[pulseaudio-discuss] USB microphone works, monitoring doesn't
gmane at colin.guthr.ie
Tue Aug 11 01:52:34 PDT 2009
'Twas brillig, and Ng Oon-Ee at 11/08/09 02:22 did gyre and gimble:
> Sorry for butting in on this, but just a comment, isn't that sort of
> hacker-ish? How would different rates and such be dealt with (perhaps
> with --rate)? Is this the recommended way of doing this?
different rates could be handled with --rate yes, but the real problem
is the deviation of clocks. i.e. even if both source and sink are 44.1,
they wont be perfectly 44.1. Either the sink will be starved of data or
it will be overwhelmed. There needs to be some kind of active link
in-between that can intelligently resample the data appropriately to
ensure that doesn't happen.
As Lennart said, there is code in module-compbine that does this kind of
resampling so could form the basis for a properly implementation.
But for many applications the hacky method works OK.
Tribalogic Limited [http://www.tribalogic.net/]
Mandriva Linux Contributor [http://www.mandriva.com/]
PulseAudio Hacker [http://www.pulseaudio.org/]
Trac Hacker [http://trac.edgewall.org/]
More information about the pulseaudio-discuss