[gmpi] Re: Decision Time: 7.1.1

  • From: Chris Grigg <gmpi-public@xxxxxxxxxxxxxx>
  • To: gmpi@xxxxxxxxxxxxx
  • Date: Tue, 10 Jun 2003 00:52:51 -0700

On Mon, Jun 09, 2003 at 03:46:23 -0700, Chris Grigg wrote: 
 Doesn't it work to say the plug -must- always support mono buffers,
 and -may- at its own option support interleave?  Then a host that
 only does mono has no problem.


Prediction: many hosts wont support this, so only a handful of plugins
will at the start, so 2nd gen hosts wont at all, so 3rd gen plugins wont
at all.

We end up with a bigish lump of spec thats bug ridden and never used, and
apparently not useful even if it was used. I call bloat.

We have the same situation in LADSPA with run() v's run_adding(), run() is
mandatory and generally more useful, so everyone uses that. I'l bet there
are plugins that have /never/ had thier run_adding() method invoked. The
API would be significantly nicer with only one of them.

- Steve

Not sure comparing interleaving with run_adding() isn't apples & oranges... so Steve, can you address Ron's previous claims about performance gains that interleaving makes possible? If he's correct -- and I'm not well-situated to verify those claims -- then interleaving would seem to add a lot more net value that what I assume run_adding() does (mix into an existing buffer?). I.e. added complexity can be acceptable if the payoff is high enough.

BTW, my suggestion is offered mainly to try to help find an accommodation between the disagreeing parties, as I don't have a dog in the interleaving race. This kind of compromise-finding is often necessary when developing public standards, because different people see the issues differently, validly, and no single person is entitled to control the whole thing. For example, sometimes there -are- people willing to do a debugged implementation of an optional chunk of the spec that other people are not sufficiently motivated to implement, so sometimes optional features -do- make sense for the industry as a whole even if some companies don't plan to use them. Specifically, I'd guess that even if Cakewalk were the only host company to implement interleaving, and it does in fact provide better performance, then market forces (Cakewalk promoting their superior performance, maybe even cutting promo deals with plug houses, etc.) would make it attractive for some plug vendors to implement interleaved plug versions. The possibility of performance gains would seem to tend to promote interleaving uptake in hosts, as a competition point, since the MI sequencer/DAW marketplace is very, very competitive these days.

-- Chris G.

----------------------------------------------------------------------
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:

  1. Home
  2. gmpi
  3. Archive
  4. 06-2003