[linux-audio-user] XRUN/XRUN callback

[Andres Cabrera]

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

Rui Nuno Capela wrote:

>Andres Cabrera wrote:
>  
>
>>Hi all,
>>I'm wondering if anyone can explain the difference between an XRUN, and
>>an XRUN callback. Is an XRUN callback not a matter of great concern? (It
>>seems to me from reports in the qjackctl message window that no time is
>>actually lost). On my system at 11ms latency i get an XRUN callback
>>every 30 minutes or so and a true XRUN when something happens (open a
>>jack application, for instance).
>>
>>    
>>
>
>Qjackctl has two (2) codepaths to detect and report XRUNs.
>
>One is optional and klunky, and probably still featured only for
>historical reasons (!). That's this method of capturing all jackd's
>stdout/stderr output and parsing it for the literal XRUN text message. As
>it seems, this must be what you call a "true" XRUN, and appears on the
>qjackctl messages windows colored in bright red (but with no timestamp).
>
>The second one uses the JACK API, by registering a proper XRUN callback
>procedure. That is the official way in which the jackd server notifies
>registerd clients of a XRUN occurrence. This one appears on the qjackctl
>messages window as a "XRUN callback" (pink?) message, prefixed by a
>timestamp and includes a callback counter, which is also shown on the main
>display--it's the number shown between the "(" and ")" if you didn't
>noticed.
>
>Cheers.
>  
>