NOT ON THE CURRENT EDITION
This blip is not on the current edition of the Radar. If it was on one of the last few editions it is likely that it is still relevant. If the blip is older it might no longer be relevant and our assessment might be different today. Unfortunately, we simply don't have the bandwidth to continuously review blips from previous editions of the RadarUnderstand more
Published: Aug 31, 2010
Last Updated: Mar 16, 2012
The use of GPUs for computing offers efficiencies and performance for certain classes of problems that would be prohibitively expensive for more traditional hardware. Problems that fit Single Instruction Multiple Data (SIMD) processing models can gain significant advantages at the cost of difficult learning curves using specialized APIs. OpenCL, CUDA from NVidia and DirectCompute from Microsoft offer developers access to General-purpose computing on graphics processing units (GPGPU).