16 20 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 15.96 15.87 15.97 MIN: 15.54 / MAX: 18.64 MIN: 15.46 / MAX: 16.57 MIN: 15.61 / MAX: 16.77 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 3 15 30 45 60 75 SE +/- 0.12, N = 3 SE +/- 0.29, N = 3 SE +/- 0.16, N = 3 67.41 67.82 67.67 MIN: 66.51 / MAX: 110.23 MIN: 66.53 / MAX: 109.38 MIN: 66.42 / MAX: 140.75 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 3 3 6 9 12 15 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 SE +/- 0.08, N = 3 10.41 10.47 10.44 MIN: 8.75 / MAX: 26.47 MIN: 8.76 / MAX: 24.1 MIN: 8.74 / MAX: 23.62 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 3 1.206 2.412 3.618 4.824 6.03 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 5.36 5.33 5.35 MIN: 5.24 / MAX: 7.13 MIN: 5.13 / MAX: 6.51 MIN: 5.21 / MAX: 7.22 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 3 3 6 9 12 15 SE +/- 0.103, N = 3 SE +/- 0.042, N = 3 SE +/- 0.038, N = 3 9.845 9.797 9.838
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: alexnet 1 2 3 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 16.13 16.18 16.18 MIN: 15.99 / MAX: 18.49 MIN: 16.03 / MAX: 16.44 MIN: 16.03 / MAX: 16.91 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: yolov4-tiny 1 2 3 7 14 21 28 35 SE +/- 0.09, N = 3 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 28.15 28.07 28.08 MIN: 27.83 / MAX: 28.94 MIN: 27.62 / MAX: 37.12 MIN: 27.74 / MAX: 30.21 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: Vulkan GPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: yolov4-tiny 1 2 3 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 2 SE +/- 0.02, N = 3 10.70 10.68 10.71 MIN: 10.47 / MAX: 11.01 MIN: 10.49 / MAX: 10.95 MIN: 10.45 / MAX: 11.19 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 3 4 8 12 16 20 SE +/- 0.20, N = 3 SE +/- 0.16, N = 3 SE +/- 0.15, N = 3 17.17 17.18 17.14 MIN: 16.63 / MAX: 121.77 MIN: 16.59 / MAX: 19.11 MIN: 16.62 / MAX: 22.64 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 3 6 12 18 24 30 SE +/- 0.08, N = 3 SE +/- 0.27, N = 3 SE +/- 0.26, N = 3 27.55 27.56 27.50 MIN: 27.15 / MAX: 40.33 MIN: 27.02 / MAX: 33.89 MIN: 27.04 / MAX: 38.34 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 3 1.0598 2.1196 3.1794 4.2392 5.299 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 4.71 4.71 4.70 MIN: 4.64 / MAX: 6.36 MIN: 4.56 / MAX: 7 MIN: 4.65 / MAX: 6.27 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 3 0.4725 0.945 1.4175 1.89 2.3625 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 2.10 2.10 2.10 MIN: 2.01 / MAX: 2.81 MIN: 2.01 / MAX: 2.87 MIN: 2.01 / MAX: 2.88 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 3 0.4298 0.8596 1.2894 1.7192 2.149 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 1.91 1.91 1.91 MIN: 1.87 / MAX: 2.43 MIN: 1.86 / MAX: 2.01 MIN: 1.85 / MAX: 2.6 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 3 2 4 6 8 10 SE +/- 0.03, N = 3 SE +/- 0.32, N = 3 SE +/- 0.14, N = 3 6.02 6.38 6.13 MIN: 5.83 / MAX: 13.11 MIN: 5.85 / MAX: 22.05 MIN: 5.85 / MAX: 23.94 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 3 0.2025 0.405 0.6075 0.81 1.0125 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.90 0.87 0.85 MIN: 0.84 / MAX: 1.56 MIN: 0.84 / MAX: 1.67 MIN: 0.83 / MAX: 1.36 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Phoronix Test Suite v10.8.5