AMD Radeon VII OpenCL GPU Compute Benchmarks

Benchmarks for Radeon VII launch day by Michael Larabel for a future article. Radeon tests with ROCm 2.0 on Linux 5.0 kernel.

HTML result view exported from: https://openbenchmarking.org/result/1902066-SP-VIICOMPUT16&sor.

AMD Radeon VII OpenCL GPU Compute BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionDisplay DriverRX 580RX 590RX Vega 56RX Vega 64Radeon VIIGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTXIntel Core i9-9900K @ 5.00GHz (8 Cores / 16 Threads)ASUS PRIME Z390-A (0602 BIOS)Intel Cannon Lake PCH Shared SRAM16384MBSamsung SSD 970 EVO 250GB + 2000GB SABRENTMSI AMD Radeon RX 470/480/570/570X/580/580X 8GB (1366/2000MHz)Realtek ALC1220Acer B286HKIntel I219-VUbuntu 18.105.0.0-050000rc4-generic (x86_64) 20190127GNOME Shell 3.30.1X Server 1.20.14.5 Mesa 19.0.0-devel padoka PPA (LLVM 9.0.0)OpenCL 2.1 AMD-APP (2783.0)1.1.90GCC 8.2.0ext43840x2160Sapphire AMD Radeon RX 470/480/570/570X/580/580X 8GB (1560/2100MHz)AMD Radeon RX 64 8GB (1590/800MHz)AMD Radeon RX 64 8GB (1630/945MHz)AMD Vega 20 16GB (1801/1000MHz)NVIDIA GeForce GTX 1060 6GB (1506/4006MHz)4.20.3-042003-generic (x86_64)NVIDIA 415.274.6.0OpenCL 1.2 CUDA 10.0.1321.1.84NVIDIA GeForce GTX 1070 8GB (1506/4006MHz)Zotac NVIDIA GeForce GTX 1070 Ti 8GB (1607/4006MHz)NVIDIA GeForce GTX 1080 8GB (1607/5005MHz)NVIDIA GeForce GTX 1080 Ti 11GB (1480/5508MHz)NVIDIA GeForce RTX 2060 6GB (1365/7000MHz)ASUS NVIDIA GeForce RTX 2070 8GB (1410/7000MHz)Zotac NVIDIA GeForce RTX 2080 8GB (1515/7000MHz)NVIDIA GeForce RTX 2080 Ti 11GB (1350/7000MHz)NVIDIA TITAN RTX 24GB (1350/7000MHz)OpenBenchmarking.orgCompiler Details- --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++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -vProcessor Details- Scaling Governor: intel_pstate performancePython Details- Python 2.7.15+ + Python 3.6.7Security Details- __user pointer sanitization + Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + SSB disabled via prctl and seccompOpenCL Details- 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 1080 Ti: GPU Compute Cores: 3584- RTX 2060: GPU Compute Cores: 1920- RTX 2070: GPU Compute Cores: 2304- RTX 2080: GPU Compute Cores: 2944- RTX 2080 Ti: GPU Compute Cores: 4352- TITAN RTX: GPU Compute Cores: 4608

AMD Radeon VII OpenCL GPU Compute Benchmarkslczero: OpenCLplaidml: Yes - Inference - Mobilenet - OpenCLplaidml: Yes - Inference - Inception V3 - OpenCLplaidml: Yes - Inference - ResNet 50 - OpenCLplaidml: Yes - Inference - VGG16 - OpenCLplaidml: No - Training - IMDB LSTM - OpenCLplaidml: No - Inference - Mobilenet - OpenCLplaidml: No - Inference - ResNet 50 - OpenCLplaidml: No - Inference - VGG16 - OpenCLluxmark: GPU - Luxball HDRclpeak: Global Memory Bandwidthclpeak: Single-Precision FloatRX 580RX 590RX Vega 56RX Vega 64Radeon VIIGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTX64056376.5912860.9813041413365.1815271206620268359982.7113764.0414445314268.69172212177069758575109.03195106.66178720228120.443107931710357856587120.03213119.99184835270137.6232565362124691391904166.30353197.632001233403224.77555708071345389955288.8518394.6213459219494.0812256145423914306641172371261567642571291728719658781478645120237128153755262131167761976776175068613226413916780129214313803222793423337741673341941878833391892168232910853148477913428214818886330915021356276663017298301573331791981008376189295323657968225989316835419221610454002032914636810270310797121145825024612915372734278750515352327597721547426424413365562894602252516386OpenBenchmarking.org

LeelaChessZero

Backend: OpenCL

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.20.1Backend: OpenCLTITAN RTXRTX 2080 TiGTX 1080 TiRTX 2080GTX 1080RTX 2070RTX 2060GTX 1070 TiGTX 1070Radeon VIIGTX 1060RX Vega 64RX Vega 56RX 590RX 5807001400210028003500SE +/- 28.14, N = 3SE +/- 20.98, N = 3SE +/- 25.14, N = 3SE +/- 22.18, N = 3SE +/- 32.06, N = 3SE +/- 20.57, N = 3SE +/- 13.06, N = 3SE +/- 13.26, N = 3SE +/- 9.63, N = 3SE +/- 8.64, N = 3SE +/- 6.27, N = 3SE +/- 5.51, N = 3SE +/- 2.66, N = 3SE +/- 2.57, N = 3SE +/- 2.68, N = 332753107233322591750172914841478143013918998567586836401. (CXX) g++ options: -lpthread

LeelaChessZero

Backend: OpenCL

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterLeelaChessZero 0.20.1Backend: OpenCLTITAN RTXRTX 2080 TiGTX 1080RTX 2080GTX 1070 TiRadeon VIIRTX 2070GTX 1080 TiGTX 1070RTX 2060GTX 1060RX Vega 56RX 580RX 590RX Vega 6451015202522.1115.1011.8110.139.459.389.278.807.777.575.974.383.743.622.73

LeelaChessZero

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterLeelaChessZero 0.20.1System Power Consumption MonitorGTX 1080TITAN RTXRadeon VIIGTX 1060GTX 1070 TiRX 580RX Vega 56GTX 1070RTX 2070RX 590RTX 2060RTX 2080 TiRTX 2080GTX 1080 TiRX Vega 6460120180240300Min: 40.2 / Avg: 148.14 / Max: 249.5Min: 97.7 / Avg: 148.16 / Max: 357.7Min: 56 / Avg: 148.33 / Max: 269Min: 41.9 / Avg: 150.59 / Max: 180Min: 80 / Avg: 156.39 / Max: 189.7Min: 69.3 / Avg: 171.21 / Max: 229.5Min: 47.2 / Avg: 173.2 / Max: 262.2Min: 42.6 / Avg: 184.04 / Max: 213.3Min: 75.1 / Avg: 186.61 / Max: 251.7Min: 75.2 / Avg: 188.72 / Max: 253.3Min: 69.4 / Avg: 196.01 / Max: 230Min: 48.2 / Avg: 205.76 / Max: 349.4Min: 88.3 / Avg: 222.99 / Max: 288.3Min: 96.9 / Avg: 265.17 / Max: 323.1Min: 50.2 / Avg: 313.27 / Max: 342.7

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCLTITAN RTXRTX 2080 TiRadeon VIIRTX 2080RTX 2070RTX 2060GTX 1080 TiGTX 1080GTX 1070GTX 1070 TiRX 590RX Vega 64RX Vega 56RX 580GTX 10602004006008001000SE +/- 3.45, N = 3SE +/- 2.70, N = 3SE +/- 4.52, N = 3SE +/- 1.63, N = 3SE +/- 1.07, N = 3SE +/- 1.27, N = 3SE +/- 3.70, N = 3SE +/- 5.02, N = 3SE +/- 4.06, N = 3SE +/- 3.15, N = 3SE +/- 1.00, N = 3SE +/- 9.53, N = 3SE +/- 6.73, N = 3SE +/- 1.03, N = 3SE +/- 0.87, N = 3977971904893830779774686664645599587575563552

PlaidML

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

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCLRadeon VIIRTX 2080 TiRTX 2070TITAN RTXRX Vega 64GTX 1080GTX 1070RTX 2080GTX 1070 TiRTX 2060RX Vega 56GTX 1080 TiRX 580GTX 1060RX 590369121511.5410.288.878.828.487.347.226.826.576.416.235.195.145.124.74

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorRX Vega 64Radeon VIIGTX 1070RX Vega 56GTX 1080RTX 2070RTX 2080 TiGTX 1070 TiGTX 1060RX 580TITAN RTXRTX 2060RX 590RTX 2080GTX 1080 Ti4080120160200Min: 54 / Avg: 69.27 / Max: 82.4Min: 58.1 / Avg: 78.39 / Max: 92.6Min: 42.8 / Avg: 91.98 / Max: 117.4Min: 49.6 / Avg: 92.35 / Max: 157.7Min: 43.7 / Avg: 93.48 / Max: 127.3Min: 46 / Avg: 93.68 / Max: 129.3Min: 48.4 / Avg: 94.48 / Max: 113.3Min: 48.7 / Avg: 98.18 / Max: 132.6Min: 42.1 / Avg: 107.84 / Max: 141.6Min: 75.6 / Avg: 109.4 / Max: 179Min: 52.9 / Avg: 110.8 / Max: 140.2Min: 45.5 / Avg: 121.65 / Max: 166.1Min: 82.3 / Avg: 126.56 / Max: 194.9Min: 56.3 / Avg: 130.87 / Max: 184.7Min: 91.6 / Avg: 149.17 / Max: 203.2

PlaidML

FP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCLTITAN RTXRTX 2080 TiRTX 2080GTX 1080 TiRadeon VIIRTX 2070RTX 2060GTX 1080RX Vega 64GTX 1070 TiGTX 1070RX Vega 56GTX 1060RX 590RX 58050100150200250SE +/- 1.30, N = 3SE +/- 0.48, N = 3SE +/- 0.40, N = 3SE +/- 0.58, N = 3SE +/- 0.20, N = 3SE +/- 0.18, N = 3SE +/- 0.44, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.30, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3215.00211.00168.00167.00166.30157.00134.00132.00120.03120.00117.00109.0388.8582.7176.59

PlaidML

FP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCLTITAN RTXRadeon VIIRTX 2080 TiRTX 2080RTX 2070RTX 2060GTX 1070 TiGTX 1080 TiGTX 1080GTX 1070RX Vega 56GTX 1060RX Vega 64RX 580RX 5900.2880.5760.8641.1521.441.281.241.121.081.070.920.900.820.780.760.740.670.660.520.51

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorGTX 1070 TiGTX 1060Radeon VIIRTX 2060RX 580RTX 2070RX Vega 56GTX 1070RTX 2080RX 590GTX 1080TITAN RTXRX Vega 64RTX 2080 TiGTX 1080 Ti50100150200250Min: 56.1 / Avg: 132.15 / Max: 155Min: 87.5 / Avg: 132.19 / Max: 156.7Min: 69.7 / Avg: 134.1 / Max: 228.1Min: 45 / Avg: 144.95 / Max: 190.7Min: 73.1 / Avg: 146.45 / Max: 195.4Min: 45.8 / Avg: 146.81 / Max: 206.6Min: 58.9 / Avg: 147.82 / Max: 249.5Min: 81.8 / Avg: 153.87 / Max: 183Min: 47.5 / Avg: 155.97 / Max: 224.6Min: 91.8 / Avg: 163.54 / Max: 221.5Min: 89.5 / Avg: 168.75 / Max: 202.8Min: 51.5 / Avg: 168.92 / Max: 275.4Min: 50.6 / Avg: 181.52 / Max: 296Min: 48.4 / Avg: 188.73 / Max: 275.6Min: 120.7 / Avg: 202.56 / Max: 256.6

PlaidML

FP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCLTITAN RTXRTX 2080 TiRTX 2080Radeon VIIGTX 1080 TiRTX 2070RTX 2060GTX 1080GTX 1070 TiGTX 1070RX Vega 64RX Vega 56GTX 1060RX 590RX 580100200300400500SE +/- 1.16, N = 3SE +/- 1.03, N = 3SE +/- 0.40, N = 3SE +/- 0.96, N = 3SE +/- 0.34, N = 3SE +/- 0.76, N = 3SE +/- 0.26, N = 3SE +/- 0.24, N = 3SE +/- 0.16, N = 3SE +/- 0.38, N = 3SE +/- 0.13, N = 3SE +/- 0.07, N = 3SE +/- 0.17, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3474458354353334333282264237237213195183137128

PlaidML

FP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCLRTX 2080 TiTITAN RTXRadeon VIIRTX 2070RTX 2080GTX 1070 TiRTX 2060GTX 1080 TiGTX 1070GTX 1080GTX 1060RX Vega 56RX Vega 64RX 580RX 5900.66381.32761.99142.65523.3192.952.922.752.512.392.081.961.781.761.621.491.381.240.910.89

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorGTX 1070 TiGTX 1060Radeon VIIRTX 2070GTX 1070RX Vega 56RX 580RTX 2060RTX 2080RX 590RTX 2080 TiGTX 1080TITAN RTXRX Vega 64GTX 1080 Ti60120180240300Min: 48.5 / Avg: 113.89 / Max: 160.8Min: 60.4 / Avg: 122.52 / Max: 164.5Min: 70.1 / Avg: 128.33 / Max: 243.5Min: 78.4 / Avg: 132.98 / Max: 222.4Min: 83.8 / Avg: 134.85 / Max: 189.4Min: 50.5 / Avg: 141.19 / Max: 254.3Min: 73.6 / Avg: 141.5 / Max: 187.8Min: 86.5 / Avg: 143.68 / Max: 202.9Min: 47.4 / Avg: 148.19 / Max: 240.2Min: 81.1 / Avg: 153.59 / Max: 208.6Min: 48.4 / Avg: 155.12 / Max: 292.5Min: 91.6 / Avg: 162.7 / Max: 211.1Min: 52.6 / Avg: 162.76 / Max: 301.7Min: 82.4 / Avg: 172.03 / Max: 313.6Min: 122.3 / Avg: 187.68 / Max: 270.3

PlaidML

FP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCLTITAN RTXRTX 2080 TiRadeon VIIGTX 1080 TiRTX 2080RTX 2070RTX 2060GTX 1080GTX 1070 TiGTX 1070RX Vega 64RX Vega 56GTX 1060RX 590RX 58060120180240300SE +/- 0.43, N = 3SE +/- 0.20, N = 3SE +/- 0.03, N = 3SE +/- 0.32, N = 3SE +/- 0.32, N = 3SE +/- 0.26, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3264.00250.00197.63194.00192.00179.00148.00139.00128.00126.00119.99106.6694.6264.0460.98

PlaidML

FP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCLRadeon VIITITAN RTXRTX 2080 TiRTX 2070RTX 2080RTX 2060GTX 1080 TiGTX 1070 TiGTX 1070GTX 1080GTX 1060RX Vega 56RX Vega 64RX 580RX 5900.34880.69761.04641.39521.7441.551.501.381.271.211.081.081.020.920.800.720.700.550.410.39

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorGTX 1070 TiRadeon VIIGTX 1060GTX 1070RTX 2060RTX 2070RX 580RX Vega 56RTX 2080RX 590GTX 1080TITAN RTXGTX 1080 TiRTX 2080 TiRX Vega 6460120180240300Min: 48.4 / Avg: 125.32 / Max: 169.8Min: 57.5 / Avg: 127.89 / Max: 264.7Min: 58.2 / Avg: 131.14 / Max: 167.2Min: 43.1 / Avg: 136.92 / Max: 191.2Min: 62.3 / Avg: 137.76 / Max: 208Min: 47.2 / Avg: 141.31 / Max: 234Min: 73.2 / Avg: 147.27 / Max: 187.1Min: 50.2 / Avg: 151.48 / Max: 258.1Min: 49.8 / Avg: 158.89 / Max: 252.8Min: 79.7 / Avg: 162.83 / Max: 210.1Min: 70.3 / Avg: 172.91 / Max: 216.6Min: 58.3 / Avg: 176.01 / Max: 322.5Min: 65.2 / Avg: 179.46 / Max: 287.9Min: 47.7 / Avg: 180.78 / Max: 312Min: 53.4 / Avg: 217.55 / Max: 319.8

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: IMDB LSTM - Device: OpenCLRTX 2080 TiTITAN RTXRTX 2080Radeon VIIRTX 2070RTX 2060GTX 1080 TiRX Vega 64RX Vega 56GTX 1080GTX 1070GTX 1070 TiRX 590GTX 1060RX 58050100150200250SE +/- 0.72, N = 3SE +/- 0.96, N = 3SE +/- 0.27, N = 3SE +/- 0.32, N = 3SE +/- 0.82, N = 3SE +/- 0.38, N = 3SE +/- 0.49, N = 3SE +/- 0.19, N = 3SE +/- 0.24, N = 3SE +/- 0.60, N = 3SE +/- 0.50, N = 3SE +/- 0.03, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 3SE +/- 0.25, N = 3246244216200198188187184178167156153144134130

PlaidML

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

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: No - Mode: Training - Network: IMDB LSTM - Device: OpenCLRTX 2060TITAN RTXRTX 2080RTX 2070RTX 2080 TiRadeon VIIGTX 1070 TiGTX 1080 TiGTX 1080GTX 1070RX Vega 64GTX 1060RX Vega 56RX 590RX 5800.36450.7291.09351.4581.82251.621.601.561.561.541.451.241.141.141.141.111.081.070.950.89

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorRTX 2060GTX 1070 TiGTX 1060RTX 2070GTX 1070Radeon VIIRTX 2080RX 580GTX 1080TITAN RTXRX 590RTX 2080 TiGTX 1080 TiRX Vega 56RX Vega 6450100150200250Min: 44.3 / Avg: 116.14 / Max: 162.7Min: 104.8 / Avg: 123.57 / Max: 135.6Min: 70.1 / Avg: 124.21 / Max: 145Min: 47.1 / Avg: 127.27 / Max: 169.5Min: 82.2 / Avg: 136.83 / Max: 164.9Min: 87.3 / Avg: 137.45 / Max: 192.4Min: 46.4 / Avg: 138.03 / Max: 185.4Min: 71.3 / Avg: 146.11 / Max: 185.3Min: 90.8 / Avg: 146.71 / Max: 176.8Min: 57.5 / Avg: 152.07 / Max: 231.9Min: 80.2 / Avg: 152.62 / Max: 212.4Min: 48.6 / Avg: 159.87 / Max: 223.7Min: 50.4 / Avg: 165.05 / Max: 214.8Min: 74.8 / Avg: 165.69 / Max: 271.3Min: 52.2 / Avg: 165.89 / Max: 305.6

PlaidML

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

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCLTITAN RTXRTX 2080 TiRadeon VIIRTX 2080RTX 2070GTX 1080 TiRTX 2060RX Vega 64GTX 1080GTX 1070GTX 1070 TiRX Vega 56GTX 1060RX 590RX 58030060090012001500SE +/- 3.01, N = 3SE +/- 1.39, N = 3SE +/- 3.74, N = 3SE +/- 0.31, N = 3SE +/- 1.43, N = 3SE +/- 4.51, N = 3SE +/- 1.86, N = 3SE +/- 7.40, N = 3SE +/- 0.61, N = 3SE +/- 1.02, N = 3SE +/- 0.31, N = 3SE +/- 9.70, N = 3SE +/- 3.04, N = 3SE +/- 0.24, N = 3SE +/- 0.26, N = 313361291123310451008883863835801764755720592453414

PlaidML

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

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCLRTX 2080Radeon VIIRTX 2060RTX 2070RTX 2080 TiGTX 1070 TiGTX 1060RX Vega 64GTX 1070GTX 1080 TiTITAN RTXRX Vega 56GTX 1080RX 580RX 590369121510.679.349.008.708.458.067.727.477.127.026.675.825.423.623.29

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorGTX 1060GTX 1070 TiRTX 2060RTX 2080GTX 1070RX Vega 64RX 580RTX 2070RX Vega 56GTX 1080 TiRadeon VIIRX 590GTX 1080RTX 2080 TiTITAN RTX50100150200250Min: 41.8 / Avg: 76.7 / Max: 94.4Min: 49.7 / Avg: 93.67 / Max: 125.7Min: 43.1 / Avg: 95.9 / Max: 118.8Min: 52 / Avg: 97.93 / Max: 121Min: 45.9 / Avg: 107.3 / Max: 155.1Min: 52.5 / Avg: 111.85 / Max: 223.9Min: 71.7 / Avg: 114.33 / Max: 178.6Min: 115.1 / Avg: 115.83 / Max: 116.5Min: 72.5 / Avg: 123.7 / Max: 259.6Min: 53.3 / Avg: 125.73 / Max: 165.2Min: 92.8 / Avg: 132.03 / Max: 241.7Min: 82.4 / Avg: 137.7 / Max: 205.5Min: 44.3 / Avg: 147.75 / Max: 195.1Min: 48.2 / Avg: 152.87 / Max: 274.6Min: 54 / Avg: 200.17 / Max: 274.2

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCLTITAN RTXRTX 2080 TiRadeon VIIRTX 2080RTX 2070GTX 1080 TiRTX 2060GTX 1080RX Vega 64GTX 1070 TiGTX 1070RX Vega 56GTX 1060RX 590RX 580120240360480600SE +/- 1.67, N = 3SE +/- 0.31, N = 3SE +/- 0.30, N = 3SE +/- 0.45, N = 3SE +/- 0.33, N = 3SE +/- 0.55, N = 3SE +/- 0.26, N = 3SE +/- 0.41, N = 3SE +/- 1.40, N = 3SE +/- 0.24, N = 3SE +/- 0.45, N = 3SE +/- 2.52, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3556537403400376339309292270262257228194142133

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCLRTX 2080 TiRadeon VIITITAN RTXRTX 2080RTX 2070RTX 2060GTX 1070 TiGTX 1080GTX 1080 TiGTX 1070GTX 1060RX Vega 64RX Vega 56RX 580RX 5900.69981.39962.09942.79923.4993.112.972.862.512.442.301.941.921.821.771.591.521.510.920.88

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorGTX 1060RTX 2060GTX 1070 TiRadeon VIIRX 580GTX 1070RX Vega 56GTX 1080RTX 2070RTX 2080RX 590RTX 2080 TiRX Vega 64GTX 1080 TiTITAN RTX60120180240300Min: 71.4 / Avg: 121.65 / Max: 172.6Min: 65 / Avg: 134.41 / Max: 208.9Min: 48.6 / Avg: 134.71 / Max: 168.9Min: 58.8 / Avg: 135.42 / Max: 277Min: 73 / Avg: 145.33 / Max: 194.5Min: 45.2 / Avg: 145.44 / Max: 201.2Min: 50.1 / Avg: 150.39 / Max: 271.1Min: 91.9 / Avg: 152.34 / Max: 225.3Min: 49.8 / Avg: 153.92 / Max: 242.3Min: 48.4 / Avg: 158.95 / Max: 257.5Min: 81 / Avg: 162.42 / Max: 220Min: 100.3 / Avg: 172.57 / Max: 323.3Min: 81.5 / Avg: 177.38 / Max: 338.2Min: 61.2 / Avg: 185.7 / Max: 280.5Min: 102.3 / Avg: 194.3 / Max: 332.7

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: OpenCLTITAN RTXRTX 2080 TiRadeon VIIRTX 2080RTX 2070GTX 1080 TiRTX 2060GTX 1080RX Vega 64GTX 1070 TiGTX 1070RX Vega 56GTX 1060RX 590RX 58060120180240300SE +/- 0.72, N = 3SE +/- 0.36, N = 3SE +/- 0.38, N = 3SE +/- 0.17, N = 3SE +/- 0.27, N = 3SE +/- 0.12, N = 3SE +/- 0.16, N = 3SE +/- 0.06, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.18, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3289.00273.00224.77203.00189.00189.00150.00143.00137.62131.00129.00120.4494.0868.6965.18

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second Per Watt, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: OpenCLRadeon VIITITAN RTXRTX 2080 TiRTX 2080RTX 2070RTX 2060GTX 1070 TiGTX 1080 TiGTX 1080GTX 1070GTX 1060RX Vega 64RX Vega 56RX 580RX 5900.29250.5850.87751.171.46251.301.291.281.081.030.900.890.820.790.790.650.640.640.400.37

PlaidML

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterPlaidMLSystem Power Consumption MonitorGTX 1060GTX 1070 TiGTX 1070RX 580RTX 2060Radeon VIIGTX 1080RTX 2070RX 590RX Vega 56RTX 2080RTX 2080 TiRX Vega 64TITAN RTXGTX 1080 Ti60120180240300Min: 72.7 / Avg: 144.61 / Max: 170.9Min: 47.9 / Avg: 146.58 / Max: 171.7Min: 43.1 / Avg: 162.78 / Max: 197.9Min: 73 / Avg: 163.55 / Max: 194.8Min: 44 / Avg: 167.51 / Max: 215.1Min: 58.2 / Avg: 172.91 / Max: 296.6Min: 45.1 / Avg: 180.6 / Max: 225Min: 68.7 / Avg: 183.65 / Max: 245.5Min: 86.9 / Avg: 184.98 / Max: 216.4Min: 78.5 / Avg: 188.36 / Max: 262.2Min: 50.9 / Avg: 188.45 / Max: 269.3Min: 47.6 / Avg: 212.98 / Max: 331.1Min: 52.7 / Avg: 216.48 / Max: 338.1Min: 55.4 / Avg: 223.68 / Max: 345.8Min: 127.7 / Avg: 231.45 / Max: 296.9

LuxMark

OpenCL Device: GPU - Scene: Luxball HDR

OpenBenchmarking.orgScore, More Is BetterLuxMark 3.1OpenCL Device: GPU - Scene: Luxball HDRRadeon VIITITAN RTXRTX 2080 TiRX Vega 64RX Vega 56RTX 2070RTX 2080GTX 1080 TiRTX 2060GTX 1070RX 590GTX 1070 TiRX 580GTX 1080GTX 106012K24K36K48K60KSE +/- 258.00, N = 3SE +/- 103.04, N = 3SE +/- 155.67, N = 3SE +/- 583.87, N = 3SE +/- 65.67, N = 3SE +/- 107.00, N = 3SE +/- 25.12, N = 3SE +/- 8.33, N = 3SE +/- 31.58, N = 3SE +/- 1.33, N = 3SE +/- 74.83, N = 3SE +/- 14.90, N = 3SE +/- 17.84, N = 3SE +/- 42.00, N = 3SE +/- 46.58, N = 3555704602242787325653107929532291462168221356172871722116776152711380312256

LuxMark

OpenCL Device: GPU - Scene: Luxball HDR

OpenBenchmarking.orgScore Per Watt, More Is BetterLuxMark 3.1OpenCL Device: GPU - Scene: Luxball HDRRadeon VIITITAN RTXRTX 2080 TiRTX 2070RTX 2080RX Vega 56GTX 1070 TiRTX 2060RX Vega 64GTX 1070GTX 1080 TiGTX 1060RX 580GTX 1080RX 5904080120160200188.36137.54133.42131.09123.99120.58114.16110.8098.8998.4889.3485.5380.5379.4278.90

LuxMark

System Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterLuxMark 3.1System Power Consumption MonitorGTX 1060GTX 1070 TiGTX 1080GTX 1070RX 580RTX 2060RX 590RTX 2070RTX 2080GTX 1080 TiRX Vega 56Radeon VIIRTX 2080 TiRX Vega 64TITAN RTX60120180240300Min: 111.3 / Avg: 143.3 / Max: 145.4Min: 47.6 / Avg: 146.95 / Max: 149.7Min: 89.3 / Avg: 173.79 / Max: 175.8Min: 82.6 / Avg: 175.54 / Max: 179.2Min: 71.5 / Avg: 189.63 / Max: 196.7Min: 44 / Avg: 192.74 / Max: 197.4Min: 82.1 / Avg: 218.27 / Max: 228.9Min: 46 / Avg: 225.28 / Max: 233.2Min: 48.9 / Avg: 235.08 / Max: 244.1Min: 161.6 / Avg: 242.69 / Max: 248.1Min: 60.2 / Avg: 257.74 / Max: 269.1Min: 87.3 / Avg: 295.02 / Max: 302.2Min: 47.7 / Avg: 320.7 / Max: 332Min: 49.5 / Avg: 329.31 / Max: 338.6Min: 50.6 / Avg: 334.61 / Max: 351.7

clpeak

OpenCL Test: Global Memory Bandwidth

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthRadeon VIITITAN RTXRTX 2080 TiRTX 2080RTX 2070RX Vega 64GTX 1080 TiRX Vega 56RTX 2060GTX 1080RX 590RX 580GTX 1070 TiGTX 1070GTX 10602004006008001000SE +/- 0.10, N = 3SE +/- 1.70, N = 3SE +/- 0.48, N = 3SE +/- 0.37, N = 3SE +/- 2.69, N = 3SE +/- 0.05, N = 3SE +/- 0.68, N = 3SE +/- 0.01, N = 3SE +/- 0.19, N = 3SE +/- 0.05, N = 3SE +/- 1.82, N = 3SE +/- 1.46, N = 3SE +/- 0.22, N = 3SE +/- 0.02, N = 3SE +/- 0.95, N = 38075255053683653623293172762222172061971961451. (CXX) g++ options: -O3 -rdynamic -lOpenCL

clpeak

OpenCL Test: Single-Precision Float

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatTITAN RTXRTX 2080 TiRadeon VIIRX Vega 64GTX 1080 TiRX Vega 56RTX 2080RTX 2070GTX 1080RX 590GTX 1070 TiRTX 2060RX 580GTX 1070GTX 10604K8K12K16K20KSE +/- 688.69, N = 3SE +/- 1065.18, N = 3SE +/- 1.83, N = 3SE +/- 41.13, N = 3SE +/- 782.31, N = 3SE +/- 38.50, N = 3SE +/- 732.21, N = 3SE +/- 495.99, N = 3SE +/- 396.02, N = 3SE +/- 15.75, N = 3SE +/- 14.43, N = 3SE +/- 657.16, N = 3SE +/- 1.21, N = 3SE +/- 393.27, N = 3SE +/- 3.56, N = 316386153521345312469108531035710270796879347069677666306202587842391. (CXX) g++ options: -O3 -rdynamic -lOpenCL

GPU Temperature Monitor

Phoronix Test Suite System Monitoring

OpenBenchmarking.orgCelsiusGPU Temperature MonitorPhoronix Test Suite System MonitoringRX 580GTX 1070 TiRX 590RX Vega 56Radeon VIIRX Vega 64RTX 2060RTX 2070GTX 1080GTX 1070RTX 2080 TiTITAN RTXGTX 1080 TiRTX 208020406080100Min: 26 / Avg: 39.71 / Max: 61Min: 38 / Avg: 45.88 / Max: 53Min: 35 / Avg: 46.18 / Max: 73Min: 28 / Avg: 47.45 / Max: 74Min: 36 / Avg: 52.37 / Max: 101Min: 45 / Avg: 52.93 / Max: 65Min: 37 / Avg: 55.81 / Max: 70Min: 32 / Avg: 56.83 / Max: 74Min: 31 / Avg: 60.14 / Max: 74Min: 32 / Avg: 61.66 / Max: 76Min: 43 / Avg: 61.88 / Max: 77Min: 33 / Avg: 62.56 / Max: 79Min: 39 / Avg: 66.36 / Max: 82Min: 48 / Avg: 67.06 / Max: 83

System Power Consumption Monitor

Phoronix Test Suite System Monitoring

OpenBenchmarking.orgWattsSystem Power Consumption MonitorPhoronix Test Suite System MonitoringGTX 1070 TiRadeon VIIRX Vega 56RX 580RX Vega 64RX 590GTX 1070GTX 1080RTX 2060RTX 2070RTX 2080GTX 1080 TiRTX 2080 TiTITAN RTX70140210280350Min: 47.2 / Avg: 123.43 / Max: 255.8Min: 55.7 / Avg: 125.32 / Max: 337.5Min: 47.1 / Avg: 125.63 / Max: 271.3Min: 69 / Avg: 125.65 / Max: 229.5Min: 49.7 / Avg: 129.09 / Max: 305.4Min: 75 / Avg: 138.2 / Max: 253.3Min: 42.3 / Avg: 142.19 / Max: 280.5Min: 40.2 / Avg: 143.79 / Max: 318.2Min: 42.5 / Avg: 144.37 / Max: 300.1Min: 43.9 / Avg: 160.13 / Max: 317.6Min: 46 / Avg: 176.62 / Max: 334.5Min: 47.2 / Avg: 198.85 / Max: 370.3Min: 47.2 / Avg: 220.83 / Max: 367.7Min: 49.4 / Avg: 228.34 / Max: 363.1

Phoronix Test Suite v10.8.4