RTX 30 Compute September

AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3801 BIOS) and NVIDIA GeForce RTX 3090 24GB on Ubuntu 21.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2109061-TJ-RTX30COMP69&sor&grs.

RTX 30 Compute SeptemberProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionRTX 3070RTX 3090GeForce RTX 3090AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (3801 BIOS)AMD Starship/Matisse16GB1000GB Western Digital WDS100T1X0E-00AFY0 + 1000GB Western Digital WD_BLACK SN850 1TB + 2000GBNVIDIA GeForce RTX 3070 8GBNVIDIA GA104 HD AudioASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 21.105.13.0-14-generic (x86_64)GNOME Shell 40.2X Server 1.20.11NVIDIA 470.63.014.6.0OpenCL 3.0 CUDA 11.4.1121.2.175GCC 11.2.0ext43840x2160NVIDIA GeForce RTX 3090 24GBNVIDIA GA102 HD AudioOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-M6DaQn/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-M6DaQn/gcc-11-11.2.0/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-build-config=bootstrap-lto-lean --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 performance (Boost: Enabled) - CPU Microcode: 0xa201016OpenCL Details- RTX 3070: GPU Compute Cores: 5888- RTX 3090: GPU Compute Cores: 10496- GeForce RTX 3090: GPU Compute Cores: 10496Python Details- Python 3.9.6Security 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: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

RTX 30 Compute Septembershoc: OpenCL - GEMM SGEMM_Ncl-mem: Readshoc: OpenCL - FFT SPshoc: OpenCL - S3Dcl-mem: Writevkresample: 2x - Singlevkpeak: fp32-scalarviennacl: OpenCL BLAS - dAXPYvkpeak: fp16-vec4vkpeak: fp32-vec4vkpeak: fp64-scalarvkpeak: int32-scalarvkpeak: fp16-scalarvkresample: 2x - Doublevkpeak: int32-vec4vkpeak: int16-scalarvkpeak: fp64-vec4viennacl: OpenCL BLAS - dGEMM-TNviennacl: OpenCL BLAS - dGEMM-NTviennacl: OpenCL BLAS - dGEMM-NNblender: Fishy Cat - NVIDIA OptiXluxcorerender: LuxCore Benchmark - GPUshoc: OpenCL - MD5 Hashshoc: OpenCL - Max SP Flopshashcat: SHA1luxcorerender: DLSC - GPUhashcat: SHA-512hashcat: MD5hashcat: TrueCrypt RIPEMD160 + XTSluxcorerender: Danish Mood - GPUhashcat: 7-Zipvkpeak: int16-vec4viennacl: OpenCL BLAS - dDOTblender: BMW27 - NVIDIA OptiXrealsr-ncnn: 4x - Yesoctanebench: Total Scorefinancebench: Black-Scholes OpenCLviennacl: OpenCL BLAS - dCOPYredshift: indigobench: OpenCL GPU - Bedroomrodinia: OpenCL Particle Filterblender: Classroom - NVIDIA OptiXblender: BMW27 - CUDAblender: Fishy Cat - CUDAluxcorerender: Rainbow Colors and Prism - GPUluxcorerender: Orange Juice - GPUblender: Classroom - CUDAindigobench: OpenCL GPU - Supercarlczero: OpenCLvkfft: arrayfire: Conjugate Gradient OpenCLrealsr-ncnn: 4x - Noviennacl: OpenCL BLAS - sAXPYblender: Barbershop - CUDAfahbench: waifu2x-ncnn: 2x - 3 - Yesviennacl: OpenCL BLAS - sCOPYncnn: Vulkan GPU - vgg16cl-mem: Copyshoc: OpenCL - Reductionncnn: Vulkan GPU - yolov4-tinyviennacl: OpenCL BLAS - dGEMV-Tviennacl: CPU BLAS - dDOTviennacl: CPU BLAS - sAXPYncnn: Vulkan GPU - blazefaceviennacl: CPU BLAS - sDOTncnn: Vulkan GPU - resnet18viennacl: OpenCL BLAS - sDOTviennacl: CPU BLAS - dCOPYncnn: Vulkan GPU-v3-v3 - mobilenet-v3viennacl: CPU BLAS - dGEMV-Tncnn: Vulkan GPU - googlenetviennacl: CPU BLAS - dGEMV-Nncnn: Vulkan GPU - mnasnetviennacl: OpenCL BLAS - dGEMV-Nncnn: Vulkan GPU - regnety_400mviennacl: CPU BLAS - sCOPYncnn: Vulkan GPU - efficientnet-b0viennacl: CPU BLAS - dAXPYnamd-cuda: ATPase Simulation - 327,506 Atomsshoc: OpenCL - Texture Read Bandwidthviennacl: CPU BLAS - dGEMM-TTncnn: Vulkan GPU - mobilenetviennacl: CPU BLAS - dGEMM-TNncnn: Vulkan GPU - alexnetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - shufflenet-v2shoc: OpenCL - Triadviennacl: CPU BLAS - dGEMM-NTncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - squeezenet_ssdviennacl: CPU BLAS - dGEMM-NNshoc: OpenCL - Bus Speed Readbackshoc: OpenCL - Bus Speed Downloadviennacl: OpenCL BLAS - dGEMM-TTwaifu2x-ncnn: 2x - 3 - NoRTX 3070RTX 3090GeForce RTX 30903732.87393.31134.82219.195382.017.52911589.5939622503.1615319.31363.7511571.3511590.48216.40611521.487644.32364.2933734234133.346.2225.408723102.9131948333336.681673633333389251666675031675.0668966710106.8339515.8246.790415.19925110.45737022612.9415.92045.7729.0054.6921.276.9976.0637.31837790315542.0897.856359507.39265.50784.1122915.43296.8326.5706.5733348.6100.80.981511.6832224.62.1489.84.1382.82.012222.4266.93.1736.50.125552130.0456.44.3158.21.901.911.7124.667153.63.744.4353.127.113526.28378174.54826.62349.19427.9357459.25421074.8972140868.9627597.6655.4220836.3220954119.82420743.813646.56655.3159860260018.9710.9244.426740344.12285370000011.532884600000667393000008575008.64115460016889.266639.5228.415683.5959996.35760713920.9683.6828.6618.3234.6532.0910.4251.1853.36453653432951.485.566501373.32344.01953.2763644.44364.3391.2117.6637643.289.60.961361.8135122.62.2183.84.3877.62.092362.57633.3534.80.131552228.01544.4856.11.881.981.7725.5289523.744.3752.427.078526.25716018105.732348.03427.07820871.3640423.4127390.08651.3120687.5520714.1420597.1913655.97651.244.194222798500000288130000066589100000859400116110016887.9228.421445675.5953.275391.09125.4496OpenBenchmarking.org

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: GEMM SGEMM_N

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NRTX 3090GeForce RTX 3090RTX 30702K4K6K8K10KSE +/- 26.69, N = 38174.548105.733732.871. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

cl-mem

Benchmark: Read

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: ReadRTX 3090RTX 30702004006008001000SE +/- 0.30, N = 3826.6393.31. (CC) gcc options: -O2 -flto -lOpenCL

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: FFT SP

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SPRTX 3090GeForce RTX 3090RTX 30705001000150020002500SE +/- 0.78, N = 32349.192348.031134.821. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: S3D

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DRTX 3090GeForce RTX 3090RTX 307090180270360450SE +/- 0.17, N = 3427.94427.08219.201. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

cl-mem

Benchmark: Write

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteRTX 3090RTX 3070160320480640800SE +/- 0.09, N = 3745.0382.01. (CC) gcc options: -O2 -flto -lOpenCL

VkResample

Upscale: 2x - Precision: Single

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRTX 3090RTX 307048121620SE +/- 0.013, N = 39.25417.5291. (CXX) g++ options: -O3 -pthread

vkpeak

fp32-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarGeForce RTX 3090RTX 3090RTX 30705K10K15K20K25KSE +/- 35.78, N = 321146.0121074.8911589.59

ViennaCL

Test: OpenCL BLAS - dAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYRTX 3090RTX 3070160320480640800SE +/- 0.00, N = 37213961. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

vkpeak

fp16-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4GeForce RTX 3090RTX 3090RTX 30709K18K27K36K45KSE +/- 71.70, N = 340869.1940868.9622503.16

vkpeak

fp32-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4GeForce RTX 3090RTX 3090RTX 30706K12K18K24K30KSE +/- 35.81, N = 327807.7527597.6015319.31

vkpeak

fp64-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarGeForce RTX 3090RTX 3090RTX 3070140280420560700SE +/- 0.36, N = 3658.56655.42363.75

vkpeak

int32-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarGeForce RTX 3090RTX 3090RTX 30704K8K12K16K20KSE +/- 16.33, N = 320925.8720836.3211571.35

vkpeak

fp16-scalar

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarRTX 3090GeForce RTX 3090RTX 30704K8K12K16K20KSE +/- 26.04, N = 320954.0020953.6511590.48

VkResample

Upscale: 2x - Precision: Double

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: DoubleRTX 3090RTX 307050100150200250SE +/- 0.03, N = 3119.82216.411. (CXX) g++ options: -O3 -pthread

vkpeak

int32-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4GeForce RTX 3090RTX 3090RTX 30704K8K12K16K20KSE +/- 18.17, N = 320746.6720743.8011521.48

vkpeak

int16-scalar

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarGeForce RTX 3090RTX 3090RTX 30703K6K9K12K15KSE +/- 11.41, N = 313752.1913646.567644.32

vkpeak

fp64-vec4

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4RTX 3090GeForce RTX 3090RTX 3070140280420560700SE +/- 0.42, N = 3655.31651.20364.29

ViennaCL

Test: OpenCL BLAS - dGEMM-TN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNRTX 3090RTX 30701302603905206505983371. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMM-NT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTRTX 3090RTX 3070130260390520650SE +/- 1.33, N = 36023421. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: OpenCL BLAS - dGEMM-NN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNRTX 3090RTX 3070130260390520650SE +/- 1.33, N = 36003411. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

Blender

Blend File: Fishy Cat - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: NVIDIA OptiXRTX 3090RTX 3070816243240SE +/- 0.01, N = 318.9733.34

LuxCoreRender

Scene: LuxCore Benchmark - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: GPURTX 3090RTX 30703691215SE +/- 0.01, N = 310.926.22MIN: 3.29 / MAX: 13.01MIN: 2.05 / MAX: 7.38

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: MD5 Hash

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashRTX 3090GeForce RTX 3090RTX 30701020304050SE +/- 0.03, N = 344.4344.1925.411. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Max SP Flops

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Max SP FlopsRTX 3090RTX 30709K18K27K36K45KSE +/- 59.20, N = 340344.123102.91. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

Hashcat

Benchmark: SHA1

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA1RTX 3090GeForce RTX 3090RTX 30705000M10000M15000M20000M25000MSE +/- 7750340.49, N = 3228537000002279850000013194833333

LuxCoreRender

Scene: DLSC - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: GPURTX 3090RTX 30703691215SE +/- 0.00, N = 311.536.68MIN: 11.29 / MAX: 11.68MIN: 6.25 / MAX: 6.94

Hashcat

Benchmark: SHA-512

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: SHA-512RTX 3090GeForce RTX 3090RTX 3070600M1200M1800M2400M3000MSE +/- 735602.55, N = 3288460000028813000001673633333

Hashcat

Benchmark: MD5

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: MD5RTX 3090GeForce RTX 3090RTX 307014000M28000M42000M56000M70000MSE +/- 41756290.28, N = 3667393000006658910000038925166667

Hashcat

Benchmark: TrueCrypt RIPEMD160 + XTS

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: TrueCrypt RIPEMD160 + XTSGeForce RTX 3090RTX 3090RTX 3070200K400K600K800K1000KSE +/- 800.69, N = 3859400857500503167

LuxCoreRender

Scene: Danish Mood - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: GPURTX 3090RTX 3070246810SE +/- 0.00, N = 38.645.06MIN: 2.95 / MAX: 10.48MIN: 1.5 / MAX: 6.12

Hashcat

Benchmark: 7-Zip

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.1.1Benchmark: 7-ZipGeForce RTX 3090RTX 3090RTX 3070200K400K600K800K1000KSE +/- 1128.91, N = 311611001154600689667

vkpeak

int16-vec4

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4GeForce RTX 3090RTX 3090RTX 30704K8K12K16K20KSE +/- 20.90, N = 317006.6116889.2610106.83

ViennaCL

Test: OpenCL BLAS - dDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTRTX 3090RTX 3070140280420560700SE +/- 0.00, N = 36633951. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

Blender

Blend File: BMW27 - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: NVIDIA OptiXRTX 3090RTX 307048121620SE +/- 0.01, N = 39.5215.82

RealSR-NCNN

Scale: 4x - TAA: Yes

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRTX 3090GeForce RTX 3090RTX 30701122334455SE +/- 0.04, N = 328.4228.4246.79

OctaneBench

Total Score

OpenBenchmarking.orgScore, More Is BetterOctaneBench 2020.1Total ScoreRTX 3090RTX 3070150300450600750683.60415.20

FinanceBench

Benchmark: Black-Scholes OpenCL

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLRTX 3090RTX 30703691215SE +/- 0.003, N = 36.35710.4571. (CXX) g++ options: -O3 -march=native -fopenmp

ViennaCL

Test: OpenCL BLAS - dCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYRTX 3090RTX 3070130260390520650SE +/- 0.33, N = 36073701. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

RedShift Demo

OpenBenchmarking.orgSeconds, Fewer Is BetterRedShift Demo 3.0RTX 3090RTX 307050100150200250SE +/- 0.33, N = 3139226

IndigoBench

Acceleration: OpenCL GPU - Scene: Bedroom

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomRTX 3090RTX 3070510152025SE +/- 0.02, N = 320.9712.94

Rodinia

Test: OpenCL Particle Filter

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle FilterRTX 3090RTX 30701.3322.6643.9965.3286.66SE +/- 0.007, N = 33.6805.9201. (CXX) g++ options: -O2 -lOpenCL

Blender

Blend File: Classroom - Compute: NVIDIA OptiX

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: NVIDIA OptiXRTX 3090RTX 30701020304050SE +/- 0.02, N = 328.6645.77

Blender

Blend File: BMW27 - Compute: CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CUDARTX 3090RTX 3070714212835SE +/- 0.04, N = 318.3229.00

Blender

Blend File: Fishy Cat - Compute: CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CUDARTX 3090RTX 30701224364860SE +/- 0.04, N = 334.6554.69

LuxCoreRender

Scene: Rainbow Colors and Prism - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: GPURTX 3090RTX 3070714212835SE +/- 0.03, N = 332.0921.27MIN: 28.96 / MAX: 34.45MIN: 20.13 / MAX: 22.99

LuxCoreRender

Scene: Orange Juice - Acceleration: GPU

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: GPURTX 3090RTX 30703691215SE +/- 0.01, N = 310.426.99MIN: 8.45 / MAX: 13.57MIN: 5.61 / MAX: 8.46

Blender

Blend File: Classroom - Compute: CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CUDARTX 3090RTX 307020406080100SE +/- 0.01, N = 351.1876.06

IndigoBench

Acceleration: OpenCL GPU - Scene: Supercar

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: SupercarRTX 3090RTX 30701224364860SE +/- 0.04, N = 353.3637.32

LeelaChessZero

Backend: OpenCL

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: OpenCLRTX 3090RTX 307011K22K33K44K55KSE +/- 178.94, N = 353653377901. (CXX) g++ options: -flto -pthread

VkFFT

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1GeForce RTX 3090RTX 3090RTX 307010K20K30K40K50KSE +/- 172.10, N = 34456743295315541. (CXX) g++ options: -O3 -pthread

ArrayFire

Test: Conjugate Gradient OpenCL

OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLRTX 3090RTX 30700.470.941.411.882.35SE +/- 0.001, N = 31.4802.0891. (CXX) g++ options: -rdynamic

RealSR-NCNN

Scale: 4x - TAA: No

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRTX 3090GeForce RTX 3090RTX 3070246810SE +/- 0.002, N = 35.5665.5957.856

ViennaCL

Test: OpenCL BLAS - sAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYRTX 3090RTX 3070110220330440550SE +/- 0.33, N = 35013591. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

Blender

Blend File: Barbershop - Compute: CUDA

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CUDARTX 3090RTX 3070110220330440550SE +/- 0.17, N = 3373.32507.39

FAHBench

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2RTX 3090RTX 307070140210280350SE +/- 0.05, N = 3344.02265.51

Waifu2x-NCNN Vulkan

Scale: 2x - Denoise: 3 - TAA: Yes

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesGeForce RTX 3090RTX 3090RTX 30700.92521.85042.77563.70084.626SE +/- 0.006, N = 33.2753.2764.112

ViennaCL

Test: OpenCL BLAS - sCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYRTX 3090RTX 307080160240320400SE +/- 0.58, N = 33642911. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: vgg16RTX 3090RTX 30701.22182.44363.66544.88726.109SE +/- 0.01, N = 34.355.43MIN: 4.02 / MAX: 18.18MIN: 5.38 / MAX: 9.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

cl-mem

Benchmark: Copy

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyRTX 3090RTX 307080160240320400SE +/- 0.45, N = 3364.3296.81. (CC) gcc options: -O2 -flto -lOpenCL

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Reduction

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionRTX 3090GeForce RTX 3090RTX 307080160240320400SE +/- 0.70, N = 3391.21391.09326.571. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: yolov4-tinyRTX 3070RTX 3090246810SE +/- 0.01, N = 36.577.20MIN: 6.33 / MAX: 6.98MIN: 6.44 / MAX: 31.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: OpenCL BLAS - dGEMV-T

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-TRTX 3090RTX 307080160240320400SE +/- 0.67, N = 33763331. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTRTX 3070RTX 30901122334455SE +/- 0.50, N = 348.643.21. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYRTX 3070RTX 309020406080100SE +/- 0.73, N = 3100.889.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: blazefaceRTX 3090RTX 30700.24080.48160.72240.96321.204SE +/- 0.00, N = 30.960.98MIN: 0.88 / MAX: 8.42MIN: 0.95 / MAX: 2.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - sDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTRTX 3070RTX 3090306090120150SE +/- 0.88, N = 31511361. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet18RTX 3090RTX 30700.40730.81461.22191.62922.0365SE +/- 0.01, N = 31.661.68MIN: 1.53 / MAX: 9.86MIN: 1.66 / MAX: 1.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: OpenCL BLAS - sDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTRTX 3090RTX 307080160240320400SE +/- 0.67, N = 33513221. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYRTX 3070RTX 3090612182430SE +/- 0.03, N = 324.622.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3RTX 3070RTX 30900.51981.03961.55942.07922.599SE +/- 0.01, N = 32.142.21MIN: 2.11 / MAX: 3.93MIN: 2.05 / MAX: 17.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - dGEMV-T

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-TRTX 3070RTX 309020406080100SE +/- 0.10, N = 389.883.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: googlenetRTX 3070RTX 30900.99231.98462.97693.96924.9615SE +/- 0.13, N = 24.134.38MIN: 3.77 / MAX: 13.51MIN: 3.61 / MAX: 17.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - dGEMV-N

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-NRTX 3070RTX 309020406080100SE +/- 0.09, N = 382.877.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mnasnetRTX 3070RTX 30900.48150.9631.44451.9262.4075SE +/- 0.01, N = 32.012.09MIN: 1.98 / MAX: 4.82MIN: 1.91 / MAX: 13.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: OpenCL BLAS - dGEMV-N

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-NRTX 3090RTX 307050100150200250SE +/- 0.58, N = 32362221. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: regnety_400mRTX 3070RTX 30900.57831.15661.73492.31322.8915SE +/- 0.01, N = 32.422.52MIN: 2.4 / MAX: 3.84MIN: 2.38 / MAX: 10.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - sCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYRTX 3070RTX 30901530456075SE +/- 0.58, N = 366.963.01. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: efficientnet-b0RTX 3070RTX 30900.75381.50762.26143.01523.769SE +/- 0.01, N = 33.173.34MIN: 3.14 / MAX: 5.06MIN: 3.09 / MAX: 9.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - dAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYRTX 3070RTX 3090816243240SE +/- 0.32, N = 336.534.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NAMD CUDA

ATPase Simulation - 327,506 Atoms

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 AtomsRTX 3070RTX 30900.02960.05920.08880.11840.148SE +/- 0.00049, N = 30.125550.13155

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Texture Read Bandwidth

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthRTX 3090RTX 30705001000150020002500SE +/- 6.58, N = 32228.012130.041. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

ViennaCL

Test: CPU BLAS - dGEMM-TT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTRTX 3070RTX 30901326395265SE +/- 0.09, N = 356.454.01. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: mobilenetRTX 3070RTX 30901.0082.0163.0244.0325.04SE +/- 0.01, N = 34.314.47MIN: 4.25 / MAX: 5.94MIN: 4.1 / MAX: 24.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - dGEMM-TN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNRTX 3070RTX 30901326395265SE +/- 0.00, N = 358.256.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: alexnetRTX 3090RTX 30700.43880.87761.31641.75522.194SE +/- 0.01, N = 31.881.90MIN: 1.77 / MAX: 6.33MIN: 1.86 / MAX: 5.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2RTX 3070RTX 30900.44550.8911.33651.7822.2275SE +/- 0.01, N = 31.911.98MIN: 1.88 / MAX: 3.42MIN: 1.84 / MAX: 7.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: shufflenet-v2RTX 3070RTX 30900.39830.79661.19491.59321.9915SE +/- 0.01, N = 31.711.77MIN: 1.68 / MAX: 3.55MIN: 1.67 / MAX: 9.471. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Triad

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadRTX 3090GeForce RTX 3090RTX 3070612182430SE +/- 0.01, N = 325.5325.4524.671. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

ViennaCL

Test: CPU BLAS - dGEMM-NT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTRTX 3070RTX 30901224364860SE +/- 0.12, N = 353.652.01. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: resnet50RTX 3090RTX 30700.84151.6832.52453.3664.2075SE +/- 0.01, N = 33.683.74MIN: 3.44 / MAX: 13.64MIN: 3.71 / MAX: 4.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20210720Target: Vulkan GPU - Model: squeezenet_ssdRTX 3090RTX 30700.99681.99362.99043.98724.984SE +/- 0.13, N = 34.374.43MIN: 3.93 / MAX: 23.52MIN: 4.1 / MAX: 7.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ViennaCL

Test: CPU BLAS - dGEMM-NN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNRTX 3070RTX 30901224364860SE +/- 0.26, N = 353.152.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Bus Speed Readback

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed ReadbackRTX 3070RTX 3090612182430SE +/- 0.01, N = 327.1127.081. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Bus Speed Download

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadRTX 3070RTX 3090612182430SE +/- 0.02, N = 326.2826.261. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

ViennaCL

Test: OpenCL BLAS - dGEMM-TT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTRTX 30901302603905206506011. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL


Phoronix Test Suite v10.8.5