[Asterisk-Users] Hardware based DSP

[Miroslav Nachev]

   Dear Shahed,

   If you use hardware DSP for encoding and decoding you will need of
less power Host CPU. There are no difference for what you will use
these codecs - I mean conference or single conversation, because for
conference all codecs are converted to G.711, then mixed and then back
to the original codec for each party of conference group. So, if you
would like to to conference with software coding you will need of very
power computer.


   Best Regards,
   Miroslav Nachev

S> Thank you all for your explanations related to my question.

S> I have one follow-up question though.

S> When I said that all dsp related stuff has to be handled
S> by software within asterisk, I was thinking of
S> conferencing at the time.

S> I mean in order to be able to conference a sip session with
S> a PSTN call, it would have to be handled by software, even
S> if both the channels had hardware dsp capabilities. Right ??

S> If you are dealing with just a single channel, then the driver
S> may handle codec/echo cancelation stuff with hardware help (??)

S> As an aside, what is the best way to go about learnig about
S> the aritecture of asterisk, other than "using the source" ??

S> The Wiki pages are great, but I have not (yet) found any info
S> about asterisks architecture itself (in depth that is).

S> Some linked websites / blogs provide good info on some topics,
S> but is there a good high level design doc available anywhere ?

S> It would stop people like me asking so many basic questions in
S> the -dev list.

S> Thanks
S> Shahed
S> _______________________________________________
S> Asterisk-Users mailing list
S> Asterisk-Users at lists.digium.com
S> http://lists.digium.com/mailman/listinfo/asterisk-users
S> To UNSUBSCRIBE or update options visit:
S>    http://lists.digium.com/mailman/listinfo/asterisk-users