Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Comparison with clFFT (performance) #114

Closed
preda opened this issue Feb 14, 2018 · 4 comments
Closed

Comparison with clFFT (performance) #114

preda opened this issue Feb 14, 2018 · 4 comments

Comments

@preda
Copy link

preda commented Feb 14, 2018

This is a documentation request:

could you please publish a performance comparison with clFFT.
@preda
Copy link
Author

preda commented Feb 15, 2018 •
edited
Loading

I did a few preliminary comparison myself, using double, complex transform, FFT sizes 2^9, 2^18, 2^21, and in all cases clFFT was much faster.

I think improving performance, to achieve at least parity with clFFT is paramount.

@bragadeesh
Copy link
Contributor

Thanks for your comment, yes we are working to get there!

malcolmroberts pushed a commit that referenced this issue Mar 27, 2020
@feizheng10 @malcolmroberts
Improved doxygen config and auto sync versioning string(#114)
e9005ed
eidenyoshida pushed a commit to eidenyoshida/rocFFT that referenced this issue Apr 17, 2020
@feizheng10
Improved doxygen config and auto sync versioning string(ROCm#114)
def2ebb
@preda
Copy link
Author

preda commented May 10, 2021

Three years later, did rocFFT reach performance parity with clFFT?

@feizheng10
Copy link
Contributor

Yes, for some of the cases! And faster for some of non-power-of-2 cases.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@preda @malcolmroberts @bragadeesh @feizheng10