OVERRIDING MIDI pt.II

[Tony Clark]

> I think his intention was the modding of existing synths to accept his
> hypothetical new protocol.  My view of this approach is that if you want to
> build a kangaroo, you don't start with a yak and then modify - you build
> fresh.

   I'll agree with this for now.  It should be rethought to a great extent.

> Well, you do have all of those continuous controllers, but a resolution of
> 0-127  is just vile.  Pitchbend is about right - if all cc's had an extra
> byte of resolution, that would go a lot further toward modeling sax wail
> subtleties.  If you have a MIDI processor or a program that does MIDI
> processing (MAX, Logic) then you can do something like this.  If you take
> the PB messages from unused channels (or from channels where PB isn't used)
> and remap the two bytes to a pair of cc's ("coarse" and "fine"), then set
> up your synth (if it allows for this) such that they both modulate the same
> parameter, the "coarse" having 128 times greater modulation strength, then
> you can get around the Insufficient Control Resolution MIDI headache.  The
> Crappy Bandwidth headache remains, but hey, it's a start.

   Yes, for any "NEW" MIDI system to work, it'll have to be at least 
10-bits of resolution.  I think that everyone would be fairly happy with 
any continuous controller operating that way.

> >So I'm all for replacing MIDI, but there haven't been much in the way
> >of concrete proposals.

   Well the USB idea is pretty darn interesting.  No, no one has posted a 
concrete proposal for the data format for such an interface, but I 
imagine that'll take some time.  Let some people tinker with it first and 
see what they can actually get to "work", and then you'll see some 
proposals start to filter through.

> What about a multiport MIDI interface *on the instrument*?  That way, you
> could maintain compatibility with MIDI while increasing throughput.  Want
> to play 32nd note triplets at 300 bpm while modulating resonance at audio
> frequencies?  Fiiiine.  Start with a pair of RS-422 serial ins so you can
> plug your computer into this hypothetical workstation instead of into an
> Opcode Studio 5 or similar.  Retain a single  MIDI in and thru just for
> those poor souls still stuck in the Stone Age.  Offer multi-port MIDI outs
> so you really *can* use it as an interface if you want.  Data controlling
> the instrument could completely bypass the MIDI bottleneck, but there would
> be no compatibility compromise.

   This would be fine and all, I suppose.

   What I wouldn't mind seeing is just some extra lines on the MIDI 
cable.  Yeah, you might have seen my 9-pin DIN post earlier.  Instead of 
sending data in parallel, lets just dedicate the extra pins to 
information that "shouldn't" be sent down the regular MIDI line.  Like 
sysex information.  That should have its own line.  Controller 
information?  You bet!  Anything that really requires a faster data 
speed and/or higher bit resolution.
   This situation would be a little bit more "backwards compatible" with 
the current MIDI system.  You could have a synth that is uploading new 
samples via the sysex line while jamming to note on/off information from 
the standard MIDI line and receiving modulation information from the 
continuous controller line.  All in real time, all without losing the 
critical timing information.
   A synth further down the line using the old MIDI type would be 
receiving the standard MIDI information but wouldn't be able to take 
advantage of the faster continuous controller information or be able to 
make use of the extra sysex line.

   How does that sound?  Just some thought to throw around.

   Tony

-------------------------------
I can't drive (my Moog) 55!
-------------------------------
Tony Clark -- clark at andrews.edu 
http://www.andrews.edu/~clark
-------------------------------