cancel
Showing results for 
Search instead for 
Did you mean: 

Archives Discussions

riza_guntur
Journeyman III

Is achieving 700Gflops on Brook+ Optimize Mat Mult sample is normal?

I wonder, does it normal or my 4850 has another fever?

I confuse, I want to benchmark the obvious (I mean about GPU is faster than normal quad core CPU setup) in various algorithm but I found the result about 700Gflops

Ow man... hard time these days

0 Likes
2 Replies
Russian
Journeyman III

Is achieving 700Gflops on Brook+ Optimize Mat Mult sample is normal?

Originally posted by: riza.guntur I wonder, does it normal or my 4850 has another fever?

 

   



 

Can you please show an example?

 

I've found, that IL compiler optimizer is very smart.

 

Thanks,

Dmitry

0 Likes
riza_guntur
Journeyman III

Is achieving 700Gflops on Brook+ Optimize Mat Mult sample is normal?

I made a program that calculate the height of a crossing section between two line

After that I want to get the smallest height along with its clusterID (reduction, comparing x then if true copying xyzw to output)

Then I want to get the biggest height on smallest I've found before (reduction, comparing x then if true copying xyzw to output)

Then if the clusterID is in the same class as expected value, then I increase the number of correct items

After experimenting with same test case, I get different results. I don't do any randomization here.

I do a lot of copy operations. What I expect from copying is to get exact same value (compared to int to float conversion) to make another comparison, mostly equivalency. Funny thing is, equivalency comparison seems to take biggest hit, not just float, int too. BUT I found CPU backend produce consistent result.

0 Likes