Re: [linux-audio-user] XRUN/XRUN callback

I don't know when you lose audio and when you don't. I actually asked
an almost identical question on the Jack Developer's list about 60
days ago and got a long thread of answers and info in return. If you
have some time then scan their archives and find it. I think it's and
interesting read.

Other than that all I can say is that xruns are never good and you
need to adjust until you never get them at critical times.

Good luck,
Mark

On Tue, 11 Jan 2005 12:34:02 -0500, Andres Cabrera
<andres@email-addr-hidden> wrote:
> Thanks Mark,
> I guess it takes a little experimentation to find suitable settings. I'm
> wondering though, that if the xrun reported says:
>
> delay of 5132.000 usecs exceeds estimated spare time of 5009.000;
> restart ...
>
> All I'm losing is 0.1 ms of audio? These short xruns I see reported
> aren't even audible on my good speakers, so I wouldn't mind (yes, I know
> it's not ideal) losing this audio to have a lower latency.
>
> Cheers,
> Andrés
>
> Mark Knecht wrote:
>
> >On Tue, 11 Jan 2005 12:11:54 -0500, Andres Cabrera
> ><andres@email-addr-hidden> wrote:
> >
> >
> >>Hi Rui,
> >>Thanks very much for your explanation. I remember there was a discussion
> >>about this not too long ago, but I'm wondering if the xruns matter when
> >>recording only. You could record at low latency, then raise it for
> >>mixing and mastering (like you usually do on any system). Would this be
> >>ok? Or would the recorded audio be missing small pieces?
> >>
> >>Cheers,
> >>Andres
> >>
> >>
> >
> >You cannot know for certain. If the xrun is Ardour not delivering
> >audio to the sound card then you get a click in the output audio. Not
> >a killer problem. If the xrun is Ardour not acquiring audio from the
> >sound card then the data is lost forever. Much bigger problem.
> >
> >I would recommend that you record with latencies just high enough to
> >not be overly painful, and then push them higher when doing mixing if
> >you are not interfacing with external signal paths for post
> >processing.
> >
> >- Mark
> >
> >
> >
> >
> >
>
>
Received on Tue Jan 11 20:15:17 2005