[asterisk-users] Re: Choice of soundfile format

Martin Joseph ast at stillnewt.org
Wed Oct 25 09:49:31 MST 2006


On 2006-10-25 08:14:56 -0700, Matthew Rubenstein <email at mattruby.com> said:

> 	What's the native soundfile format for SIP?
??? I think you might need to do some research (the above is a nonsense 
question I think).
>  Any idea which soundfile
> takes the least CPU for mixing together in conferences?
Probably slinear for prerecorded stuff, as that will only have to 
transcoded one way (the direction of your output devices). Unless all 
your devices are using ulaw or alaw, then the previous posters advice 
is correct.
> 
> 	How about whether the CPU load for conferencing native data is
> greater/less than the CPU load for transcoding non-native data that is
> "CPU lighter" in the conference mixing phase?
Transcoding is a bigger hit then mixing as i understand it.

If all the conference members are using ulaw for example, then having 
the playback material encoded in ulaw is the big winner.  If there are 
different codecs connecting, then there is a lot of 
decoding/mixing/recoding that will need to occur.

Marty




More information about the asterisk-users mailing list