VkFFT + Caffe + Other GPU Tests - AMD vs. NVIDIA

AMD Ryzen 9 3950X 16-Core testing with a ASUS ROG CROSSHAIR VIII HERO (WI-FI) (1302 BIOS) and Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2010013-PTS-VKFFTAMD84&sor.

VkFFT + Caffe + Other GPU Tests - AMD vs. NVIDIAProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionGTX 1060GTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTXRX Vega 56RX 5600 XTRX 5700RX 5700 XTRadeon VIIAMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (1302 BIOS)AMD Starship/Matisse16GB2000GB Corsair Force MP600 + 2000GBNVIDIA GeForce GTX 1060 6GB (1506/4006MHz)NVIDIA GP106 HD AudioDELL P2415QRealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.045.4.0-48-generic (x86_64)GNOME Shell 3.36.4X Server 1.20.8NVIDIA 450.664.6.0OpenCL 2.0 AMD-APP (3182.0) + OpenCL 1.2 CUDA 11.0.2281.2.133GCC 9.3.0 + CUDA 11.0ext43840x2160NVIDIA GeForce GTX 1080 8GB (1607/5005MHz)NVIDIA GP104 HD AudioASUS NVIDIA GeForce GTX 1650 4GB (525/405MHz)NVIDIA Device 10faASUS NVIDIA GeForce GTX 1650 SUPER 4GB (1530/6000MHz)NVIDIA TU116 HD AudioASUS NVIDIA GeForce GTX 1660 6GB (405/405MHz)eVGA NVIDIA GeForce GTX 1660 SUPER 6GB (435/405MHz)eVGA NVIDIA GeForce GTX 1660 Ti 6GB (390/405MHz)NVIDIA GeForce RTX 2060 6GB (435/405MHz)NVIDIA TU106 HD AudioNVIDIA GeForce RTX 2060 SUPER 8GB (1470/7000MHz)ASUS NVIDIA GeForce RTX 2070 8GB (420/405MHz)NVIDIA GeForce RTX 2070 SUPER 8GB (1605/7000MHz)NVIDIA TU104 HD AudioZotac NVIDIA GeForce RTX 2080 8GB (1515/7000MHz)NVIDIA GeForce RTX 2080 SUPER 8GB (390/405MHz)NVIDIA GeForce RTX 2080 Ti 11GB (420/405MHz)NVIDIA TU102 HD AudioNVIDIA TITAN RTX 24GB (420/405MHz)AMD Radeon RX 56/64 8GB (1590/800MHz)AMD Vega 10 HDMI Audio5.9.0-050900rc7daily20201001-generic (x86_64) 202009304.6 Mesa 20.3.0-devel (git-3173367 2020-09-25 focal-oibaf-ppa) (LLVM 10.0.1)OpenCL 2.0 AMD-APP (3182.0)1.2.145Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz)AMD Navi 10 HDMI AudioAMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz)AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (2100/875MHz)AMD Radeon VII 16GB (1801/1000MHz)AMD Vega 20 HDMI AudioOpenBenchmarking.orgCompiler 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 -vProcessor Details- Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8701013Python Details- GTX 1060, GTX 1080, GTX 1650, GTX 1650 SUPER, GTX 1660, GTX 1660 SUPER, GTX 1660 Ti, RTX 2060, RTX 2060 SUPER, RTX 2070, RTX 2070 SUPER, RTX 2080, RTX 2080 SUPER, RTX 2080 Ti, TITAN RTX, RX Vega 56, RX 5600 XT, RX 5700, RX 5700 XT: Python 3.8.2Security 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

VkFFT + Caffe + Other GPU Tests - AMD vs. NVIDIAvkfft: realsr-ncnn: 4x - Norealsr-ncnn: 4x - Yesglmark2: 1920 x 1080glmark2: 2560 x 1440glmark2: 3840 x 2160ncnn: Vulkan GPU - squeezenetncnn: 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-tinyGTX 1060GTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTXRX Vega 56RX 5600 XTRX 5700RX 5700 XTRadeon VII129162088794951307313765187331821321217270182704428070289873136437847402101869610.26666.0987209459523154.916.252.613.712.352.8010.560.875.7411.612.194.406.577.681659312.02980.0817399450720623.606.582.143.091.832.237.160.764.2315.311.735.205.008.51197499.41359.2558831556327073.576.602.153.131.862.247.450.774.2715.051.715.455.048.50207398.99455.7349308580527983.316.561.952.821.712.056.210.733.8712.971.564.794.448.29332527.25741.29910659734139035.079.872.904.082.653.2011.210.947.1810.272.073.996.749.73OpenBenchmarking.org

VkFFT

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 2020-09-29TITAN RTXRTX 2080 TiRadeon VIIRTX 2080 SUPERRTX 2080RTX 2070 SUPERRTX 2070RTX 2060 SUPERRTX 2060GTX 1080RX 5700 XTRX 5700GTX 1660 SUPERRX Vega 56GTX 1660 TiRX 5600 XTGTX 1660GTX 1650 SUPERGTX 1060GTX 16509K18K27K36K45KSE +/- 130.48, N = 3SE +/- 62.20, N = 3SE +/- 44.69, N = 3SE +/- 9.64, N = 3SE +/- 94.29, N = 3SE +/- 41.73, N = 3SE +/- 70.64, N = 3SE +/- 39.01, N = 3SE +/- 47.07, N = 3SE +/- 41.53, N = 3SE +/- 110.33, N = 3SE +/- 1.86, N = 3SE +/- 50.89, N = 3SE +/- 30.14, N = 3SE +/- 28.10, N = 3SE +/- 20.99, N = 3SE +/- 20.28, N = 3SE +/- 12.06, N = 3SE +/- 30.94, N = 3SE +/- 12.41, N = 3402103784733252313642898728070270442701821217208872073919749187331869618213165931376513073129169495

RealSR-NCNN

Scale: 4x - TAA: No

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRadeon VIIRX 5700 XTRX 5700RX Vega 56RX 5600 XT3691215SE +/- 0.006, N = 3SE +/- 0.035, N = 3SE +/- 0.019, N = 3SE +/- 0.025, N = 3SE +/- 0.013, N = 37.2578.9949.41310.26612.029

RealSR-NCNN

Scale: 4x - TAA: Yes

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRadeon VIIRX 5700 XTRX 5700RX Vega 56RX 5600 XT20406080100SE +/- 0.08, N = 3SE +/- 0.21, N = 3SE +/- 0.32, N = 3SE +/- 0.14, N = 3SE +/- 0.10, N = 341.3055.7359.2666.1080.08

GLmark2

Resolution: 1920 x 1080

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 1920 x 1080Radeon VIIRX 5700 XTRX 5700RX 5600 XTRX Vega 562K4K6K8K10K106599308883173997209

GLmark2

Resolution: 2560 x 1440

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 2560 x 1440Radeon VIIRX 5700 XTRX 5700RX Vega 56RX 5600 XT1600320048006400800073415805556345954507

GLmark2

Resolution: 3840 x 2160

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 3840 x 2160Radeon VIIRX 5700 XTRX 5700RX Vega 56RX 5600 XT800160024003200400039032798270723152062

NCNN

Target: Vulkan GPU - Model: squeezenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: squeezenetRX 5700 XTRX 5700RX 5600 XTRX Vega 56Radeon VII1.14082.28163.42244.56325.704SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 33.313.573.604.915.07MIN: 3.21 / MAX: 3.5MIN: 3.46 / MAX: 3.77MIN: 3.48 / MAX: 5.3MIN: 4.76 / MAX: 5.78MIN: 4.71 / MAX: 9.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mobilenetRX Vega 56RX 5700 XTRX 5600 XTRX 5700Radeon VII3691215SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 36.256.566.586.609.87MIN: 6.17 / MAX: 6.44MIN: 6.49 / MAX: 6.71MIN: 6.54 / MAX: 6.77MIN: 6.54 / MAX: 6.76MIN: 6.98 / MAX: 25.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2RX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII0.65251.3051.95752.613.2625SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 31.952.142.152.612.90MIN: 1.92 / MAX: 2.19MIN: 2.09 / MAX: 2.36MIN: 2.11 / MAX: 3.05MIN: 2.57 / MAX: 3.31MIN: 2.6 / MAX: 4.841. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3RX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII0.9181.8362.7543.6724.59SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 32.823.093.133.714.08MIN: 2.78 / MAX: 3.06MIN: 3.05 / MAX: 3.31MIN: 3.08 / MAX: 3.74MIN: 3.67 / MAX: 4.44MIN: 4.04 / MAX: 9.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: shufflenet-v2RX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII0.59631.19261.78892.38522.9815SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 31.711.831.862.352.65MIN: 1.69 / MAX: 2MIN: 1.81 / MAX: 2.02MIN: 1.84 / MAX: 2.07MIN: 2.31 / MAX: 3.34MIN: 2.63 / MAX: 2.881. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mnasnetRX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII0.721.442.162.883.6SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 32.052.232.242.803.20MIN: 2.01 / MAX: 2.47MIN: 2.18 / MAX: 4.11MIN: 2.21 / MAX: 2.47MIN: 2.75 / MAX: 3.88MIN: 3.17 / MAX: 4.841. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: efficientnet-b0RX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII3691215SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.18, N = 3SE +/- 0.30, N = 3SE +/- 0.21, N = 36.217.167.4510.5611.21MIN: 6.07 / MAX: 11.02MIN: 6.68 / MAX: 24.27MIN: 6.81 / MAX: 27.08MIN: 9.2 / MAX: 27.19MIN: 9.77 / MAX: 33.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: blazefaceRX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII0.21150.4230.63450.8461.0575SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 30.730.760.770.870.94MIN: 0.7 / MAX: 1.11MIN: 0.74 / MAX: 1.1MIN: 0.75 / MAX: 1.11MIN: 0.85 / MAX: 1.56MIN: 0.92 / MAX: 1.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: googlenetRX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII246810SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.28, N = 33.874.234.275.747.18MIN: 3.83 / MAX: 9.58MIN: 4.21 / MAX: 4.44MIN: 4.24 / MAX: 5.75MIN: 5.7 / MAX: 6.62MIN: 6.67 / MAX: 29.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: vgg16Radeon VIIRX Vega 56RX 5700 XTRX 5700RX 5600 XT48121620SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.13, N = 3SE +/- 0.13, N = 3SE +/- 0.08, N = 310.2711.6112.9715.0515.31MIN: 8.89 / MAX: 31.17MIN: 10.68 / MAX: 28.02MIN: 12.15 / MAX: 32.38MIN: 13.68 / MAX: 35.48MIN: 14.47 / MAX: 34.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet18RX 5700 XTRX 5700RX 5600 XTRadeon VIIRX Vega 560.49280.98561.47841.97122.464SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 31.561.711.732.072.19MIN: 1.54 / MAX: 3.04MIN: 1.68 / MAX: 2MIN: 1.7 / MAX: 6.56MIN: 2.02 / MAX: 2.26MIN: 2.11 / MAX: 2.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: alexnetRadeon VIIRX Vega 56RX 5700 XTRX 5600 XTRX 57001.22632.45263.67894.90526.1315SE +/- 0.19, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 33.994.404.795.205.45MIN: 3.62 / MAX: 21.47MIN: 4.29 / MAX: 5.04MIN: 4.55 / MAX: 5.54MIN: 4.91 / MAX: 13.16MIN: 5.1 / MAX: 15.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet50RX 5700 XTRX 5600 XTRX 5700RX Vega 56Radeon VII246810SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.16, N = 34.445.005.046.576.74MIN: 4.4 / MAX: 4.62MIN: 4.9 / MAX: 20.02MIN: 4.94 / MAX: 20.04MIN: 6.13 / MAX: 20.05MIN: 6.05 / MAX: 28.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: yolov4-tinyRX Vega 56RX 5700 XTRX 5700RX 5600 XTRadeon VII3691215SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.15, N = 37.688.298.508.519.73MIN: 7.59 / MAX: 17.7MIN: 8.2 / MAX: 8.46MIN: 8.43 / MAX: 9.12MIN: 8.45 / MAX: 9.17MIN: 9.25 / MAX: 17.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread


Phoronix Test Suite v10.8.4