Re: [Jack-Devel] xruns on NVidia HMDI output

PrevNext  Index
DateFri, 23 Sep 2011 00:49:36 -0400
From Tim E. Real <[hidden] at rogers dot com>
To[hidden] at lists dot jackaudio dot org
In-Reply-ToJoshua N Pritikin Re: [Jack-Devel] xruns on NVidia HMDI output
Follow-UpTim E. Real Re: [Jack-Devel] xruns on NVidia HMDI output
Follow-UpJeremy Jongepier Re: [Jack-Devel] xruns on NVidia HMDI output
On September 22, 2011 09:58:13 pm you wrote:
> On Thu, Sep 22, 2011 at 07:17:44PM -0400, Tim E. Real wrote:
> > Especially now that I know how to use Jack's -S (synchronous) setting.
> > You can try that setting if you get a lot of xruns or "jack activation
> > count" errors.
>
> Sync does the trick. That is the only change I made and it works great
> now.
>
> What does sync do?

Was hoping the experts would answer...
I saw the official explanation somewhere yesterday.

For now here's a thread with some info.
http://comments.gmane.org/gmane.comp.audio.ardour.user/11471

Basically Jack-2 normally doesn't wait for clients now. But by using 
 the -S switch it can, just like Jack-1.
The drawback is that like Jack-1, there will be a 1-cycle latency. 
I think using -S makes Jack-2 behave almost exactly like Jack-1.
Hope that's right...

Forgot to mention: I was curious about that audio.conf security limits 
 file because the generic kernel did not install that file for me.
The -lowlatency kernel installed that file for me. 
But curiously, the first time installing the OS, it was called audio.conf.
Then after some botches, and tearing down and re-installing the OS,
 it was called audio.conf.disabled, and I had to change that. Strange.

Tim.
PrevNext  Index

1316755168.9266_0.ltw:2,a <201109230049.37500.termtech at rogers dot com>