OpenCL Benchmarking Radeon vs. Intel Skylake

Intel Core i5-6600K tests with Beignet compared to some open-source Radeon dGPU results.

HTML result view exported from: https://openbenchmarking.org/result/1602222-GA-PEAKING9549&rdt.

OpenCL Benchmarking Radeon vs. Intel SkylakeProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 6950Radeon HD 7950Core i5 6600K SKLIntel Core i7-5960X @ 3.50GHz (16 Cores)Gigabyte X99-UD4-CFIntel Xeon E7 v3/Xeon16384MB2 x 120GB SanDisk SDSSDA12 + 120GB Samsung SSD 850XFX AMD Radeon R9 285/380 2048MBRealtek ALC1150DELL P2415QIntel ConnectionUbuntu 15.104.5.0-rc2amdgpu-pp+ (x86_64) 20160201UnityX Server 1.17.2amdgpu 1.0.994.1 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4GCC 5.2.1 20151010ext43840x2160XFX AMD Radeon R9 290 4096MBradeon 7.6.99MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MBSapphire AMD Radeon HD 6870 1024MB3.3 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4Sapphire AMD Radeon HD 6950 2048MB4.1 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4XFX AMD Radeon HD 7950/8950 OEM / R9 280 3072MBIntel Core i5-6600K @ 3.90GHz (4 Cores)MSI Z170A GAMING PRO (MS-7984) v1.0Intel Sky Lake15360MB256GB TS256GSSD370SIntel Sky Lake (1150MHz)4.5.0-999-generic (x86_64) 20160218intel 2.99.9173.3 Mesa 11.2.0-devel (padoka PPA)GCC 5.2.1 20151010 + Clang 3.7.1-1ubuntu3~gd~w + LLVM 3.7.1OpenBenchmarking.orgKernel Details- Radeon R9 285, Radeon R9 290, Radeon R7 370, Radeon HD 6870, Radeon HD 6950, Radeon HD 7950: amdgpu.powerplay=1Compiler Details- Radeon R9 285, Radeon R9 290, Radeon R7 370, Radeon HD 6870, Radeon HD 6950, Radeon HD 7950: --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 -v Processor Details- Radeon R9 285: Scaling Governor: intel_pstate performance- Radeon R9 290: Scaling Governor: intel_pstate performance- Radeon R7 370: Scaling Governor: intel_pstate performance- Radeon HD 6870: Scaling Governor: intel_pstate performance- Radeon HD 6950: Scaling Governor: intel_pstate performance- Radeon HD 7950: Scaling Governor: intel_pstate performance- Core i5 6600K SKL: Scaling Governor: intel_pstate powersaveGraphics Details- Radeon R9 285: GLAMOR- Radeon R9 290: GLAMOR- Radeon R7 370: GLAMOR- Radeon HD 6870: EXA- Radeon HD 6950: EXA- Radeon HD 7950: GLAMOR

OpenCL Benchmarking Radeon vs. Intel Skylakeshoc: OpenCL - Bus Speed Downloadshoc: OpenCL - Bus Speed Readbackclpeak: Kernel Latencyclpeak: Integer Compute INTclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Transfer Bandwidth enqueueReadBufferclpeak: Transfer Bandwidth enqueueWriteBufferclpeak: Global Memory BandwidthRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 6950Radeon HD 7950Core i5 6600K SKL10.145.94143.39650.803102.60204.7510.0914.827.723.02118.711578.014608.55596.559.8212.3210.215.05110.24418.742031.91131.4910.4115.77145.85182.44360.923.585.2610.215.59105.54491.622345.12547.0310.4015.7225.8233.9923.71389.9714.1726.7631.55OpenBenchmarking.org

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Bus Speed Download

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Bus Speed DownloadRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 7950Core i5 6600K SKL612182430SE +/- 0.04, N = 3SE +/- 0.17, N = 6SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.10, N = 310.147.7210.2110.2125.82-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc1. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Bus Speed Readback

OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Bus Speed ReadbackRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 7950Core i5 6600K SKL816243240SE +/- 0.01, N = 3SE +/- 0.04, N = 6SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.25, N = 35.943.025.055.5933.99-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc-lSHOCCommonMPI -pthread -lmpi_cxx -lmpi -ldl -lhwloc1. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

clpeak

OpenCL Test: Kernel Latency

OpenBenchmarking.orgus, More Is BetterclpeakOpenCL Test: Kernel LatencyRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 7950Core i5 6600K SKL306090120150SE +/- 14.32, N = 6SE +/- 2.08, N = 6SE +/- 3.85, N = 6SE +/- 8.80, N = 6SE +/- 1.78, N = 6SE +/- 0.06, N = 3143.39118.71110.24145.85105.5423.71

clpeak

OpenCL Test: Integer Compute INT

OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 795030060090012001500SE +/- 0.05, N = 3SE +/- 0.49, N = 3SE +/- 0.01, N = 3SE +/- 0.16, N = 3SE +/- 7.09, N = 6650.801578.01418.74182.44491.62

clpeak

OpenCL Test: Single-Precision Float

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 7950Core i5 6600K SKL10002000300040005000SE +/- 1.52, N = 3SE +/- 2.13, N = 3SE +/- 0.22, N = 3SE +/- 0.08, N = 3SE +/- 47.92, N = 6SE +/- 0.05, N = 33102.604608.552031.91360.922345.12389.97

clpeak

OpenCL Test: Double-Precision Double

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 7950130260390520650SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.00, N = 3SE +/- 16.28, N = 6204.75596.55131.49547.03

clpeak

OpenCL Test: Transfer Bandwidth enqueueReadBuffer

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueReadBufferRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 7950Core i5 6600K SKL48121620SE +/- 0.30, N = 6SE +/- 0.25, N = 6SE +/- 0.02, N = 3SE +/- 0.10, N = 6SE +/- 0.03, N = 3SE +/- 0.08, N = 310.099.8210.413.5810.4014.17

clpeak

OpenCL Test: Transfer Bandwidth enqueueWriteBuffer

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferRadeon R9 285Radeon R9 290Radeon R7 370Radeon HD 6870Radeon HD 7950Core i5 6600K SKL612182430SE +/- 0.59, N = 6SE +/- 0.06, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 314.8212.3215.775.2615.7226.76

clpeak

OpenCL Test: Global Memory Bandwidth

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory BandwidthCore i5 6600K SKL714212835SE +/- 0.02, N = 331.55

Phoronix Test Suite v10.8.5