> < ^ Date: Wed, 14 Aug 2002 14:06:40 -0400
< ^ From: Markus Pueschel <pueschel@earthlink.net >
< ^ Subject: Re: 32-point dct. (fwd)

Dear Srini,

I suggest you get SPIRAL
http://www.ece.cmu.edu/~spiral
With SPIRAL you can generate many different algorithms for
the DCT 32, you can automatically translate them into C or
Fortran code, and you can search for the best implementation
for your computing platform.

I hope this helps,
Markus

--
Markus Pueschel
Research Faculty
Dept. of Electrical and Computer Engineering
Carnegie Mellon University
http://www.ece.cmu.edu/~pueschel

On Wednesday 14 August 2002 05:21, Joachim Neubueser wrote:
> Dear GAP Forum members,
>
> We received the appended question, being sent to the address
> 'gap'. This is certainly the wrong place, being reserved for
> confirmation of installation and general questions about the
> availability of GAP.
>
> However in this case, at least to my knowledge, we do not have the
> wanted code in GAP or one of its packages, and again to my knowledge,
> none of the 'gap-trouble shooters' works in this field. Therefore I
> forward this question to the Forum in the hope that someone among the
> GAP users may be able to help.
>
> Thanks for your attention Joachim Neubueser
>
>
> ----- Forwarded message from Srinivas Patil -----
>
> Date: Mon, 12 Aug 2002 17:13:20 +0530
> From: Srinivas Patil <srinivas.patil@analog.com>
> Organization: Analog Devices Inc.
> To: gap@dcs.st-and.ac.uk
> Subject: 32-point dct.
>
> Hi,
> is there any simple code to get matrices of efficient implementation of
> 32-point discrete cosine transform.
>
> waiting for response,
> thanks
> srini
>
>
> ----- End of forwarded message from Srinivas Patil -----


> < [top]