OpenCL Radeon Open-Source vs. Catalyst, hd7970

Radeon R600g and RadeonSI Gallium3D Clover OpenCL benchmarks with clpeak using Linux 4.5 + Mesa 11.2-devel. Compared to Catalyst. Benchmark results by Michael Larabel of Phoronix.com for a future article. hd7970: Intel Core i5-3330 testing with a ASRock H61 Pro BTC and Intel HD 2500 3072MB on SUSE LINUX 42.1 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/1602026-GA-OPENCLAMD14,1602036-HA-HD797015261&sro&gru.

OpenCL Radeon Open-Source vs. Catalyst, hd7970ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionOpen-Source: HD 6870Open-Source: HD 7950Open-Source: R9 285Open-Source: R9 290Open-Source: R7 370Catalyst: HD 6870Catalyst: HD 7950Catalyst: R9 285Catalyst: R9 290Catalyst: R7 370hd7970Intel Core i7-5960X @ 3.50GHz (16 Cores)Gigabyte X99-UD4-CFIntel Xeon E7 v3/Xeon16384MB2 x 120GB SanDisk SDSSDA12 + 120GB Samsung SSD 850Sapphire AMD Radeon HD 6870 1024MBRealtek ALC1150DELL P2415QIntel ConnectionUbuntu 15.104.5.0-rc2amdgpu-pp+ (x86_64) 20160201UnityX Server 1.17.2radeon 7.6.993.3 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4GCC 5.2.1 20151010ext43840x2160XFX AMD Radeon HD 7950/8950 OEM / R9 280 3072MB4.1 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4XFX AMD Radeon R9 285/380 2048MBamdgpu 1.0.99XFX AMD Radeon R9 290 4096MBradeon 7.6.99MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MBSapphire AMD Radeon HD 6800 1024MB4.2.0-27-generic (x86_64)fglrx 15.20.34.5.13399XFX AMD Radeon HD 7900 3072MBSupported device 6939 2048MBSupported device 67B1 4096MBMSI AMD Radeon R7 370 4096MBIntel Core i5-3330 @ 3.20GHz (4 Cores)ASRock H61 Pro BTCIntel Xeon E3-1200 v2/3rd8192MB65GBIntel HD 2500 3072MB (1050MHz)VIA VT1705H225WRealtek RTL8111/8168/8411SUSE LINUX 42.14.5.0-rc1-3-default (x86_64)KDE Frameworks 5X Server 1.18.99.1intel 2.99.9174.1 Mesa 11.2.0-devel (git-43a401a pontostroy:X11) Gallium 0.4GCC 4.8.5 + Clang 3.9.0 + LLVM 3.9.0svnxfs1920x1080OpenBenchmarking.orgKernel Details- Open-Source: HD 6870, Open-Source: HD 7950, Open-Source: R9 285, Open-Source: R9 290, Open-Source: R7 370: amdgpu.powerplay=1Compiler Details- Open-Source: HD 6870, Open-Source: HD 7950, Open-Source: R9 285, Open-Source: R9 290, Open-Source: R7 370, Catalyst: HD 6870, Catalyst: HD 7950, Catalyst: R9 285, Catalyst: R9 290, Catalyst: R7 370: --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- Open-Source: HD 6870: Scaling Governor: intel_pstate performance- Open-Source: HD 7950: Scaling Governor: intel_pstate performance- Open-Source: R9 285: Scaling Governor: intel_pstate performance- Open-Source: R9 290: Scaling Governor: intel_pstate performance- Open-Source: R7 370: Scaling Governor: intel_pstate performance- Catalyst: HD 6870: Scaling Governor: intel_pstate performance- Catalyst: HD 7950: Scaling Governor: intel_pstate performance- Catalyst: R9 285: Scaling Governor: intel_pstate performance- Catalyst: R9 290: Scaling Governor: intel_pstate performance- Catalyst: R7 370: Scaling Governor: intel_pstate performance- hd7970: Scaling Governor: intel_pstate powersaveGraphics Details- Open-Source: HD 6870: EXA- Open-Source: HD 7950: GLAMOR- Open-Source: R9 285: GLAMOR- Open-Source: R9 290: GLAMOR- Open-Source: R7 370: GLAMOREnvironment Details- Catalyst: HD 6870, Catalyst: HD 7950, Catalyst: R9 285, Catalyst: R9 290, Catalyst: R7 370: LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib/x86_64-linux-gnu/dri:/usr/lib/dri:/usr/lib32/fglrx/dri:/usr/lib/i386-linux-gnu/dri

OpenCL Radeon Open-Source vs. Catalyst, hd7970clpeak: Transfer Bandwidth enqueueReadBufferclpeak: Transfer Bandwidth enqueueWriteBufferclpeak: Global Memory Bandwidthclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Integer Compute INTclpeak: Kernel LatencyOpen-Source: HD 6870Open-Source: HD 7950Open-Source: R9 285Open-Source: R9 290Open-Source: R7 370Catalyst: HD 6870Catalyst: HD 7950Catalyst: R9 285Catalyst: R9 290Catalyst: R7 370hd79703.585.26360.92182.44145.8510.4015.722345.12547.03491.62105.5410.0914.823102.60204.75650.80143.399.8212.324608.55596.551578.01118.7110.4115.772031.91131.49418.74110.248.809.49398.29199.3824.9010.4713.882292.54536.84516.7434.039.0013.213177.62202.71647.0826.7810.6414.814721.18598.811591.7437.7710.6714.972095.91128.73415.6735.425.837.99168.854149.221058.38847.7069.45OpenBenchmarking.org

clpeak

OpenCL Test: Transfer Bandwidth enqueueReadBuffer

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueReadBufferCatalyst: HD 6870Catalyst: HD 7950Catalyst: R7 370Catalyst: R9 285Catalyst: R9 290Open-Source: HD 6870Open-Source: HD 7950Open-Source: R7 370Open-Source: R9 285Open-Source: R9 290hd79703691215SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 6SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.30, N = 6SE +/- 0.25, N = 6SE +/- 0.01, N = 38.8010.4710.679.0010.643.5810.4010.4110.099.825.83

clpeak

OpenCL Test: Transfer Bandwidth enqueueWriteBuffer

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferCatalyst: HD 6870Catalyst: HD 7950Catalyst: R7 370Catalyst: R9 285Catalyst: R9 290Open-Source: HD 6870Open-Source: HD 7950Open-Source: R7 370Open-Source: R9 285Open-Source: R9 290hd797048121620SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.59, N = 6SE +/- 0.06, N = 3SE +/- 0.01, N = 39.4913.8814.9713.2114.815.2615.7215.7714.8212.327.99

clpeak

OpenCL Test: Global Memory Bandwidth

OpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Global Memory Bandwidthhd79704080120160200SE +/- 0.73, N = 3168.85

clpeak

OpenCL Test: Single-Precision Float

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatCatalyst: HD 6870Catalyst: HD 7950Catalyst: R7 370Catalyst: R9 285Catalyst: R9 290Open-Source: HD 6870Open-Source: HD 7950Open-Source: R7 370Open-Source: R9 285Open-Source: R9 290hd797010002000300040005000SE +/- 0.00, N = 3SE +/- 44.66, N = 6SE +/- 2.25, N = 3SE +/- 0.24, N = 3SE +/- 0.95, N = 3SE +/- 0.08, N = 3SE +/- 47.92, N = 6SE +/- 0.22, N = 3SE +/- 1.52, N = 3SE +/- 2.13, N = 3SE +/- 2.09, N = 3398.292292.542095.913177.624721.18360.922345.122031.913102.604608.554149.22

clpeak

OpenCL Test: Double-Precision Double

OpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleCatalyst: HD 7950Catalyst: R7 370Catalyst: R9 285Catalyst: R9 290Open-Source: HD 7950Open-Source: R7 370Open-Source: R9 285Open-Source: R9 290hd79702004006008001000SE +/- 11.37, N = 6SE +/- 0.36, N = 3SE +/- 0.22, N = 3SE +/- 0.01, N = 3SE +/- 16.28, N = 6SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.12, N = 3536.84128.73202.71598.81547.03131.49204.75596.551058.38

clpeak

OpenCL Test: Integer Compute INT

OpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTCatalyst: HD 6870Catalyst: HD 7950Catalyst: R7 370Catalyst: R9 285Catalyst: R9 290Open-Source: HD 6870Open-Source: HD 7950Open-Source: R7 370Open-Source: R9 285Open-Source: R9 290hd797030060090012001500SE +/- 0.02, N = 3SE +/- 21.03, N = 6SE +/- 3.65, N = 3SE +/- 1.77, N = 3SE +/- 0.05, N = 3SE +/- 0.16, N = 3SE +/- 7.09, N = 6SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.49, N = 3SE +/- 0.06, N = 3199.38516.74415.67647.081591.74182.44491.62418.74650.801578.01847.70

clpeak

OpenCL Test: Kernel Latency

OpenBenchmarking.orgus, Fewer Is BetterclpeakOpenCL Test: Kernel LatencyCatalyst: HD 6870Catalyst: HD 7950Catalyst: R7 370Catalyst: R9 285Catalyst: R9 290Open-Source: HD 6870Open-Source: HD 7950Open-Source: R7 370Open-Source: R9 285Open-Source: R9 290hd7970306090120150SE +/- 0.28, N = 3SE +/- 0.68, N = 3SE +/- 0.52, N = 5SE +/- 0.38, N = 3SE +/- 2.12, N = 6SE +/- 8.80, N = 6SE +/- 1.78, N = 6SE +/- 3.85, N = 6SE +/- 14.32, N = 6SE +/- 2.08, N = 6SE +/- 0.81, N = 324.9034.0335.4226.7837.77145.85105.54110.24143.39118.7169.45


Phoronix Test Suite v10.8.4