[gmpi] Re: Reqs 3.9. Time - opening arguments.1

  • From: "Koen Tanghe" <koen@xxxxxxxxxxxxxxxxxxx>
  • To: <gmpi@xxxxxxxxxxxxx>
  • Date: Fri, 6 Feb 2004 22:03:11 +0100

On Friday, February 06, 2004 12:56 AM [GMT+1=CET],
RonKuper@xxxxxxxxxxxx <xxxRonKuper@xxxxxxxxxxxxxxx> wrote:

> And, as David said: it can already be done with VST.
> <<<
> A VST plugin can send tempo change events to the host, and the host
> will follow without editing or undo implications?

Well, no VST hosts currently support plugins sending time info back to the
host (well, not that I know of...) But a lot of hosts won't even allow
plugins to send midi back to the host for re-routing to other plugins or
tracks either...
The only exception (regarding time info) is Bidule, where I was said by
David that it will accept and acknowledge VstTimeInfo from a plugin. Don't
know about editing or undo implications though... Sorry.

But since there had been some talking about use cases:
Say you have a host that allows plugins to send back tempo info to the host,
which will adjust a "timeline" (as it was called), and several plugins are
taking their time info from that timeline.
Let's take the example of a plugin that analyzes an audio stream and sends
out tempo changes, and a delay plugin that is synchronized to a musical
Now, if a tempo change is detected by a plugin analyzing streaming audio,
the new tempo will be sent to the host which will adjust the timeline
(question: can it go directly to plugins too? I'd guess not), and the
depending plugins will change their delay times (probably expressed in
samples internally) accordingly.
Your question is (if I understood correctly): will the tempo change sent out
by the analyzing plugin and accepted by "the one who controls the timeline"
(probably the host, but maybe a plugin?) be undoable?
Well, I'm not an expert at all, but I'd guess it should not be undoable,
unless you record the tempo change events onto a tempo or automation track
(whatever). This feels the same to me as when playing a virtual instrument
while the song is playing on the one hand, and recording the played MIDI
events on a track on the other hand. What the host stores should be
undoable, what just happens while playing not. If in Cubase or Sonar I move
a slider in a mixer or in a plugin (without recording that movement), it can
not be undone either. Just my feelings about this. Maybe more experienced
people have better ideas?

> Of course, I'm not a host writer, but I can understand your concerns
> as it might add complexity over many of the existing systems of today.
> <<<
> Tremendous complexity, unless the host was designed from the
> beginning to support this sort of thing.  I think the popular
> commercial sequences have one thing in common: they've been around a
> while, before plugins existed on the host.  So I'm pretty sure this
> will be similarly challenging for nearly any host.

Challenging? Probably, yes. But why not?
I know some live musicians that always complain that they can not play
together with a sequencer, because they can't vary their timing "as they
feel it". They're slaves to the machine. They just want the machine to
follow *their* timing and tempo fluctuations. And this IS becoming possible

Anyway, maybe this is something most people here don't want to deal with for
now. But I'd rather try to get the possibility to do this in there now
instead of waiting another two years for a revision/extension of GMPI. How
do other list subscribers feel about this?


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

Other related posts: