Linux GPU Compute

Intel Core i9-12900K testing with a ASUS ROG STRIX Z690-E GAMING WIFI (1601 BIOS) and NVIDIA GeForce RTX 3070 Ti 8GB on Ubuntu 22.04 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 2208115-NE-GPUCOMPUT74
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

CPU Massive 4 Tests
Creator Workloads 4 Tests
HPC - High Performance Computing 3 Tests
Machine Learning 2 Tests
Multi-Core 4 Tests
NVIDIA GPU Compute 18 Tests
OpenCL 9 Tests
Renderers 3 Tests
Vulkan Compute 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
RTX 3060
July 10 2022
  3 Hours, 51 Minutes
RTX 3060 Ti
July 12 2022
  3 Hours, 39 Minutes
RTX 3070
July 12 2022
  3 Hours, 20 Minutes
RTX 3070 Ti
July 12 2022
  3 Hours, 17 Minutes
RTX 3080
July 12 2022
  3 Hours, 14 Minutes
RTX 3080 Ti
July 12 2022
  3 Hours, 43 Minutes
RTX 3090
July 08 2022
  3 Hours, 39 Minutes
Invert Hiding All Results Option
  3 Hours, 32 Minutes

Only show results where is faster than
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):


Linux GPU ComputeProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090Intel Core i9-12900K @ 5.20GHz (16 Cores / 24 Threads)ASUS ROG STRIX Z690-E GAMING WIFI (1601 BIOS)Intel Alder Lake-S PCH32GB1000GB Western Digital WDS100T1X0E-00AFY0eVGA NVIDIA GeForce RTX 3060 12GBIntel Alder Lake-S HD AudioASUS VP28UIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 22.045.18.0-051800-generic (x86_64)GNOME Shell 42.2X Server 1.21.1.3NVIDIA 515.49.064.6.0OpenCL 3.0 CUDA 11.7.991.3.217GCC 11.2.0ext43840x21601000GB Western Digital WDS100T1X0E-00AFY0 + 2000GBNVIDIA GeForce RTX 3060 Ti 8GBNVIDIA GeForce RTX 3070 8GBNVIDIA GeForce RTX 3070 Ti 8GBNVIDIA GeForce RTX 3080 10GBNVIDIA GeForce RTX 3080 Ti 12GBIntel Device 7aa7NVIDIA GeForce RTX 3090 24GBIntel Device 7ad0OpenBenchmarking.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-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/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: intel_pstate performance (EPP: performance) - CPU Microcode: 0x1f - Thermald 2.4.9Graphics Details- RTX 3060: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 94.06.14.40.46- RTX 3060 Ti: BAR1 / Visible vRAM Size: 8192 MiB - vBIOS Version: 94.04.25.00.2c- RTX 3070: BAR1 / Visible vRAM Size: 8192 MiB - vBIOS Version: 94.04.25.00.2b- RTX 3070 Ti: BAR1 / Visible vRAM Size: 8192 MiB - vBIOS Version: 94.04.5b.00.02- RTX 3080: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 94.02.20.00.07- RTX 3080 Ti: BAR1 / Visible vRAM Size: 16384 MiB - vBIOS Version: 94.02.71.00.01- RTX 3090: BAR1 / Visible vRAM Size: 32768 MiB - vBIOS Version: 94.02.27.00.02OpenCL Details- RTX 3060: GPU Compute Cores: 3584- RTX 3060 Ti: GPU Compute Cores: 4864- RTX 3070: GPU Compute Cores: 5888- RTX 3070 Ti: GPU Compute Cores: 6144- RTX 3080: GPU Compute Cores: 8704- RTX 3080 Ti: GPU Compute Cores: 10240- RTX 3090: GPU Compute Cores: 10496Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected Python Details- RTX 3090: Python 3.10.4

RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090Result OverviewPhoronix Test Suite100%151%202%253%303%vkpeakHashcatVkResampleLuxCoreRenderLeelaChessZeroChaos Group V-RAYMandelGPUOctaneBenchIndigoBenchFinanceBenchRealSR-NCNNcl-memFAHBenchViennaCLclpeakSHOC Scalable HeterOgeneous ComputingDarktableRodiniaWaifu2x-NCNN VulkanJuliaGPUSmallPT GPU

RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090Logarithmic Per Watt Result OverviewPhoronix Test SuiteMeta Performance Per WattsSmallPT GPUJuliaGPUSHOC Scalable HeterOgeneous ComputingclpeakFAHBenchHashcatMandelGPUChaos Group V-RAYIndigoBenchLuxCoreRenderLeelaChessZeroOctaneBenchcl-memP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

Linux GPU Computeshoc: OpenCL - Bus Speed Readbackshoc: OpenCL - Triadshoc: OpenCL - Bus Speed Downloadshoc: OpenCL - Texture Read Bandwidthcl-mem: Copycl-mem: Writeviennacl: OpenCL BLAS - sCOPYshoc: OpenCL - Reductioncl-mem: Readviennacl: OpenCL BLAS - sAXPYviennacl: OpenCL BLAS - sDOTviennacl: OpenCL BLAS - dCOPYviennacl: OpenCL BLAS - dAXPYviennacl: OpenCL BLAS - dDOTviennacl: OpenCL BLAS - dGEMV-Nviennacl: OpenCL BLAS - dGEMV-Tclpeak: Global Memory Bandwidthclpeak: Transfer Bandwidth enqueueReadBufferclpeak: Transfer Bandwidth enqueueWriteBuffervkpeak: fp32-scalarvkpeak: fp32-vec4vkpeak: fp16-scalarvkpeak: fp16-vec4vkpeak: fp64-scalarvkpeak: fp64-vec4shoc: OpenCL - S3Dshoc: OpenCL - FFT SPshoc: OpenCL - GEMM SGEMM_Nclpeak: Single-Precision Floatclpeak: Double-Precision Doubleviennacl: OpenCL BLAS - dGEMM-NNviennacl: OpenCL BLAS - dGEMM-NTviennacl: OpenCL BLAS - dGEMM-TNviennacl: OpenCL BLAS - dGEMM-TTshoc: OpenCL - MD5 Hashvkpeak: int32-scalarvkpeak: int32-vec4vkpeak: int16-scalarvkpeak: int16-vec4clpeak: Integer Compute INThashcat: MD5hashcat: SHA1hashcat: 7-Ziphashcat: SHA-512hashcat: TrueCrypt RIPEMD160 + XTSindigobench: OpenCL GPU - Bedroomindigobench: OpenCL GPU - Supercarluxcorerender: DLSC - GPUluxcorerender: Danish Mood - GPUluxcorerender: Orange Juice - GPUluxcorerender: LuxCore Benchmark - GPUluxcorerender: Rainbow Colors and Prism - GPUlczero: OpenCLfahbench: juliagpu: GPUmandelgpu: GPUsmallpt-gpu: GPU - Complexsmallpt-gpu: GPU - Cornellsmallpt-gpu: GPU - Caustic3octanebench: Total Scorev-ray: NVIDIA CUDA GPUv-ray: NVIDIA RTX GPUvkresample: 2x - Singlefinancebench: Monte-Carlo OpenCLfinancebench: Black-Scholes OpenCLrealsr-ncnn: 4x - Norealsr-ncnn: 4x - Yeswaifu2x-ncnn: 2x - 3 - Yesrodinia: OpenCL Myocyterodinia: OpenCL Particle Filterdarktable: Boat - OpenCLdarktable: Masskrug - OpenCLdarktable: Server Rack - OpenCLdarktable: Server Room - OpenCLclpeak: Kernel LatencyRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309026.352223.352425.10021269.69270.3299.9276307.567318.0307292316327331223304315.4318.2621.376808.219036.986820.2613484.60212.35210.82171.946758.3972778.3412563.04212.0420420420315.12466799.666769.024494.325955.726347.142488596666777354666674137899949833332876388.50725.9284.293.264.874.0513.5025149186.7246378413490.2215226018.8165761985116576199841657620126294.862019855116323.184622.23199516.24446610.45168.6924.89928.5659.5212.5732.6260.1261.8653.7726.354623.689024.69272137.68297.3385.9297325.496392.9358330372399402224335389.1518.1721.809750.7512877.559725.9019166.40304.18304.44212.470962.4173435.6316043.12305.2429129229121.56409705.339663.576413.898487.358397.043522180000010987733333578250141213333341794411.50434.0815.734.356.145.3918.0737359245.7954411524637.1285067246.1165766026116576603941657660536387.1688671263163217.689465.27842912.8508.32851.5554.19027.6426.9272.5712.6040.1221.8573.8126.354323.661724.72052150.10296.4382.6285327.410392.9359328368397401222335389.3918.1021.9211587.4115337.2611579.7122752.71361.35362.02219.1051136.263802.5320056.49361.7634034134025.326911562.5911509.587639.1810095.6310206.344136915714312963333333669744166498333349377512.97037.3946.615.116.986.2220.3340342274.5694426397082.5321865318.9165762835916576284931657628635417.9146851416178717.558465.29671410.3797.79247.3014.02628.1395.9562.5782.6060.1261.8473.7526.354624.036024.71252102.79338.6518.0337381.559539.5437370482529527238371529.2618.1821.7711705.3915490.3411766.5023227.90366.40367.88285.7871292.094697.6721518.74370.1134734834726.213411695.9911651.047705.9310244.7510945.904267337142913271300000712544170406666750487813.79839.1287.005.427.216.6121.0847753275.1210427836038.7329682482.6165767034716576704811657670624461.8634091505199013.537474.0894349.7307.05741.3263.71828.4045.7712.5792.6050.1241.8443.7226.354224.250624.69812227.53354.6647.6359390.626673.3477375558632605243379662.2518.1721.8817334.5322809.3617271.0534090.93539.84539.15338.6891763.586356.2329616.92545.5850450650250637.534617172.3717105.4011320.9014594.4515397.275999948571418975633333976022242603333371207817.71846.9999.527.329.039.1426.5456188333.5891459314633.0428359837.3165763791916576380551657638199574.8658831788243611.272388.8207147.3526.01732.8533.33727.9804.1892.5772.6120.1261.8493.7826.353924.388724.70122182.89356.2738.8359383.712794.4492367591707654232369794.6618.1321.8919739.0326094.9719662.1038772.75614.23612.78419.8672016.867981.9033795.91625.4256156355956541.872019552.2219459.2712873.8416331.7617218.7167689533333214214833331086667272865000080005020.28851.90310.988.4510.2310.5130.4861726345.5999472565820.3461789875.5165764873816576488731657649017677.708281206428409.570326.0891426.6815.64129.8193.26828.5583.7572.5792.6180.1281.8703.8626.352724.436725.03302246.05357.8730.2362389.582793.7498374602717644235374795.9818.2621.3620867.7827459.9120781.4140968.90651.77653.38429.4262103.738197.2734509.53640.1659259459059343.548420754.4820669.2213699.5716879.0617830.0069887233333221008500001109178282036666779817520.68352.71211.288.5410.3410.8531.0161151353.4483492011904.2501044446.7165733889216573390281657339172682.29681208428699.422325.2667156.3875.58729.1853.26628.9403.9571.1181.9210.1240.7074.09OpenBenchmarking.org

GPU Temperature Monitor

OpenBenchmarking.orgCelsiusGPU Temperature MonitorPhoronix Test Suite System MonitoringRTX 3080 TiRTX 3080RTX 3070 TiRTX 3090RTX 3070RTX 3060RTX 3060 Ti1428425670Min: 28 / Avg: 60.86 / Max: 75Min: 25 / Avg: 60.01 / Max: 73Min: 34 / Avg: 59.76 / Max: 75Min: 34 / Avg: 56.12 / Max: 68Min: 27 / Avg: 55.86 / Max: 72Min: 42 / Avg: 54.3 / Max: 65Min: 26 / Avg: 54.14 / Max: 70

GPU Power Consumption Monitor

OpenBenchmarking.orgWattsGPU Power Consumption MonitorPhoronix Test Suite System MonitoringRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3070RTX 3060 TiRTX 306060120180240300Min: 13.97 / Avg: 211.78 / Max: 349.8Min: 26.71 / Avg: 190.85 / Max: 348.7Min: 8.28 / Avg: 188.35 / Max: 323.12Min: 13.41 / Avg: 151.19 / Max: 289.69Min: 8.61 / Avg: 115.13 / Max: 219.94Min: 9.47 / Avg: 107.18 / Max: 199.84Min: 18.7 / Avg: 88.76 / Max: 170.06

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.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed ReadbackRTX 3060RTX 3090RTX 3080 TiRTX 3080RTX 3070RTX 3060 TiRTX 3070 Ti612182430SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 15SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 1326.3526.3526.3526.3526.3526.3526.351. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed ReadbackRTX 3060RTX 3090RTX 3080 TiRTX 3080RTX 3070RTX 3060 TiRTX 3070 Ti612182430Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 26.35 / Avg: 26.35 / Max: 26.35Min: 26.35 / Avg: 26.35 / Max: 26.351. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3060 TiRTX 3060RTX 30700.18090.36180.54270.72360.90450.3500.3760.4380.4940.7720.7820.804

OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3060 TiRTX 3060RTX 30702468103.4053.6014.2264.9566.3306.3736.445

OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3060 TiRTX 3070RTX 30600.17640.35280.52920.70560.8820.3090.3150.3560.4320.6800.6930.784

OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthRTX 3090RTX 3080 TiRTX 3080RTX 3070 TiRTX 3060RTX 3070RTX 3060 Ti5101520259.99910.05112.21514.90115.83619.84721.686

cl-mem

OpenBenchmarking.orgGB/s Per Watt, More Is Bettercl-mem 2017-01-13Benchmark: CopyRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3070RTX 3060 TiRTX 30600.93021.86042.79063.72084.6512.3622.3872.7413.0983.8823.9544.134

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.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 3090612182430SE +/- 0.01, N = 13SE +/- 0.01, N = 13SE +/- 0.01, N = 13SE +/- 0.01, N = 15SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.03, N = 1323.3523.6623.6924.0424.2524.3924.441. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: TriadRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 3090612182430Min: 23.28 / Avg: 23.35 / Max: 23.4Min: 23.6 / Avg: 23.66 / Max: 23.68Min: 23.65 / Avg: 23.69 / Max: 23.73Min: 23.97 / Avg: 24.04 / Max: 24.06Min: 24.21 / Avg: 24.25 / Max: 24.27Min: 24.36 / Avg: 24.39 / Max: 24.41Min: 24.22 / Avg: 24.44 / Max: 24.531. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadRTX 3060 TiRTX 3080RTX 3080 TiRTX 3070 TiRTX 3070RTX 3090RTX 3060612182430SE +/- 0.01, N = 13SE +/- 0.01, N = 13SE +/- 0.02, N = 13SE +/- 0.02, N = 13SE +/- 0.03, N = 13SE +/- 0.00, N = 13SE +/- 0.01, N = 1324.6924.7024.7024.7124.7225.0325.101. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Bus Speed DownloadRTX 3060 TiRTX 3080RTX 3080 TiRTX 3070 TiRTX 3070RTX 3090RTX 3060612182430Min: 24.67 / Avg: 24.69 / Max: 24.76Min: 24.67 / Avg: 24.7 / Max: 24.79Min: 24.67 / Avg: 24.7 / Max: 24.87Min: 24.67 / Avg: 24.71 / Max: 24.84Min: 24.66 / Avg: 24.72 / Max: 24.99Min: 25.02 / Avg: 25.03 / Max: 25.05Min: 25.07 / Avg: 25.1 / Max: 25.211. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthRTX 3060RTX 3070 TiRTX 3060 TiRTX 3070RTX 3080 TiRTX 3080RTX 30905001000150020002500SE +/- 1.37, N = 4SE +/- 4.43, N = 4SE +/- 4.32, N = 4SE +/- 5.40, N = 4SE +/- 4.90, N = 3SE +/- 7.52, N = 3SE +/- 6.59, N = 31269.692102.792137.682150.102182.892227.532246.051. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: Texture Read BandwidthRTX 3060RTX 3070 TiRTX 3060 TiRTX 3070RTX 3080 TiRTX 3080RTX 3090400800120016002000Min: 1266.4 / Avg: 1269.69 / Max: 1272.99Min: 2092.97 / Avg: 2102.79 / Max: 2111.94Min: 2128.52 / Avg: 2137.68 / Max: 2145.94Min: 2140.33 / Avg: 2150.1 / Max: 2162.59Min: 2173.21 / Avg: 2182.89 / Max: 2189.09Min: 2216.04 / Avg: 2227.53 / Max: 2241.67Min: 2233.1 / Avg: 2246.05 / Max: 2254.651. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

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: CopyRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 309080160240320400SE +/- 0.05, N = 7SE +/- 0.16, N = 8SE +/- 0.11, N = 8SE +/- 0.11, N = 9SE +/- 0.03, N = 9SE +/- 0.02, N = 10SE +/- 0.24, N = 9270.3296.4297.3338.6354.6356.2357.81. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: CopyRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 309060120180240300Min: 270.2 / Avg: 270.3 / Max: 270.6Min: 295.9 / Avg: 296.38 / Max: 297Min: 296.9 / Avg: 297.25 / Max: 297.8Min: 338.1 / Avg: 338.6 / Max: 339Min: 354.5 / Avg: 354.64 / Max: 354.8Min: 356.1 / Avg: 356.15 / Max: 356.2Min: 357.1 / Avg: 357.84 / Max: 359.11. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti160320480640800SE +/- 0.04, N = 7SE +/- 0.06, N = 8SE +/- 0.05, N = 8SE +/- 0.04, N = 9SE +/- 0.09, N = 9SE +/- 0.77, N = 9SE +/- 0.10, N = 10299.9382.6385.9518.0647.6730.2738.81. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: WriteRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti130260390520650Min: 299.8 / Avg: 299.9 / Max: 300Min: 382.3 / Avg: 382.56 / Max: 382.8Min: 385.7 / Avg: 385.85 / Max: 386.1Min: 517.9 / Avg: 518.04 / Max: 518.2Min: 647.1 / Avg: 647.56 / Max: 647.8Min: 724.4 / Avg: 730.17 / Max: 732.1Min: 738.1 / Avg: 738.75 / Max: 739.21. (CC) gcc options: -O2 -flto -lOpenCL

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: OpenCL BLAS - sCOPYRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 309080160240320400SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.58, N = 32762852923373593593621. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sCOPYRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 309060120180240300Min: 276 / Avg: 276 / Max: 276Min: 285 / Avg: 285.33 / Max: 286Min: 291 / Avg: 292 / Max: 293Min: 337 / Avg: 337.33 / Max: 338Min: 359 / Avg: 359.33 / Max: 360Min: 359 / Avg: 359 / Max: 359Min: 361 / Avg: 362 / Max: 3631. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

cl-mem

OpenBenchmarking.orgGB/s Per Watt, More Is Bettercl-mem 2017-01-13Benchmark: ReadRTX 3070 TiRTX 3060RTX 3070RTX 3060 TiRTX 3080RTX 3080 TiRTX 30901.19272.38543.57814.77085.96354.8024.8155.1565.2075.2475.2945.301

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.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080 TiRTX 3090RTX 308080160240320400SE +/- 0.09, N = 12SE +/- 0.51, N = 12SE +/- 0.31, N = 12SE +/- 0.13, N = 13SE +/- 0.17, N = 13SE +/- 0.07, N = 13SE +/- 0.08, N = 13307.57325.50327.41381.56383.71389.58390.631. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: ReductionRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080 TiRTX 3090RTX 308070140210280350Min: 306.91 / Avg: 307.57 / Max: 308.04Min: 323.11 / Avg: 325.5 / Max: 327.92Min: 325.62 / Avg: 327.41 / Max: 329.74Min: 380.56 / Avg: 381.56 / Max: 382.25Min: 382.31 / Avg: 383.71 / Max: 384.26Min: 389.3 / Avg: 389.58 / Max: 389.93Min: 390.18 / Avg: 390.63 / Max: 391.011. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

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: ReadRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti2004006008001000SE +/- 0.09, N = 7SE +/- 0.11, N = 8SE +/- 0.13, N = 8SE +/- 0.02, N = 9SE +/- 0.08, N = 9SE +/- 0.29, N = 9SE +/- 0.13, N = 10318.0392.9392.9539.5673.3793.7794.41. (CC) gcc options: -O2 -flto -lOpenCL
OpenBenchmarking.orgGB/s, More Is Bettercl-mem 2017-01-13Benchmark: ReadRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti140280420560700Min: 317.7 / Avg: 317.96 / Max: 318.4Min: 392.6 / Avg: 392.85 / Max: 393.6Min: 392.5 / Avg: 392.93 / Max: 393.7Min: 539.3 / Avg: 539.46 / Max: 539.5Min: 672.7 / Avg: 673.31 / Max: 673.5Min: 791.8 / Avg: 793.68 / Max: 794.7Min: 793.7 / Avg: 794.43 / Max: 795.11. (CC) gcc options: -O2 -flto -lOpenCL

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: OpenCL BLAS - sAXPYRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090110220330440550SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 33073583594364774924981. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sAXPYRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309090180270360450Min: 307 / Avg: 307 / Max: 307Min: 358 / Avg: 358.33 / Max: 359Min: 359 / Avg: 359.33 / Max: 360Min: 436 / Avg: 436.33 / Max: 437Min: 476 / Avg: 476.67 / Max: 477Min: 491 / Avg: 492.33 / Max: 493Min: 497 / Avg: 497.67 / Max: 4981. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTRTX 3060RTX 3070RTX 3060 TiRTX 3080 TiRTX 3070 TiRTX 3090RTX 308080160240320400SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 32923283303673703743751. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - sDOTRTX 3060RTX 3070RTX 3060 TiRTX 3080 TiRTX 3070 TiRTX 3090RTX 308070140210280350Min: 292 / Avg: 292.33 / Max: 293Min: 328 / Avg: 328.33 / Max: 329Min: 329 / Avg: 330 / Max: 331Min: 366 / Avg: 366.67 / Max: 367Min: 370 / Avg: 370 / Max: 370Min: 373 / Avg: 373.67 / Max: 374Min: 373 / Avg: 374.67 / Max: 3761. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 3090130260390520650SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.67, N = 33163673724825585916021. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dCOPYRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 3090110220330440550Min: 316 / Avg: 316 / Max: 316Min: 367 / Avg: 367 / Max: 367Min: 372 / Avg: 372.33 / Max: 373Min: 482 / Avg: 482 / Max: 482Min: 558 / Avg: 558.33 / Max: 559Min: 590 / Avg: 590.67 / Max: 592Min: 601 / Avg: 602.33 / Max: 6031. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 3090150300450600750SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 33273973995296327077171. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dAXPYRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3080 TiRTX 3090130260390520650Min: 327 / Avg: 327 / Max: 327Min: 397 / Avg: 397 / Max: 397Min: 399 / Avg: 399 / Max: 399Min: 529 / Avg: 529.33 / Max: 530Min: 632 / Avg: 632 / Max: 632Min: 707 / Avg: 707.33 / Max: 708Min: 717 / Avg: 717.33 / Max: 7181. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti140280420560700SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.67, N = 33314014025276056446541. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dDOTRTX 3060RTX 3070RTX 3060 TiRTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti120240360480600Min: 331 / Avg: 331 / Max: 331Min: 400 / Avg: 400.67 / Max: 401Min: 402 / Avg: 402.33 / Max: 403Min: 527 / Avg: 527.33 / Max: 528Min: 604 / Avg: 604.67 / Max: 605Min: 644 / Avg: 644 / Max: 644Min: 653 / Avg: 653.67 / Max: 6551. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-NRTX 3070RTX 3060 TiRTX 3060RTX 3080 TiRTX 3090RTX 3070 TiRTX 308050100150200250SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 32222232232322352382431. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-NRTX 3070RTX 3060 TiRTX 3060RTX 3080 TiRTX 3090RTX 3070 TiRTX 30804080120160200Min: 221 / Avg: 221.67 / Max: 222Min: 223 / Avg: 223.33 / Max: 224Min: 223 / Avg: 223 / Max: 223Min: 231 / Avg: 231.67 / Max: 232Min: 234 / Avg: 234.67 / Max: 235Min: 238 / Avg: 238.33 / Max: 239Min: 242 / Avg: 242.67 / Max: 2431. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-TRTX 3060RTX 3060 TiRTX 3070RTX 3080 TiRTX 3070 TiRTX 3090RTX 308080160240320400SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 33033343353693713743791. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMV-TRTX 3060RTX 3060 TiRTX 3070RTX 3080 TiRTX 3070 TiRTX 3090RTX 308070140210280350Min: 303 / Avg: 303.33 / Max: 304Min: 333 / Avg: 334 / Max: 335Min: 334 / Avg: 334.67 / Max: 335Min: 369 / Avg: 369.33 / Max: 370Min: 371 / Avg: 371.33 / Max: 372Min: 373 / Avg: 374 / Max: 375Min: 378 / Avg: 379.33 / Max: 3801. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

clpeak

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

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30902004006008001000SE +/- 0.00, N = 11SE +/- 0.02, N = 13SE +/- 0.13, N = 15SE +/- 0.03, N = 13SE +/- 0.03, N = 12SE +/- 0.04, N = 13SE +/- 0.22, N = 15315.43389.15389.39529.26662.25794.66795.981. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090140280420560700Min: 315.41 / Avg: 315.43 / Max: 315.46Min: 389.07 / Avg: 389.15 / Max: 389.32Min: 387.58 / Avg: 389.39 / Max: 389.61Min: 529.02 / Avg: 529.26 / Max: 529.44Min: 662.09 / Avg: 662.25 / Max: 662.41Min: 794.54 / Avg: 794.66 / Max: 794.91Min: 795.36 / Avg: 795.98 / Max: 798.841. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueReadBufferRTX 3070RTX 3080 TiRTX 3060 TiRTX 3080RTX 3070 TiRTX 3060RTX 309048121620SE +/- 0.05, N = 8SE +/- 0.05, N = 5SE +/- 0.07, N = 8SE +/- 0.05, N = 5SE +/- 0.08, N = 8SE +/- 0.03, N = 5SE +/- 0.03, N = 818.1018.1318.1718.1718.1818.2618.261. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueReadBufferRTX 3070RTX 3080 TiRTX 3060 TiRTX 3080RTX 3070 TiRTX 3060RTX 3090510152025Min: 17.9 / Avg: 18.1 / Max: 18.32Min: 18.02 / Avg: 18.13 / Max: 18.3Min: 17.84 / Avg: 18.17 / Max: 18.39Min: 18.03 / Avg: 18.17 / Max: 18.34Min: 17.65 / Avg: 18.18 / Max: 18.33Min: 18.17 / Avg: 18.26 / Max: 18.32Min: 18.09 / Avg: 18.26 / Max: 18.391. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferRTX 3090RTX 3060RTX 3070 TiRTX 3060 TiRTX 3080RTX 3080 TiRTX 3070510152025SE +/- 0.08, N = 8SE +/- 0.06, N = 5SE +/- 0.06, N = 8SE +/- 0.06, N = 8SE +/- 0.10, N = 5SE +/- 0.08, N = 5SE +/- 0.07, N = 821.3621.3721.7721.8021.8821.8921.921. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferRTX 3090RTX 3060RTX 3070 TiRTX 3060 TiRTX 3080RTX 3080 TiRTX 3070510152025Min: 21.01 / Avg: 21.36 / Max: 21.63Min: 21.16 / Avg: 21.37 / Max: 21.49Min: 21.54 / Avg: 21.77 / Max: 22.01Min: 21.59 / Avg: 21.8 / Max: 22.06Min: 21.63 / Avg: 21.88 / Max: 22.25Min: 21.74 / Avg: 21.89 / Max: 22.12Min: 21.66 / Avg: 21.92 / Max: 22.241. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterclpeakOpenCL Test: Single-Precision FloatRTX 3070 TiRTX 3060 TiRTX 3080 TiRTX 3060RTX 3080RTX 3090RTX 307090180270360450311.99336.23337.79338.88339.06359.46435.01

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterclpeakOpenCL Test: Double-Precision DoubleRTX 3070 TiRTX 3080 TiRTX 3080RTX 3090RTX 3060RTX 3060 TiRTX 30701.08882.17763.26644.35525.4443.4453.7043.7713.8623.9184.3324.839

vkpeak

OpenBenchmarking.orgGFLOPS Per Watt, More Is Bettervkpeak 20210424fp64-vec4RTX 30600.55711.11421.67132.22842.78552.476

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KSE +/- 41.90, N = 3SE +/- 44.58, N = 3SE +/- 58.23, N = 3SE +/- 22.27, N = 3SE +/- 159.53, N = 3SE +/- 200.38, N = 3SE +/- 73.54, N = 36808.219750.7511587.4111705.3917334.5319739.0320867.78
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KMin: 6735.57 / Avg: 6808.21 / Max: 6880.73Min: 9698.6 / Avg: 9750.75 / Max: 9839.46Min: 11501.5 / Avg: 11587.41 / Max: 11698.47Min: 11682.78 / Avg: 11705.39 / Max: 11749.93Min: 17117.28 / Avg: 17334.53 / Max: 17645.52Min: 19537.95 / Avg: 19739.03 / Max: 20139.79Min: 20793.27 / Avg: 20867.78 / Max: 21014.85

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DRTX 3070 TiRTX 3060RTX 3080RTX 3080 TiRTX 3090RTX 3060 TiRTX 30702468104.8845.3525.3885.5385.8356.1796.650

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-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30906K12K18K24K30KSE +/- 2.30, N = 3SE +/- 45.93, N = 3SE +/- 65.69, N = 3SE +/- 0.80, N = 3SE +/- 50.25, N = 3SE +/- 179.42, N = 3SE +/- 66.30, N = 39036.9812877.5515337.2615490.3422809.3626094.9727459.91
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp32-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30905K10K15K20K25KMin: 9034.52 / Avg: 9036.98 / Max: 9041.58Min: 12805.16 / Avg: 12877.55 / Max: 12962.73Min: 15222.21 / Avg: 15337.26 / Max: 15449.74Min: 15488.75 / Avg: 15490.34 / Max: 15491.15Min: 22733.17 / Avg: 22809.36 / Max: 22904.2Min: 25914.94 / Avg: 26094.97 / Max: 26453.82Min: 27393.26 / Avg: 27459.91 / Max: 27592.51

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KSE +/- 16.13, N = 3SE +/- 16.37, N = 3SE +/- 30.65, N = 3SE +/- 43.21, N = 3SE +/- 86.43, N = 3SE +/- 135.01, N = 3SE +/- 0.16, N = 36820.269725.9011579.7111766.5017271.0519662.1020781.41
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KMin: 6788 / Avg: 6820.26 / Max: 6836.71Min: 9693.53 / Avg: 9725.9 / Max: 9746.28Min: 11518.85 / Avg: 11579.71 / Max: 11616.45Min: 11722.98 / Avg: 11766.5 / Max: 11852.91Min: 17183.62 / Avg: 17271.05 / Max: 17443.91Min: 19526.81 / Avg: 19662.1 / Max: 19932.12Min: 20781.11 / Avg: 20781.41 / Max: 20781.65

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30909K18K27K36K45KSE +/- 1.19, N = 3SE +/- 30.18, N = 3SE +/- 11.98, N = 3SE +/- 84.42, N = 3SE +/- 170.86, N = 3SE +/- 100.05, N = 3SE +/- 101.51, N = 313484.6019166.4022752.7123227.9034090.9338772.7540968.90
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp16-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30907K14K21K28K35KMin: 13482.26 / Avg: 13484.6 / Max: 13486.14Min: 19133.92 / Avg: 19166.4 / Max: 19226.7Min: 22740.33 / Avg: 22752.71 / Max: 22776.67Min: 23142.5 / Avg: 23227.9 / Max: 23396.74Min: 33917.88 / Avg: 34090.93 / Max: 34432.64Min: 38671.71 / Avg: 38772.75 / Max: 38972.84Min: 40865.77 / Avg: 40968.9 / Max: 41171.92

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090140280420560700SE +/- 0.74, N = 3SE +/- 0.72, N = 3SE +/- 0.34, N = 3SE +/- 0.74, N = 3SE +/- 2.71, N = 3SE +/- 1.72, N = 3SE +/- 1.62, N = 3212.35304.18361.35366.40539.84614.23651.77
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090110220330440550Min: 210.86 / Avg: 212.35 / Max: 213.12Min: 302.73 / Avg: 304.18 / Max: 304.9Min: 360.66 / Avg: 361.35 / Max: 361.7Min: 365.64 / Avg: 366.4 / Max: 367.89Min: 537.12 / Avg: 539.84 / Max: 545.25Min: 612.23 / Avg: 614.23 / Max: 617.66Min: 648.52 / Avg: 651.77 / Max: 653.42

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090140280420560700SE +/- 0.00, N = 2SE +/- 0.47, N = 3SE +/- 0.42, N = 3SE +/- 0.01, N = 3SE +/- 2.58, N = 3SE +/- 1.42, N = 3SE +/- 0.03, N = 3210.82304.44362.02367.88539.15612.78653.38
OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20210424fp64-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090120240360480600Min: 210.81 / Avg: 210.82 / Max: 210.82Min: 303.51 / Avg: 304.44 / Max: 304.91Min: 361.18 / Avg: 362.02 / Max: 362.44Min: 367.86 / Avg: 367.88 / Max: 367.9Min: 536.06 / Avg: 539.15 / Max: 544.27Min: 610.79 / Avg: 612.78 / Max: 615.52Min: 653.34 / Avg: 653.38 / Max: 653.43

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: S3DRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309090180270360450SE +/- 0.05, N = 13SE +/- 0.04, N = 13SE +/- 0.07, N = 14SE +/- 0.05, N = 14SE +/- 0.15, N = 14SE +/- 0.09, N = 14SE +/- 0.12, N = 13171.95212.47219.11285.79338.69419.87429.431. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: S3DRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309080160240320400Min: 171.67 / Avg: 171.95 / Max: 172.27Min: 212.11 / Avg: 212.47 / Max: 212.64Min: 218.57 / Avg: 219.11 / Max: 219.46Min: 285.49 / Avg: 285.79 / Max: 286.15Min: 337.94 / Avg: 338.69 / Max: 339.75Min: 419.26 / Avg: 419.87 / Max: 420.25Min: 428.61 / Avg: 429.43 / Max: 430.121. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SPRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30905001000150020002500SE +/- 0.18, N = 12SE +/- 0.25, N = 12SE +/- 0.29, N = 12SE +/- 0.04, N = 12SE +/- 0.07, N = 13SE +/- 0.14, N = 14SE +/- 0.07, N = 12758.40962.421136.261292.091763.582016.862103.731. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: FFT SPRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090400800120016002000Min: 757.59 / Avg: 758.4 / Max: 759.56Min: 961.82 / Avg: 962.42 / Max: 964.26Min: 1134.88 / Avg: 1136.26 / Max: 1138.19Min: 1291.88 / Avg: 1292.09 / Max: 1292.48Min: 1763.43 / Avg: 1763.58 / Max: 1764.29Min: 2015.63 / Avg: 2016.86 / Max: 2017.35Min: 2103.32 / Avg: 2103.73 / Max: 2104.071. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30902K4K6K8K10KSE +/- 5.73, N = 9SE +/- 3.49, N = 10SE +/- 16.49, N = 10SE +/- 6.05, N = 11SE +/- 23.13, N = 11SE +/- 24.71, N = 12SE +/- 29.85, N = 122778.343435.633802.534697.676356.237981.908197.271. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: GEMM SGEMM_NRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309014002800420056007000Min: 2757.05 / Avg: 2778.34 / Max: 2806.26Min: 3417.86 / Avg: 3435.63 / Max: 3452.5Min: 3711.35 / Avg: 3802.53 / Max: 3880.68Min: 4667.08 / Avg: 4697.67 / Max: 4733.93Min: 6212.13 / Avg: 6356.23 / Max: 6465.43Min: 7837.46 / Avg: 7981.9 / Max: 8110.78Min: 8064.58 / Avg: 8197.27 / Max: 8413.681. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

clpeak

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

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30907K14K21K28K35KSE +/- 11.99, N = 14SE +/- 35.84, N = 14SE +/- 12.68, N = 14SE +/- 1.58, N = 14SE +/- 43.31, N = 13SE +/- 46.12, N = 15SE +/- 35.71, N = 1412563.0416043.1220056.4921518.7429616.9233795.9134509.531. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30906K12K18K24K30KMin: 12446.47 / Avg: 12563.04 / Max: 12595.21Min: 15856.39 / Avg: 16043.12 / Max: 16374.26Min: 19921.58 / Avg: 20056.49 / Max: 20100.86Min: 21507.77 / Avg: 21518.74 / Max: 21529.11Min: 29426.19 / Avg: 29616.92 / Max: 29866.56Min: 33678.31 / Avg: 33795.91 / Max: 34399.59Min: 34366.04 / Avg: 34509.53 / Max: 34797.611. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090140280420560700SE +/- 0.85, N = 3SE +/- 0.47, N = 3SE +/- 1.02, N = 3SE +/- 0.96, N = 3SE +/- 1.73, N = 3SE +/- 2.30, N = 3SE +/- 1.62, N = 3212.04305.24361.76370.11545.58625.42640.161. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090110220330440550Min: 211.18 / Avg: 212.04 / Max: 213.73Min: 304.31 / Avg: 305.24 / Max: 305.72Min: 360.73 / Avg: 361.76 / Max: 363.79Min: 368.19 / Avg: 370.11 / Max: 371.08Min: 542.12 / Avg: 545.58 / Max: 547.34Min: 620.82 / Avg: 625.42 / Max: 627.74Min: 638.5 / Avg: 640.16 / Max: 643.391. (CXX) g++ options: -O3 -rdynamic -lOpenCL

ViennaCL

OpenBenchmarking.orgGFLOPs/s Per Watt, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTRTX 3080 TiRTX 3080RTX 30900.63251.2651.89752.533.16252.6212.6722.811

OpenBenchmarking.orgGFLOPs/s Per Watt, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTRTX 30600.51661.03321.54982.06642.5832.296

OpenBenchmarking.orgGFLOPs/s Per Watt, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNRTX 3070 TiRTX 3060RTX 3060 TiRTX 30700.6661.3321.9982.6643.332.2452.2922.6352.934

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090130260390520650SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 1.00, N = 3SE +/- 1.00, N = 3SE +/- 1.33, N = 3SE +/- 1.00, N = 32032903383475045615921. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NNRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090100200300400500Min: 203 / Avg: 203.33 / Max: 204Min: 289 / Avg: 289.67 / Max: 290Min: 337 / Avg: 338.33 / Max: 340Min: 346 / Avg: 347 / Max: 349Min: 502 / Avg: 504 / Max: 505Min: 560 / Avg: 561.33 / Max: 564Min: 590 / Avg: 592 / Max: 5931. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090130260390520650SE +/- 0.00, N = 2SE +/- 0.50, N = 2SE +/- 0.88, N = 3SE +/- 1.00, N = 3SE +/- 1.33, N = 3SE +/- 1.33, N = 3SE +/- 1.00, N = 32042913393485065635941. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-NTRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090100200300400500Min: 204 / Avg: 204 / Max: 204Min: 290 / Avg: 290.5 / Max: 291Min: 338 / Avg: 339.33 / Max: 341Min: 347 / Avg: 348 / Max: 350Min: 503 / Avg: 505.67 / Max: 507Min: 562 / Avg: 563.33 / Max: 566Min: 592 / Avg: 594 / Max: 5951. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090130260390520650SE +/- 1.20, N = 3SE +/- 1.33, N = 3SE +/- 0.88, N = 32032893373475025595901. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TNRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090100200300400500Min: 500 / Avg: 502.33 / Max: 504Min: 558 / Avg: 559.33 / Max: 562Min: 588 / Avg: 589.67 / Max: 5911. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTRTX 3080RTX 3080 TiRTX 3090130260390520650SE +/- 0.00, N = 2SE +/- 1.00, N = 35065655931. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: OpenCL BLAS - dGEMM-TTRTX 3080RTX 3080 TiRTX 3090100200300400500Min: 506 / Avg: 506 / Max: 506Min: 591 / Avg: 593 / Max: 5941. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgGHash/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashRTX 3060RTX 3070 TiRTX 3060 TiRTX 3080 TiRTX 3080RTX 3070RTX 30900.10870.21740.32610.43480.54350.2840.3500.3970.4310.4400.4800.483

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30901020304050SE +/- 0.00, N = 15SE +/- 0.01, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.03, N = 14SE +/- 0.05, N = 14SE +/- 0.03, N = 1415.1221.5625.3326.2137.5341.8743.551. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi
OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2020-04-17Target: OpenCL - Benchmark: MD5 HashRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090918273645Min: 15.06 / Avg: 15.12 / Max: 15.13Min: 21.48 / Avg: 21.56 / Max: 21.65Min: 25.3 / Avg: 25.33 / Max: 25.33Min: 26.2 / Avg: 26.21 / Max: 26.23Min: 37.41 / Avg: 37.53 / Max: 37.72Min: 41.66 / Avg: 41.87 / Max: 42.34Min: 43.23 / Avg: 43.55 / Max: 43.641. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -lmpi_cxx -lmpi

clpeak

OpenBenchmarking.orgGIOPS Per Watt, More Is BetterclpeakOpenCL Test: Integer Compute INTRTX 3080RTX 3070 TiRTX 3060 TiRTX 3080 TiRTX 3090RTX 3060RTX 30704080120160200124.79125.03130.32132.96137.87139.61165.20

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.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KSE +/- 0.32, N = 3SE +/- 18.16, N = 3SE +/- 15.80, N = 3SE +/- 0.79, N = 3SE +/- 86.56, N = 3SE +/- 51.14, N = 3SE +/- 0.42, N = 36799.669705.3311562.5911695.9917172.3719552.2220754.48
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KMin: 6799.08 / Avg: 6799.66 / Max: 6800.18Min: 9686.91 / Avg: 9705.33 / Max: 9741.64Min: 11546.57 / Avg: 11562.59 / Max: 11594.19Min: 11694.46 / Avg: 11695.99 / Max: 11697.12Min: 17083.27 / Avg: 17172.37 / Max: 17345.46Min: 19500.8 / Avg: 19552.22 / Max: 19654.51Min: 20753.75 / Avg: 20754.48 / Max: 20755.22

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KSE +/- 0.10, N = 3SE +/- 18.74, N = 3SE +/- 16.41, N = 3SE +/- 0.41, N = 3SE +/- 84.25, N = 3SE +/- 49.71, N = 3SE +/- 0.25, N = 36769.029663.5711509.5811651.0417105.4019459.2720669.22
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int32-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KMin: 6768.86 / Avg: 6769.02 / Max: 6769.19Min: 9643.92 / Avg: 9663.57 / Max: 9701.03Min: 11493.14 / Avg: 11509.58 / Max: 11542.4Min: 11650.59 / Avg: 11651.04 / Max: 11651.87Min: 17018.2 / Avg: 17105.4 / Max: 17273.87Min: 19408.98 / Avg: 19459.27 / Max: 19558.69Min: 20668.91 / Avg: 20669.22 / Max: 20669.72

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903K6K9K12K15KSE +/- 0.20, N = 3SE +/- 11.82, N = 3SE +/- 11.50, N = 3SE +/- 0.41, N = 3SE +/- 28.42, N = 3SE +/- 0.07, N = 3SE +/- 0.10, N = 34494.326413.897639.187705.9311320.9012873.8413699.57
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-scalarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30902K4K6K8K10KMin: 4493.97 / Avg: 4494.32 / Max: 4494.65Min: 6401.93 / Avg: 6413.89 / Max: 6437.54Min: 7627.32 / Avg: 7639.18 / Max: 7662.18Min: 7705.11 / Avg: 7705.93 / Max: 7706.34Min: 11292.4 / Avg: 11320.9 / Max: 11377.74Min: 12873.75 / Avg: 12873.84 / Max: 12873.99Min: 13699.44 / Avg: 13699.57 / Max: 13699.76

OpenBenchmarking.orgGIOPS Per Watt, More Is Bettervkpeak 20210424int16-vec4RTX 3080 TiRTX 3090RTX 3070 TiRTX 3080RTX 3060 TiRTX 30702040608010062.6964.6465.5568.2680.4085.39

OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KSE +/- 21.63, N = 3SE +/- 13.25, N = 3SE +/- 24.75, N = 3SE +/- 0.41, N = 3SE +/- 15.08, N = 3SE +/- 2.65, N = 3SE +/- 1.27, N = 35955.728487.3510095.6310244.7514594.4516331.7616879.06
OpenBenchmarking.orgGIOPS, More Is Bettervkpeak 20210424int16-vec4RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903K6K9K12K15KMin: 5912.47 / Avg: 5955.72 / Max: 5977.41Min: 8473.58 / Avg: 8487.35 / Max: 8513.84Min: 10069.7 / Avg: 10095.63 / Max: 10145.11Min: 10243.92 / Avg: 10244.75 / Max: 10245.17Min: 14579.17 / Avg: 14594.45 / Max: 14624.62Min: 16328.4 / Avg: 16331.76 / Max: 16336.98Min: 16876.85 / Avg: 16879.06 / Max: 16881.25

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 INTRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904K8K12K16K20KSE +/- 6.64, N = 13SE +/- 65.78, N = 15SE +/- 41.26, N = 13SE +/- 42.92, N = 13SE +/- 119.12, N = 15SE +/- 79.28, N = 12SE +/- 120.91, N = 136347.148397.0410206.3410945.9015397.2717218.7117830.001. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903K6K9K12K15KMin: 6309.17 / Avg: 6347.14 / Max: 6383.21Min: 8040.63 / Avg: 8397.04 / Max: 8740.38Min: 10068.35 / Avg: 10206.34 / Max: 10513.41Min: 10810.96 / Avg: 10945.9 / Max: 11205.12Min: 14688.93 / Avg: 15397.27 / Max: 16045.46Min: 16986.58 / Avg: 17218.71 / Max: 17575.46Min: 17324.53 / Avg: 17830 / Max: 18482.581. (CXX) g++ options: -O3 -rdynamic -lOpenCL

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: MD5RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309015000M30000M45000M60000M75000MSE +/- 23113714.06, N = 6SE +/- 35135391.19, N = 7SE +/- 34107189.98, N = 7SE +/- 6723155.24, N = 7SE +/- 47721835.02, N = 7SE +/- 63727349.79, N = 6SE +/- 35085900.94, N = 624885966667352218000004136915714342673371429599994857146768953333369887233333
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: MD5RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309012000M24000M36000M48000M60000MMin: 24789500000 / Avg: 24885966666.67 / Max: 24925400000Min: 35119800000 / Avg: 35221800000 / Max: 35345500000Min: 41234400000 / Avg: 41369157142.86 / Max: 41504900000Min: 42654000000 / Avg: 42673371428.57 / Max: 42693300000Min: 59857300000 / Avg: 59999485714.29 / Max: 60186200000Min: 67440700000 / Avg: 67689533333.33 / Max: 67883600000Min: 69784600000 / Avg: 69887233333.33 / Max: 69981100000

OpenBenchmarking.orgH/s Per Watt, More Is BetterHashcat 6.2.4Benchmark: SHA1RTX 3070 TiRTX 3060RTX 3060 TiRTX 3080RTX 3080 TiRTX 3070RTX 309020M40M60M80M100M84457906.1194634461.4297823664.4498663614.59100043686.93104004535.65106101460.43

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30905000M10000M15000M20000M25000MSE +/- 8351553.42, N = 6SE +/- 5257481.44, N = 6SE +/- 11473321.71, N = 6SE +/- 3912288.33, N = 6SE +/- 14809381.41, N = 6SE +/- 21246072.84, N = 6SE +/- 24085870.69, N = 67735466667109877333331296333333313271300000189756333332142148333322100850000
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA1RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30904000M8000M12000M16000M20000MMin: 7717500000 / Avg: 7735466666.67 / Max: 7761600000Min: 10972400000 / Avg: 10987733333.33 / Max: 11011500000Min: 12930300000 / Avg: 12963333333.33 / Max: 12998800000Min: 13266100000 / Avg: 13271300000 / Max: 13290700000Min: 18928800000 / Avg: 18975633333.33 / Max: 19039100000Min: 21329000000 / Avg: 21421483333.33 / Max: 21465000000Min: 22030600000 / Avg: 22100850000 / Max: 22177800000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-ZipRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090200K400K600K800K1000KSE +/- 371.72, N = 9SE +/- 419.32, N = 10SE +/- 1441.17, N = 9SE +/- 348.45, N = 9SE +/- 612.10, N = 9SE +/- 955.83, N = 9SE +/- 3011.22, N = 941378957825066974471254497602210866671109178
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: 7-ZipRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090200K400K600K800K1000KMin: 411900 / Avg: 413788.89 / Max: 415400Min: 575500 / Avg: 578250 / Max: 580400Min: 661700 / Avg: 669744.44 / Max: 674200Min: 710500 / Avg: 712544.44 / Max: 713900Min: 971900 / Avg: 976022.22 / Max: 978400Min: 1081500 / Avg: 1086666.67 / Max: 1090700Min: 1088900 / Avg: 1109177.78 / Max: 1117200

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090600M1200M1800M2400M3000MSE +/- 1211725.12, N = 6SE +/- 832532.95, N = 6SE +/- 2532115.94, N = 6SE +/- 313758.58, N = 6SE +/- 3109197.82, N = 6SE +/- 2958462.44, N = 6SE +/- 2677519.58, N = 6994983333141213333316649833331704066667242603333327286500002820366667
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: SHA-512RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090500M1000M1500M2000M2500MMin: 992500000 / Avg: 994983333.33 / Max: 1000300000Min: 1409600000 / Avg: 1412133333.33 / Max: 1414500000Min: 1658300000 / Avg: 1664983333.33 / Max: 1676000000Min: 1703400000 / Avg: 1704066666.67 / Max: 1705400000Min: 2416400000 / Avg: 2426033333.33 / Max: 2435400000Min: 2722100000 / Avg: 2728650000 / Max: 2742200000Min: 2815200000 / Avg: 2820366666.67 / Max: 2831200000

OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti200K400K600K800K1000KSE +/- 2289.57, N = 8SE +/- 1142.14, N = 9SE +/- 752.08, N = 8SE +/- 1390.42, N = 9SE +/- 1903.85, N = 9SE +/- 2665.37, N = 8SE +/- 757.11, N = 8287638417944493775504878712078798175800050
OpenBenchmarking.orgH/s, More Is BetterHashcat 6.2.4Benchmark: TrueCrypt RIPEMD160 + XTSRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti140K280K420K560K700KMin: 281600 / Avg: 287637.5 / Max: 299200Min: 409800 / Avg: 417944.44 / Max: 422900Min: 491100 / Avg: 493775 / Max: 496100Min: 500600 / Avg: 504877.78 / Max: 511500Min: 701600 / Avg: 712077.78 / Max: 717400Min: 787800 / Avg: 798175 / Max: 814400Min: 796200 / Avg: 800050 / Max: 804000

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: BedroomRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090510152025SE +/- 0.006, N = 3SE +/- 0.009, N = 3SE +/- 0.019, N = 3SE +/- 0.014, N = 3SE +/- 0.022, N = 3SE +/- 0.028, N = 3SE +/- 0.048, N = 38.50711.50412.97013.79817.71820.28820.683
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: BedroomRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090510152025Min: 8.5 / Avg: 8.51 / Max: 8.52Min: 11.49 / Avg: 11.5 / Max: 11.52Min: 12.95 / Avg: 12.97 / Max: 13.01Min: 13.78 / Avg: 13.8 / Max: 13.83Min: 17.69 / Avg: 17.72 / Max: 17.76Min: 20.25 / Avg: 20.29 / Max: 20.34Min: 20.63 / Avg: 20.68 / Max: 20.78

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: SupercarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30901224364860SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 325.9334.0837.3939.1347.0051.9052.71
OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.4Acceleration: OpenCL GPU - Scene: SupercarRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30901122334455Min: 25.91 / Avg: 25.93 / Max: 25.94Min: 34.07 / Avg: 34.08 / Max: 34.09Min: 37.35 / Avg: 37.39 / Max: 37.44Min: 39.11 / Avg: 39.13 / Max: 39.16Min: 46.94 / Avg: 47 / Max: 47.03Min: 51.89 / Avg: 51.9 / Max: 51.94Min: 52.66 / Avg: 52.71 / Max: 52.76

LuxCoreRender

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: GPURTX 3080 TiRTX 3070 TiRTX 3080RTX 3060 TiRTX 3070RTX 3090RTX 30600.01190.02380.03570.04760.05950.0360.0370.0370.0410.0430.0500.053

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPURTX 3070 TiRTX 3080 TiRTX 3090RTX 3080RTX 3060RTX 3060 TiRTX 30700.0090.0180.0270.0360.0450.0330.0330.0330.0340.0360.0370.040

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPURTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3060 TiRTX 3070RTX 30600.01060.02120.03180.04240.0530.0350.0350.0360.0390.0440.0450.047

OpenBenchmarking.orgM samples/sec Per Watt, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPURTX 3070 TiRTX 3080RTX 3080 TiRTX 3060 TiRTX 3070RTX 3090RTX 30600.01240.02480.03720.04960.0620.0390.0390.0390.0430.0460.0530.055

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 34.295.736.617.009.5210.9811.28MIN: 4.11 / MAX: 4.53MIN: 5.31 / MAX: 5.98MIN: 6.05 / MAX: 6.91MIN: 6.87 / MAX: 7.16MIN: 8.4 / MAX: 9.74MIN: 10.11 / MAX: 11.41MIN: 10.49 / MAX: 11.58
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: DLSC - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215Min: 4.28 / Avg: 4.29 / Max: 4.3Min: 5.73 / Avg: 5.73 / Max: 5.74Min: 6.6 / Avg: 6.61 / Max: 6.62Min: 7 / Avg: 7 / Max: 7.01Min: 9.5 / Avg: 9.52 / Max: 9.56Min: 10.94 / Avg: 10.98 / Max: 11.04Min: 11.23 / Avg: 11.28 / Max: 11.31

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090246810SE +/- 0.02, N = 15SE +/- 0.05, N = 3SE +/- 0.06, N = 4SE +/- 0.01, N = 3SE +/- 0.05, N = 12SE +/- 0.06, N = 3SE +/- 0.06, N = 33.264.355.115.427.328.458.54MIN: 0.87 / MAX: 3.88MIN: 1.49 / MAX: 5.3MIN: 1.22 / MAX: 6.27MIN: 1.62 / MAX: 6.55MIN: 1.64 / MAX: 9.08MIN: 2.17 / MAX: 10.28MIN: 2.14 / MAX: 10.54
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215Min: 3.15 / Avg: 3.26 / Max: 3.38Min: 4.29 / Avg: 4.35 / Max: 4.44Min: 5.01 / Avg: 5.11 / Max: 5.25Min: 5.41 / Avg: 5.42 / Max: 5.43Min: 7.11 / Avg: 7.32 / Max: 7.7Min: 8.34 / Avg: 8.45 / Max: 8.54Min: 8.47 / Avg: 8.54 / Max: 8.65

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.09, N = 5SE +/- 0.03, N = 3SE +/- 0.01, N = 34.876.146.987.219.0310.2310.34MIN: 3.84 / MAX: 6.13MIN: 4.37 / MAX: 7.27MIN: 5.6 / MAX: 8.45MIN: 5.79 / MAX: 8.76MIN: 7.24 / MAX: 11.66MIN: 8.3 / MAX: 13.23MIN: 8.41 / MAX: 13.65
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215Min: 4.82 / Avg: 4.87 / Max: 4.92Min: 6.05 / Avg: 6.14 / Max: 6.19Min: 6.92 / Avg: 6.98 / Max: 7.06Min: 7.17 / Avg: 7.21 / Max: 7.25Min: 8.86 / Avg: 9.03 / Max: 9.34Min: 10.18 / Avg: 10.23 / Max: 10.29Min: 10.33 / Avg: 10.34 / Max: 10.35

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 34.055.396.226.619.1410.5110.85MIN: 1.14 / MAX: 4.78MIN: 1.78 / MAX: 6.36MIN: 2.17 / MAX: 7.35MIN: 2.19 / MAX: 7.8MIN: 2.47 / MAX: 10.83MIN: 2.79 / MAX: 12.46MIN: 3.57 / MAX: 12.86
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215Min: 4.05 / Avg: 4.05 / Max: 4.06Min: 5.39 / Avg: 5.39 / Max: 5.39Min: 6.18 / Avg: 6.22 / Max: 6.24Min: 6.58 / Avg: 6.61 / Max: 6.62Min: 9.14 / Avg: 9.14 / Max: 9.15Min: 10.5 / Avg: 10.51 / Max: 10.52Min: 10.83 / Avg: 10.85 / Max: 10.89

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090714212835SE +/- 0.06, N = 4SE +/- 0.11, N = 5SE +/- 0.12, N = 5SE +/- 0.18, N = 5SE +/- 0.13, N = 6SE +/- 0.47, N = 15SE +/- 0.32, N = 613.5018.0720.3321.0826.5430.4831.01MIN: 12.02 / MAX: 14.47MIN: 15.25 / MAX: 19.46MIN: 18.07 / MAX: 21.88MIN: 19.53 / MAX: 22.8MIN: 23.12 / MAX: 29.17MIN: 4.31 / MAX: 33.71MIN: 28.41 / MAX: 33.92
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090714212835Min: 13.39 / Avg: 13.5 / Max: 13.68Min: 17.64 / Avg: 18.07 / Max: 18.26Min: 19.96 / Avg: 20.33 / Max: 20.56Min: 20.49 / Avg: 21.08 / Max: 21.56Min: 26.07 / Avg: 26.54 / Max: 26.88Min: 24.65 / Avg: 30.48 / Max: 32.17Min: 29.69 / Avg: 31.01 / Max: 32.05

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: OpenCLRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti13K26K39K52K65KSE +/- 205.27, N = 3SE +/- 136.70, N = 3SE +/- 66.16, N = 3SE +/- 164.81, N = 3SE +/- 199.06, N = 3SE +/- 514.58, N = 3SE +/- 231.66, N = 3251493735940342477535618861151617261. (CXX) g++ options: -flto -pthread
OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: OpenCLRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti11K22K33K44K55KMin: 24831 / Avg: 25149.33 / Max: 25533Min: 37217 / Avg: 37358.67 / Max: 37632Min: 40240 / Avg: 40342 / Max: 40466Min: 47455 / Avg: 47753 / Max: 48024Min: 55905 / Avg: 56188 / Max: 56572Min: 60428 / Avg: 61151.33 / Max: 62147Min: 61333 / Avg: 61726 / Max: 621351. (CXX) g++ options: -flto -pthread

FAHBench

OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterFAHBench 2.3.2RTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3060RTX 3060 TiRTX 30700.56031.12061.68092.24122.80151.7671.8251.8931.9092.2982.4032.490

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309080160240320400SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.15, N = 3SE +/- 0.51, N = 3SE +/- 0.03, N = 3SE +/- 0.37, N = 3186.72245.80274.57275.12333.59345.60353.45
OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2RTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309060120180240300Min: 186.63 / Avg: 186.72 / Max: 186.8Min: 245.71 / Avg: 245.8 / Max: 245.85Min: 274.49 / Avg: 274.57 / Max: 274.72Min: 274.86 / Avg: 275.12 / Max: 275.38Min: 332.61 / Avg: 333.59 / Max: 334.31Min: 345.54 / Avg: 345.6 / Max: 345.66Min: 352.81 / Avg: 353.45 / Max: 354.08

Meta Performance Per Watts

OpenBenchmarking.orgPerformance Per Watts, More Is BetterMeta Performance Per WattsPerformance Per WattsRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30902K4K6K8K10K1454.125548.856071.236518.358566.059408.699893.96

MandelGPU

OpenBenchmarking.orgSamples/sec Per Watt, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPURTX 3070 TiRTX 3060RTX 3080 TiRTX 3080RTX 3060 TiRTX 3090RTX 3070800K1600K2400K3200K4000K3034056.003121625.403227841.883420743.433530918.963558069.623909868.25

JuliaGPU

JuliaGPU is an OpenCL benchmark with this version containing various PTS-specific enhancements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples/sec, More Is BetterJuliaGPU 1.2pts1OpenCL Device: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090110M220M330M440M550MSE +/- 642390.88, N = 8SE +/- 1945793.57, N = 8SE +/- 3561096.98, N = 8SE +/- 3560374.60, N = 8SE +/- 2361749.05, N = 8SE +/- 2408858.67, N = 9SE +/- 2811742.88, N = 8378413490.2411524637.1426397082.5427836038.7459314633.0472565820.3492011904.21. (CC) gcc options: -O3 -march=native -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL -lm
OpenBenchmarking.orgSamples/sec, More Is BetterJuliaGPU 1.2pts1OpenCL Device: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309090M180M270M360M450MMin: 375669869.5 / Avg: 378413490.15 / Max: 380695700.3Min: 405090313.3 / Avg: 411524637.06 / Max: 419870752Min: 410192585.4 / Avg: 426397082.45 / Max: 435387059.6Min: 414003147.4 / Avg: 427836038.73 / Max: 440102465.5Min: 447613931.1 / Avg: 459314633.04 / Max: 469223604Min: 461098942.6 / Avg: 472565820.28 / Max: 482086538.3Min: 473628884.3 / Avg: 492011904.16 / Max: 498432176.41. (CC) gcc options: -O3 -march=native -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL -lm

SmallPT GPU

OpenBenchmarking.orgSamples/sec Per Watt, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: ComplexRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3070RTX 3060 TiRTX 30605M10M15M20M25M7635643.237787782.188820802.2411758663.2115696067.1717485060.8023066499.50

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: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090110M220M330M440M550MSE +/- 242876.80, N = 7SE +/- 1279497.74, N = 8SE +/- 422788.81, N = 9SE +/- 530024.12, N = 9SE +/- 572727.71, N = 10SE +/- 1175920.80, N = 10SE +/- 1780706.00, N = 10215226018.8285067246.1321865318.9329682482.6428359837.3461789875.5501044446.71. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309090M180M270M360M450MMin: 214368831.4 / Avg: 215226018.83 / Max: 215957299.1Min: 282688119.9 / Avg: 285067246.14 / Max: 293648655.2Min: 319457420.4 / Avg: 321865318.9 / Max: 323144689.8Min: 327167075.8 / Avg: 329682482.6 / Max: 331570362.2Min: 425956724.2 / Avg: 428359837.29 / Max: 430710759.7Min: 454494273.8 / Avg: 461789875.5 / Max: 466118347.6Min: 488239486.4 / Avg: 501044446.72 / Max: 509127248.31. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

SmallPT GPU

OpenBenchmarking.orgSamples/sec Per Watt, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: CornellRTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3070RTX 3060 TiRTX 30602M4M6M8M10M5194201.295250171.615691498.286302630.538377067.469162824.8811552400.60

OpenBenchmarking.orgSamples/sec, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: ComplexRTX 3090RTX 3060RTX 3070RTX 3080RTX 3080 TiRTX 3060 TiRTX 3070 Ti400M800M1200M1600M2000MSE +/- 21.07, N = 3SE +/- 19.92, N = 3SE +/- 19.92, N = 3SE +/- 21.36, N = 3SE +/- 20.21, N = 3SE +/- 19.34, N = 3SE +/- 19.92, N = 316573388921657619851165762835916576379191657648738165766026116576703471. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: ComplexRTX 3090RTX 3060RTX 3070RTX 3080RTX 3080 TiRTX 3060 TiRTX 3070 Ti300M600M900M1200M1500MMin: 1657338856 / Avg: 1657338892.33 / Max: 1657338929Min: 1657619816 / Avg: 1657619850.67 / Max: 1657619885Min: 1657628324 / Avg: 1657628358.67 / Max: 1657628393Min: 1657637882 / Avg: 1657637919 / Max: 1657637956Min: 1657648703 / Avg: 1657648738.33 / Max: 1657648773Min: 1657660228 / Avg: 1657660261.33 / Max: 1657660295Min: 1657670312 / Avg: 1657670346.67 / Max: 16576703811. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

OpenBenchmarking.orgSamples/sec Per Watt, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: Caustic3RTX 3080 TiRTX 3090RTX 3080RTX 3070 TiRTX 3070RTX 3060 TiRTX 30602M4M6M8M10M5178820.325283173.025659717.136277002.498304302.539128614.0510690789.95

OpenBenchmarking.orgSamples/sec, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: CornellRTX 3090RTX 3060RTX 3070RTX 3080RTX 3080 TiRTX 3060 TiRTX 3070 Ti400M800M1200M1600M2000MSE +/- 24.54, N = 3SE +/- 24.25, N = 3SE +/- 24.54, N = 3SE +/- 24.54, N = 3SE +/- 24.54, N = 3SE +/- 24.25, N = 3SE +/- 24.54, N = 316573390281657619984165762849316576380551657648873165766039416576704811. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: CornellRTX 3090RTX 3060RTX 3070RTX 3080RTX 3080 TiRTX 3060 TiRTX 3070 Ti300M600M900M1200M1500MMin: 1657338986 / Avg: 1657339028.33 / Max: 1657339071Min: 1657619942 / Avg: 1657619984 / Max: 1657620026Min: 1657628450 / Avg: 1657628492.67 / Max: 1657628535Min: 1657638013 / Avg: 1657638055.33 / Max: 1657638098Min: 1657648831 / Avg: 1657648873.33 / Max: 1657648916Min: 1657660352 / Avg: 1657660394 / Max: 1657660436Min: 1657670438 / Avg: 1657670480.67 / Max: 16576705231. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

OpenBenchmarking.orgSamples/sec, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: Caustic3RTX 3090RTX 3060RTX 3070RTX 3080RTX 3080 TiRTX 3060 TiRTX 3070 Ti400M800M1200M1600M2000MSE +/- 25.12, N = 3SE +/- 24.83, N = 3SE +/- 24.54, N = 3SE +/- 24.83, N = 3SE +/- 24.83, N = 3SE +/- 24.54, N = 3SE +/- 24.83, N = 316573391721657620126165762863516576381991657649017165766053616576706241. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL
OpenBenchmarking.orgSamples/sec, More Is BetterSmallPT GPU 1.6pts1OpenCL Device: GPU - Scene: Caustic3RTX 3090RTX 3060RTX 3070RTX 3080RTX 3080 TiRTX 3060 TiRTX 3070 Ti300M600M900M1200M1500MMin: 1657339128 / Avg: 1657339171.67 / Max: 1657339215Min: 1657620083 / Avg: 1657620126 / Max: 1657620169Min: 1657628593 / Avg: 1657628635.33 / Max: 1657628678Min: 1657638156 / Avg: 1657638199 / Max: 1657638242Min: 1657648974 / Avg: 1657649017 / Max: 1657649060Min: 1657660494 / Avg: 1657660536.33 / Max: 1657660579Min: 1657670581 / Avg: 1657670624 / Max: 16576706671. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

OctaneBench

OpenBenchmarking.orgScore Per Watt, More Is BetterOctaneBench 2020.1Total ScoreRTX 3080RTX 3070 TiRTX 3090RTX 3080 TiRTX 3060RTX 3060 TiRTX 30700.54431.08861.63292.17722.72151.9592.0422.0472.0732.4102.4112.419

OpenBenchmarking.orgScore, More Is BetterOctaneBench 2020.1Total ScoreRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090150300450600750294.86387.17417.91461.86574.87677.71682.30

Chaos Group V-RAY

OpenBenchmarking.orgvpaths Per Watt, More Is BetterChaos Group V-RAY 5.02Mode: NVIDIA CUDA GPURTX 3080RTX 3080 TiRTX 3090RTX 3070 TiRTX 3060RTX 3060 TiRTX 307036912157.4727.5607.8037.8498.5659.2859.584

OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 5.02Mode: NVIDIA CUDA GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090400800120016002000SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 3.51, N = 3SE +/- 1.53, N = 3SE +/- 2.40, N = 3855126314161505178820642084
OpenBenchmarking.orgvpaths, More Is BetterChaos Group V-RAY 5.02Mode: NVIDIA CUDA GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090400800120016002000Min: 855 / Avg: 855 / Max: 855Min: 1262 / Avg: 1262.67 / Max: 1264Min: 1415 / Avg: 1416.33 / Max: 1418Min: 1505 / Avg: 1505.33 / Max: 1506Min: 1784 / Avg: 1788 / Max: 1795Min: 2062 / Avg: 2064 / Max: 2067Min: 2081 / Avg: 2084.33 / Max: 2089

OpenBenchmarking.orgvrays, More Is BetterChaos Group V-RAY 5.02Mode: NVIDIA RTX GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30906001200180024003000SE +/- 1.67, N = 3SE +/- 0.67, N = 3SE +/- 0.00, N = 3SE +/- 1.33, N = 3SE +/- 4.91, N = 3SE +/- 6.17, N = 3SE +/- 4.93, N = 31163163217871990243628402869
OpenBenchmarking.orgvrays, More Is BetterChaos Group V-RAY 5.02Mode: NVIDIA RTX GPURTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30905001000150020002500Min: 1160 / Avg: 1163.33 / Max: 1165Min: 1631 / Avg: 1632.33 / Max: 1633Min: 1787 / Avg: 1787 / Max: 1787Min: 1989 / Avg: 1990.33 / Max: 1993Min: 2426 / Avg: 2435.67 / Max: 2442Min: 2831 / Avg: 2840.33 / Max: 2852Min: 2861 / Avg: 2869 / Max: 2878

RealSR-NCNN

MinAvgMaxRTX 3070 Ti51.060.969.0RTX 3080 Ti56.059.063.0RTX 308054.058.563.0RTX 307050.057.964.0RTX 3060 Ti48.057.064.0RTX 309053.056.561.0RTX 306043.053.661.0OpenBenchmarking.orgCelsius, Fewer Is BetterRealSR-NCNN 20200818GPU Temperature Monitor20406080100

MinAvgMaxRTX 3070 Ti50.070.675.0RTX 307049.067.071.0RTX 308052.066.272.0RTX 3060 Ti47.065.870.0RTX 3080 Ti52.065.171.0RTX 306042.062.065.0RTX 309050.061.766.0OpenBenchmarking.orgCelsius, Fewer Is BetterRealSR-NCNN 20200818GPU Temperature Monitor20406080100

Waifu2x-NCNN Vulkan

MinAvgMaxRTX 3070 Ti53.059.365.0RTX 308055.058.263.0RTX 3080 Ti55.058.262.0RTX 307052.056.061.0RTX 3060 Ti51.055.761.0RTX 309050.053.957.0RTX 306043.051.159.0OpenBenchmarking.orgCelsius, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818GPU Temperature Monitor20406080100

VkResample

MinAvgMaxRTX 306043.048.259.0RTX 3070 Ti44.048.159.0RTX 3080 Ti44.047.854.0RTX 308044.046.654.0RTX 309043.045.853.0RTX 307040.043.051.0RTX 3060 Ti37.040.250.0OpenBenchmarking.orgCelsius, Fewer Is BetterVkResample 1.0GPU Temperature Monitor1632486480

Rodinia

MinAvgMaxRTX 3080 Ti53.057.360.0RTX 308053.056.559.0RTX 3070 Ti51.055.957.0RTX 307048.051.954.0RTX 3060 Ti47.051.154.0RTX 306043.050.756.0RTX 309043.047.649.0OpenBenchmarking.orgCelsius, Fewer Is BetterRodinia 3.1GPU Temperature Monitor1632486480

MinAvgMaxRTX 3070 Ti48.054.558.0RTX 3080 Ti50.053.956.0RTX 308050.053.756.0RTX 306043.051.159.0RTX 307045.049.152.0RTX 309044.048.251.0RTX 3060 Ti43.048.151.0OpenBenchmarking.orgCelsius, Fewer Is BetterRodinia 3.1GPU Temperature Monitor1632486480

FinanceBench

MinAvgMaxRTX 3080 Ti49.052.660.0RTX 308048.051.456.0RTX 309048.051.056.0RTX 3070 Ti46.050.359.0RTX 307044.047.755.0RTX 306043.046.052.0RTX 3060 Ti42.046.054.0OpenBenchmarking.orgCelsius, Fewer Is BetterFinanceBench 2016-07-25GPU Temperature Monitor1632486480

MinAvgMaxRTX 306047.048.450.0RTX 309043.045.148.0RTX 3080 Ti42.045.148.0RTX 308042.044.347.0RTX 3070 Ti41.043.847.0RTX 307039.040.843.0RTX 3060 Ti37.038.441.0OpenBenchmarking.orgCelsius, Fewer Is BetterFinanceBench 2016-07-25GPU Temperature Monitor1428425670

Darktable

MinAvgMaxRTX 309047.049.354.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktableGPU Temperature Monitor1530456075

MinAvgMaxRTX 309044.045.948.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktableGPU Temperature Monitor1428425670

MinAvgMaxRTX 309040.042.044.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktableGPU Temperature Monitor1224364860

MinAvgMaxRTX 309039.041.143.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktableGPU Temperature Monitor1224364860

clpeak

MinAvgMaxRTX 3080 Ti50.054.861.0RTX 308050.054.460.0RTX 309049.053.359.0RTX 3070 Ti47.051.957.0RTX 307045.048.755.0RTX 3060 Ti43.047.954.0RTX 306043.045.448.0OpenBenchmarking.orgCelsius, Fewer Is BetterclpeakGPU Temperature Monitor20406080100

Darktable

MinAvgMaxRTX 3070 Ti45.049.954.0RTX 3080 Ti46.049.552.0RTX 308045.048.051.0RTX 306043.045.548.0RTX 307041.044.347.0RTX 3060 Ti39.041.845.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktable 3.8.1GPU Temperature Monitor1530456075

MinAvgMaxRTX 306046.048.751.0RTX 3070 Ti41.044.747.0RTX 3080 Ti43.044.446.0RTX 308041.043.145.0RTX 307038.039.341.0RTX 3060 Ti35.036.938.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktable 3.8.1GPU Temperature Monitor1530456075

MinAvgMaxRTX 306049.050.352.0RTX 3080 Ti39.040.943.0RTX 3070 Ti38.040.543.0RTX 308038.039.541.0RTX 307035.036.338.0RTX 3060 Ti33.034.435.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktable 3.8.1GPU Temperature Monitor1530456075

MinAvgMaxRTX 306050.052.354.0RTX 3080 Ti38.039.841.0RTX 3070 Ti37.039.241.0RTX 308037.037.839.0RTX 3060 Ti33.035.237.0RTX 307034.035.136.0OpenBenchmarking.orgCelsius, Fewer Is BetterDarktable 3.8.1GPU Temperature Monitor1530456075

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: SingleRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090612182430SE +/- 0.006, N = 5SE +/- 0.006, N = 5SE +/- 0.007, N = 5SE +/- 0.002, N = 5SE +/- 0.003, N = 5SE +/- 0.003, N = 5SE +/- 0.009, N = 523.18417.68917.55813.53711.2729.5709.4221. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: SingleRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090510152025Min: 23.18 / Avg: 23.18 / Max: 23.2Min: 17.68 / Avg: 17.69 / Max: 17.71Min: 17.54 / Avg: 17.56 / Max: 17.59Min: 13.53 / Avg: 13.54 / Max: 13.54Min: 11.26 / Avg: 11.27 / Max: 11.28Min: 9.56 / Avg: 9.57 / Max: 9.57Min: 9.39 / Avg: 9.42 / Max: 9.441. (CXX) g++ options: -O3

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: Monte-Carlo OpenCLRTX 3060RTX 3070 TiRTX 3070RTX 3060 TiRTX 3080RTX 3080 TiRTX 3090130260390520650SE +/- 0.29, N = 7SE +/- 0.04, N = 7SE +/- 0.28, N = 7SE +/- 0.34, N = 7SE +/- 0.29, N = 7SE +/- 0.41, N = 7SE +/- 0.69, N = 7622.23474.09465.30465.28388.82326.09325.271. (CXX) g++ options: -O3 -march=native -fopenmp
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Monte-Carlo OpenCLRTX 3060RTX 3070 TiRTX 3070RTX 3060 TiRTX 3080RTX 3080 TiRTX 3090110220330440550Min: 621.02 / Avg: 622.23 / Max: 623.21Min: 473.95 / Avg: 474.09 / Max: 474.24Min: 464.44 / Avg: 465.3 / Max: 466.49Min: 463.91 / Avg: 465.28 / Max: 466.44Min: 388.02 / Avg: 388.82 / Max: 389.75Min: 324.71 / Avg: 326.09 / Max: 327.36Min: 322.78 / Avg: 325.27 / Max: 326.951. (CXX) g++ options: -O3 -march=native -fopenmp

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309048121620SE +/- 0.030644, N = 15SE +/- 0.000768, N = 15SE +/- 0.005828, N = 14SE +/- 0.004661, N = 14SE +/- 0.054610, N = 15SE +/- 0.074354, N = 15SE +/- 0.066100, N = 1516.24446612.85000010.3790009.7300007.3520006.6810006.3870001. (CXX) g++ options: -O3 -march=native -fopenmp
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Black-Scholes OpenCLRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 309048121620Min: 16.2 / Avg: 16.24 / Max: 16.65Min: 12.85 / Avg: 12.85 / Max: 12.86Min: 10.36 / Avg: 10.38 / Max: 10.44Min: 9.71 / Avg: 9.73 / Max: 9.78Min: 7.25 / Avg: 7.35 / Max: 7.89Min: 6.4 / Avg: 6.68 / Max: 7.11Min: 6.17 / Avg: 6.39 / Max: 6.91. (CXX) g++ options: -O3 -march=native -fopenmp

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: NoRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215SE +/- 0.004, N = 5SE +/- 0.008, N = 6SE +/- 0.006, N = 6SE +/- 0.008, N = 6SE +/- 0.012, N = 7SE +/- 0.012, N = 7SE +/- 0.014, N = 710.4518.3287.7927.0576.0175.6415.587
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30903691215Min: 10.44 / Avg: 10.45 / Max: 10.46Min: 8.31 / Avg: 8.33 / Max: 8.35Min: 7.77 / Avg: 7.79 / Max: 7.81Min: 7.04 / Avg: 7.06 / Max: 7.09Min: 5.99 / Avg: 6.02 / Max: 6.07Min: 5.61 / Avg: 5.64 / Max: 5.69Min: 5.54 / Avg: 5.59 / Max: 5.64

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30901530456075SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 368.6951.5647.3041.3332.8529.8229.19
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30901326395265Min: 68.54 / Avg: 68.69 / Max: 68.78Min: 51.43 / Avg: 51.56 / Max: 51.63Min: 47.17 / Avg: 47.3 / Max: 47.4Min: 41.22 / Avg: 41.33 / Max: 41.41Min: 32.72 / Avg: 32.85 / Max: 32.95Min: 29.71 / Avg: 29.82 / Max: 29.94Min: 29.1 / Avg: 29.19 / Max: 29.25

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.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 30901.10232.20463.30694.40925.5115SE +/- 0.010, N = 7SE +/- 0.003, N = 8SE +/- 0.004, N = 8SE +/- 0.005, N = 9SE +/- 0.007, N = 9SE +/- 0.005, N = 9SE +/- 0.004, N = 94.8994.1904.0263.7183.3373.2683.266
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3080 TiRTX 3090246810Min: 4.87 / Avg: 4.9 / Max: 4.95Min: 4.18 / Avg: 4.19 / Max: 4.2Min: 4.01 / Avg: 4.03 / Max: 4.05Min: 3.7 / Avg: 3.72 / Max: 3.75Min: 3.31 / Avg: 3.34 / Max: 3.37Min: 3.25 / Avg: 3.27 / Max: 3.29Min: 3.24 / Avg: 3.27 / Max: 3.29

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 MyocyteRTX 3090RTX 3060RTX 3080 TiRTX 3070 TiRTX 3070RTX 3080RTX 3060 Ti714212835SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 328.9428.5728.5628.4028.1427.9827.641. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL MyocyteRTX 3090RTX 3060RTX 3080 TiRTX 3070 TiRTX 3070RTX 3080RTX 3060 Ti612182430Min: 28.84 / Avg: 28.94 / Max: 29.13Min: 28.48 / Avg: 28.56 / Max: 28.68Min: 28.54 / Avg: 28.56 / Max: 28.57Min: 28.38 / Avg: 28.4 / Max: 28.43Min: 27.91 / Avg: 28.14 / Max: 28.31Min: 27.91 / Avg: 27.98 / Max: 28.04Min: 27.6 / Avg: 27.64 / Max: 27.71. (CXX) g++ options: -O2 -lOpenCL

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle FilterRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti3691215SE +/- 0.028, N = 5SE +/- 0.008, N = 6SE +/- 0.008, N = 7SE +/- 0.002, N = 7SE +/- 0.003, N = 8SE +/- 0.023, N = 8SE +/- 0.009, N = 89.5216.9275.9565.7714.1893.9573.7571. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle FilterRTX 3060RTX 3060 TiRTX 3070RTX 3070 TiRTX 3080RTX 3090RTX 3080 Ti3691215Min: 9.46 / Avg: 9.52 / Max: 9.61Min: 6.9 / Avg: 6.93 / Max: 6.95Min: 5.93 / Avg: 5.96 / Max: 5.99Min: 5.76 / Avg: 5.77 / Max: 5.78Min: 4.18 / Avg: 4.19 / Max: 4.2Min: 3.91 / Avg: 3.96 / Max: 4.11Min: 3.73 / Avg: 3.76 / Max: 3.811. (CXX) g++ options: -O2 -lOpenCL

Darktable

Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Boat - Acceleration: OpenCLRTX 3080 TiRTX 3070 TiRTX 3070RTX 3080RTX 3060RTX 3060 TiRTX 30900.58031.16061.74092.32122.9015SE +/- 0.009, N = 9SE +/- 0.013, N = 9SE +/- 0.016, N = 9SE +/- 0.011, N = 9SE +/- 0.008, N = 9SE +/- 0.010, N = 9SE +/- 0.006, N = 102.5792.5792.5782.5772.5732.5711.118
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Boat - Acceleration: OpenCLRTX 3080 TiRTX 3070 TiRTX 3070RTX 3080RTX 3060RTX 3060 TiRTX 3090246810Min: 2.54 / Avg: 2.58 / Max: 2.62Min: 2.54 / Avg: 2.58 / Max: 2.64Min: 2.51 / Avg: 2.58 / Max: 2.67Min: 2.53 / Avg: 2.58 / Max: 2.65Min: 2.55 / Avg: 2.57 / Max: 2.61Min: 2.53 / Avg: 2.57 / Max: 2.62Min: 1.1 / Avg: 1.12 / Max: 1.16

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Masskrug - Acceleration: OpenCLRTX 3060RTX 3080 TiRTX 3080RTX 3070RTX 3070 TiRTX 3060 TiRTX 30900.59091.18181.77272.36362.9545SE +/- 0.008, N = 9SE +/- 0.010, N = 9SE +/- 0.007, N = 9SE +/- 0.006, N = 9SE +/- 0.008, N = 9SE +/- 0.005, N = 9SE +/- 0.005, N = 102.6262.6182.6122.6062.6052.6041.921
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Masskrug - Acceleration: OpenCLRTX 3060RTX 3080 TiRTX 3080RTX 3070RTX 3070 TiRTX 3060 TiRTX 3090246810Min: 2.6 / Avg: 2.63 / Max: 2.67Min: 2.58 / Avg: 2.62 / Max: 2.67Min: 2.59 / Avg: 2.61 / Max: 2.65Min: 2.57 / Avg: 2.61 / Max: 2.63Min: 2.56 / Avg: 2.61 / Max: 2.64Min: 2.58 / Avg: 2.6 / Max: 2.63Min: 1.9 / Avg: 1.92 / Max: 1.95

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Server Rack - Acceleration: OpenCLRTX 3080 TiRTX 3080RTX 3070RTX 3060RTX 3090RTX 3070 TiRTX 3060 Ti0.02880.05760.08640.11520.144SE +/- 0.004, N = 15SE +/- 0.004, N = 15SE +/- 0.004, N = 15SE +/- 0.004, N = 15SE +/- 0.001, N = 15SE +/- 0.004, N = 15SE +/- 0.003, N = 150.1280.1260.1260.1260.1240.1240.122
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Server Rack - Acceleration: OpenCLRTX 3080 TiRTX 3080RTX 3070RTX 3060RTX 3090RTX 3070 TiRTX 3060 Ti12345Min: 0.12 / Avg: 0.13 / Max: 0.17Min: 0.12 / Avg: 0.13 / Max: 0.16Min: 0.12 / Avg: 0.13 / Max: 0.17Min: 0.12 / Avg: 0.13 / Max: 0.17Min: 0.12 / Avg: 0.12 / Max: 0.14Min: 0.12 / Avg: 0.12 / Max: 0.16Min: 0.12 / Avg: 0.12 / Max: 0.16

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Server Room - Acceleration: OpenCLRTX 3080 TiRTX 3060RTX 3060 TiRTX 3080RTX 3070RTX 3070 TiRTX 30900.42080.84161.26241.68322.104SE +/- 0.006, N = 10SE +/- 0.008, N = 10SE +/- 0.006, N = 10SE +/- 0.007, N = 10SE +/- 0.009, N = 10SE +/- 0.011, N = 10SE +/- 0.002, N = 121.8701.8651.8571.8491.8471.8440.707
OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.8.1Test: Server Room - Acceleration: OpenCLRTX 3080 TiRTX 3060RTX 3060 TiRTX 3080RTX 3070RTX 3070 TiRTX 3090246810Min: 1.85 / Avg: 1.87 / Max: 1.91Min: 1.83 / Avg: 1.87 / Max: 1.9Min: 1.83 / Avg: 1.86 / Max: 1.9Min: 1.8 / Avg: 1.85 / Max: 1.87Min: 1.81 / Avg: 1.85 / Max: 1.9Min: 1.81 / Avg: 1.84 / Max: 1.93Min: 0.7 / Avg: 0.71 / Max: 0.72

clpeak

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

OpenBenchmarking.orgus, Fewer Is BetterclpeakOpenCL Test: Kernel LatencyRTX 3090RTX 3080 TiRTX 3060 TiRTX 3080RTX 3060RTX 3070RTX 3070 Ti0.92031.84062.76093.68124.6015SE +/- 0.01, N = 15SE +/- 0.00, N = 15SE +/- 0.01, N = 15SE +/- 0.01, N = 15SE +/- 0.00, N = 15SE +/- 0.01, N = 15SE +/- 0.00, N = 154.093.863.813.783.773.753.721. (CXX) g++ options: -O3 -rdynamic -lOpenCL
OpenBenchmarking.orgus, Fewer Is BetterclpeakOpenCL Test: Kernel LatencyRTX 3090RTX 3080 TiRTX 3060 TiRTX 3080RTX 3060RTX 3070RTX 3070 Ti246810Min: 4.04 / Avg: 4.09 / Max: 4.14Min: 3.83 / Avg: 3.86 / Max: 3.91Min: 3.77 / Avg: 3.81 / Max: 3.84Min: 3.75 / Avg: 3.78 / Max: 3.83Min: 3.74 / Avg: 3.77 / Max: 3.79Min: 3.73 / Avg: 3.75 / Max: 3.8Min: 3.7 / Avg: 3.72 / Max: 3.751. (CXX) g++ options: -O3 -rdynamic -lOpenCL

123 Results Shown

GPU Temperature Monitor:
  Phoronix Test Suite System Monitoring:
    Celsius
    Watts
SHOC Scalable HeterOgeneous Computing
SHOC Scalable HeterOgeneous Computing:
  OpenCL - Triad
  OpenCL - Reduction
  OpenCL - Bus Speed Download
  OpenCL - Texture Read Bandwidth
  Copy
SHOC Scalable HeterOgeneous Computing:
  OpenCL - Triad
  OpenCL - Bus Speed Download
  OpenCL - Texture Read Bandwidth
cl-mem:
  Copy
  Write
ViennaCL
cl-mem
SHOC Scalable HeterOgeneous Computing
cl-mem
ViennaCL:
  OpenCL BLAS - sAXPY
  OpenCL BLAS - sDOT
  OpenCL BLAS - dCOPY
  OpenCL BLAS - dAXPY
  OpenCL BLAS - dDOT
  OpenCL BLAS - dGEMV-N
  OpenCL BLAS - dGEMV-T
clpeak:
  Global Memory Bandwidth
  Transfer Bandwidth enqueueReadBuffer
  Transfer Bandwidth enqueueWriteBuffer
clpeak:
  Single-Precision Float
  Double-Precision Double
  fp64-vec4
vkpeak
SHOC Scalable HeterOgeneous Computing
vkpeak:
  fp32-vec4
  fp16-scalar
  fp16-vec4
  fp64-scalar
  fp64-vec4
SHOC Scalable HeterOgeneous Computing:
  OpenCL - S3D
  OpenCL - FFT SP
  OpenCL - GEMM SGEMM_N
clpeak:
  Single-Precision Float
  Double-Precision Double
ViennaCL:
  OpenCL BLAS - dGEMM-TT
  OpenCL BLAS - dGEMM-NT
  OpenCL BLAS - dGEMM-TN
ViennaCL:
  OpenCL BLAS - dGEMM-NN
  OpenCL BLAS - dGEMM-NT
  OpenCL BLAS - dGEMM-TN
  OpenCL BLAS - dGEMM-TT
SHOC Scalable HeterOgeneous Computing
SHOC Scalable HeterOgeneous Computing
clpeak
vkpeak:
  int32-scalar
  int32-vec4
  int16-scalar
vkpeak
vkpeak
clpeak
Hashcat
Hashcat
Hashcat:
  SHA1
  7-Zip
  SHA-512
  TrueCrypt RIPEMD160 + XTS
IndigoBench:
  OpenCL GPU - Bedroom
  OpenCL GPU - Supercar
LuxCoreRender:
  DLSC - GPU
  Danish Mood - GPU
  Orange Juice - GPU
  LuxCore Benchmark - GPU
LuxCoreRender:
  DLSC - GPU
  Danish Mood - GPU
  Orange Juice - GPU
  LuxCore Benchmark - GPU
  Rainbow Colors and Prism - GPU
LeelaChessZero
FAHBench
FAHBench
Meta Performance Per Watts:
  Performance Per Watts
  GPU
JuliaGPU
SmallPT GPU
MandelGPU
SmallPT GPU
SmallPT GPU
SmallPT GPU
SmallPT GPU:
  GPU - Cornell
  GPU - Caustic3
OctaneBench
OctaneBench
Chaos Group V-RAY
Chaos Group V-RAY:
  NVIDIA CUDA GPU
  NVIDIA RTX GPU
RealSR-NCNN:
  GPU Temp Monitor:
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
    Celsius
VkResample
FinanceBench:
  Monte-Carlo OpenCL
  Black-Scholes OpenCL
RealSR-NCNN:
  4x - No
  4x - Yes
Waifu2x-NCNN Vulkan
Rodinia:
  OpenCL Myocyte
  OpenCL Particle Filter
Darktable:
  Boat - OpenCL
  Masskrug - OpenCL
  Server Rack - OpenCL
  Server Room - OpenCL
clpeak