On 2023-12-04 12:59, Roger wrote:
On 5/12/23 00:42, Peter P. wrote:
* Paul Davis <paul@xxxxxxxxxxxxxxxxxxxxx> [2023-12-04 14:21]:
They are not synced.
No HDA nor HDMI audio devices offer any mechanism for sync.
Thanks Paul for clarifying this, much appreciated! is there a way to
tell which of the two will eventually skip samples? Or would it simply
be the
one card whose word clock is slightly late (which I can't really
measure)? Would such a drop-out be reported by jackd(2) in its
"messages"?
Does it even matter if one is input and one is output?
The only situation I can imagine sync would be important is if you are
using
separate cards to have extra channels for recording.
If Jack is syncing from the output device and has an xrun on the
input... you record that xrun. Depending on where your recording is
going, demo to show the band what it sounds like may not matter but song
to distribute you want zero xruns.
While it is true that HDA does not (the docs say it does but it
"sloppy", not a hard lock) support for sync, many mother board audio
systems split their internal audio into more than one device which
should mean they are not synced but because they are cheap they all use
the same clock anyway. However, in the case of hdmi, which is always 48k
SR, it always has it's own clock so this can be enforced.
So use your input as jack master and alsa_out for your hdmi. You want
your input to be jack master because you are recording it and you
probably don't want that going through SRC if it can be avoided. Output
doesn't matter till you do mixdown... then use it as jack master. That
is restart jack for mixing. I personally have not had success getting
HDMI to work with reasonable latency but that may not matter for most
recording.
--
Len Ovens
www.OvenWerks.net
_______________________________________________
Linux-audio-user mailing list -- linux-audio-user@xxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to linux-audio-user-leave@xxxxxxxxxxxxxxxxxxxx
