vulkan rem

AMD Ryzen 7 PRO 6850U testing with a LENOVO 21CM0001US (R22ET51W 1.21 BIOS) and AMD Radeon 680M 1GB on Ubuntu 23.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2308012-NE-VULKANREM18&grs.

vulkan remProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionabcAMD Ryzen 7 PRO 6850U @ 2.70GHz (8 Cores / 16 Threads)LENOVO 21CM0001US (R22ET51W 1.21 BIOS)AMD 17h-19h PCIe Root Complex16GB512GB Micron MTFDKBA512TFKAMD Radeon 680M 1GB (2200/400MHz)AMD Rembrandt Radeon HD AudioQualcomm QCNFA765Ubuntu 23.046.2.0-23-generic (x86_64)GNOME Shell 44.0X Server + Wayland4.6 Mesa 23.0.2 (LLVM 15.0.7 DRM 3.49)GCC 12.2.0ext41920x1200OpenBenchmarking.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) - Platform Profile: balanced - CPU Microcode: 0xa404102 - ACPI Profile: balanced Graphics Details- BAR1 / Visible vRAM Size: 1024 MB - vBIOS Version: 113-REMBRANDT-036Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + 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 IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

vulkan remvkpeak: fp32-scalarvkpeak: fp16-vec4vkpeak: int32-scalarvkpeak: fp16-scalarvkpeak: int16-scalarvkpeak: int16-vec4vkpeak: int32-vec4ncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - mnasnetncnn: CPU - blazefacencnn: Vulkan GPU - efficientnet-b0ncnn: CPU - shufflenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: CPU - FastestDetvkpeak: fp64-vec4ncnn: CPU - mnasnetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - mobilenetncnn: CPU - resnet18ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - regnety_400mncnn: CPU - efficientnet-b0ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - alexnetncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet50ncnn: CPU - googlenetncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - yolov4-tinyvkpeak: fp64-scalarncnn: CPU - squeezenet_ssdvkresample: 2x - Singlencnn: CPU - resnet50vkfft: FFT + iFFT C2C 1D batched in half precisionncnn: CPU - yolov4-tinyncnn: CPU - mobilenetncnn: CPU - vision_transformerncnn: Vulkan GPU - vision_transformerncnn: CPU - vgg16ncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - vgg16vkfft: FFT + iFFT R2C / C2Rncnn: Vulkan GPU - FastestDetvkpeak: fp32-vec4abc2944.083807.71524.702642.572593.433663.43574.370.686.231.959.672.620.684.821.932.612.59211.332.613.1112.967.253.136.324.762.557.527.5716.919.787.1722.16212.108.4856.22416.941994321.9612.84126.10126.5150.178.4250.2462832.702243.243103.054015.65529.132740.422710.103870.38562.500.696.372.009.762.670.694.831.962.622.56211.422.623.1512.837.273.146.354.802.597.497.4617.079.827.1322.00211.748.4456.46216.911981422.0912.81126.33126.3450.328.4350.2862832.692351.213197.24177.12579.332787.132745.023988.17597.520.656.031.99.282.540.664.651.892.532.63211.792.563.0812.687.123.086.244.722.557.417.4616.849.697.0821.92212.18.4156.00216.831990421.9612.77125.84126.0450.158.4450.2262822.852617.93OpenBenchmarking.org

vkpeak

fp32-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-scalarabc7001400210028003500SE +/- 14.11, N = 3SE +/- 23.68, N = 32718.373103.053197.20

vkpeak

fp16-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-vec4abc9001800270036004500SE +/- 6.67, N = 3SE +/- 8.25, N = 33644.384015.654177.12

vkpeak

int32-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-scalarabc130260390520650SE +/- 0.31, N = 3SE +/- 5.09, N = 3508.88529.13579.33

vkpeak

fp16-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp16-scalarabc6001200180024003000SE +/- 3.99, N = 3SE +/- 5.74, N = 32479.902740.422787.13

vkpeak

int16-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-scalarabc6001200180024003000SE +/- 1.78, N = 3SE +/- 2.74, N = 32469.932710.102745.02

vkpeak

int16-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int16-vec4abc9001800270036004500SE +/- 3.59, N = 3SE +/- 1.88, N = 33628.613870.383988.17

vkpeak

int32-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20230730int32-vec4abc130260390520650SE +/- 0.23, N = 3SE +/- 0.16, N = 3572.29562.50597.52

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefaceabc0.15530.31060.46590.62120.7765SE +/- 0.01, N = 3SE +/- 0.01, N = 30.680.690.65MIN: 0.65 / MAX: 1.07MIN: 0.65 / MAX: 1.64MIN: 0.62 / MAX: 0.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mabc246810SE +/- 0.04, N = 3SE +/- 0.03, N = 36.236.376.03MIN: 5.87 / MAX: 8.11MIN: 5.88 / MAX: 8.18MIN: 5.54 / MAX: 7.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2abc0.450.91.351.82.25SE +/- 0.01, N = 3SE +/- 0.02, N = 31.952.001.90MIN: 1.85 / MAX: 3MIN: 1.67 / MAX: 3.67MIN: 1.84 / MAX: 2.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetabc3691215SE +/- 0.09, N = 3SE +/- 0.10, N = 39.679.769.28MIN: 9.02 / MAX: 11.84MIN: 9.08 / MAX: 20.73MIN: 8.96 / MAX: 11.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetabc0.60081.20161.80242.40323.004SE +/- 0.01, N = 3SE +/- 0.03, N = 32.622.672.54MIN: 2.35 / MAX: 3.91MIN: 2.51 / MAX: 4.2MIN: 2.43 / MAX: 3.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: blazefaceabc0.15530.31060.46590.62120.7765SE +/- 0.02, N = 3SE +/- 0.01, N = 30.680.690.66MIN: 0.61 / MAX: 1.04MIN: 0.62 / MAX: 1.56MIN: 0.58 / MAX: 1.591. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0abc1.08682.17363.26044.34725.434SE +/- 0.03, N = 3SE +/- 0.03, N = 34.824.834.65MIN: 4.4 / MAX: 18.07MIN: 4.39 / MAX: 7.05MIN: 4.49 / MAX: 6.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: shufflenet-v2abc0.4410.8821.3231.7642.205SE +/- 0.01, N = 3SE +/- 0.02, N = 31.931.961.89MIN: 1.78 / MAX: 3.55MIN: 1.76 / MAX: 2.37MIN: 1.75 / MAX: 2.771. (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-v3abc0.58951.1791.76852.3582.9475SE +/- 0.02, N = 3SE +/- 0.02, N = 32.612.622.53MIN: 2.39 / MAX: 14.7MIN: 2.46 / MAX: 4.3MIN: 2.32 / MAX: 3.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: FastestDet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: FastestDetabc0.59181.18361.77542.36722.959SE +/- 0.02, N = 32.592.562.63MIN: 2.42 / MAX: 4.18MIN: 2.44 / MAX: 3MIN: 2.54 / MAX: 3.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

vkpeak

fp64-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-vec4abc50100150200250SE +/- 1.84, N = 3SE +/- 0.01, N = 3206.80211.42211.79

NCNN

Target: CPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mnasnetabc0.58951.1791.76852.3582.9475SE +/- 0.01, N = 3SE +/- 0.02, N = 32.612.622.56MIN: 2.4 / MAX: 13.39MIN: 2.41 / MAX: 4.05MIN: 2.43 / MAX: 3.861. (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-v2abc0.70881.41762.12642.83523.544SE +/- 0.01, N = 3SE +/- 0.03, N = 33.113.153.08MIN: 2.87 / MAX: 4.89MIN: 2.86 / MAX: 4.58MIN: 2.8 / MAX: 4.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mobilenetabc3691215SE +/- 0.17, N = 3SE +/- 0.01, N = 312.9612.8312.68MIN: 12.39 / MAX: 63.67MIN: 12.51 / MAX: 14.26MIN: 12.47 / MAX: 13.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet18abc246810SE +/- 0.03, N = 3SE +/- 0.06, N = 37.257.277.12MIN: 6.86 / MAX: 18.35MIN: 6.79 / MAX: 8.94MIN: 6.81 / MAX: 9.181. (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-v2abc0.70651.4132.11952.8263.5325SE +/- 0.01, N = 3SE +/- 0.02, N = 33.133.143.08MIN: 2.84 / MAX: 4.95MIN: 2.82 / MAX: 6.88MIN: 2.87 / MAX: 5.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: regnety_400mabc246810SE +/- 0.06, N = 3SE +/- 0.07, N = 36.326.356.24MIN: 5.79 / MAX: 8.1MIN: 5.75 / MAX: 8.02MIN: 5.75 / MAX: 17.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: efficientnet-b0abc1.082.163.244.325.4SE +/- 0.01, N = 3SE +/- 0.05, N = 34.764.804.72MIN: 4.52 / MAX: 6.74MIN: 4.43 / MAX: 16.71MIN: 4.48 / MAX: 6.951. (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-v3abc0.58281.16561.74842.33122.914SE +/- 0.01, N = 3SE +/- 0.02, N = 32.552.592.55MIN: 2.33 / MAX: 4.5MIN: 2.37 / MAX: 4.44MIN: 2.36 / MAX: 4.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: alexnetabc246810SE +/- 0.02, N = 3SE +/- 0.01, N = 37.527.497.41MIN: 7.23 / MAX: 18.52MIN: 7.23 / MAX: 9.71MIN: 7.16 / MAX: 8.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetabc246810SE +/- 0.01, N = 3SE +/- 0.04, N = 37.577.467.46MIN: 7.1 / MAX: 9.67MIN: 7.1 / MAX: 10.13MIN: 7.2 / MAX: 9.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50abc48121620SE +/- 0.02, N = 3SE +/- 0.07, N = 316.9117.0716.84MIN: 16.43 / MAX: 27.99MIN: 16.47 / MAX: 28.45MIN: 16.47 / MAX: 19.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: googlenetabc3691215SE +/- 0.13, N = 3SE +/- 0.12, N = 39.789.829.69MIN: 8.97 / MAX: 12.44MIN: 9.05 / MAX: 20.33MIN: 9.1 / MAX: 14.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18abc246810SE +/- 0.12, N = 3SE +/- 0.07, N = 37.177.137.08MIN: 6.62 / MAX: 9.23MIN: 6.57 / MAX: 9.32MIN: 6.61 / MAX: 11.11. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinyabc510152025SE +/- 0.14, N = 3SE +/- 0.05, N = 322.1622.0021.92MIN: 21.38 / MAX: 32.76MIN: 21.31 / MAX: 25.03MIN: 21.18 / MAX: 24.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

vkpeak

fp64-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp64-scalarabc50100150200250SE +/- 0.31, N = 3SE +/- 0.02, N = 3209.84211.74212.10

NCNN

Target: CPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: squeezenet_ssdabc246810SE +/- 0.05, N = 3SE +/- 0.00, N = 38.488.448.41MIN: 7.97 / MAX: 10.91MIN: 8.13 / MAX: 10.76MIN: 8.13 / MAX: 10.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VkResample

Upscale: 2x - Precision: Single

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Singleabc1326395265SE +/- 0.11, N = 3SE +/- 0.06, N = 356.2256.4656.001. (CXX) g++ options: -O3

NCNN

Target: CPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet50abc48121620SE +/- 0.06, N = 3SE +/- 0.05, N = 316.9416.9116.83MIN: 16.46 / MAX: 19.7MIN: 16.24 / MAX: 28.07MIN: 16.52 / MAX: 17.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 precisionabc4K8K12K16K20KSE +/- 100.68, N = 3SE +/- 39.59, N = 31994319814199041. (CXX) g++ options: -O3

NCNN

Target: CPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: yolov4-tinyabc510152025SE +/- 0.04, N = 3SE +/- 0.03, N = 321.9622.0921.96MIN: 21.38 / MAX: 24.78MIN: 21.39 / MAX: 34.39MIN: 21.3 / MAX: 22.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mobilenetabc3691215SE +/- 0.07, N = 3SE +/- 0.05, N = 312.8412.8112.77MIN: 12.39 / MAX: 24.16MIN: 12.4 / MAX: 24.01MIN: 12.53 / MAX: 13.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vision_transformer

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vision_transformerabc306090120150SE +/- 0.07, N = 3SE +/- 0.11, N = 3126.10126.33125.84MIN: 122.74 / MAX: 136.47MIN: 123.54 / MAX: 135.11MIN: 123.61 / MAX: 158.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vision_transformer

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerabc306090120150SE +/- 0.07, N = 3SE +/- 0.19, N = 3126.51126.34126.04MIN: 123.01 / MAX: 136.01MIN: 123.22 / MAX: 135.57MIN: 123.54 / MAX: 133.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: CPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vgg16abc1122334455SE +/- 0.05, N = 3SE +/- 0.14, N = 350.1750.3250.15MIN: 49.14 / MAX: 61.46MIN: 48.97 / MAX: 62.15MIN: 49.73 / MAX: 62.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdabc246810SE +/- 0.02, N = 3SE +/- 0.01, N = 38.428.438.44MIN: 8.05 / MAX: 10.16MIN: 8.12 / MAX: 11.62MIN: 8.08 / MAX: 19.621. (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 = 350.2450.2850.22MIN: 49.26 / MAX: 62.9MIN: 49.14 / MAX: 62.5MIN: 49.37 / MAX: 63.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VkFFT

Test: FFT + iFFT R2C / C2R

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.2.31Test: FFT + iFFT R2C / C2Rabc13002600390052006500SE +/- 14.84, N = 3SE +/- 22.70, N = 36283628362821. (CXX) g++ options: -O3

NCNN

Target: Vulkan GPU - Model: FastestDet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetabc0.64131.28261.92392.56523.2065SE +/- 0.09, N = 3SE +/- 0.12, N = 32.702.692.85MIN: 2.48 / MAX: 5.38MIN: 2.46 / MAX: 14.64MIN: 2.76 / MAX: 3.261. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

vkpeak

fp32-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-vec4abc6001200180024003000SE +/- 30.23, N = 3SE +/- 123.18, N = 32246.902351.212617.93

Phoronix Test Suite v10.8.5