Radeon ROCm vs. NVIDIA OpenCL August 2017

Radeon ROCm and NVIDIA OpenCL Linux testing by Michael Larabel for a future article on Phoronix.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1708103-TY-1708107TY43
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:

NVIDIA GPU Compute 5 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

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

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
Radeon R9 285
August 08 2017
 
Radeon R9 290
August 08 2017
 
Radeon RX 480
August 08 2017
 
Radeon RX 560
August 08 2017
 
Radeon RX 580
August 08 2017
 
Radeon R9 Fury
August 08 2017
 
GeForce GTX 780 Ti
August 02 2017
 
GeForce GTX 960
August 06 2017
 
GeForce GTX 970
August 02 2017
 
GeForce GTX 980
August 01 2017
 
GeForce GTX 980 Ti
August 01 2017
 
GeForce GTX 1050
August 01 2017
 
GeForce GTX 1060
August 01 2017
 
GeForce GTX 1070
August 01 2017
 
GeForce GTX 1080
August 01 2017
 
GeForce GTX 1080 Ti
August 02 2017
 
Radeon Vega FE
August 10 2017
 
Invert Hiding All Results Option
 

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):


Radeon ROCm vs. NVIDIA OpenCL August 2017ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkMonitorOSKernelDesktopDisplay DriverOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon R9 285Radeon Vega FEIntel Core i7-7740K @ 4.50GHz (8 Cores)ASUS PRIME X299-AIntel Device 591f16384MB525GB Crucial_CT525MX3 + Samsung SSD 950 PRO 256GBNVIDIA GeForce GTX 980 4096MB (1126/3505MHz)Realtek ALC1220Intel ConnectionUbuntu 16.044.13.0-999-generic (x86_64) 20170730Unity 7.4.0NVIDIA 384.594.5.0OpenCL 1.2 CUDA 9.0.1301.0.42GCC 5.4.0 20160609ext43840x2160NVIDIA GeForce GTX 980 Ti 6144MB (999/3505MHz)NVIDIA GeForce GTX 1060 6GB 6144MB (1506/4006MHz)NVIDIA GeForce GTX 1070 8192MB (1506/4006MHz)NVIDIA GeForce GTX 1080 8192MB (1607/5005MHz)Zotac NVIDIA GeForce GTX 1050 2048MB (1354/3504MHz)NVIDIA GeForce GTX 1080 Ti 11264MB (1480/5508MHz)NVIDIA GeForce GTX 780 Ti 3072MB (875/3500MHz)eVGA NVIDIA GeForce GTX 970 4096MB (1163/3505MHz)eVGA NVIDIA GeForce GTX 960 2048MB (1277/3505MHz)XFX AMD HAWAII 4096MBRealtek GenericAcer B286HK4.11.0-kfd-compute-rocm-rel-1.6-127 (x86_64)modesetting 1.19.34.5 Mesa 17.3.0-devel- padoka PPA (LLVM 6.0.0)OpenCL 2.0 AMD-APP (2450.0)Sapphire AMD FIJI 4096MBMSI AMD POLARIS10 8192MBAMD POLARIS11 4096MBamdgpu 1.3.0AMD POLARIS10 8192MBXFX AMD TONGA 2048MBmodesetting 1.19.3AMD Ryzen 5 1600 Six-Core @ 3.20GHz (12 Cores)Gigabyte AB350M-D3H-CFAMD Device 1450Samsung SSD 960 EVO 500GBamdgpudrmfbAMD Device aaf8HP 27erRealtek RTL8111/8168/8411OpenCL 2.0 AMD-APP (2450.0)1920x1080OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-tune=generic -vProcessor Details- GeForce GTX 980: Scaling Governor: intel_pstate performance- GeForce GTX 980 Ti: Scaling Governor: intel_pstate performance- GeForce GTX 1060: Scaling Governor: intel_pstate performance- GeForce GTX 1070: Scaling Governor: intel_pstate performance- GeForce GTX 1080: Scaling Governor: intel_pstate performance- GeForce GTX 1050: Scaling Governor: intel_pstate performance- GeForce GTX 1080 Ti: Scaling Governor: intel_pstate performance- GeForce GTX 780 Ti: Scaling Governor: intel_pstate performance- GeForce GTX 970: Scaling Governor: intel_pstate performance- GeForce GTX 960: Scaling Governor: intel_pstate performance- Radeon R9 290: Scaling Governor: intel_pstate performance- Radeon R9 Fury: Scaling Governor: intel_pstate performance- Radeon RX 580: Scaling Governor: intel_pstate performance- Radeon RX 560: Scaling Governor: intel_pstate performance- Radeon RX 480: Scaling Governor: intel_pstate performance- Radeon R9 285: Scaling Governor: intel_pstate performance- Radeon Vega FE: Scaling Governor: acpi-cpufreq ondemandOpenCL Details- GeForce GTX 980: GPU Compute Cores: 2048- GeForce GTX 980 Ti: GPU Compute Cores: 2816- GeForce GTX 1060: GPU Compute Cores: 1280- GeForce GTX 1070: GPU Compute Cores: 1920- GeForce GTX 1080: GPU Compute Cores: 2560- GeForce GTX 1050: GPU Compute Cores: 640- GeForce GTX 1080 Ti: GPU Compute Cores: 3584- GeForce GTX 780 Ti: GPU Compute Cores: 2880- GeForce GTX 970: GPU Compute Cores: 1664- GeForce GTX 960: GPU Compute Cores: 1024System Details- GeForce GTX 980: GPU Compute Cores: 2048.- GeForce GTX 980 Ti: GPU Compute Cores: 2816.- GeForce GTX 1060: GPU Compute Cores: 1280.- GeForce GTX 1070: GPU Compute Cores: 1920.- GeForce GTX 1080: GPU Compute Cores: 2560.- GeForce GTX 1050: GPU Compute Cores: 640.- GeForce GTX 1080 Ti: GPU Compute Cores: 3584.- GeForce GTX 780 Ti: GPU Compute Cores: 2880.- GeForce GTX 970: GPU Compute Cores: 1664.- GeForce GTX 960: GPU Compute Cores: 1024.

Radeon ROCm vs. NVIDIA OpenCL August 2017mixbench: Single Precisionclpeak: Double-Precision Doublemixbench: Double Precisionluxmark: GPU - Hotelshoc: OpenCL - Max SP Flopsdarktable: Boat - OpenCLclpeak: Integer Compute INTshoc: OpenCL - MD5 Hashclpeak: Single-Precision Floatcl-mem: Writeclpeak: Global Memory Bandwidthviennacl: OpenCL LU Factorizationshoc: OpenCL - Texture Read Bandwidthclpeak: Transfer Bandwidth enqueueWriteBuffershoc: OpenCL - FFT SPcl-mem: Copyluxmark: GPU - Luxball HDRcl-mem: Readfahbench: shoc: OpenCL - Triaddarktable: Server Room - OpenCLmixbench: Integerclpeak: Kernel Latencydarktable: Masskrug - OpenCLGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon R9 285Radeon Vega FE4700.91159.68159.8617565051.8915.231296.117.544288.05152.30164.1054.96332.1112.48447.62142.6011959164.4797.3812.030.211402.274.070.215563.69195.67194.7221426208.654.191583.859.275292.12238.40263.3256.91351.4512.21693.44216.3714811266.17108.1512.290.221717.514.330.224389.15150.41152.1617424829.994.661223.587.364157.55138.70146.2354.03382.2112.37322.16137.7011572151.6097.2411.960.191366.844.100.186402.60225.45223.8722867125.883.781631.0310.616359.20191.43196.3658.95454.4712.61470.12186.6316186205.43132.7912.200.192027.963.550.188493.48295.22295.3427259446.743.672349.8614.298249.62213.13218.1561.26520.1312.59597.69206.5312776227.20145.3812.310.182662.464.060.172040.4666.8866.7610222125.7818.18568.643.241937.0085.7392.4341.80274.916.44204.3687.5065699549.786.120.23614.523.820.2289.23415.066.39353213274.173.133200.3019.7411780.26335.47329.4363.67596.8212.61974.37316.8019662338.07186.7412.510.1827.883.550.174263.71246.21245.1111994944.7215.24961.134.663847.88252.00252.9854.59287.3912.36429.87237.109516271.6772.7712.350.25968.935.560.274125.24137.56136.1816494361.6516.261128.646.543728.55129.70143.4153.03288.8712.49382.02125.2710704143.5585.4511.910.221220.804.050.212765.1392.4492.7411302960.4919.30781.164.462429.1070.8081.1247.64277.1112.46209.1570.60614881.4058.3511.240.25835.013.880.24599.0610134790.313.851595.416.114755.89217.70272.1720.97253.6430.33192.4010400123.000.156.670.176508.34447.26440.8513677144.883.421429.329.097068.81391.50431.2621.76245.6830.74550.79206.1313212123.3010.170.131386.505.940.135863.73391.77389.9712186260.484.081251.637.966175.41180.70208.9611.98207.9730.38498.10183.4710400159.739.680.171227.325.630.182465.62164.50164.005092634.536.91524.703.342588.6680.9089.2512.08115.8130.38209.7681.03446893.504.950.27508.695.820.275450.06364.04362.2910705812.644.161162.887.335737.62179.70209.3912.05193.4330.64462.35185.379885157.879.610.131140.245.860.1492.7412.6612653.803.7816.00362.8030.68422.09885.04211.93154.2011.8219.70OpenBenchmarking.org

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2016-06-06Benchmark: Single PrecisionGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE2K4K6K8K10KSE +/- 5.25, N = 3SE +/- 255.36, N = 3SE +/- 2.95, N = 3SE +/- 54.41, N = 3SE +/- 48.08, N = 3SE +/- 0.81, N = 3SE +/- 3.63, N = 3SE +/- 5.00, N = 3SE +/- 2.66, N = 3SE +/- 0.91, N = 3SE +/- 1.46, N = 3SE +/- 4.20, N = 3SE +/- 0.14, N = 3SE +/- 2.06, N = 3SE +/- 0.06, N = 34700.915563.694389.156402.608493.482040.4689.234263.714125.242765.136508.345863.732465.625450.0692.741. (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.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480130260390520650SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.30, N = 3SE +/- 1.12, N = 3SE +/- 0.76, N = 3SE +/- 0.08, N = 3SE +/- 1.65, N = 3SE +/- 0.29, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3159.68195.67150.41225.45295.2266.88415.06246.21137.5692.44599.06447.26391.77164.50364.04

OpenBenchmarking.orgWatts, Fewer Is BetterclpeakSystem Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48060120180240300Min: 254.1 / Avg: 254.4 / Max: 254.7Min: 63.5 / Avg: 152.7 / Max: 241.9Min: 50.7 / Avg: 88 / Max: 125.3Min: 51.8 / Avg: 159 / Max: 266.2Min: 44.6 / Avg: 55.5 / Max: 66.4Min: 59.7 / Avg: 109 / Max: 158.3Min: 82.9 / Avg: 130.65 / Max: 178.4Min: 53.3 / Avg: 128.25 / Max: 203.2Min: 158.9 / Avg: 196.69 / Max: 311.4Min: 91.2 / Avg: 171.7 / Max: 293.2Min: 62.1 / Avg: 135.37 / Max: 215.7Min: 57.7 / Avg: 83.39 / Max: 110.6Min: 74.5 / Avg: 139.09 / Max: 198.1

Mixbench

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

OpenBenchmarking.orgGFLOPS, More Is BetterMixbench 2016-06-06Benchmark: Double PrecisionGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE100200300400500SE +/- 0.01, N = 3SE +/- 2.32, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.92, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 3SE +/- 0.71, N = 3SE +/- 0.25, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3159.86194.72152.16223.87295.3466.766.39245.11136.1892.74440.85389.97164.00362.2912.661. (CXX) g++ options: -lm -lstdc++ -lOpenCL -lrt -O2

LuxMark

LuxMark is a multi-platform OpenGL benchmark using LuxRender / SLG2. LuxMark supports targeting different OpenCL devices and has multiple scenes available for rendering. LuxMark is a fully open-source OpenCL program with real-world rendering examples. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: HotelGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4808001600240032004000SE +/- 55.00, N = 3SE +/- 36.99, N = 3SE +/- 9.21, N = 3SE +/- 6.67, N = 3SE +/- 28.15, N = 3SE +/- 1.00, N = 3SE +/- 54.49, N = 3SE +/- 5.33, N = 3SE +/- 36.71, N = 3SE +/- 7.64, N = 3SE +/- 0.67, N = 3SE +/- 3.84, N = 3SE +/- 3.33, N = 3SE +/- 2.33, N = 3SE +/- 0.67, N = 317562142174222862725102235321199164911301013136712185091070

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48060120180240300Min: 55.9 / Avg: 163.16 / Max: 251Min: 151.8 / Avg: 198.76 / Max: 304.5Min: 73.6 / Avg: 118.38 / Max: 167.3Min: 78.5 / Avg: 140.86 / Max: 210.3Min: 127.9 / Avg: 156.13 / Max: 262.7Min: 65.4 / Avg: 91.36 / Max: 114.7Min: 52.6 / Avg: 203.91 / Max: 304.8Min: 59.2 / Avg: 226.01 / Max: 336.8Min: 57.8 / Avg: 149.11 / Max: 234.2Min: 117.1 / Avg: 124.83 / Max: 188.6Min: 109.8 / Avg: 204.67 / Max: 296.3Min: 121.6 / Avg: 215.47 / Max: 344.8Min: 61.4 / Avg: 165.94 / Max: 235Min: 50.6 / Avg: 92.86 / Max: 123.4Min: 66.7 / Avg: 159.05 / Max: 205.4

LuxMark

OpenBenchmarking.orgWatts, Fewer Is BetterLuxMark 3.0System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon R9 28550100150200250Min: 56.4 / Avg: 203.63 / Max: 213.1Min: 61 / Avg: 242.75 / Max: 254.4Min: 61.8 / Avg: 152.59 / Max: 156.6Min: 52.1 / Avg: 183.39 / Max: 188.4Min: 61.4 / Avg: 203.35 / Max: 209.7Min: 44.6 / Avg: 107.9 / Max: 115.3Min: 75.8 / Avg: 261.7 / Max: 278.5Min: 79.8 / Avg: 279.87 / Max: 294.8Min: 57.1 / Avg: 191.79 / Max: 197.1Min: 68.3 / Avg: 154.46 / Max: 158.8Min: 164 / Avg: 223.01 / Max: 246.2Min: 87.9 / Avg: 202.29 / Max: 248.6Min: 62.8 / Avg: 167.57 / Max: 197.1Min: 51.4 / Avg: 94.7 / Max: 102.8Min: 103.4 / Avg: 161.24 / Max: 183.2

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48050100150200250Min: 55.7 / Avg: 159.95 / Max: 185.7Min: 143.3 / Avg: 204.53 / Max: 231.7Min: 49.9 / Avg: 117.9 / Max: 137.8Min: 50.7 / Avg: 138.56 / Max: 159.1Min: 52.2 / Avg: 141.33 / Max: 169.7Min: 44.8 / Avg: 89.24 / Max: 101.5Min: 66.8 / Avg: 190.32 / Max: 229.1Min: 94.2 / Avg: 240.33 / Max: 287.5Min: 58 / Avg: 146.29 / Max: 167.8Min: 96.7 / Avg: 124.57 / Max: 141.9Min: 111.2 / Avg: 180.14 / Max: 214.7Min: 167.4 / Avg: 178.62 / Max: 202Min: 62.8 / Avg: 129.51 / Max: 166.9Min: 51.3 / Avg: 83.15 / Max: 94.2Min: 69.5 / Avg: 134.82 / Max: 175

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Max SP FlopsGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE3K6K9K12K15KSE +/- 3.16, N = 3SE +/- 15.98, N = 3SE +/- 6.12, N = 3SE +/- 32.37, N = 3SE +/- 38.82, N = 3SE +/- 0.09, N = 3SE +/- 65.79, N = 3SE +/- 19.58, N = 3SE +/- 1.82, N = 3SE +/- 7.79, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 1.35, N = 3SE +/- 0.11, N = 3SE +/- 1.92, N = 3SE +/- 8.00, N = 35051.896208.654829.997125.889446.742125.7813274.174944.724361.652960.494790.317144.886260.482634.535812.6412653.801. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

LuxMark

OpenBenchmarking.orgWatts, Fewer Is BetterLuxMark 3.0System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon R9 28560120180240300Min: 136.6 / Avg: 207.25 / Max: 212.8Min: 120.1 / Avg: 253.78 / Max: 259Min: 50 / Avg: 151.57 / Max: 157.1Min: 78.2 / Avg: 186.9 / Max: 196.9Min: 81.4 / Avg: 184.66 / Max: 189.3Min: 62.1 / Avg: 109.06 / Max: 119.8Min: 103.3 / Avg: 258.5 / Max: 265.6Min: 127.9 / Avg: 295.03 / Max: 310.7Min: 88.4 / Avg: 191.17 / Max: 194.5Min: 73.2 / Avg: 145.52 / Max: 149.6Min: 108.1 / Avg: 253.37 / Max: 262.7Min: 60.8 / Avg: 231.32 / Max: 244.7Min: 61.7 / Avg: 198.14 / Max: 205.1Min: 50.4 / Avg: 104.54 / Max: 107.3Min: 71.5 / Avg: 186.43 / Max: 197.7

Darktable

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.0.3Test: Boat - Acceleration: OpenCLGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE510152025SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 315.234.194.663.783.6718.183.1315.2416.2619.303.853.424.086.914.163.78

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 INTGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4807001400210028003500SE +/- 0.62, N = 3SE +/- 16.78, N = 3SE +/- 52.05, N = 3SE +/- 19.23, N = 3SE +/- 29.30, N = 3SE +/- 15.64, N = 3SE +/- 117.22, N = 3SE +/- 17.95, N = 3SE +/- 22.00, N = 3SE +/- 6.09, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 31296.111583.851223.581631.032349.86568.643200.30961.131128.64781.161595.411429.321251.63524.701162.88

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGHash/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: MD5 HashGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.01, 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 +/- 0.00, N = 37.549.277.3610.6114.293.2419.744.666.544.466.119.097.963.347.3316.001. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -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 FloatGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4803K6K9K12K15KSE +/- 0.64, N = 3SE +/- 13.41, N = 3SE +/- 74.08, N = 3SE +/- 0.57, N = 3SE +/- 157.37, N = 3SE +/- 0.29, N = 3SE +/- 0.82, N = 3SE +/- 0.62, N = 3SE +/- 0.43, N = 3SE +/- 70.89, N = 3SE +/- 0.30, N = 3SE +/- 0.51, N = 3SE +/- 0.38, N = 3SE +/- 0.20, N = 3SE +/- 0.06, N = 34288.055292.124157.556359.208249.621937.0011780.263847.883728.552429.104755.897068.816175.412588.665737.62

cl-mem

OpenBenchmarking.orgWatts, Fewer Is Bettercl-mem 2017-01-13System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48050100150200250Min: 102 / Avg: 140.4 / Max: 165.1Min: 63.5 / Avg: 169.7 / Max: 225.2Min: 77.6 / Avg: 112.65 / Max: 130.4Min: 50.4 / Avg: 132.24 / Max: 154.3Min: 99.7 / Avg: 126.93 / Max: 164.8Min: 44.6 / Avg: 91.7 / Max: 98.4Min: 213.3 / Avg: 215.1 / Max: 216.9Min: 149 / Avg: 212.7 / Max: 266.1Min: 86.1 / Avg: 140.75 / Max: 165.4Min: 53 / Avg: 112.52 / Max: 126.6Min: 208.1 / Avg: 219.88 / Max: 227.1Min: 111.7 / Avg: 167.53 / Max: 194.9Min: 61.5 / Avg: 128.3 / Max: 177.9Min: 61.8 / Avg: 94.24 / Max: 99.4Min: 67.2 / Avg: 149.58 / Max: 172.7

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgWatts, Fewer Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48060120180240300Min: 104 / Avg: 105.7 / Max: 107.4Min: 82.3 / Avg: 132.4 / Max: 182.5Min: 116.8 / Avg: 117.17 / Max: 117.4Min: 130.1 / Avg: 172.7 / Max: 215.3Min: 98.4 / Avg: 109.4 / Max: 120.4Min: 168.7 / Avg: 174.7 / Max: 180.7Min: 109.8 / Avg: 199.5 / Max: 289.2Min: 63.4 / Avg: 204.7 / Max: 346Min: 61.6 / Avg: 65.85 / Max: 70.1Min: 114.5 / Avg: 114.7 / Max: 114.9Min: 112.9 / Avg: 157.65 / Max: 202.4

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: WriteGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE80160240320400SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.31, N = 3SE +/- 0.09, N = 3SE +/- 0.87, N = 3SE +/- 0.03, N = 3SE +/- 0.22, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 2SE +/- 1.08, N = 3SE +/- 4.06, N = 3SE +/- 0.80, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 3SE +/- 0.21, N = 3152.30238.40138.70191.43213.1385.73335.47252.00129.7070.80217.70391.50180.7080.90179.70362.801. (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 BandwidthGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48090180270360450SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.45, N = 3SE +/- 0.15, N = 3SE +/- 3.85, N = 3SE +/- 0.18, N = 3SE +/- 0.82, N = 3SE +/- 10.22, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.90, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3164.10263.32146.23196.36218.1592.43329.43252.98143.4181.12272.17431.26208.9689.25209.39

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 FactorizationGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE1428425670SE +/- 0.03, N = 3SE +/- 0.34, N = 3SE +/- 0.37, N = 3SE +/- 0.02, N = 3SE +/- 0.14, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 2.80, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.27, N = 3SE +/- 0.75, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.20, N = 354.9656.9154.0358.9561.2641.8063.6754.5953.0347.6420.9721.7611.9812.0812.0530.681. (CXX) g++ options: -rdynamic -lOpenCL

cl-mem

OpenBenchmarking.orgWatts, Fewer Is Bettercl-mem 2017-01-13System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48050100150200250Min: 55.9 / Avg: 137.83 / Max: 165.7Min: 164 / Avg: 209.78 / Max: 231Min: 113.9 / Avg: 126.18 / Max: 129.8Min: 50.8 / Avg: 133.6 / Max: 156.7Min: 120.4 / Avg: 154.88 / Max: 168.9Min: 51 / Avg: 88.33 / Max: 98.2Min: 131.4 / Avg: 163.5 / Max: 225.5Min: 205.2 / Avg: 232.33 / Max: 267.5Min: 145.8 / Avg: 161.02 / Max: 165.8Min: 75.2 / Avg: 119.81 / Max: 126.4Min: 110.3 / Avg: 194.92 / Max: 227.7Min: 63.2 / Avg: 162.7 / Max: 196.7Min: 61.6 / Avg: 134.4 / Max: 178.4Min: 50.7 / Avg: 89.22 / Max: 99.8Min: 158.5 / Avg: 168.35 / Max: 173.8

OpenBenchmarking.orgWatts, Fewer Is Bettercl-mem 2017-01-13System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48050100150200250Min: 110.7 / Avg: 156.45 / Max: 166Min: 136.5 / Avg: 208.38 / Max: 235Min: 123.2 / Avg: 128.26 / Max: 130.1Min: 51.8 / Avg: 133.92 / Max: 156.5Min: 99.9 / Avg: 143.28 / Max: 167.2Min: 97.2 / Avg: 97.62 / Max: 98.3Min: 215.6 / Avg: 216.75 / Max: 217.9Min: 59.1 / Avg: 193.83 / Max: 264Min: 57.9 / Avg: 143.07 / Max: 165.3Min: 75.3 / Avg: 117.34 / Max: 125.9Min: 110.7 / Avg: 173.45 / Max: 211.5Min: 63.6 / Avg: 161.35 / Max: 231Min: 126.3 / Avg: 165.33 / Max: 179.6Min: 51.1 / Avg: 91.53 / Max: 100.1Min: 73.8 / Avg: 137.13 / Max: 171.7

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Texture Read BandwidthGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE130260390520650SE +/- 1.03, N = 3SE +/- 0.74, N = 3SE +/- 0.59, N = 3SE +/- 0.27, N = 3SE +/- 0.88, N = 3SE +/- 1.17, N = 3SE +/- 0.23, N = 3SE +/- 0.11, N = 3SE +/- 0.24, N = 3SE +/- 0.45, N = 3SE +/- 1.41, N = 3SE +/- 0.89, N = 3SE +/- 0.08, N = 3SE +/- 0.14, N = 3SE +/- 0.23, N = 3SE +/- 1.01, N = 3332.11351.45382.21454.47520.13274.91596.82287.39288.87277.11253.64245.68207.97115.81193.43422.091. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

clpeak

OpenBenchmarking.orgWatts, Fewer Is BetterclpeakSystem Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4804080120160200Min: 57 / Avg: 152.7 / Max: 162.3Min: 139.5 / Avg: 189.78 / Max: 196.9Min: 49.1 / Avg: 105.58 / Max: 118Min: 133.6 / Avg: 138.18 / Max: 139.4Min: 52.7 / Avg: 144.57 / Max: 160.3Min: 44.7 / Avg: 85.97 / Max: 90.4Min: 57.5 / Avg: 179.46 / Max: 201.8Min: 220.2 / Avg: 220.42 / Max: 221Min: 61.8 / Avg: 134.1 / Max: 144.6Min: 85.3 / Avg: 121.06 / Max: 124.6Min: 110.3 / Avg: 178.68 / Max: 199.3Min: 63.9 / Avg: 177.34 / Max: 225.5Min: 95.5 / Avg: 138.34 / Max: 170Min: 51 / Avg: 87.81 / Max: 93.9Min: 73.8 / Avg: 140.82 / Max: 162.3

OpenBenchmarking.orgGBPS Per Watt, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 970Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.08780.17560.26340.35120.4390.110.080.130.110.110.080.080.120.180.290.320.390.31

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480714212835SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.33, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 312.4812.2112.3712.6112.596.4412.6112.3612.4912.4630.3330.7430.3830.3830.64

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: FFT SPGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE2004006008001000SE +/- 0.99, N = 3SE +/- 21.12, N = 3SE +/- 7.21, N = 3SE +/- 8.94, N = 3SE +/- 6.81, N = 3SE +/- 12.49, N = 3SE +/- 5.16, N = 3SE +/- 21.22, N = 3SE +/- 6.83, N = 3SE +/- 0.45, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.88, N = 3447.62693.44322.16470.12597.69204.36974.37429.87382.02209.15550.79498.10209.76462.35885.041. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -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: CopyGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE70140210280350SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.40, N = 3SE +/- 0.03, N = 3SE +/- 0.84, N = 3SE +/- 0.15, N = 3SE +/- 0.10, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 2SE +/- 0.64, N = 3SE +/- 0.48, N = 3SE +/- 0.38, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.32, N = 3142.60216.37137.70186.63206.5387.50316.80237.10125.2770.60192.40206.13183.4781.03185.37211.931. (CC) gcc options: -O2 -flto -lOpenCL

clpeak

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterclpeakOpenCL Test: Double-Precision DoubleGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.75381.50762.26143.01523.7691.051.031.421.632.040.782.311.121.030.763.352.522.831.872.59

LuxMark

LuxMark is a multi-platform OpenGL benchmark using LuxRender / SLG2. LuxMark supports targeting different OpenCL devices and has multiple scenes available for rendering. LuxMark is a fully open-source OpenCL program with real-world rendering examples. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: Luxball HDRGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4804K8K12K16K20KSE +/- 26.67, N = 3SE +/- 118.82, N = 3SE +/- 38.33, N = 3SE +/- 3.84, N = 3SE +/- 51.02, N = 3SE +/- 15.33, N = 3SE +/- 13.57, N = 3SE +/- 21.08, N = 3SE +/- 20.00, N = 3SE +/- 3.00, N = 3SE +/- 66.33, N = 3SE +/- 3.33, N = 3SE +/- 8.69, N = 3SE +/- 3.67, N = 31195914811115721618612776656919662951610704614810400132121040044689885

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: ReadGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE70140210280350SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.21, N = 3SE +/- 0.03, N = 3SE +/- 1.03, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 2SE +/- 0.00, N = 2SE +/- 0.30, N = 3SE +/- 0.12, N = 3SE +/- 0.41, N = 3SE +/- 0.00, N = 3SE +/- 2.42, N = 3SE +/- 0.10, N = 3164.47266.17151.60205.43227.2095.00338.07271.67143.5581.40123.00123.30159.7393.50157.87154.201. (CC) gcc options: -O2 -flto -lOpenCL

OpenBenchmarking.orgGB/s Per Watt, More Is Bettercl-mem 2017-01-13Benchmark: WriteGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.54231.08461.62692.16922.71151.111.141.101.431.380.972.051.080.810.591.122.411.340.911.07

FAHBench

OpenBenchmarking.orgNs Per Day, More Is BetterFAHBench 2.3.2GeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 9604080120160200SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.26, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 397.38108.1597.24132.79145.3849.78186.7472.7785.4558.35

clpeak

OpenBenchmarking.orgWatts, Fewer Is BetterclpeakSystem Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480306090120150Min: 90.6 / Avg: 117.27 / Max: 130.7Min: 134.8 / Avg: 156.02 / Max: 164.4Min: 75.5 / Avg: 92.25 / Max: 105.6Min: 103.2 / Avg: 114.13 / Max: 119.3Min: 99 / Avg: 118.32 / Max: 126.1Min: 65.2 / Avg: 76.8 / Max: 89.4Min: 149.6 / Avg: 155.28 / Max: 157.1Min: 170.7 / Avg: 175.55 / Max: 180.4Min: 97.8 / Avg: 103.53 / Max: 108Min: 146.4 / Avg: 167.81 / Max: 175Min: 63.2 / Avg: 106.3 / Max: 120.3Min: 61.8 / Avg: 95.67 / Max: 107Min: 51.2 / Avg: 76.94 / Max: 84.1Min: 76 / Avg: 99.48 / Max: 105.5

ViennaCL

OpenBenchmarking.orgWatts, Fewer Is BetterViennaCL 1.4.2System Power Consumption MonitorGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480306090120150Min: 56 / Avg: 80.3 / Max: 104.6Min: 134.4 / Avg: 142.1 / Max: 149.8Min: 68.9 / Avg: 80.05 / Max: 91.2Min: 91.5 / Avg: 111.55 / Max: 131.6Min: 79.6 / Avg: 86.2 / Max: 92.8Min: 130.6 / Avg: 146.5 / Max: 162.4Min: 58.3 / Avg: 105.5 / Max: 152.7Min: 98.7 / Avg: 125.25 / Max: 151.8Min: 85.8 / Avg: 108.05 / Max: 130.3Min: 109.8 / Avg: 132.4 / Max: 158.7Min: 63.1 / Avg: 86.03 / Max: 126.9Min: 62.2 / Avg: 78.77 / Max: 96.1Min: 50.6 / Avg: 72.03 / Max: 82.1Min: 72.7 / Avg: 93.3 / Max: 104.2

cl-mem

OpenBenchmarking.orgGB/s Per Watt, More Is Bettercl-mem 2017-01-13Benchmark: ReadGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.40280.80561.20841.61122.0141.171.571.351.551.791.041.281.020.720.560.741.240.991.06

LuxMark

OpenBenchmarking.orgScore Per Watt, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: HotelGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48036912158.628.8211.4212.4713.409.4713.504.288.607.324.546.767.275.376.64

SHOC Scalable HeterOgeneous Computing

OpenBenchmarking.orgGB/s Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Texture Read BandwidthGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.8281.6562.4843.3124.142.081.723.243.283.683.083.141.201.972.221.411.381.611.391.43

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Max SP FlopsGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480153045607530.9631.2440.8050.5960.5023.2765.1021.8829.2523.7223.4033.1637.7328.3736.55

FAHBench

OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterFAHBench 2.3.2GeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 9600.23180.46360.69540.92721.1590.640.610.820.950.980.551.030.360.600.49

LuxMark

OpenBenchmarking.orgScore Per Watt, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: Luxball HDRGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4802040608010057.7058.3676.3586.6069.1960.2476.0632.2555.9942.2541.0557.1252.4942.7453.02

SHOC Scalable HeterOgeneous Computing

The CUDA and OpenCL version of Vetter's Scalable HeterOgeneous Computing benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: TriadGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, 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 +/- 0.00, N = 3SE +/- 0.22, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 312.0312.2911.9612.2012.316.1212.5112.3511.9111.2410.179.684.959.6111.821. (CXX) g++ options: -O2 -lSHOCCommonMPI -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt -pthread -lmpi_cxx -lmpi

cl-mem

OpenBenchmarking.orgGB/s Per Watt, More Is Bettercl-mem 2017-01-13Benchmark: CopyGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.3240.6480.9721.2961.620.911.041.071.391.440.901.220.880.601.111.281.110.891.35

Mixbench

OpenBenchmarking.orgGFLOPS Per Watt, More Is BetterMixbench 2016-06-06Benchmark: Single PrecisionGeForce GTX 980GeForce GTX 1060GeForce GTX 780 TiRadeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480163248648045.6349.6331.1958.6070.4830.6953.37

Darktable

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.0.3Test: Server Room - Acceleration: OpenCLGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.06080.12160.18240.24320.304SE +/- 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 +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 30.210.220.190.190.180.230.180.250.220.250.150.130.170.270.13

Mixbench

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

OpenBenchmarking.orgGIOPS, More Is BetterMixbench 2016-06-06Benchmark: IntegerGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480Radeon Vega FE6001200180024003000SE +/- 0.90, N = 3SE +/- 1.33, N = 3SE +/- 0.54, N = 3SE +/- 2.60, N = 3SE +/- 3.22, N = 3SE +/- 0.97, N = 3SE +/- 1.30, N = 3SE +/- 0.88, N = 3SE +/- 0.50, N = 3SE +/- 2.57, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 31402.271717.511366.842027.962662.46614.5227.88968.931220.80835.011386.501227.32508.691140.2419.701. (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.orgus, Fewer Is BetterclpeakOpenCL Test: Kernel LatencyGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 480246810SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.36, N = 3SE +/- 0.01, N = 3SE +/- 0.27, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.39, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 34.074.334.103.554.063.823.555.564.053.886.675.945.635.825.86

System Power Consumption Monitor

OpenBenchmarking.orgWattsSystem Power Consumption MonitorPhoronix Test Suite System MonitoringGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 48060120180240300Min: 54.8 / Avg: 152.53 / Max: 254.7Min: 60.5 / Avg: 188.98 / Max: 336.2Min: 48.3 / Avg: 116.84 / Max: 190.4Min: 47.8 / Avg: 138.1 / Max: 211.9Min: 48.9 / Avg: 149.85 / Max: 266.2Min: 44.2 / Avg: 92.91 / Max: 132Min: 51.8 / Avg: 191.22 / Max: 330.8Min: 57.6 / Avg: 202.67 / Max: 336.8Min: 55.7 / Avg: 145.83 / Max: 234.2Min: 52.1 / Avg: 123.58 / Max: 210.4Min: 108.1 / Avg: 170.72 / Max: 311.4Min: 60.8 / Avg: 177.77 / Max: 346Min: 61.1 / Avg: 144.94 / Max: 235Min: 50.2 / Avg: 92.38 / Max: 145.4Min: 66.5 / Avg: 142.76 / Max: 205.4

Darktable

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.0.3Test: Masskrug - Acceleration: OpenCLGeForce GTX 980GeForce GTX 980 TiGeForce GTX 1060GeForce GTX 1070GeForce GTX 1080GeForce GTX 1050GeForce GTX 1080 TiGeForce GTX 780 TiGeForce GTX 970GeForce GTX 960Radeon R9 290Radeon R9 FuryRadeon RX 580Radeon RX 560Radeon RX 4800.06080.12160.18240.24320.304SE +/- 0.00, N = 3SE +/- 0.01, 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 +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 30.210.220.180.180.170.220.170.270.210.240.170.130.180.270.14