[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Denemo-devel] SOC - Some remarks 1

From: Dominic Sacré
Subject: Re: [Denemo-devel] SOC - Some remarks 1
Date: Thu, 4 Aug 2011 21:37:40 +0200
User-agent: KMail/1.13.5 (Linux/2.6.32-33-generic; KDE/4.4.5; i686; ; )

On Thursday 04 August 2011 20:31:43 Ralf Mattes wrote:
> On Thu, 04 Aug 2011 15:21:42 +0000, Ralf Mattes wrote:
> > I got some strange cracks/jack xruns after playback when playing
> > back to an external app (aeolus in this case). After some
> > debugging I think that is related to the call to
> > fluidsynth_all_notes_off() in process_audio(). In fact, commenting
> > out that call fixes the xrun problem. Now, why is fluidsynth
> > called at all even so I don't use it?
> Ah, I see you already did comment on this in fluid.c ...
> BTW, I just updated my fluidsynth library (from subversion) and the
> xruns disappeared. But that doesn't mean that xruns won't happen.

I'll soon remove the call to fluidsynth_all_notes_off() at the end of 
playback. It was just a very cheap way to make sure that all notes are 
stopped (including the case when you stop playback in the middle of a 
note), but there are better ways to do this.

As far as I can see, we'll still need to use the non-realtime-safe 
fluid_synth_system_reset() when the user hits the panic button. Maybe I 
can find a clean way to do that outside the JACK process callback.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]