[gmpi] Re: Topic 7: Audio packaging, Process-In-Place

> >That's bass-ackwards to me.  Only the plugin knows what it can optimize.
> 
> I touched on this earlier. To optimize buffer management in a graph, you 
> label each input and output of each node with a buffer starting from the 
> inputs to the graph. Any cascade of operations is best done in-place. A Y 
> branch is dealt with by doing one branch as an in-place operation, and the 
> other branch as not-in-place to temporary buffers. This general approach 

yes, but NOTHING in the baton-passing model precludes this.

> This assumes all operations can be done in-place or not-in-place. If we 
> have flags that identify capability, then the host can use this to decide 
> which branch of a Y is done which way.

OK, the baton-passing model doesn't allow the host to decide which output is
in-place and which is duplicated.  Does it REALLY matter? (if you say yes, I
want to hear a reason :).

As long as the buffer gets re-used, there is a net savings in memory usage
and bandwidth.  It doesn't matter if it goes to the first, second, or 111th
output.

Or maybe it does and I just don't see it yet - in which case, I keep asking
- illuminate me!?

----------------------------------------------------------------------
Generalized Music Plugin Interface (GMPI) public discussion list
Participation in this list is contingent upon your abiding by the
following rules:  Please stay on topic.  You are responsible for your own
words.  Please respect your fellow subscribers.  Please do not
redistribute anyone else's words without their permission.

Archive: http://www.freelists.org/archives/gmpi
Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe

Other related posts: