GPUBench: summary_intro.html - File Exchange

Code covered by the BSD License

Highlights from
GPUBench

GPUBench Compare GPUs using standard numerical benchmarks in MATLAB.
gpuBench() GPUBENCH MATLAB GPU Benchmark
gpuBenchApp() A wrapper around gpuBench for use in the app
gpuBenchReport( varargin ) gpuBenchReport create an HTML report based on some GPU performance data
gpubench.addAreaShadows( axh ) addAreaShadows add shadows to any patches in the axes
gpubench.addGradientToAxes( a... addGradientToAxes Add some shading to the axes
gpubench.captureFigure( figh,... captureFigure: use print to capture a figure with over-sampling
gpubench.checkMATLABVersion() checkMATLABVersion error if the MATLAB version is not suitable
gpubench.checkPCT() checkPCT check that Parallel Computing Toolbox is installed and licensed
gpubench.cpuinfo() CPUINFO read CPU configuration
gpubench.getDataDir() getDataDir get the full path to the folder containing the GPUBench data
gpubench.gpuinfo() GPUINFO get basic details about the GPU being used
gpubench.legend( varargin ) legend Create a legend and make it match the background
gpubench.makeFigure( name ) makeFigure create a figure suitable for printing in a GPUBench report
gpubench.multiWaitbar( label,... multiWaitbar: add, remove or update an entry on the multi waitbar
gpubench.plotFLOPS( data, col... plotFLOPS plot some gpuBench data as floating-point operations per second
gpubench.saveResults( newData... saveResults add some new results to the stored data
gpubench.sizeof( type ) SIZEOF Get the size in bytes for an element of the specified type
gpubench.trimImage( filename,... trimImage Trim white-space from around an image
gpubench.version() version Return the GPUBench version as a string.
gpubench.PerformanceData Split a long name 'fcn (datatype)' into its component name and type
gpubench.SummaryData
View all files

from GPUBench by Ben Tordoff
Compare GPUs using standard numerical benchmarks in MATLAB.

summary_intro.html

The table and chart below show the peak performance of various GPUs using the same MATLAB version. Your results (if any) are highlighted in bold in the table and on the chart. All other results are from pre-stored data. The peak performance shown is usually achieved when dealing with extremely large arrays. Typical performance in day-to-day use will usually be much lower.

Since MATLAB works mostly in double precision the devices are ranked according to how well they perform double-precision calculations. Single precision results are included for completeness. For all results, higher is better.

Highlights from GPUBench

Highlights from
GPUBench