[Numpy-discussion] FFTS for numpy's FFTs (was: Re: Choosing between NumPy and SciPy functions)

Henry Gomersall heng at cantab.net
Tue Oct 28 05:50:43 EDT 2014

On 28/10/14 04:28, Nathaniel Smith wrote:
> - not sure if it can handle non-power-of-two problems at all, or at 
> all efficiently. (FFTPACK isn't great here either but major 
> regressions would be bad.)

 From my reading, this seems to be the biggest issue with FFTS (from my 
reading as well) and where FFTW really wins.

Having a faster algorithm used when it will work, with fallback to 
fftpack (or something else) is a good solution IMO.


More information about the NumPy-Discussion mailing list