berathebrain

Reduction kernel sample not working with float4

Discussion created by berathebrain on Jan 10, 2010
Latest reply on Jan 18, 2010 by genaganna

I have tried to change the reduction kernel sample to work with float4 numbers. It turns out that the kernel produces some minor errors when adding numbers. The reference CPU reduction is different from OpenCL reduction. It differs by 0.01% but if I have like a million values then the error is pretty huge.

Can someone confirm this, or am I doing something wrong.

With float works but with float4 it doesn't;

Tried the CPU context and GPU.

I have Radeon 4870, Windows 7 64bit.

Outcomes