RTX 3080 NN Vulkan AMD Ryzen 9 3950X 16-Core testing with a ASUS ROG CROSSHAIR VIII HERO (WI-FI) (1302 BIOS) and NVIDIA GeForce RTX 3080 10GB on Ubuntu 20.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2010075-PTS-RTX3080N35 .
RTX 3080 NN Vulkan Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL OpenCL Vulkan Compiler File-System Screen Resolution 1 2 3 4 AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR VIII HERO (WI-FI) (1302 BIOS) AMD Starship/Matisse 16GB 2000GB Corsair Force MP600 + 2000GB NVIDIA GeForce RTX 3080 10GB (1710/9501MHz) NVIDIA Device 1aef DELL P2415Q Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 20.04 5.4.0-48-generic (x86_64) GNOME Shell 3.36.4 X Server 1.20.8 NVIDIA 455.23.05 4.6.0 OpenCL 1.2 CUDA 11.1.70 1.2.142 GCC 9.3.0 + CUDA 11.1 ext4 3840x2160 NVIDIA GeForce RTX 3080 10GB (360/5001MHz) NVIDIA GeForce RTX 3080 10GB (1710/9501MHz) NVIDIA GeForce RTX 3080 10GB (375/810MHz) 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 performance - CPU Microcode: 0x8701013 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
RTX 3080 NN Vulkan realsr-ncnn: 4x - No realsr-ncnn: 4x - Yes ncnn: Vulkan GPU - squeezenet ncnn: Vulkan GPU - mobilenet ncnn: Vulkan GPU-v2-v2 - mobilenet-v2 ncnn: Vulkan GPU-v3-v3 - mobilenet-v3 ncnn: Vulkan GPU - shufflenet-v2 ncnn: Vulkan GPU - mnasnet ncnn: Vulkan GPU - efficientnet-b0 ncnn: Vulkan GPU - blazeface ncnn: Vulkan GPU - googlenet ncnn: Vulkan GPU - vgg16 ncnn: Vulkan GPU - resnet18 ncnn: Vulkan GPU - alexnet ncnn: Vulkan GPU - resnet50 ncnn: Vulkan GPU - yolov4-tiny 1 2 3 4 8.792 36.341 3.49 4.26 1.48 1.74 1.33 1.57 2.74 0.58 3.28 4.11 1.25 1.46 3.10 6.99 6.896 34.742 3.48 4.27 1.48 1.75 1.34 1.58 2.73 0.59 3.28 4.10 1.26 1.47 3.09 6.99 6.862 34.688 3.47 4.26 1.47 1.74 1.33 1.57 2.72 0.57 3.26 4.10 1.25 1.46 3.10 6.95 6.920 34.735 3.48 4.27 1.47 1.74 1.33 1.57 2.72 0.57 3.27 4.09 1.25 1.47 3.11 6.96 OpenBenchmarking.org
RealSR-NCNN Scale: 4x - TAA: No OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No 1 2 3 4 2 4 6 8 10 SE +/- 0.311, N = 12 SE +/- 0.039, N = 3 SE +/- 0.044, N = 3 SE +/- 0.045, N = 3 8.792 6.896 6.862 6.920
RealSR-NCNN Scale: 4x - TAA: Yes OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes 1 2 3 4 8 16 24 32 40 SE +/- 0.41, N = 15 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.10, N = 3 36.34 34.74 34.69 34.74
NCNN Target: Vulkan GPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: Vulkan GPU - Model: squeezenet 1 2 3 4 0.7853 1.5706 2.3559 3.1412 3.9265 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 3.49 3.48 3.47 3.48 MIN: 3.36 / MAX: 4.27 MIN: 3.34 / MAX: 3.7 MIN: 3.29 / MAX: 4.32 MIN: 3.34 / MAX: 3.72 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 4 0.9608 1.9216 2.8824 3.8432 4.804 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 4.26 4.27 4.26 4.27 MIN: 4.17 / MAX: 4.66 MIN: 4.16 / MAX: 4.62 MIN: 4.19 / MAX: 4.53 MIN: 4.2 / MAX: 4.69 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 4 0.333 0.666 0.999 1.332 1.665 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.48 1.48 1.47 1.47 MIN: 1.46 / MAX: 2.64 MIN: 1.46 / MAX: 2.51 MIN: 1.46 / MAX: 2.61 MIN: 1.46 / MAX: 2.38 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 3 4 0.3938 0.7876 1.1814 1.5752 1.969 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 1.74 1.75 1.74 1.74 MIN: 1.72 / MAX: 2.53 MIN: 1.72 / MAX: 5.64 MIN: 1.72 / MAX: 2.24 MIN: 1.71 / MAX: 2.91 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 4 0.3015 0.603 0.9045 1.206 1.5075 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.33 1.34 1.33 1.33 MIN: 1.31 / MAX: 1.92 MIN: 1.31 / MAX: 2.49 MIN: 1.31 / MAX: 2.35 MIN: 1.31 / MAX: 2.54 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 4 0.3555 0.711 1.0665 1.422 1.7775 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.57 1.58 1.57 1.57 MIN: 1.56 / MAX: 1.74 MIN: 1.55 / MAX: 2.59 MIN: 1.55 / MAX: 1.77 MIN: 1.56 / MAX: 1.74 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 4 0.6165 1.233 1.8495 2.466 3.0825 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 2.74 2.73 2.72 2.72 MIN: 2.7 / MAX: 7.8 MIN: 2.7 / MAX: 4.03 MIN: 2.69 / MAX: 3.24 MIN: 2.68 / MAX: 4.21 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 4 0.1328 0.2656 0.3984 0.5312 0.664 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.58 0.59 0.57 0.57 MIN: 0.56 / MAX: 0.76 MIN: 0.55 / MAX: 1.31 MIN: 0.55 / MAX: 0.74 MIN: 0.55 / MAX: 0.78 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 4 0.738 1.476 2.214 2.952 3.69 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.28 3.28 3.26 3.27 MIN: 3.24 / MAX: 4.17 MIN: 3.24 / MAX: 3.95 MIN: 3.24 / MAX: 3.66 MIN: 3.24 / MAX: 4.41 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 3 4 0.9248 1.8496 2.7744 3.6992 4.624 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 4.11 4.10 4.10 4.09 MIN: 4.06 / MAX: 7.85 MIN: 4.06 / MAX: 5.4 MIN: 4.04 / MAX: 10.03 MIN: 4.06 / MAX: 6.3 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 4 0.2835 0.567 0.8505 1.134 1.4175 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.25 1.26 1.25 1.25 MIN: 1.23 / MAX: 1.82 MIN: 1.23 / MAX: 2.21 MIN: 1.23 / MAX: 1.54 MIN: 1.23 / MAX: 1.41 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 4 0.3308 0.6616 0.9924 1.3232 1.654 SE +/- 0.01, N = 3 SE +/- 0.01, N = 2 SE +/- 0.00, N = 2 SE +/- 0.01, N = 2 1.46 1.47 1.46 1.47 MIN: 1.44 / MAX: 1.62 MIN: 1.45 / MAX: 1.87 MIN: 1.45 / MAX: 1.62 MIN: 1.44 / MAX: 1.83 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 4 0.6998 1.3996 2.0994 2.7992 3.499 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 3.10 3.09 3.10 3.11 MIN: 3.07 / MAX: 3.63 MIN: 3.07 / MAX: 3.51 MIN: 3.06 / MAX: 4.18 MIN: 3.07 / MAX: 10.12 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 4 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 6.99 6.99 6.95 6.96 MIN: 6.7 / MAX: 11.21 MIN: 6.63 / MAX: 10.85 MIN: 6.65 / MAX: 10.85 MIN: 6.66 / MAX: 19.83 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Phoronix Test Suite v10.8.4