ncnnxa

ARMv8 rev 0 testing with a Jetson-AGX and NVIDIA Tegra Xavier on Ubuntu 18.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2103010-HA-NCNNXA16277
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table
Only show results matching title/arguments (delimit multiple options with a comma):


ncnnxaOpenBenchmarking.orgPhoronix Test Suite 10.4.0m1ARMv8 rev 0 @ 1.19GHz (8 Cores)Jetson-AGX32GB31GB HBG4a2NVIDIA Tegra XavierPL2493HUbuntu 18.044.9.140-tegra (aarch64)GNOME Shell 3.28.4X Server 1.19.6NVIDIA 32.4.44.6.01.2.131GCC 7.5.0 + CUDA 10.2ext41920x1080ProcessorMotherboardMemoryDiskGraphicsMonitorOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionNcnnxa BenchmarksSystem Logs- Transparent Huge Pages: always- --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only -v - Scaling Governor: tegra_cpufreq schedutil

ncnnxancnn: CPU - mobilenetncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - shufflenet-v2ncnn: CPU - mnasnetncnn: CPU - efficientnet-b0ncnn: CPU - blazefacencnn: CPU - googlenetncnn: CPU - vgg16ncnn: CPU - resnet18ncnn: CPU - alexnetncnn: CPU - resnet50ncnn: CPU - yolov4-tinyncnn: CPU - squeezenet_ssdncnn: CPU - regnety_400mncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - googlenetncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - regnety_400mARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX67.2926.3925.9226.6727.9356.3013.7759.08116.2038.3432.61114.3884.7581.11147.2561.9922.4222.9322.9623.4234.3812.6855.65107.037.2333.1792.3782.4067.02147.30OpenBenchmarking.org

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX1530456075SE +/- 2.85, N = 667.29MIN: 49.46 / MAX: 324.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX612182430SE +/- 2.31, N = 626.39MIN: 13.89 / MAX: 203.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX612182430SE +/- 1.63, N = 625.92MIN: 14.8 / MAX: 267.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX612182430SE +/- 1.72, N = 626.67MIN: 15.07 / MAX: 248.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX714212835SE +/- 2.52, N = 627.93MIN: 16.59 / MAX: 288.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX1326395265SE +/- 22.75, N = 656.30MIN: 24.19 / MAX: 576.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX48121620SE +/- 0.79, N = 613.77MIN: 8.93 / MAX: 97.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX1326395265SE +/- 3.02, N = 659.08MIN: 37.45 / MAX: 154.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX306090120150SE +/- 8.89, N = 6116.20MIN: 87.85 / MAX: 333.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX918273645SE +/- 2.06, N = 638.34MIN: 22.64 / MAX: 85.161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX816243240SE +/- 1.36, N = 632.61MIN: 22.44 / MAX: 60.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX306090120150SE +/- 20.79, N = 6114.38MIN: 65.79 / MAX: 551.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX20406080100SE +/- 1.80, N = 684.75MIN: 60.43 / MAX: 236.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX20406080100SE +/- 11.20, N = 681.11MIN: 46.11 / MAX: 486.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400mARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX306090120150SE +/- 5.46, N = 6147.25MIN: 99.03 / MAX: 513.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX1428425670SE +/- 0.43, N = 361.99MIN: 49.37 / MAX: 135.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX510152025SE +/- 0.50, N = 322.42MIN: 13.92 / MAX: 78.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX510152025SE +/- 0.43, N = 322.93MIN: 15.95 / MAX: 91.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v2ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX612182430SE +/- 0.83, N = 322.96MIN: 18.21 / MAX: 143.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX612182430SE +/- 0.37, N = 323.42MIN: 19.93 / MAX: 86.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b0ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX816243240SE +/- 0.48, N = 334.38MIN: 28.79 / MAX: 101.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazefaceARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX3691215SE +/- 0.12, N = 312.68MIN: 9.9 / MAX: 39.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX1326395265SE +/- 0.66, N = 355.65MIN: 39.62 / MAX: 158.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX20406080100SE +/- 3.42, N = 3107.0MIN: 90.69 / MAX: 209.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX918273645SE +/- 0.40, N = 337.23MIN: 30.2 / MAX: 74.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnetARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX816243240SE +/- 0.20, N = 333.17MIN: 28.12 / MAX: 54.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet50ARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX20406080100SE +/- 1.55, N = 392.37MIN: 67.3 / MAX: 223.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tinyARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX20406080100SE +/- 0.67, N = 382.40MIN: 73.67 / MAX: 134.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssdARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX1530456075SE +/- 0.17, N = 367.02MIN: 57.19 / MAX: 154.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400mARMv8 rev 0 - NVIDIA Tegra Xavier - Jetson-AGX306090120150SE +/- 1.08, N = 3147.30MIN: 95.5 / MAX: 420.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread