pts-nvidia-gpu-compute-02

Intel Core i7-9700K testing with a MSI MPG Z390 GAMING PRO CARBON AC (MS-7B17) v1.0 (1.20 BIOS) and eVGA NVIDIA GeForce GTX 1070 8GB on Ubuntu 20.04 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 2210206-NE-PTSNVIDIA76
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

Run Management

Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
second
October 19 2022
  7 Hours, 23 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


pts-nvidia-gpu-compute-02OpenBenchmarking.orgPhoronix Test SuiteIntel Core i7-9700K @ 4.90GHz (8 Cores)MSI MPG Z390 GAMING PRO CARBON AC (MS-7B17) v1.0 (1.20 BIOS)Intel Cannon Lake PCH32GB1000GB Samsung SSD 970 EVO 1TB + 2000GB Seagate ST2000DL003-9VT1 + 1000GB Western Digital WD10EZEX-75M + 60GB Corsair Force 3eVGA NVIDIA GeForce GTX 1070 8GBRealtek ALC1220LG Ultra HDIntel I219-V + Intel-AC 9560Ubuntu 20.045.4.0-131-generic (x86_64)KDE Plasma 5.18.8X Server 1.20.13NVIDIA 515.764.6.0OpenCL 3.0 CUDA 11.7.1021.3.205GCC 9.4.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionPts-nvidia-gpu-compute-02 BenchmarksSystem Logs- Transparent Huge Pages: madvise- NVM_CD_FLAGS=- --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=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,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 - Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1- BAR1 / Visible vRAM Size: 256 MiB - vBIOS Version: 86.04.50.00.72- GPU Compute Cores: 1920- Python 2.7.18 + Python 3.8.10- itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Mitigation of Clear buffers; SMT disabled + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT disabled + 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 Retpolines IBPB: conditional IBRS_FW STIBP: disabled RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of Clear buffers; SMT disabled

pts-nvidia-gpu-compute-02vkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4vkpeak: int32-scalarvkpeak: int32-vec4realsr-ncnn: 4x - Norealsr-ncnn: 4x - Yeswaifu2x-ncnn: 2x - 3 - Yesvkfft: hashcat: MD5hashcat: SHA1hashcat: 7-Ziphashcat: SHA-512hashcat: TrueCrypt RIPEMD160 + XTSshoc: OpenCL - S3Dshoc: OpenCL - Triadshoc: OpenCL - FFT SPshoc: OpenCL - MD5 Hashshoc: OpenCL - Reductionshoc: OpenCL - GEMM SGEMM_Nshoc: OpenCL - Max SP Flopsshoc: OpenCL - Bus Speed Downloadshoc: OpenCL - Bus Speed Readbackshoc: OpenCL - Texture Read Bandwidthcl-mem: Copycl-mem: Readcl-mem: Writenamd-cuda: ATPase Simulation - 327,506 Atomsbetsy: ETC1 - Highestbetsy: ETC2 RGB - Highestvkresample: 2x - Doublevkresample: 2x - Singleoctanebench: Total Scorefahbench: lczero: OpenCLrodinia: OpenCL Particle Filterarrayfire: Conjugate Gradient OpenCLluxcorerender: DLSC - GPUluxcorerender: Danish Mood - GPUluxcorerender: Orange Juice - GPUluxcorerender: LuxCore Benchmark - GPUluxcorerender: Rainbow Colors and Prism - GPUfinancebench: Black-Scholes OpenCLviennacl: CPU BLAS - sCOPYviennacl: CPU BLAS - sAXPYviennacl: CPU BLAS - sDOTviennacl: CPU BLAS - dCOPYviennacl: CPU BLAS - dAXPYviennacl: CPU BLAS - dDOTviennacl: CPU BLAS - dGEMV-Nviennacl: CPU BLAS - dGEMV-Tviennacl: CPU BLAS - dGEMM-NNviennacl: CPU BLAS - dGEMM-NTviennacl: CPU BLAS - dGEMM-TNviennacl: CPU BLAS - dGEMM-TTviennacl: OpenCL BLAS - sCOPYviennacl: OpenCL BLAS - sAXPYviennacl: OpenCL BLAS - sDOTviennacl: OpenCL BLAS - dAXPYviennacl: OpenCL BLAS - dDOTviennacl: OpenCL BLAS - dGEMV-Nviennacl: OpenCL BLAS - dGEMV-Tviennacl: OpenCL BLAS - dGEMM-NNviennacl: OpenCL BLAS - dGEMM-TNviennacl: OpenCL BLAS - dGEMM-TTviennacl: OpenCL BLAS - dGEMM-NTncnn: 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 - FastestDetblender: BMW27 - CUDAblender: Classroom - CUDAblender: Fishy Cat - CUDAblender: Barbershop - CUDAblender: BMW27 - NVIDIA OptiXblender: Classroom - NVIDIA OptiXblender: Fishy Cat - NVIDIA OptiXblender: Barbershop - NVIDIA OptiXblender: Pabellon Barcelona - CUDAblender: Pabellon Barcelona - NVIDIA OptiXindigobench: OpenCL GPU - Bedroomindigobench: OpenCL GPU - Supercarmandelgpu: GPUclpeak: Integer Compute INTclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Global Memory Bandwidthneatbench: GPUsecond7375.407328.67236.47237.212509.602323.6714.87099.4237.01820254194252666676652833333294700739500000204467113.91312.2815558.39111.3564213.4551812.747530.2413.079613.1798479.106184.4207.3204.40.2238911.55515.140307.04036.541121.189221144.6554190967.9483.7771.660.962.051.248.5111.48819.629.229.618.727.928.630.229.925.624.626.625.91872011962042141732031921921941904.601.611.811.431.683.020.763.726.792.202.484.557.594.452.24244.321.8449.36100.24167.17619.0860.47123.55117.79519.01268.50297.115.08814.784190621068.91946.946347.52237.12196.301070OpenBenchmarking.org

vkpeak

Vkpeak is a Vulkan compute benchmark inspired by OpenCL's clpeak. Vkpeak provides Vulkan compute performance measurements for FP16 / FP32 / FP64 / INT16 / INT32 scalar and vec4 performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarsecond16003200480064008000SE +/- 13.45, N = 37375.40

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4second16003200480064008000SE +/- 14.95, N = 37328.67

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarsecond50100150200250SE +/- 0.02, N = 3236.47

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4second50100150200250SE +/- 0.12, N = 3237.21

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarsecond5001000150020002500SE +/- 0.90, N = 32509.60

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4second5001000150020002500SE +/- 0.39, N = 32323.67

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Nosecond48121620SE +/- 0.02, N = 314.87

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: Yessecond20406080100SE +/- 0.02, N = 399.42

Waifu2x-NCNN Vulkan

Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.

Scale: 2x - Denoise: 3 - TAA: No

second: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: Yessecond246810SE +/- 0.014, N = 37.018

VkFFT

VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1second4K8K12K16K20KSE +/- 5.17, N = 3202541. (CXX) g++ options: -O3 -pthread

Hashcat

Hashcat is an open-source, advanced password recovery tool supporting GPU acceleration with OpenCL, NVIDIA CUDA, and Radeon ROCm. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: MD5second4000M8000M12000M16000M20000MSE +/- 11663952.07, N = 319425266667

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1second1400M2800M4200M5600M7000MSE +/- 4267838.38, N = 36652833333

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-Zipsecond60K120K180K240K300KSE +/- 2400.00, N = 3294700

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512second160M320M480M640M800MSE +/- 152752.52, N = 3739500000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSsecond40K80K120K160K200KSE +/- 2923.66, N = 3204467

Mixbench

A benchmark suite for GPUs on mixed operational intensity kernels. Learn more via the OpenBenchmarking.org test page.

Backend: OpenCL - Benchmark: Integer

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: NVIDIA CUDA - Benchmark: Integer

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: OpenCL - Benchmark: Double Precision

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: OpenCL - Benchmark: Single Precision

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-ocl-ro: not found

Backend: NVIDIA CUDA - Benchmark: Half Precision

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: NVIDIA CUDA - Benchmark: Double Precision

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

Backend: NVIDIA CUDA - Benchmark: Single Precision

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./mixbench: 3: ./mixbench-cuda-ro: not found

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. SHOC provides a number of different benchmark programs for evaluating the performance and stability of compute devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3Dsecond306090120150SE +/- 0.02, N = 3113.911. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Triadsecond3691215SE +/- 0.00, N = 312.281. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SPsecond120240360480600SE +/- 1.72, N = 3558.391. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 Hashsecond3691215SE +/- 0.03, N = 311.361. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Reductionsecond50100150200250SE +/- 0.05, N = 3213.461. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_Nsecond400800120016002000SE +/- 3.49, N = 31812.741. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP Flopssecond16003200480064008000SE +/- 1.15, N = 37530.241. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed Downloadsecond3691215SE +/- 0.00, N = 313.081. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed Readbacksecond3691215SE +/- 0.00, N = 313.181. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read Bandwidthsecond100200300400500SE +/- 0.72, N = 3479.111. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

Libplacebo

Libplacebo is a multimedia rendering library based on the core rendering code of the MPV player. The libplacebo benchmark relies on the Vulkan API and tests various primitives. Learn more via the OpenBenchmarking.org test page.

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./libplacebo: 3: ./src/bench: not found

cl-mem

A basic OpenCL memory benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Copysecond4080120160200SE +/- 0.07, N = 3184.41. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Readsecond50100150200250SE +/- 0.06, N = 3207.31. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: Writesecond4080120160200SE +/- 0.07, N = 3204.41. (CC) gcc options: -O2 -flto -lOpenCL

NAMD CUDA

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. This version of the NAMD test profile uses CUDA GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 Atomssecond0.05040.10080.15120.20160.252SE +/- 0.00088, N = 30.22389

Betsy GPU Compressor

Betsy is an open-source GPU compressor of various GPU compression techniques. Betsy is written in GLSL for Vulkan/OpenGL (compute shader) support for GPU-based texture compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC1 - Quality: Highestsecond3691215SE +/- 0.43, N = 1211.561. (CXX) g++ options: -O3 -O2 -lpthread -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterBetsy GPU Compressor 1.1 BetaCodec: ETC2 RGB - Quality: Highestsecond48121620SE +/- 0.22, N = 1515.141. (CXX) g++ options: -O3 -O2 -lpthread -ldl

VkResample

VkResample is a Vulkan-based image upscaling library based on VkFFT. The sample input file is upscaling a 4K image to 8K using Vulkan-based GPU acceleration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Doublesecond70140210280350SE +/- 0.20, N = 3307.041. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Singlesecond816243240SE +/- 0.07, N = 336.541. (CXX) g++ options: -O3 -pthread

OctaneBench

OctaneBench is a test of the OctaneRender on the GPU and requires the use of NVIDIA CUDA. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterOctaneBench 2020.1Total Scoresecond306090120150121.19

RedShift Demo

This is a test of MAXON's RedShift demo build that currently requires NVIDIA GPU acceleration. Learn more via the OpenBenchmarking.org test page.

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./redshift: 3: /usr/redshift/bin/redshiftBenchmark: not found

FAHBench

FAHBench is a Folding@Home benchmark on the GPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2second306090120150SE +/- 0.05, N = 3144.66

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: OpenCLsecond4K8K12K16K20KSE +/- 114.24, N = 3190961. (CXX) g++ options: -flto -pthread

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle Filtersecond246810SE +/- 0.093, N = 37.9481. (CXX) g++ options: -O2 -lOpenCL

ArrayFire

ArrayFire is an GPU and CPU numeric processing library, this test uses the built-in CPU and OpenCL ArrayFire benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLsecond0.84981.69962.54943.39924.249SE +/- 0.008, N = 33.7771. (CXX) g++ options: -rdynamic

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: GPUsecond0.37350.7471.12051.4941.8675SE +/- 0.01, N = 31.66MIN: 1.56 / MAX: 1.71

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPUsecond0.2160.4320.6480.8641.08SE +/- 0.01, N = 30.96MIN: 0.15 / MAX: 1.34

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPUsecond0.46130.92261.38391.84522.3065SE +/- 0.00, N = 32.05MIN: 1.6 / MAX: 2.37

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPUsecond0.2790.5580.8371.1161.395SE +/- 0.01, N = 31.24MIN: 0.26 / MAX: 1.64

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: GPUsecond246810SE +/- 0.03, N = 38.51MIN: 7.98 / MAX: 8.93

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLsecond3691215SE +/- 0.00, N = 311.491. (CXX) g++ options: -O3 -march=native -fopenmp

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYsecond510152025SE +/- 0.09, N = 319.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYsecond714212835SE +/- 0.03, N = 329.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTsecond714212835SE +/- 0.00, N = 329.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYsecond510152025SE +/- 0.00, N = 318.71. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYsecond714212835SE +/- 0.00, N = 327.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTsecond714212835SE +/- 0.00, N = 328.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Nsecond714212835SE +/- 0.00, N = 330.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Tsecond714212835SE +/- 0.06, N = 329.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNsecond612182430SE +/- 0.03, N = 325.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTsecond612182430SE +/- 0.00, N = 324.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNsecond612182430SE +/- 0.00, N = 326.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTsecond612182430SE +/- 0.03, N = 325.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYsecond40801201602001871. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYsecond40801201602002011. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTsecond40801201602001961. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYsecond40801201602002041. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTsecond501001502002502141. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-Nsecond40801201602001731. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-Tsecond40801201602002031. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNsecond4080120160200SE +/- 0.67, N = 31921. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNsecond40801201602001921. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTsecond40801201602001941. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTsecond40801201602001901. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

Implementation: NVIDIA CUDA GPU - Input: water_GMX50_bare

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ./gromacs: 5: /cuda-build/run-gromacs: not found

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 100

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f84a80546ef google::LogMessageFatal::~LogMessageFatal()

Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 200

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f7960a5c6ef google::LogMessageFatal::~LogMessageFatal()

Model: AlexNet - Acceleration: NVIDIA CUDA - Iterations: 1000

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f8ece7276ef google::LogMessageFatal::~LogMessageFatal()

Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 100

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7fbb9919d6ef google::LogMessageFatal::~LogMessageFatal()

Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 200

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7feb04d276ef google::LogMessageFatal::~LogMessageFatal()

Model: GoogleNet - Acceleration: NVIDIA CUDA - Iterations: 1000

second: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: @ 0x7f9c9c40c6ef google::LogMessageFatal::~LogMessageFatal()

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 20220729Target: Vulkan GPU - Model: mobilenetsecond1.0352.073.1054.145.175SE +/- 0.01, N = 34.60MIN: 4.5 / MAX: 14.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2second0.36230.72461.08691.44921.8115SE +/- 0.00, N = 31.61MIN: 1.57 / MAX: 6.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3second0.40730.81461.22191.62922.0365SE +/- 0.01, N = 31.81MIN: 1.77 / MAX: 4.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: shufflenet-v2second0.32180.64360.96541.28721.609SE +/- 0.01, N = 31.43MIN: 1.39 / MAX: 3.031. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: mnasnetsecond0.3780.7561.1341.5121.89SE +/- 0.01, N = 31.68MIN: 1.63 / MAX: 5.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: efficientnet-b0second0.67951.3592.03852.7183.3975SE +/- 0.01, N = 33.02MIN: 2.97 / MAX: 3.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: blazefacesecond0.1710.3420.5130.6840.855SE +/- 0.01, N = 30.76MIN: 0.7 / MAX: 2.841. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: googlenetsecond0.8371.6742.5113.3484.185SE +/- 0.01, N = 33.72MIN: 3.64 / MAX: 6.731. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: vgg16second246810SE +/- 0.01, N = 36.79MIN: 6.67 / MAX: 15.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: resnet18second0.4950.991.4851.982.475SE +/- 0.02, N = 32.20MIN: 2.15 / MAX: 10.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: alexnetsecond0.5581.1161.6742.2322.79SE +/- 0.01, N = 32.48MIN: 2.36 / MAX: 11.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: resnet50second1.02382.04763.07144.09525.119SE +/- 0.01, N = 34.55MIN: 4.44 / MAX: 12.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: yolov4-tinysecond246810SE +/- 0.10, N = 37.59MIN: 7.21 / MAX: 15.731. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: squeezenet_ssdsecond1.00132.00263.00394.00525.0065SE +/- 0.23, N = 34.45MIN: 4.13 / MAX: 47.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: regnety_400msecond0.5041.0081.5122.0162.52SE +/- 0.00, N = 32.24MIN: 2.17 / MAX: 8.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: vision_transformersecond50100150200250SE +/- 0.57, N = 3244.32MIN: 228.48 / MAX: 273.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: Vulkan GPU - Model: FastestDetsecond0.4140.8281.2421.6562.07SE +/- 0.01, N = 31.84MIN: 1.53 / MAX: 9.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL

second: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL

second: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL

second: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL

second: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL

second: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: BMW27 - Compute: CUDAsecond1122334455SE +/- 0.01, N = 349.36

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Classroom - Compute: CUDAsecond20406080100SE +/- 0.01, N = 3100.24

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Fishy Cat - Compute: CUDAsecond4080120160200SE +/- 0.03, N = 3167.17

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Barbershop - Compute: CUDAsecond130260390520650SE +/- 0.16, N = 3619.08

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: BMW27 - Compute: NVIDIA OptiXsecond1428425670SE +/- 0.39, N = 360.47

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Classroom - Compute: NVIDIA OptiXsecond306090120150SE +/- 0.04, N = 3123.55

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Fishy Cat - Compute: NVIDIA OptiXsecond306090120150SE +/- 0.30, N = 3117.79

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Barbershop - Compute: NVIDIA OptiXsecond110220330440550SE +/- 0.18, N = 3519.01

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Pabellon Barcelona - Compute: CUDAsecond60120180240300SE +/- 0.07, N = 3268.50

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.3Blend File: Pabellon Barcelona - Compute: NVIDIA OptiXsecond60120180240300SE +/- 0.02, N = 3297.11

IndigoBench

This is a test of Indigo Renderer's IndigoBench benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Bedroomsecond1.14482.28963.43444.57925.724SE +/- 0.000, N = 35.088

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: Supercarsecond48121620SE +/- 0.00, N = 314.78

MandelGPU

MandelGPU is an OpenCL benchmark and this test runs with the OpenCL rendering float4 kernel with a maximum of 4096 iterations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPUsecond40M80M120M160M200MSE +/- 1064714.78, N = 3190621068.91. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

clpeak

Clpeak is designed to test the peak capabilities of OpenCL devices. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTsecond400800120016002000SE +/- 21.64, N = 151946.941. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision Floatsecond14002800420056007000SE +/- 110.51, N = 156347.521. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision Doublesecond50100150200250SE +/- 0.28, N = 3237.121. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory Bandwidthsecond4080120160200SE +/- 0.14, N = 3196.301. (CXX) g++ options: -O3 -rdynamic -lOpenCL

NeatBench

NeatBench is a benchmark of the cross-platform Neat Video software on the CPU and optional GPU (OpenCL / CUDA) support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPUsecond20040060080010001070

102 Results Shown

vkpeak:
  fp32-scalar
  fp32-vec4
  fp64-scalar
  fp64-vec4
  int32-scalar
  int32-vec4
RealSR-NCNN:
  4x - No
  4x - Yes
Waifu2x-NCNN Vulkan
VkFFT
Hashcat:
  MD5
  SHA1
  7-Zip
  SHA-512
  TrueCrypt RIPEMD160 + XTS
SHOC Scalable HeterOgeneous Computing:
  OpenCL - S3D
  OpenCL - Triad
  OpenCL - FFT SP
  OpenCL - MD5 Hash
  OpenCL - Reduction
  OpenCL - GEMM SGEMM_N
  OpenCL - Max SP Flops
  OpenCL - Bus Speed Download
  OpenCL - Bus Speed Readback
  OpenCL - Texture Read Bandwidth
cl-mem:
  Copy
  Read
  Write
NAMD CUDA
Betsy GPU Compressor:
  ETC1 - Highest
  ETC2 RGB - Highest
VkResample:
  2x - Double
  2x - Single
OctaneBench
FAHBench
LeelaChessZero
Rodinia
ArrayFire
LuxCoreRender:
  DLSC - GPU
  Danish Mood - GPU
  Orange Juice - GPU
  LuxCore Benchmark - GPU
  Rainbow Colors and Prism - GPU
FinanceBench
ViennaCL:
  CPU BLAS - sCOPY
  CPU BLAS - sAXPY
  CPU BLAS - sDOT
  CPU BLAS - dCOPY
  CPU BLAS - dAXPY
  CPU BLAS - dDOT
  CPU BLAS - dGEMV-N
  CPU BLAS - dGEMV-T
  CPU BLAS - dGEMM-NN
  CPU BLAS - dGEMM-NT
  CPU BLAS - dGEMM-TN
  CPU BLAS - dGEMM-TT
  OpenCL BLAS - sCOPY
  OpenCL BLAS - sAXPY
  OpenCL BLAS - sDOT
  OpenCL BLAS - dAXPY
  OpenCL BLAS - dDOT
  OpenCL BLAS - dGEMV-N
  OpenCL BLAS - dGEMV-T
  OpenCL BLAS - dGEMM-NN
  OpenCL BLAS - dGEMM-TN
  OpenCL BLAS - dGEMM-TT
  OpenCL BLAS - dGEMM-NT
NCNN:
  Vulkan GPU - mobilenet
  Vulkan GPU-v2-v2 - mobilenet-v2
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU - shufflenet-v2
  Vulkan GPU - mnasnet
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - blazeface
  Vulkan GPU - googlenet
  Vulkan GPU - vgg16
  Vulkan GPU - resnet18
  Vulkan GPU - alexnet
  Vulkan GPU - resnet50
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - squeezenet_ssd
  Vulkan GPU - regnety_400m
  Vulkan GPU - vision_transformer
  Vulkan GPU - FastestDet
Blender:
  BMW27 - CUDA
  Classroom - CUDA
  Fishy Cat - CUDA
  Barbershop - CUDA
  BMW27 - NVIDIA OptiX
  Classroom - NVIDIA OptiX
  Fishy Cat - NVIDIA OptiX
  Barbershop - NVIDIA OptiX
  Pabellon Barcelona - CUDA
  Pabellon Barcelona - NVIDIA OptiX
IndigoBench:
  OpenCL GPU - Bedroom
  OpenCL GPU - Supercar
MandelGPU
clpeak:
  Integer Compute INT
  Single-Precision Float
  Double-Precision Double
  Global Memory Bandwidth
NeatBench