[LV2] plugin bypass

[David Robillard]

On Thu, 2014-05-08 at 12:49 +0000, Fons Adriaensen wrote:
> On Tue, May 06, 2014 at 09:50:07PM +0200, Robin Gareus wrote:
>  
> > A host supporting 'bypassFeature' must always add any plugin which does
> > have a port tagged 'bypass' in bypassed mode and un-bypass it after the
> > first process callback. That will allow for clickless insertion of the
> > plugin. (same for removal, first bypass, then remove.)
> 
> This could be problematic as the first process callback could be
> too short to allow the plugin to do a clean transition, defeating
> the original purpose.
> 
> Ideally the plugin should indicate to the host if it is in bypass
> mode or not. So when a plugin is to be removed, the host puts in
> bypass, waits until the plugin confirms this, and then removes it.

Couldn't one argue that, in the removal case, the plugin will be totally
removed at some point, so if the bypassed-by-plugin output is not
identical to running without the plugin at all, there would be
artifacts?

That would mean that having the host crossfade from wet to dry is the
way to go for clickless insertion/removal.

-- 
dr