memory bandwidth on 5800

Discussion created by BarnacleJunior on Jan 8, 2010

What kind of memory bandwidth should we expect on 5800?  I'm doing D3D11 dev primarily (OpenCL has crippling perf problems for me) and I'm only seeing 42GB/s read/write.  This is for a shader of 512 threads where each thread reads 8 DWORDs in sequence from a structured UAV.  Is there some better access pattern to use?  Is it bad to write to a UAV you just read from - should I try ping ponging?  The test array is 16M elements and I'm running the loop 10000 times for that 42GB/s figure.