NVIDIA GPU Compute OpenCL CUDA Benchmarks

Some benchmarks by Michael Larabel tests for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2009112-FI-ROCMGPUCO10
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
Show Result Confidence Charts

Limit displaying results to tests within:

BLAS (Basic Linear Algebra Sub-Routine) Tests 2 Tests
CPU Massive 3 Tests
HPC - High Performance Computing 4 Tests
Machine Learning 2 Tests
Multi-Core 3 Tests
NVIDIA GPU Compute 14 Tests
OpenCL 5 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
GTX 1060
September 08 2020
  1 Hour, 10 Minutes
GTX 1070
September 07 2020
  1 Hour, 6 Minutes
GTX 1070 Ti
September 06 2020
  1 Hour, 11 Minutes
GTX 1080
September 07 2020
  1 Hour, 4 Minutes
GTX 1650
September 09 2020
  1 Hour, 10 Minutes
GTX 1650 SUPER
September 08 2020
  1 Hour, 8 Minutes
GTX 1660
September 09 2020
  1 Hour, 5 Minutes
GTX 1660 SUPER
September 07 2020
  1 Hour, 5 Minutes
GTX 1660 Ti
September 09 2020
  1 Hour, 3 Minutes
RTX 2060
September 08 2020
  1 Hour, 3 Minutes
RTX 2060 SUPER
September 09 2020
  1 Hour, 1 Minute
RTX 2070
September 09 2020
  1 Hour, 1 Minute
RTX 2070 SUPER
September 08 2020
  59 Minutes
RTX 2080
September 08 2020
  1 Hour
RTX 2080 SUPER
September 08 2020
  59 Minutes
RTX 2080 Ti
September 07 2020
  1 Hour
TITAN RTX
September 06 2020
  1 Hour, 1 Minute
Invert Hiding All Results Option
  1 Hour, 4 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):


NVIDIA GPU Compute OpenCL CUDA BenchmarksOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (1302 BIOS)AMD Starship/Matisse16GB2000GB Corsair Force MP600NVIDIA GeForce GTX 1060 6GB (1506/4006MHz)NVIDIA GeForce GTX 1070 8GB (1506/4006MHz)Zotac NVIDIA GeForce GTX 1070 Ti 8GB (1607/4006MHz)NVIDIA GeForce GTX 1080 8GB (1607/5005MHz)ASUS NVIDIA GeForce GTX 1650 4GB (1485/4001MHz)ASUS NVIDIA GeForce GTX 1650 SUPER 4GB (615/405MHz)ASUS NVIDIA GeForce GTX 1660 6GB (1530/4001MHz)eVGA NVIDIA GeForce GTX 1660 SUPER 6GB (1530/7000MHz)eVGA NVIDIA GeForce GTX 1660 Ti 6GB (1500/6000MHz)NVIDIA GeForce RTX 2060 6GB (1365/7000MHz)NVIDIA GeForce RTX 2060 SUPER 8GB (1470/7000MHz)ASUS NVIDIA GeForce RTX 2070 8GB (1410/7000MHz)NVIDIA GeForce RTX 2070 SUPER 8GB (450/405MHz)Zotac NVIDIA GeForce RTX 2080 8GB (1515/7000MHz)NVIDIA GeForce RTX 2080 SUPER 8GB (1650/7750MHz)NVIDIA GeForce RTX 2080 Ti 11GB (1350/7000MHz)NVIDIA TITAN RTX 24GB (420/405MHz)DELL P2415QRealtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.045.4.0-45-generic (x86_64)GNOME Shell 3.36.4X Server 1.20.8NVIDIA 450.664.6.0OpenCL 1.2 CUDA 11.0.2281.2.133GCC 9.3.0 + CUDA 11.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionNVIDIA GPU Compute OpenCL CUDA Benchmarks PerformanceSystem Logs- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x8701013- GTX 1060: GPU Compute Cores: 1280- GTX 1070: GPU Compute Cores: 1920- GTX 1070 Ti: GPU Compute Cores: 2432- GTX 1080: GPU Compute Cores: 2560- GTX 1650: GPU Compute Cores: 896- GTX 1650 SUPER: GPU Compute Cores: 1280- GTX 1660: GPU Compute Cores: 1408- GTX 1660 SUPER: GPU Compute Cores: 1408- GTX 1660 Ti: GPU Compute Cores: 1536- RTX 2060: GPU Compute Cores: 1920- RTX 2060 SUPER: GPU Compute Cores: 2176- RTX 2070: GPU Compute Cores: 2304- RTX 2070 SUPER: GPU Compute Cores: 2560- RTX 2080: GPU Compute Cores: 2944- RTX 2080 SUPER: GPU Compute Cores: 3072- RTX 2080 Ti: GPU Compute Cores: 4352- TITAN RTX: GPU Compute Cores: 4608- Python 3.8.2- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

GTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTXLogarithmic Result OverviewPhoronix Test SuiteMixbenchLeelaChessZeroclpeakcl-memFinanceBenchRodiniaOctaneBenchArrayFireFAHBenchPlaidMLMandelGPULuxCoreRender OpenCLGROMACSNAMD CUDAViennaCL

NVIDIA GPU Compute OpenCL CUDA Benchmarkslczero: OpenCLclpeak: Single-Precision Floatcl-mem: Readclpeak: Global Memory Bandwidthclpeak: Double-Precision Doublemixbench: NVIDIA CUDA - Half Precisionmixbench: OpenCL - Double Precisioncl-mem: Writemixbench: OpenCL - Single Precisionluxcorerender-cl: Rainbow Colors and Prismmixbench: NVIDIA CUDA - Double Precisionplaidml: No - Inference - Mobilenet - OpenCLmixbench: NVIDIA CUDA - Single Precisionfinancebench: Black-Scholes OpenCLrodinia: OpenCL Particle Filteroctanebench: Total Scorearrayfire: Conjugate Gradient OpenCLplaidml: No - Inference - IMDB LSTM - OpenCLfahbench: plaidml: No - Inference - DenseNet 201 - OpenCLneatbench: GPUluxcorerender-cl: DLSCmandelgpu: GPUluxcorerender-cl: LuxCore Benchmarkplaidml: Yes - Inference - Mobilenet - OpenCLluxcorerender-cl: Foodcl-mem: Copyplaidml: No - Training - Mobilenet - OpenCLgromacs-gpu: Water Benchmarkmixbench: OpenCL - Integermixbench: NVIDIA CUDA - Integerclpeak: Integer Compute INTnamd-cuda: ATPase Simulation - 327,506 Atomsviennacl: OpenCL LU FactorizationGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX59334110.71154.4147.66150.6875.77150.48148.14355.285.41151.63775.734167.4113.34012.00291.7329065.000227.23102.268379.131.78130290433.81.53987.070.65138.583.523.1811354.141187.721275.120.3095761.653097506093.02206.2197.20223.60113.32224.82199.56476.926.62225.051052.646174.808.9688.205132.8393544.077314.83139.4974108.4621.52.46184077179.72.121392.460.92183.3103.794.0522019.081764.631695.790.2402266.206899476788.40206.2198.18242.70121.42241.99198.27018.905.38240.921070.906647.947.2987.791141.6676553.814331.84137.1393108.232.55194875210.82.241485.780.98184.199.644.2592157.311887.612076.450.2230069.1838126628218.62229.6223.14298.46149.35296.66225.68520.974.23296.091154.138182.726.9816.457148.4720093.387398.82151.5679112.2824.62.45231826397.82.131665.720.99207.499.824.6342655.582326.092430.550.2242670.688031012673.64108.7105.31109.206948.3391.77104.43441.363.9392.37573.673474.1724.14528517.82377.4102186.808201.2775.813559.3712.81.73132266243.91.51814.500.6799.265.593.2533466.723108.842641.380.3450362.960751514123.66167.0156.86157.9810057.46132.75145.84943.062.73133.88740.174993.9216.57778512.48792.3678394.445269.1782.052173.171.61186025762.61.361080.430.81143.877.104.1224975.184492.324128.390.2621167.933058954543.64163.9158.15168.5610568.46141.25157.35259.295.58141.36858.905313.3315.25911.893119.1052654.541289.42125.615887.2518.62.56193795979.52.201199.660.99145.997.944.4125303.514753.114338.670.2429067.820073664440.23296.6276.35172.4910862.90144.30240.75462.595.61144.141080.115440.7415.27111.482135.6571342.642298.77135.6073100.512.63200846066.42.261380.681.09240.2111.394.6295432.404866.614280.550.2317768.153277334778.23251.2234.91184.3411787.37155.78223.95821.195.01157.141038.865888.7814.38510.847133.3453.068313.08138.557898.992.51210891296.22.111427.431.07211.1108.064.7895803.635271.344734.360.2280070.0287102905417.55297.4276.94232.1714646.46194.80258.77222.287.97196.301285.587339.3912.5558.836166.0985712.659390.07178.4712125.2328.33.34252924048.72.791719.711.17239.4125.335.4167275.336539.815356.840.2063573.3091123357034.77396.8368.94261.8916507.06222.15341.38159.1511.32221.801585.768264.6610.4987.914206.0789432.083438.26200.7664146.904.20281931310.03.542012.371.33288.0154.815.7898218.027348.916930.210.2005772.5010123117265.91396.9369.30267.7916838.72226.65328.58357.9610.98226.211575.658435.2310.2547.767207.5382862.097444.84199.8719147.8735.84.17284631717.43.491996.601.30284.4153.035.8658362.867540.287133.770.2036272.5699147308540.73396.8369.62308.8619501.67259.11322.99630.8310.21260.971712.759754.998.2636.837221.1937482.071500.26221.7284149.694.44323061985.43.692287.031.90292.1154.746.2499696.388722.288484.410.1962574.4473154429161.18396.5369.23345.1121764.40290.13338.110812.529.24291.561726.6010845.367.6576.196224.3872112.075544.54235.5341151.9938.34.32345359936.43.552322.391.87290.0154.356.53610930.569732.489141.230.1926574.72981750110264.38437.6405.92376.3223628.78315.03352.911758.659.14315.611815.9211685.956.8115.718234.0664461.900588.04248.0118159.794.38374422420.63.622432.731.93301.7158.826.57211752.1110562.9910266.720.1932976.10162183913536.25545.7506.90522.2231492.24436.81449.516061.0912.33409.462406.1614734.036.0204.414310.8143991.677753.75290.7249212.115.67448915216.64.752751.852.23324.0187.737.13616323.9414103.7913629.370.1908275.78712242414178.85567.7530.79548.2632084.79454.37491.715032.4712.73413.662544.6715103.615.7014.250324.5467691.666775.69292.1593227.3048.15.98480798376.64.992856.832.23322.7198.377.11716979.9914426.1113783.930.1919576.6410OpenBenchmarking.org

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 0.26Backend: OpenCLGTX 1650GTX 1650 SUPERGTX 1660GTX 1060GTX 1660 SUPERGTX 1660 TiGTX 1070GTX 1070 TiRTX 2060RTX 2070RTX 2060 SUPERGTX 1080RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX5K10K15K20K25KSE +/- 6.17, N = 3SE +/- 19.16, N = 3SE +/- 3.38, N = 3SE +/- 28.32, N = 3SE +/- 54.30, N = 3SE +/- 56.77, N = 3SE +/- 85.75, N = 3SE +/- 64.75, N = 3SE +/- 70.73, N = 3SE +/- 61.94, N = 3SE +/- 57.21, N = 3SE +/- 28.58, N = 3SE +/- 77.45, N = 3SE +/- 79.08, N = 3SE +/- 55.16, N = 3SE +/- 20.50, N = 3310151515895593373667733975099471029012311123351266214730154421750121839224241. (CXX) g++ options: -flto -pthread

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 FloatGTX 1650GTX 1060GTX 1650 SUPERGTX 1660 SUPERGTX 1660GTX 1660 TiRTX 2060GTX 1070GTX 1070 TiRTX 2060 SUPERRTX 2070GTX 1080RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX3K6K9K12K15KSE +/- 21.41, N = 13SE +/- 76.28, N = 15SE +/- 49.35, N = 15SE +/- 43.47, N = 15SE +/- 27.06, N = 12SE +/- 56.79, N = 15SE +/- 67.09, N = 15SE +/- 112.12, N = 15SE +/- 3.15, N = 13SE +/- 85.70, N = 15SE +/- 92.93, N = 15SE +/- 104.59, N = 15SE +/- 77.74, N = 15SE +/- 73.37, N = 15SE +/- 38.81, N = 13SE +/- 177.79, N = 15SE +/- 206.22, N = 152673.644110.714123.664440.234543.644778.235417.556093.026788.407034.777265.918218.628540.739161.1810264.3813536.2514178.851. (CXX) g++ options: -O3 -rdynamic -lOpenCL

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: ReadGTX 1650GTX 1060GTX 1660GTX 1650 SUPERGTX 1070GTX 1070 TiGTX 1080GTX 1660 TiGTX 1660 SUPERRTX 2060RTX 2080RTX 2060 SUPERRTX 2070 SUPERRTX 2070RTX 2080 SUPERRTX 2080 TiTITAN RTX120240360480600SE +/- 0.04, N = 4SE +/- 0.02, N = 5SE +/- 0.02, N = 5SE +/- 0.04, N = 5SE +/- 0.03, N = 6SE +/- 0.03, N = 6SE +/- 0.08, N = 6SE +/- 0.07, N = 6SE +/- 0.07, N = 7SE +/- 0.08, N = 7SE +/- 0.42, N = 8SE +/- 0.13, N = 8SE +/- 0.09, N = 8SE +/- 0.10, N = 8SE +/- 0.11, N = 8SE +/- 0.13, N = 8SE +/- 0.13, N = 9108.7154.4163.9167.0206.2206.2229.6251.2296.6297.4396.5396.8396.8396.9437.6545.7567.71. (CC) gcc options: -O2 -flto -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 BandwidthGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1070GTX 1070 TiGTX 1080GTX 1660 TiGTX 1660 SUPERRTX 2060RTX 2060 SUPERRTX 2080RTX 2070RTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiTITAN RTX110220330440550SE +/- 0.06, N = 12SE +/- 0.04, N = 10SE +/- 0.02, N = 13SE +/- 0.06, N = 11SE +/- 0.04, N = 11SE +/- 0.01, N = 11SE +/- 0.11, N = 15SE +/- 0.12, N = 12SE +/- 0.16, N = 12SE +/- 0.15, N = 12SE +/- 0.06, N = 13SE +/- 0.07, N = 13SE +/- 0.05, N = 13SE +/- 0.10, N = 13SE +/- 0.14, N = 13SE +/- 0.46, N = 12SE +/- 0.05, N = 13105.31147.66156.86158.15197.20198.18223.14234.91276.35276.94368.94369.23369.30369.62405.92506.90530.791. (CXX) g++ options: -O3 -rdynamic -lOpenCL

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiGTX 1070RTX 2060GTX 1070 TiRTX 2060 SUPERRTX 2070GTX 1080RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX120240360480600SE +/- 0.27, N = 3SE +/- 0.23, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.46, N = 3SE +/- 0.38, N = 4SE +/- 0.58, N = 3SE +/- 0.06, N = 3SE +/- 0.69, N = 3SE +/- 0.71, N = 3SE +/- 0.60, N = 3SE +/- 0.00, N = 3SE +/- 0.91, N = 3SE +/- 0.03, N = 3SE +/- 1.30, N = 3SE +/- 1.32, N = 3109.20150.68157.98168.56172.49184.34223.60232.17242.70261.89267.79298.46308.86345.11376.32522.22548.261. (CXX) g++ options: -O3 -rdynamic -lOpenCL

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Half PrecisionGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX7K14K21K28K35KSE +/- 0.02, N = 9SE +/- 0.02, N = 11SE +/- 0.07, N = 11SE +/- 0.10, N = 11SE +/- 0.56, N = 12SE +/- 0.94, N = 13SE +/- 53.59, N = 15SE +/- 15.42, N = 13SE +/- 0.43, N = 13SE +/- 7.32, N = 13SE +/- 1.95, N = 14SE +/- 1.36, N = 14SE +/- 1.10, N = 14SE +/- 10.25, N = 14SE +/- 15.07, N = 15SE +/- 601.15, N = 15SE +/- 939.79, N = 1575.77113.32121.42149.356948.3310057.4610568.4610862.9011787.3714646.4616507.0616838.7219501.6721764.4023628.7831492.2432084.791. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Double PrecisionGTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1060GTX 1660 TiRTX 2060RTX 2060 SUPERGTX 1070RTX 2070GTX 1070 TiRTX 2070 SUPERRTX 2080GTX 1080RTX 2080 SUPERRTX 2080 TiTITAN RTX100200300400500SE +/- 0.01, N = 12SE +/- 0.15, N = 13SE +/- 0.14, N = 13SE +/- 0.22, N = 13SE +/- 0.23, N = 13SE +/- 0.12, N = 13SE +/- 0.00, N = 13SE +/- 0.01, N = 13SE +/- 0.06, N = 13SE +/- 0.11, N = 15SE +/- 0.08, N = 14SE +/- 0.00, N = 14SE +/- 0.56, N = 14SE +/- 0.15, N = 14SE +/- 0.08, N = 15SE +/- 0.56, N = 15SE +/- 1.13, N = 1491.77132.75141.25144.30150.48155.78194.80222.15224.82226.65241.99259.11290.13296.66315.03436.81454.371. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

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: WriteGTX 1650GTX 1650 SUPERGTX 1060GTX 1660GTX 1070 TiGTX 1070GTX 1660 TiGTX 1080GTX 1660 SUPERRTX 2060RTX 2070 SUPERRTX 2070RTX 2080RTX 2060 SUPERRTX 2080 SUPERRTX 2080 TiTITAN RTX110220330440550SE +/- 0.02, N = 4SE +/- 0.04, N = 5SE +/- 0.02, N = 5SE +/- 0.04, N = 5SE +/- 0.03, N = 6SE +/- 0.03, N = 6SE +/- 0.14, N = 6SE +/- 0.09, N = 6SE +/- 0.36, N = 7SE +/- 0.46, N = 7SE +/- 0.29, N = 8SE +/- 0.29, N = 8SE +/- 0.51, N = 8SE +/- 0.29, N = 8SE +/- 0.27, N = 8SE +/- 0.21, N = 8SE +/- 0.97, N = 9104.4145.8148.1157.3198.2199.5223.9225.6240.7258.7322.9328.5338.1341.3352.9449.5491.71. (CC) gcc options: -O2 -flto -lOpenCL

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: Single PrecisionGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiGTX 1070GTX 1070 TiRTX 2060RTX 2060 SUPERRTX 2070GTX 1080RTX 2070 SUPERRTX 2080RTX 2080 SUPERTITAN RTXRTX 2080 Ti3K6K9K12K15KSE +/- 0.54, N = 12SE +/- 2.22, N = 13SE +/- 1.39, N = 13SE +/- 5.87, N = 13SE +/- 8.38, N = 13SE +/- 6.62, N = 13SE +/- 12.17, N = 15SE +/- 2.95, N = 12SE +/- 6.45, N = 13SE +/- 2.35, N = 13SE +/- 10.00, N = 13SE +/- 11.17, N = 14SE +/- 4.39, N = 14SE +/- 14.62, N = 14SE +/- 8.02, N = 14SE +/- 619.26, N = 15SE +/- 66.33, N = 153441.364355.284943.065259.295462.595821.196476.927018.907222.288159.158357.968520.979630.8310812.5211758.6515032.4716061.091. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

LuxCoreRender OpenCL

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on OpenCL accelerators/GPUs. The alternative luxcorerender test profile is for CPU execution due to a difference in tests, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: Rainbow Colors and PrismGTX 1650 SUPERGTX 1650GTX 1080GTX 1660 TiGTX 1070 TiGTX 1060GTX 1660GTX 1660 SUPERGTX 1070RTX 2060RTX 2080 SUPERRTX 2080RTX 2070 SUPERRTX 2070RTX 2060 SUPERRTX 2080 TiTITAN RTX3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 12SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 4SE +/- 0.02, N = 4SE +/- 0.09, N = 4SE +/- 0.02, N = 4SE +/- 0.18, N = 42.733.934.235.015.385.415.585.616.627.979.149.2410.2110.9811.3212.3312.73MIN: 2.15 / MAX: 2.83MIN: 3.46 / MAX: 4.06MIN: 3.57 / MAX: 4.38MIN: 3.24 / MAX: 5.25MIN: 1.63 / MAX: 5.69MIN: 4.66 / MAX: 5.64MIN: 4.65 / MAX: 5.8MIN: 5.07 / MAX: 5.86MIN: 5.17 / MAX: 6.91MIN: 6.98 / MAX: 8.23MIN: 7.66 / MAX: 9.61MIN: 8.34 / MAX: 9.72MIN: 9.44 / MAX: 10.58MIN: 9.7 / MAX: 11.42MIN: 10.46 / MAX: 11.82MIN: 11.33 / MAX: 12.81MIN: 11.95 / MAX: 14

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Double PrecisionGTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1060GTX 1660 TiRTX 2060RTX 2060 SUPERGTX 1070RTX 2070GTX 1070 TiRTX 2070 SUPERRTX 2080GTX 1080RTX 2080 SUPERRTX 2080 TiTITAN RTX90180270360450SE +/- 0.06, N = 12SE +/- 0.13, N = 13SE +/- 0.10, N = 13SE +/- 0.66, N = 15SE +/- 0.00, N = 9SE +/- 0.08, N = 13SE +/- 0.01, N = 13SE +/- 0.23, N = 14SE +/- 0.10, N = 11SE +/- 0.27, N = 14SE +/- 0.35, N = 11SE +/- 0.15, N = 15SE +/- 0.22, N = 15SE +/- 0.03, N = 11SE +/- 0.96, N = 15SE +/- 10.76, N = 15SE +/- 15.67, N = 1592.37133.88141.36144.14151.63157.14196.30221.80225.05226.21240.92260.97291.56296.09315.61409.46413.661. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCLGTX 1650GTX 1650 SUPERGTX 1060GTX 1660GTX 1660 TiGTX 1070GTX 1070 TiGTX 1660 SUPERGTX 1080RTX 2060RTX 2070RTX 2060 SUPERRTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX5001000150020002500SE +/- 0.31, N = 7SE +/- 0.61, N = 7SE +/- 0.68, N = 7SE +/- 0.61, N = 7SE +/- 0.87, N = 8SE +/- 1.14, N = 8SE +/- 0.27, N = 8SE +/- 1.11, N = 8SE +/- 0.60, N = 8SE +/- 2.01, N = 8SE +/- 1.79, N = 8SE +/- 1.35, N = 8SE +/- 1.66, N = 8SE +/- 1.77, N = 8SE +/- 1.80, N = 9SE +/- 1.58, N = 8SE +/- 2.36, N = 8573.67740.17775.73858.901038.861052.641070.901080.111154.131285.581575.651585.761712.751726.601815.922406.162544.67

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: Single PrecisionGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiGTX 1070GTX 1070 TiRTX 2060GTX 1080RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX3K6K9K12K15KSE +/- 0.12, N = 12SE +/- 1.19, N = 9SE +/- 0.29, N = 13SE +/- 0.26, N = 13SE +/- 0.25, N = 13SE +/- 2.28, N = 13SE +/- 30.05, N = 15SE +/- 2.31, N = 11SE +/- 3.59, N = 13SE +/- 3.04, N = 11SE +/- 0.94, N = 14SE +/- 0.92, N = 14SE +/- 5.98, N = 14SE +/- 56.09, N = 15SE +/- 134.69, N = 15SE +/- 556.93, N = 15SE +/- 595.40, N = 153474.174167.414993.925313.335440.745888.786174.806647.947339.398182.728264.668435.239754.9910845.3611685.9514734.0315103.611. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-06-06Benchmark: Black-Scholes OpenCLGTX 1650GTX 1650 SUPERGTX 1660 SUPERGTX 1660GTX 1660 TiGTX 1060RTX 2060RTX 2060 SUPERRTX 2070GTX 1070RTX 2070 SUPERRTX 2080GTX 1070 TiGTX 1080RTX 2080 SUPERRTX 2080 TiTITAN RTX612182430SE +/- 0.002491, N = 14SE +/- 0.001228, N = 14SE +/- 0.001295, N = 14SE +/- 0.001244, N = 14SE +/- 0.001932, N = 14SE +/- 0.317433, N = 15SE +/- 0.000959, N = 14SE +/- 0.015324, N = 15SE +/- 0.001589, N = 14SE +/- 0.209293, N = 15SE +/- 0.014259, N = 15SE +/- 0.001547, N = 14SE +/- 0.003359, N = 15SE +/- 0.003290, N = 15SE +/- 0.003249, N = 15SE +/- 0.001717, N = 14SE +/- 0.001983, N = 1524.14528516.57778515.27100015.25900014.38500013.34000012.55500010.49800010.2540008.9680008.2630007.6570007.2980006.9810006.8110006.0200005.7010001. (CXX) g++ options: -O3 -lOpenCL

Rodinia

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

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenCL Particle FilterGTX 1650GTX 1650 SUPERGTX 1060GTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060GTX 1070RTX 2060 SUPERGTX 1070 TiRTX 2070RTX 2070 SUPERGTX 1080RTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX48121620SE +/- 0.004, N = 3SE +/- 0.021, N = 4SE +/- 0.012, N = 4SE +/- 0.015, N = 4SE +/- 0.009, N = 5SE +/- 0.016, N = 5SE +/- 0.004, N = 5SE +/- 0.019, N = 6SE +/- 0.008, N = 6SE +/- 0.006, N = 6SE +/- 0.005, N = 6SE +/- 0.011, N = 6SE +/- 0.018, N = 6SE +/- 0.008, N = 7SE +/- 0.006, N = 7SE +/- 0.009, N = 8SE +/- 0.008, N = 817.82312.48712.00211.89311.48210.8478.8368.2057.9147.7917.7676.8376.4576.1965.7184.4144.2501. (CXX) g++ options: -m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl

OctaneBench

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

OpenBenchmarking.orgScore, More Is BetterOctaneBench 4.00cTotal ScoreGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1070GTX 1660 TiGTX 1660 SUPERGTX 1070 TiGTX 1080RTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX7014021028035077.4191.7392.37119.11132.84133.35135.66141.67148.47166.10206.08207.54221.19224.39234.07310.81324.55

ArrayFire

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

OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient OpenCLGTX 1650GTX 1060GTX 1660GTX 1650 SUPERGTX 1070GTX 1070 TiGTX 1080GTX 1660 TiRTX 2060GTX 1660 SUPERRTX 2070RTX 2060 SUPERRTX 2080RTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiTITAN RTX246810SE +/- 0.003, N = 10SE +/- 0.003, N = 10SE +/- 0.002, N = 10SE +/- 0.002, N = 10SE +/- 0.004, N = 10SE +/- 0.001, N = 10SE +/- 0.002, N = 10SE +/- 0.001, N = 10SE +/- 0.002, N = 10SE +/- 0.001, N = 10SE +/- 0.002, N = 10SE +/- 0.001, N = 10SE +/- 0.001, N = 10SE +/- 0.001, N = 10SE +/- 0.001, N = 10SE +/- 0.005, N = 10SE +/- 0.001, N = 106.8085.0004.5414.4454.0773.8143.3873.0682.6592.6422.0972.0832.0752.0711.9001.6771.6661. (CXX) g++ options: -rdynamic

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCLGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiGTX 1070GTX 1070 TiRTX 2060GTX 1080RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX2004006008001000SE +/- 0.34, N = 4SE +/- 0.09, N = 4SE +/- 0.08, N = 4SE +/- 0.07, N = 4SE +/- 0.17, N = 4SE +/- 0.06, N = 4SE +/- 0.44, N = 4SE +/- 0.07, N = 4SE +/- 0.19, N = 4SE +/- 0.35, N = 4SE +/- 0.28, N = 5SE +/- 0.29, N = 5SE +/- 0.28, N = 5SE +/- 0.58, N = 5SE +/- 0.50, N = 5SE +/- 1.41, N = 5SE +/- 0.80, N = 5201.27227.23269.17289.42298.77313.08314.83331.84390.07398.82438.26444.84500.26544.54588.04753.75775.69

FAHBench

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

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2GTX 1650GTX 1650 SUPERGTX 1060GTX 1660GTX 1660 SUPERGTX 1070 TiGTX 1660 TiGTX 1070GTX 1080RTX 2060RTX 2070RTX 2060 SUPERRTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX60120180240300SE +/- 0.05, N = 3SE +/- 0.21, N = 3SE +/- 0.16, N = 3SE +/- 0.10, N = 3SE +/- 0.17, N = 3SE +/- 0.06, N = 3SE +/- 0.42, N = 3SE +/- 0.46, N = 3SE +/- 0.41, N = 3SE +/- 0.54, N = 3SE +/- 0.34, N = 3SE +/- 0.32, N = 3SE +/- 1.06, N = 3SE +/- 0.92, N = 3SE +/- 0.82, N = 3SE +/- 1.17, N = 3SE +/- 0.85, N = 375.8182.05102.27125.62135.61137.14138.56139.50151.57178.47199.87200.77221.73235.53248.01290.72292.16

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCLGTX 1650GTX 1650 SUPERGTX 1060GTX 1660GTX 1660 TiGTX 1660 SUPERGTX 1070 TiGTX 1070GTX 1080RTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX50100150200250SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.25, N = 359.3773.1779.1387.2598.99100.51108.23108.46112.28125.23146.90147.87149.69151.99159.79212.11227.30

NeatBench

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

OpenBenchmarking.orgFPS, More Is BetterNeatBench 5Acceleration: GPUGTX 1650GTX 1660GTX 1070GTX 1080RTX 2060RTX 2070RTX 2080TITAN RTX1122334455SE +/- 0.05, N = 13SE +/- 0.06, N = 13SE +/- 0.13, N = 13SE +/- 0.04, N = 13SE +/- 0.23, N = 14SE +/- 0.26, N = 13SE +/- 0.11, N = 15SE +/- 0.02, N = 1312.818.621.524.628.335.838.348.1

LuxCoreRender OpenCL

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on OpenCL accelerators/GPUs. The alternative luxcorerender test profile is for CPU execution due to a difference in tests, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: DLSCGTX 1650 SUPERGTX 1650GTX 1060GTX 1080GTX 1070GTX 1660 TiGTX 1070 TiGTX 1660GTX 1660 SUPERRTX 2060RTX 2070RTX 2060 SUPERRTX 2080RTX 2080 SUPERRTX 2070 SUPERRTX 2080 TiTITAN RTX1.34552.6914.03655.3826.7275SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 31.611.731.782.452.462.512.552.562.633.344.174.204.324.384.445.675.98MIN: 1.43 / MAX: 1.65MIN: 1.64 / MAX: 1.78MIN: 1.64 / MAX: 1.82MIN: 2.29 / MAX: 2.5MIN: 2.29 / MAX: 2.51MIN: 2.26 / MAX: 2.58MIN: 2.31 / MAX: 2.62MIN: 2.31 / MAX: 2.64MIN: 2.55 / MAX: 2.68MIN: 3.13 / MAX: 3.41MIN: 4.09 / MAX: 4.28MIN: 4.1 / MAX: 4.33MIN: 4.11 / MAX: 4.45MIN: 3.78 / MAX: 4.5MIN: 4.11 / MAX: 4.55MIN: 5.22 / MAX: 5.74MIN: 4.3 / MAX: 6.07

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: GPUGTX 1060GTX 1650GTX 1070GTX 1650 SUPERGTX 1660GTX 1070 TiGTX 1660 SUPERGTX 1660 TiGTX 1080RTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX100M200M300M400M500MSE +/- 339240.54, N = 6SE +/- 92918.17, N = 6SE +/- 327094.24, N = 7SE +/- 421595.14, N = 7SE +/- 481764.73, N = 7SE +/- 459096.61, N = 7SE +/- 173522.04, N = 7SE +/- 340263.98, N = 7SE +/- 528438.95, N = 8SE +/- 1119740.36, N = 8SE +/- 1074135.38, N = 8SE +/- 2064512.42, N = 8SE +/- 1663935.40, N = 9SE +/- 1245853.73, N = 9SE +/- 2703218.20, N = 9SE +/- 2765042.11, N = 10SE +/- 2829613.77, N = 10130290433.8132266243.9184077179.7186025762.6193795979.5194875210.8200846066.4210891296.2231826397.8252924048.7281931310.0284631717.4323061985.4345359936.4374422420.6448915216.6480798376.61. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

LuxCoreRender OpenCL

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on OpenCL accelerators/GPUs. The alternative luxcorerender test profile is for CPU execution due to a difference in tests, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: LuxCore BenchmarkGTX 1650 SUPERGTX 1650GTX 1060GTX 1660 TiGTX 1070GTX 1080GTX 1660GTX 1070 TiGTX 1660 SUPERRTX 2060RTX 2070RTX 2060 SUPERRTX 2080RTX 2080 SUPERRTX 2070 SUPERRTX 2080 TiTITAN RTX1.12282.24563.36844.49125.614SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 31.361.511.532.112.122.132.202.242.262.793.493.543.553.623.694.754.99MIN: 0.23 / MAX: 1.51MIN: 0.27 / MAX: 1.68MIN: 0.23 / MAX: 1.72MIN: 0.23 / MAX: 2.4MIN: 0.27 / MAX: 2.4MIN: 0.27 / MAX: 2.39MIN: 0.32 / MAX: 2.47MIN: 0.27 / MAX: 2.52MIN: 0.27 / MAX: 2.54MIN: 0.27 / MAX: 3.18MIN: 0.27 / MAX: 3.96MIN: 0.27 / MAX: 4.03MIN: 0.27 / MAX: 4.09MIN: 0.27 / MAX: 4.17MIN: 0.27 / MAX: 4.23MIN: 0.32 / MAX: 5.39MIN: 0.32 / MAX: 5.69

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCLGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1070GTX 1660 TiGTX 1070 TiGTX 1080RTX 2060RTX 2070RTX 2060 SUPERRTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX6001200180024003000SE +/- 0.63, N = 7SE +/- 2.10, N = 6SE +/- 0.88, N = 6SE +/- 1.27, N = 7SE +/- 0.83, N = 7SE +/- 3.20, N = 7SE +/- 1.12, N = 7SE +/- 0.31, N = 7SE +/- 1.79, N = 7SE +/- 0.92, N = 7SE +/- 3.47, N = 8SE +/- 3.83, N = 8SE +/- 2.51, N = 8SE +/- 2.80, N = 7SE +/- 6.26, N = 8SE +/- 1.82, N = 7SE +/- 1.27, N = 8814.50987.071080.431199.661380.681392.461427.431485.781665.721719.711996.602012.372287.032322.392432.732751.852856.83

LuxCoreRender OpenCL

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on OpenCL accelerators/GPUs. The alternative luxcorerender test profile is for CPU execution due to a difference in tests, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender OpenCL 2.3Scene: FoodGTX 1060GTX 1650GTX 1650 SUPERGTX 1070GTX 1070 TiGTX 1080GTX 1660GTX 1660 TiGTX 1660 SUPERRTX 2060RTX 2070RTX 2060 SUPERRTX 2080RTX 2070 SUPERRTX 2080 SUPERRTX 2080 TiTITAN RTX0.50181.00361.50542.00722.509SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 4SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 4SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 30.650.670.810.920.980.990.991.071.091.171.301.331.871.901.932.232.23MIN: 0.22 / MAX: 0.74MIN: 0.22 / MAX: 0.8MIN: 0.24 / MAX: 0.93MIN: 0.23 / MAX: 1.06MIN: 0.24 / MAX: 1.14MIN: 0.24 / MAX: 1.15MIN: 0.24 / MAX: 1.16MIN: 0.25 / MAX: 1.23MIN: 0.25 / MAX: 1.26MIN: 0.26 / MAX: 1.39MIN: 0.29 / MAX: 1.49MIN: 0.29 / MAX: 1.53MIN: 0.25 / MAX: 2.25MIN: 0.28 / MAX: 2.24MIN: 0.29 / MAX: 2.33MIN: 0.32 / MAX: 2.65MIN: 0.25 / MAX: 2.7

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: CopyGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1070GTX 1070 TiGTX 1080GTX 1660 TiRTX 2060GTX 1660 SUPERRTX 2070RTX 2060 SUPERRTX 2080RTX 2070 SUPERRTX 2080 SUPERTITAN RTXRTX 2080 Ti70140210280350SE +/- 0.00, N = 4SE +/- 0.02, N = 5SE +/- 0.00, N = 5SE +/- 0.17, N = 5SE +/- 0.02, N = 6SE +/- 0.03, N = 6SE +/- 0.02, N = 6SE +/- 0.06, N = 6SE +/- 0.06, N = 7SE +/- 0.05, N = 7SE +/- 0.09, N = 8SE +/- 0.06, N = 8SE +/- 0.06, N = 8SE +/- 0.07, N = 8SE +/- 0.05, N = 8SE +/- 0.08, N = 9SE +/- 0.10, N = 899.2138.5143.8145.9183.3184.1207.4211.1239.4240.2284.4288.0290.0292.1301.7322.7324.01. (CC) gcc options: -O2 -flto -lOpenCL

PlaidML

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

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Training - Network: Mobilenet - Device: OpenCLGTX 1650GTX 1650 SUPERGTX 1060GTX 1660GTX 1070 TiGTX 1080GTX 1070GTX 1660 TiGTX 1660 SUPERRTX 2060RTX 2070RTX 2080RTX 2070 SUPERRTX 2060 SUPERRTX 2080 SUPERRTX 2080 TiTITAN RTX4080120160200SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.17, N = 3SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.14, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.21, N = 365.5977.1083.5297.9499.6499.82103.79108.06111.39125.33153.03154.35154.74154.81158.82187.73198.37

GROMACS

The CUDA version of the Gromacs molecular dynamics package. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkGTX 1060GTX 1650GTX 1070GTX 1650 SUPERGTX 1070 TiGTX 1660GTX 1660 SUPERGTX 1080GTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERTITAN RTXRTX 2080 Ti246810SE +/- 0.003, N = 3SE +/- 0.005, N = 3SE +/- 0.003, N = 3SE +/- 0.004, N = 3SE +/- 0.002, N = 3SE +/- 0.010, N = 3SE +/- 0.007, N = 3SE +/- 0.004, N = 3SE +/- 0.006, N = 3SE +/- 0.008, N = 3SE +/- 0.016, N = 3SE +/- 0.013, N = 3SE +/- 0.008, N = 3SE +/- 0.015, N = 3SE +/- 0.041, N = 3SE +/- 0.005, N = 3SE +/- 0.020, N = 23.1813.2534.0524.1224.2594.4124.6294.6344.7895.4165.7895.8656.2496.5366.5727.1177.1361. (CXX) g++ options: -O3 -lpthread -ldl -lrt -lm

Mixbench

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

OpenBenchmarking.orgGIOPS, More Is BetterMixbench 2020-06-23Backend: OpenCL - Benchmark: IntegerGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX4K8K12K16K20KSE +/- 2.70, N = 14SE +/- 0.96, N = 13SE +/- 0.83, N = 14SE +/- 6.51, N = 15SE +/- 0.52, N = 12SE +/- 0.38, N = 13SE +/- 0.61, N = 13SE +/- 7.30, N = 13SE +/- 50.53, N = 15SE +/- 14.13, N = 13SE +/- 2.40, N = 13SE +/- 1.82, N = 13SE +/- 1.47, N = 14SE +/- 1.20, N = 14SE +/- 11.48, N = 14SE +/- 11.06, N = 14SE +/- 63.55, N = 151354.142019.082157.312655.583466.724975.185303.515432.405803.637275.338218.028362.869696.3810930.5611752.1116323.9416979.991. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

OpenBenchmarking.orgGIOPS, More Is BetterMixbench 2020-06-23Backend: NVIDIA CUDA - Benchmark: IntegerGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1650GTX 1650 SUPERGTX 1660GTX 1660 SUPERGTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX3K6K9K12K15KSE +/- 2.01, N = 9SE +/- 3.28, N = 11SE +/- 0.87, N = 11SE +/- 1.56, N = 11SE +/- 0.11, N = 12SE +/- 0.38, N = 13SE +/- 0.22, N = 13SE +/- 4.94, N = 15SE +/- 0.35, N = 13SE +/- 4.65, N = 13SE +/- 49.03, N = 15SE +/- 0.55, N = 14SE +/- 0.98, N = 14SE +/- 7.76, N = 14SE +/- 1.56, N = 14SE +/- 264.51, N = 15SE +/- 458.60, N = 151187.721764.631887.612326.093108.844492.324753.114866.615271.346539.817348.917540.288722.289732.4810562.9914103.7914426.111. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

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 INTGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1650GTX 1650 SUPERGTX 1660 SUPERGTX 1660GTX 1660 TiRTX 2060RTX 2060 SUPERRTX 2070RTX 2070 SUPERRTX 2080RTX 2080 SUPERRTX 2080 TiTITAN RTX3K6K9K12K15KSE +/- 10.83, N = 15SE +/- 6.03, N = 12SE +/- 0.30, N = 10SE +/- 7.00, N = 10SE +/- 1.93, N = 13SE +/- 41.07, N = 15SE +/- 6.44, N = 12SE +/- 10.20, N = 12SE +/- 49.36, N = 15SE +/- 65.66, N = 15SE +/- 81.72, N = 15SE +/- 83.63, N = 15SE +/- 66.40, N = 12SE +/- 51.79, N = 12SE +/- 58.52, N = 13SE +/- 23.37, N = 12SE +/- 169.21, N = 151275.121695.792076.452430.552641.384128.394280.554338.674734.365356.846930.217133.778484.419141.2310266.7213629.3713783.931. (CXX) g++ options: -O3 -rdynamic -lOpenCL

NAMD CUDA

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

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD CUDA 2.14ATPase Simulation - 327,506 AtomsGTX 1650GTX 1060GTX 1650 SUPERGTX 1660GTX 1070GTX 1660 SUPERGTX 1660 TiGTX 1080GTX 1070 TiRTX 2060RTX 2070RTX 2060 SUPERRTX 2070 SUPERRTX 2080 SUPERRTX 2080TITAN RTXRTX 2080 Ti0.07760.15520.23280.31040.388SE +/- 0.00088, N = 3SE +/- 0.00077, N = 3SE +/- 0.00061, N = 4SE +/- 0.00142, N = 4SE +/- 0.00276, N = 4SE +/- 0.00026, N = 4SE +/- 0.00217, N = 4SE +/- 0.00174, N = 4SE +/- 0.00101, N = 4SE +/- 0.00144, N = 4SE +/- 0.00191, N = 4SE +/- 0.00060, N = 4SE +/- 0.00025, N = 4SE +/- 0.00097, N = 4SE +/- 0.00054, N = 4SE +/- 0.00134, N = 4SE +/- 0.00046, N = 40.345030.309570.262110.242900.240220.231770.228000.224260.223000.206350.203620.200570.196250.193290.192650.191950.19082

ViennaCL

ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile uses ViennaCL OpenCL support and runs the included computational benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterViennaCL 1.4.2OpenCL LU FactorizationGTX 1060GTX 1650GTX 1070GTX 1660GTX 1650 SUPERGTX 1660 SUPERGTX 1070 TiGTX 1660 TiGTX 1080RTX 2060 SUPERRTX 2070RTX 2060RTX 2070 SUPERRTX 2080RTX 2080 TiRTX 2080 SUPERTITAN RTX20406080100SE +/- 0.32, N = 11SE +/- 0.21, N = 11SE +/- 0.23, N = 11SE +/- 0.25, N = 11SE +/- 0.27, N = 11SE +/- 0.37, N = 11SE +/- 0.38, N = 11SE +/- 0.37, N = 11SE +/- 0.21, N = 11SE +/- 0.24, N = 11SE +/- 0.47, N = 11SE +/- 0.29, N = 11SE +/- 0.44, N = 11SE +/- 0.38, N = 11SE +/- 0.39, N = 11SE +/- 0.41, N = 11SE +/- 0.57, N = 1161.6562.9666.2167.8267.9368.1569.1870.0370.6972.5072.5773.3174.4574.7375.7976.1076.641. (CXX) g++ options: -rdynamic -lOpenCL

GPU Temperature Monitor

OpenBenchmarking.orgCelsiusGPU Temperature MonitorPhoronix Test Suite System MonitoringRTX 2080GTX 1080GTX 1070RTX 2080 TiTITAN RTXGTX 1660 SUPERRTX 2070RTX 2060 SUPERGTX 1060RTX 2060GTX 1660GTX 1650 SUPERRTX 2070 SUPERGTX 1650RTX 2080 SUPERGTX 1660 TiGTX 1070 Ti1632486480Min: 28 / Avg: 62.06 / Max: 83Min: 28 / Avg: 61.17 / Max: 83Min: 29 / Avg: 59.81 / Max: 78Min: 29 / Avg: 56.11 / Max: 78Min: 35 / Avg: 55.61 / Max: 78Min: 33 / Avg: 55.48 / Max: 68Min: 29 / Avg: 54.87 / Max: 76Min: 33 / Avg: 53.72 / Max: 74Min: 32 / Avg: 52.77 / Max: 69Min: 29 / Avg: 51.56 / Max: 73Min: 30 / Avg: 51.27 / Max: 76Min: 30 / Avg: 50.85 / Max: 66Min: 29 / Avg: 50.76 / Max: 71Min: 27 / Avg: 49.62 / Max: 67Min: 28 / Avg: 49.61 / Max: 73Min: 29 / Avg: 48.98 / Max: 66Min: 32 / Avg: 45.46 / Max: 58

GPU Power Consumption Monitor

OpenBenchmarking.orgWattsGPU Power Consumption MonitorPhoronix Test Suite System MonitoringTITAN RTXRTX 2080 TiRTX 2080RTX 2070 SUPERGTX 1080RTX 2080 SUPERRTX 2060 SUPERRTX 2070GTX 1070RTX 2060GTX 1660 SUPERGTX 1060GTX 1660 TiGTX 1650 SUPERGTX 1660GTX 165060120180240300Min: 24.02 / Avg: 126.58 / Max: 312.98Min: 5.74 / Avg: 124.11 / Max: 273.84Min: 13.71 / Avg: 103.45 / Max: 219.93Min: 14.36 / Avg: 96.17 / Max: 217.58Min: 6.47 / Avg: 92.36 / Max: 196.24Min: 8.85 / Avg: 91.93 / Max: 248.16Min: 9.27 / Avg: 88.74 / Max: 180.97Min: 7.57 / Avg: 85.38 / Max: 180.92Min: 6.84 / Avg: 82.21 / Max: 159.92Min: 9.59 / Avg: 74.07 / Max: 165.3Min: 10.68 / Avg: 67.65 / Max: 128.36Min: 6.24 / Avg: 64.43 / Max: 128.04Min: 7.22 / Avg: 59.54 / Max: 132.74Min: 6.47 / Avg: 47.87 / Max: 99.24Min: 6.72 / Avg: 45.57 / Max: 122.32Min: 5.55 / Avg: 37.62 / Max: 70.94