Compare Backends

AMD Ryzen 5 4500U testing with a LENOVO LNVNB161216 (EECN20WW BIOS) and AMD Renoir 512MB on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2010011-PTS-COMPAREB12&grs.

Compare BackendsProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionAMDGPU LLVMACOAMD Ryzen 5 4500U @ 2.38GHz (6 Cores)LENOVO LNVNB161216 (EECN20WW BIOS)AMD Renoir Root Complex16GB256GB SKHynix_HFM256GDHTNI-87A0BAMD Renoir 512MB (1500/400MHz)AMD Device 1637Realtek RTL8822CE 802.11ac PCIeUbuntu 20.045.8.10-050810-generic (x86_64)GNOME Shell 3.36.4X Server 1.20.8modesetting 1.20.84.6 Mesa 20.2.0 kisak-mesa PPA (LLVM 10.0.1)1.2.131GCC 9.3.0 + Clang 10.0.1-1~kisak~fTarget:ext41920x1080OpenBenchmarking.orgEnvironment Details- AMDGPU LLVM: RADV_DEBUG=llvmCompiler 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 ondemand - CPU Microcode: 0x8600102 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 IBRS_FW STIBP: disabled RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Compare Backendsncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet50realsr-ncnn: 4x - Noncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - squeezenetncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - blazefaceAMDGPU LLVMACO359.8093.96413.05243.9945.1641.1414.8488.6913.9360.0114.8841.208.7226.413.0660.7519.2185.4459.2911.8914.025.1532.075.2024.066.2018.164.0513.211.88OpenBenchmarking.org

NCNN

Target: Vulkan GPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: vgg16AMDGPU LLVMACO80160240320400SE +/- 0.01, N = 3SE +/- 0.13, N = 3359.8060.75MIN: 358.11 / MAX: 361.96MIN: 59.53 / MAX: 98.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet50AMDGPU LLVMACO20406080100SE +/- 0.01, N = 3SE +/- 0.08, N = 393.9619.21MIN: 92.81 / MAX: 95.16MIN: 18.45 / MAX: 40.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

RealSR-NCNN

Scale: 4x - TAA: No

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoAMDGPU LLVMACO90180270360450SE +/- 0.10, N = 3SE +/- 0.06, N = 3413.0585.45

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: resnet18AMDGPU LLVMACO1020304050SE +/- 0.01, N = 3SE +/- 0.16, N = 343.999.29MIN: 43.11 / MAX: 65.82MIN: 8.44 / MAX: 28.471. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: googlenetAMDGPU LLVMACO1020304050SE +/- 0.03, N = 3SE +/- 0.06, N = 345.1611.89MIN: 44.68 / MAX: 67.37MIN: 11.19 / MAX: 31.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: alexnetAMDGPU LLVMACO918273645SE +/- 0.14, N = 3SE +/- 0.04, N = 341.1414.02MIN: 38.17 / MAX: 63.98MIN: 13.63 / MAX: 32.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mnasnetAMDGPU LLVMACO48121620SE +/- 0.03, N = 3SE +/- 0.01, N = 314.845.15MIN: 14.49 / MAX: 15.64MIN: 4.81 / MAX: 8.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: yolov4-tinyAMDGPU LLVMACO20406080100SE +/- 0.28, N = 3SE +/- 0.55, N = 388.6932.07MIN: 82 / MAX: 113.9MIN: 26.4 / MAX: 54.951. (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-v2AMDGPU LLVMACO48121620SE +/- 0.04, N = 3SE +/- 0.07, N = 313.935.20MIN: 13.54 / MAX: 14.75MIN: 4.75 / MAX: 20.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: squeezenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: squeezenetAMDGPU LLVMACO1326395265SE +/- 0.24, N = 3SE +/- 0.07, N = 360.0124.06MIN: 56.01 / MAX: 82.48MIN: 19.86 / MAX: 44.621. (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-v3AMDGPU LLVMACO48121620SE +/- 0.06, N = 3SE +/- 0.01, N = 314.886.20MIN: 14.35 / MAX: 15.92MIN: 5.74 / MAX: 6.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: mobilenetAMDGPU LLVMACO918273645SE +/- 0.26, N = 3SE +/- 0.59, N = 341.2018.16MIN: 36.01 / MAX: 69.78MIN: 12.79 / MAX: 42.221. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: shufflenet-v2AMDGPU LLVMACO246810SE +/- 0.07, N = 3SE +/- 0.01, N = 38.724.05MIN: 8.18 / MAX: 9.77MIN: 3.84 / MAX: 5.241. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: efficientnet-b0AMDGPU LLVMACO612182430SE +/- 0.09, N = 3SE +/- 0.11, N = 326.4113.21MIN: 25.85 / MAX: 27.61MIN: 12.48 / MAX: 32.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20200916Target: Vulkan GPU - Model: blazefaceAMDGPU LLVMACO0.68851.3772.06552.7543.4425SE +/- 0.05, N = 3SE +/- 0.04, N = 33.061.88MIN: 2.83 / MAX: 10.7MIN: 1.62 / MAX: 9.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Phoronix Test Suite v10.8.4