[linux-audio-dev] XAP spec - early scribbles

[Dave Griffiths]

On Thu, 6 Feb 2003 10:40:30 +0000, Steve Harris wrote
> On Thu, Feb 06, 2003 at 01:27:39 -0500, Paul Davis wrote:
> > >Again, I think we are speaking of slightly different things.  I am talking
> > >about the time when (for example) the synth at the head of the chain has
> > >stopped playing notes.  A reverb with this as it's input would be told 'your
> > >input is now silent'.  It has a tail, of course.  It can be marked silent
> > >when it's tail is done.  Once it is silent, it does not need to be
> > >processed._
> > 
> > this sounds very dangerous to me. 
> > 
> > your description sounds OK. but ... my kawai k5000 is an additive
> > synth with a moderately interesting FX bank. its still producing sound
> > for most patches long after all of its "voices" are off. if someone
> > were to write a plugin modelled on this kind of design, you can't
> > really know when the top of the chain is silent.
> 
> I agree, and more than that its not a usful optimisation (in RT
> processing), I dont htink the offline case is important enough to warrent
> this level of complexit and debugging nightmare.
> 
> If you can point me to a working (not pure research) implemntation I 
> will consider that it might not be total insanity ;)

I had a vague attempt at doing something like this (after noticing that
filters filtering silence uses up a lot of cpu). Each sample buffer object
that ssm uses would record whether it had been written to, and setting them to
zero (by calling memset via a Zero() function in the sample object) would
clear this flag. 
In this way plugins would know whether there was any information worth
processing in the buffer. I ended up coming to the same conclusion too
complex/not enough of a win to be worth the effort.

dave