Hi Dave,
If you're concerned with power consumption, then I would definitely
recommend using a CPU/DSP with no FPU and use the Speex fixed-point
port. The exact requirement depend on what sampling rate and bit-rate
you want to use. If you tell me what mode of operation you're targeting,
then I can give you a better idea of what would do the job.
As for the FLOP/MIPS rating, I wouldn't take those too seriously as
there are so many ways to count and in practice, there are very few
applications that achieve the values in these benchmarks.
Jean-Marc
Le mercredi 08 juin 2005 ? 17:14 -0300, Dave Bardsley a ?crit
:> Jean-Marc,
>
> I'm an embedded systems designer and am in the process of selecting a
> hardware
> platform to run a Linux-Speex VoIP application. Being concerned with size
> and
> power consumption I must choose a processor with a minimal amount of
> computational power. To that end I've narrowed my search to number of
> different products that reference the popular "SiSoft Sandra(TM)"
wetstone
> bench-mark test for floating point performance.
>
> Given that hardware manufacturers shamelessly choose the most favorable
> test conditions to maximize their product's apparent performance, for
> example
> operands may be stored and fetched solely from cache, do the Speex FLOP
> requirements given in the table four of the manual correlate well to a
> stated hardware manufacturer's FLOP performance? By comparison, would
you
> share the wet/drystone performance figures for any of the test and
> development
> systems used during Speex development and test?
>
> Sincerely,
>
> Dave
>
>
> David Bardsley
> Electrical Engineer
>
> Product Development
> Pantel International Inc.
> Tel: 902 468 5998 x210
> Fax: 902 468 6577
> Web: www.InterTalk.ca
>
>
> _______________________________________________
> Speex-dev mailing list
> Speex-dev@xiph.org
> http://lists.xiph.org/mailman/listinfo/speex-dev
--
Jean-Marc Valin <Jean-Marc.Valin@USherbrooke.ca>
Universite de Sherbrooke