Two things: data parallel code is best, and you need to have a large enough amou...

		scott_s on Aug 23, 2009 \| parent \| context \| favorite \| on: Intel acquires two multicore software firms, Cilk ... Two things: data parallel code is best, and you need to have a large enough amount of data to amortize the high cost of transferring data to the GPU. "Large enough" depends on your data access patterns. A good place to start: http://gpgpu.org/developer

Thanks, that's the sort of thing I was looking for.