[hpsdr] FFT benchmarking

[Philip Covington]

That the benchmarks were performed by the FFTW team should tell what you
need to do.  A search on Google will give you other results performed by
truly objective parties that show the difference is much less.

Also, I am talking about the Ooura code that is being optimized by the JIT
in .NET.  The comparisons that you linked to are statically compiled so I
would expect FFTW's hand tuned routines to come out a little better.  The
joy of the JIT is that it produces code that rivals hand tuned stuff like in
FFTW.

Phil N8VB


On 10/12/06, Alberto I2PHD <i2phd at weaksignals.com> wrote:
>
> ***** High Performance Software Defined Radio Discussion List *****
>
>
>
>  Puzzled by the claim that the Ooura FFT seems to be faster, or as fast as
> the FFTW package, I looked for benchmarks on Internet, and found the
> following page (and others related)
>
> http://www.fftw.org/speed/opteron-2.2GHz-32bit/
>
>
> The graph on that page is here below. Admittedly, the measures were
> performed by the FFTW team, but I think they are honest and unbiased
> persons.
> For a 1024 points transform, the FFTW seems to 1.3 times faster than
> Ooura's. The platform was Linux. Maybe on the .NET the situation is
> reversed...
>
> 73  Alberto  I2PHD
>
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at hpsdr.org
> Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
> HPSDR web page: http://hpsdr.org
> Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20061013/f552acf0/attachment-0004.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fft.gif
Type: image/gif
Size: 53299 bytes
Desc: not available
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20061013/f552acf0/attachment-0004.gif>