vulkan tests

AMD Ryzen Threadripper 3990X 64-Core testing with a Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS) and AMD Radeon RX 5700 8GB on Ubuntu 23.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2308016-PTS-VULKANTE10&gru&sro.

vulkan testsProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionabcAMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads)Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS)AMD Starship/Matisse128GBSamsung SSD 970 EVO Plus 500GBAMD Radeon RX 5700 8GB (1750/875MHz)AMD Navi 10 HDMI AudioDELL P2415QIntel I211 + Intel Wi-Fi 6 AX200Ubuntu 23.046.2.0-26-generic (x86_64)GNOME Shell 44.0X Server + Wayland4.6 Mesa 23.0.2 (LLVM 15.0.7 DRM 3.49)GCC 12.2.0ext43840x2160OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Pa930Z/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Pa930Z/gcc-12-12.2.0/debian/tmp-gcn/usr --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 schedutil (Boost: Enabled) - CPU Microcode: 0x830107aGraphics Details- BAR1 / Visible vRAM Size: 256 MB - vBIOS Version: 113-D1820201-101Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

vulkan testsvkfft: FFT + iFFT R2C / C2Rvkfft: FFT + iFFT C2C 1D batched in half precisionvkfft: FFT + iFFT C2C Bluestein in single precisionvkfft: FFT + iFFT C2C 1D batched in double precisionvkfft: FFT + iFFT C2C 1D batched in single precisionvkfft: FFT + iFFT C2C multidimensional in single precisionvkfft: FFT + iFFT C2C Bluestein benchmark in double precisionvkfft: FFT + iFFT C2C 1D batched in single precision, no reshufflingvkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarvkpeak: int16-vec4vkresample: 2x - Singlencnn: 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: CPU - vision_transformerncnn: CPU - FastestDetncnn: 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_400mncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - FastestDetabc2552510090372681843158944169933066641407836.617781.347853.6612516.94493.36493.281214.411574.827856.2112531.3512.77025.2113.8312.7314.0813.1318.005.8930.8745.6419.0711.7635.2037.7222.3045.4373.4714.6624.6615.7112.9714.7514.1818.427.2231.4347.5219.2015.2238.9136.7221.8146.7375.1714.252544710106372581842858947169763064641397856.197808.567860.2912550.91493.01492.961213.481574.607852.0312550.3812.77125.7814.4913.4514.3812.9019.697.1231.5547.5319.4013.9739.0337.3822.2748.0374.1113.7025.2514.7813.2514.2713.7219.477.0931.2448.1319.4215.3439.8437.7922.0146.2272.2614.532545110083972591842658938169673064641317866.837808.487866.3312559.31493.51493.571214.401575.917857.0112563.1412.77525.9616.0113.6014.1215.2719.066.9531.6446.8818.9112.3738.1837.1422.8947.0275.9616.0325.3713.6312.5713.2912.9318.026.0831.4647.5019.0413.0738.4337.3621.7846.1072.7013.88OpenBenchmarking.org

VkFFT

Test: FFT + iFFT R2C / C2R

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT R2C / C2Rabc5K10K15K20K25KSE +/- 73.86, N = 3SE +/- 4.33, N = 3SE +/- 2.03, N = 32552525447254511. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in half precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in half precisionabc20K40K60K80K100KSE +/- 98.36, N = 3SE +/- 43.64, N = 3SE +/- 219.62, N = 31009031010631008391. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C Bluestein in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C Bluestein in single precisionabc16003200480064008000SE +/- 15.24, N = 3SE +/- 2.67, N = 3SE +/- 4.70, N = 37268725872591. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in double precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in double precisionabc4K8K12K16K20KSE +/- 6.17, N = 3SE +/- 3.61, N = 3SE +/- 1.86, N = 31843118428184261. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in single precisionabc13K26K39K52K65KSE +/- 1.33, N = 3SE +/- 3.33, N = 3SE +/- 1.20, N = 35894458947589381. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C multidimensional in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C multidimensional in single precisionabc4K8K12K16K20KSE +/- 51.10, N = 3SE +/- 12.01, N = 3SE +/- 9.53, N = 31699316976169671. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C Bluestein benchmark in double precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C Bluestein benchmark in double precisionabc7001400210028003500SE +/- 4.33, N = 3SE +/- 1.00, N = 3SE +/- 1.53, N = 33066306430641. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT C2C 1D batched in single precision, no reshufflingabc14K28K42K56K70KSE +/- 6.43, N = 3SE +/- 5.46, N = 3SE +/- 7.69, N = 36414064139641311. (CXX) g++ options: -O3

vkpeak

fp32-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-scalarabc2K4K6K8K10KSE +/- 0.62, N = 3SE +/- 1.09, N = 3SE +/- 1.93, N = 37836.617856.197866.83

vkpeak

fp32-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-vec4abc2K4K6K8K10KSE +/- 7.57, N = 3SE +/- 7.17, N = 3SE +/- 8.31, N = 37781.347808.567808.48

vkpeak

fp16-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-scalarabc2K4K6K8K10KSE +/- 1.36, N = 3SE +/- 0.75, N = 3SE +/- 1.85, N = 37853.667860.297866.33

vkpeak

fp16-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-vec4abc3K6K9K12K15KSE +/- 1.96, N = 3SE +/- 1.15, N = 3SE +/- 1.25, N = 312516.9412550.9112559.31

vkpeak

fp64-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-scalarabc110220330440550SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3493.36493.01493.51

vkpeak

fp64-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-vec4abc110220330440550SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3493.28492.96493.57

vkpeak

int32-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-scalarabc30060090012001500SE +/- 0.27, N = 3SE +/- 0.05, N = 3SE +/- 0.32, N = 31214.411213.481214.40

vkpeak

int32-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-vec4abc30060090012001500SE +/- 0.17, N = 3SE +/- 0.05, N = 3SE +/- 0.21, N = 31574.821574.601575.91

vkpeak

int16-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-scalarabc2K4K6K8K10KSE +/- 0.44, N = 3SE +/- 0.51, N = 3SE +/- 0.96, N = 37856.217852.037857.01

vkpeak

int16-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-vec4abc3K6K9K12K15KSE +/- 0.58, N = 3SE +/- 0.77, N = 3SE +/- 1.54, N = 312531.3512550.3812563.14

VkResample

Upscale: 2x - Precision: Single

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Singleabc3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 312.7712.7712.781. (CXX) g++ options: -O3

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mobilenetabc612182430SE +/- 0.05, N = 3SE +/- 0.28, N = 3SE +/- 0.21, N = 325.2125.7825.96MIN: 21.72 / MAX: 91.8MIN: 22.57 / MAX: 80.73MIN: 22.99 / MAX: 30.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v2-v2 - Model: mobilenet-v2abc48121620SE +/- 1.66, N = 3SE +/- 1.52, N = 3SE +/- 1.00, N = 313.8314.4916.01MIN: 9.93 / MAX: 25.72MIN: 10.12 / MAX: 23.01MIN: 11.37 / MAX: 25.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v3-v3 - Model: mobilenet-v3abc3691215SE +/- 1.00, N = 3SE +/- 0.98, N = 3SE +/- 0.59, N = 312.7313.4513.60MIN: 9.51 / MAX: 20.96MIN: 9.44 / MAX: 19.16MIN: 9.51 / MAX: 51.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: shufflenet-v2abc48121620SE +/- 1.17, N = 3SE +/- 0.93, N = 3SE +/- 1.04, N = 314.0814.3814.12MIN: 10.99 / MAX: 18.54MIN: 11.05 / MAX: 20.51MIN: 11.12 / MAX: 22.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mnasnetabc48121620SE +/- 1.54, N = 3SE +/- 1.05, N = 3SE +/- 0.65, N = 313.1312.9015.27MIN: 9.27 / MAX: 22.76MIN: 9.19 / MAX: 19.69MIN: 10.33 / MAX: 25.751. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: efficientnet-b0abc510152025SE +/- 1.70, N = 3SE +/- 1.68, N = 3SE +/- 1.07, N = 318.0019.6919.06MIN: 13.54 / MAX: 69.84MIN: 13.76 / MAX: 27.29MIN: 13.85 / MAX: 27.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: blazefaceabc246810SE +/- 0.59, N = 3SE +/- 0.58, N = 3SE +/- 0.44, N = 35.897.126.95MIN: 4.78 / MAX: 9.52MIN: 5.21 / MAX: 13.65MIN: 5.22 / MAX: 9.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: googlenetabc714212835SE +/- 0.21, N = 3SE +/- 0.22, N = 3SE +/- 0.41, N = 330.8731.5531.64MIN: 27.29 / MAX: 90.04MIN: 29.53 / MAX: 39MIN: 26.58 / MAX: 37.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vgg16abc1122334455SE +/- 0.74, N = 3SE +/- 0.34, N = 3SE +/- 0.66, N = 345.6447.5346.88MIN: 42.07 / MAX: 55.26MIN: 44.76 / MAX: 58.97MIN: 43.93 / MAX: 59.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet18abc510152025SE +/- 0.16, N = 3SE +/- 0.21, N = 3SE +/- 0.03, N = 319.0719.4018.91MIN: 16.37 / MAX: 82.83MIN: 16.25 / MAX: 94.41MIN: 16.19 / MAX: 66.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: alexnetabc48121620SE +/- 0.62, N = 3SE +/- 0.45, N = 3SE +/- 0.24, N = 311.7613.9712.37MIN: 9.49 / MAX: 19.16MIN: 11.94 / MAX: 26.76MIN: 10.09 / MAX: 15.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet50abc918273645SE +/- 0.28, N = 3SE +/- 0.53, N = 3SE +/- 1.02, N = 335.2039.0338.18MIN: 30.71 / MAX: 97.13MIN: 34.39 / MAX: 46.86MIN: 32.56 / MAX: 45.611. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: yolov4-tinyabc918273645SE +/- 0.32, N = 3SE +/- 0.58, N = 3SE +/- 0.31, N = 337.7237.3837.14MIN: 34.14 / MAX: 48.53MIN: 33.26 / MAX: 45.88MIN: 33.68 / MAX: 43.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: squeezenet_ssdabc510152025SE +/- 0.47, N = 3SE +/- 0.39, N = 3SE +/- 0.30, N = 322.3022.2722.89MIN: 20.67 / MAX: 28.29MIN: 20.98 / MAX: 27.69MIN: 20.98 / MAX: 29.541. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: regnety_400mabc1122334455SE +/- 1.49, N = 3SE +/- 0.70, N = 3SE +/- 0.33, N = 345.4348.0347.02MIN: 39.85 / MAX: 90.63MIN: 41.67 / MAX: 101.38MIN: 41.08 / MAX: 93.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vision_transformer

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vision_transformerabc20406080100SE +/- 2.54, N = 3SE +/- 2.67, N = 3SE +/- 2.22, N = 373.4774.1175.96MIN: 70.04 / MAX: 83.77MIN: 70.44 / MAX: 113.62MIN: 71.06 / MAX: 93.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: FastestDet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: FastestDetabc48121620SE +/- 1.04, N = 3SE +/- 0.53, N = 3SE +/- 0.49, N = 314.6613.7016.03MIN: 11.69 / MAX: 19.64MIN: 11.46 / MAX: 17.73MIN: 11.61 / MAX: 24.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mobilenetabc612182430SE +/- 0.20, N = 3SE +/- 0.39, N = 12SE +/- 0.04, N = 324.6625.2525.37MIN: 22.49 / MAX: 28.96MIN: 21.74 / MAX: 83.58MIN: 22.3 / MAX: 30.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2abc48121620SE +/- 3.00, N = 3SE +/- 0.85, N = 12SE +/- 0.74, N = 315.7114.7813.63MIN: 10.21 / MAX: 26.19MIN: 10.2 / MAX: 70.67MIN: 10.24 / MAX: 27.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3abc3691215SE +/- 1.36, N = 3SE +/- 0.62, N = 12SE +/- 0.32, N = 312.9713.2512.57MIN: 9.71 / MAX: 20.92MIN: 9.52 / MAX: 53.34MIN: 9.6 / MAX: 17.011. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2abc48121620SE +/- 1.24, N = 3SE +/- 0.46, N = 12SE +/- 0.21, N = 314.7514.2713.29MIN: 11.39 / MAX: 23.44MIN: 10.94 / MAX: 22.32MIN: 11.24 / MAX: 22.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetabc48121620SE +/- 2.23, N = 3SE +/- 0.74, N = 12SE +/- 1.40, N = 314.1813.7212.93MIN: 9.68 / MAX: 23.38MIN: 9.31 / MAX: 23.03MIN: 9.34 / MAX: 22.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0abc510152025SE +/- 0.44, N = 3SE +/- 0.83, N = 12SE +/- 0.52, N = 318.4219.4718.02MIN: 14.24 / MAX: 27.56MIN: 13.59 / MAX: 97.32MIN: 13.61 / MAX: 27.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefaceabc246810SE +/- 0.32, N = 3SE +/- 0.38, N = 12SE +/- 0.20, N = 37.227.096.08MIN: 5.7 / MAX: 10.52MIN: 4.88 / MAX: 78.29MIN: 5.01 / MAX: 7.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetabc714212835SE +/- 0.18, N = 3SE +/- 0.16, N = 12SE +/- 0.10, N = 331.4331.2431.46MIN: 26.65 / MAX: 84.75MIN: 27.01 / MAX: 86.35MIN: 26.64 / MAX: 38.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vgg16abc1122334455SE +/- 0.06, N = 3SE +/- 0.15, N = 12SE +/- 0.08, N = 347.5248.1347.50MIN: 45.43 / MAX: 59.56MIN: 44.54 / MAX: 88.13MIN: 45.06 / MAX: 62.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18abc510152025SE +/- 0.06, N = 3SE +/- 0.09, N = 12SE +/- 0.17, N = 319.2019.4219.04MIN: 16.68 / MAX: 25.67MIN: 16.13 / MAX: 88.96MIN: 16.24 / MAX: 22.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetabc48121620SE +/- 0.36, N = 3SE +/- 0.17, N = 12SE +/- 0.28, N = 315.2215.3413.07MIN: 11.93 / MAX: 21.72MIN: 12.01 / MAX: 62MIN: 10.71 / MAX: 16.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50abc918273645SE +/- 0.49, N = 3SE +/- 0.24, N = 12SE +/- 0.59, N = 338.9139.8438.43MIN: 33.52 / MAX: 45.34MIN: 32.55 / MAX: 100.04MIN: 33.37 / MAX: 44.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinyabc918273645SE +/- 0.55, N = 3SE +/- 0.27, N = 12SE +/- 0.31, N = 336.7237.7937.36MIN: 32.78 / MAX: 42.58MIN: 33.15 / MAX: 109.38MIN: 33.45 / MAX: 45.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdabc510152025SE +/- 0.21, N = 3SE +/- 0.08, N = 12SE +/- 0.20, N = 321.8122.0121.78MIN: 20.68 / MAX: 25.84MIN: 20.55 / MAX: 81.08MIN: 20.62 / MAX: 26.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mabc1122334455SE +/- 0.37, N = 3SE +/- 0.51, N = 12SE +/- 0.73, N = 346.7346.2246.10MIN: 39.79 / MAX: 151.9MIN: 39.54 / MAX: 121.82MIN: 39.52 / MAX: 107.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vision_transformer

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerabc20406080100SE +/- 2.14, N = 3SE +/- 0.80, N = 12SE +/- 1.92, N = 375.1772.2672.70MIN: 69.97 / MAX: 147.06MIN: 69.21 / MAX: 182.16MIN: 69.83 / MAX: 1641. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: FastestDet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetabc48121620SE +/- 1.49, N = 3SE +/- 0.63, N = 12SE +/- 0.66, N = 314.2514.5313.88MIN: 11.35 / MAX: 22.25MIN: 11.05 / MAX: 27.1MIN: 11.52 / MAX: 18.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread


Phoronix Test Suite v10.8.5