On 26/03/2013 6:28 PM, Ross Bencina wrote:
CoreAudio does a lot of this under the hood I think. If you're on Mac
you get relatively stabe timestamps for free.
I should correct that:
My uninformed belief is that CoreAudio does this type of clock recovery
under the hood. This is based mostly on observing the relative stability
of audio callback timestamps (compared to Windows) and some discussions
I've seen about how CoreAudio synchronises driver timestamps and
schedules client callbacks.
Some low latency audio driver stacks propagate driver interrupts to
userspace more or less directly (one way or another). The alternative is
for the driver to poll the hardware position and then call user space --
which introduces callback jitter unless you have (1) really accurate
timer callbacks, and (2) a clock recovery mechanism that can work out
when to schedule the really accurate timer callback so that it is
synchronised with the hardware position.
Ross.
--
dupswapdrop -- the music-dsp mailing list and website:
subscription info, FAQ, source code archive, list archive, book reviews, dsp
links
http://music.columbia.edu/cmc/music-dsp
http://music.columbia.edu/mailman/listinfo/music-dsp
