Re: Mutitimbral - A clarification, sort of
Re: Mutitimbral - A clarification, sort of
- Subject: Re: Mutitimbral - A clarification, sort of
- From: Frank Hoffmann <email@hidden>
- Date: Wed, 16 Jul 2003 20:34:36 +0200
On Mittwoch, Juli 16, 2003, at 07:25 Uhr, Michael Olsen wrote:
A: (As Angus says) you don't have to instantiate 16 plugins
Where is the problem with instantiate 16 plugins? We are talking about
shared code and data here. The overhead is minimal.
The problem is from the users point of view... Managing 16 windows is
not optimal, when it can be done in 1.
Sure. That is true. But there can be a big confusion which part belongs
to which plugin. Isn't it nicer when the user clicks somewhere and the
required editor with the settings the user wants to edit comes up
instead of the right instrument, but with the totally wrong part? That
can be also confusing in times. Window management is a host decision as
well.
B: The multiple parts can share the same voice manager. (This is very
important in my opinion).
In which terms? To manage in a way a part of the overall cpu load?
So that the 16-parts together (i.e. the "full sound module") can share
the same polyphony managenet. This way you can, for instance, set all
16-parts to 32-notes, and let each part have priorities, as you see it
in hardware modules.
This
is only of use if you use only one instance of one plugin in the whole
system, isn't it?
No, but I admit it makes more sense then.
C: The multiple parts can share the same (build-into-instrument)
effects (Admittedly, we don't do that in SampleTank, but some
developers might want to).
It would be more flexible when the send effect you described and the
synthesis parts are separate Audio Units.
The problem about doing that is that then we wouldn't be able to make
presets which covers both instruments and effects.
This is more a limitation of the host, isn't it? For me the additional
freedom one earns is of bigger advantage here.
The other issues aside, the really important point, I feel, is (B).
And please don't tell me it doesn't make sense because of other
instruments, because it does (maybe I have plug A set to 32 notes, and
plug B set to 64 notes, and I know that this will never cause cpu
overload).
Excuse that I have a different opinion. I believe the cpu load and
voice management should be controlled by the host and not the plugin.
To set up every plugin by the user for his needs of cpu and voice
management is simply not convenient.
Rgds,
Michael Olsen
email@hidden
Frank
------------------------------------------------------------------------
--------------------------
frank hoffmann mailto: email@hidden
ableton ag
http://www.ableton.com
_______________________________________________
coreaudio-api mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/coreaudio-api
Do not post admin requests to the list. They will be ignored.