Re: Mutitimbral - philosophy
Re: Mutitimbral - philosophy
- Subject: Re: Mutitimbral - philosophy
- From: Scott Ruda <email@hidden>
- Date: Thu, 17 Jul 2003 13:56:18 -0700
On Thursday, July 17, 2003, at 04:12 AM, "Angus F. Hewlett"
<email@hidden> wrote:
At 11:06 PM 7/16/2003 +0100, Richard Dobson wrote:
One interesting use of a multi-timbral synth is as a guitar
synthesizer.
In many
such systems, each string is synthesized on a different MIDI Channel,
in
order
that string-bending can be made independent for each string (many
controls
in
MIDI are channel-wide, most significantly pitchbend).
That strikes me as something of a hack... I know it's how it is
commonly
done today, but the best way of accomplishing what you want there is
for
the API to support note-level parameters rather than sending the notes
on
different MIDI channels.
This kind of mode is actually part of the MIDI spec. It's called MONO
mode and was specifically for use with guitar synths. In MONO mode the
user defines a 'base channel' and then N channels above that to
dedicate to a controller with each channel explicitly in monophonic
mode (now commonly referred to as solo mode). The original MIDI spec
did not have what is now commonly referred to as Multi Mode where all
16 channels are active simultaneously. Now most synths have at least
32-128 voices, can do multi mode and can be set to do solo mode on a
per preset basis. Back in the day (circa 1982-85) most synths were
either mono or at most 4 or 8 voices (and you typically used 2-4 voices
per tone), so I guess it didn't seem likely that one piece of gear
would be doing a full orchestration using multi-mode. I believe the
E-mu EII was the first device to do Multi Mode (called SuperMode at the
time. Machine came out in 1984, but SM was added about 1987).
Multi-channel for a single controller is still how guitar controllers
work today. It gives the most amount of flexibility in control, albeit
difficult to set up. Yes, given a control protocol other than MIDI (can
anyone say Lone Wolf or Zeta or X-MIDI or ZIPI or GWIZ or ...) it would
be nice to have full controllers per key, but MIDI is still today's
protocol for physical controllers. Poly key pressure would help in some
cases, but that is still only one per-key controller (other than key
velocity, but that can not be used after the note-on). If you want to
modulate things like pitch, volume, pan, filtering, attack time etc.
per-key (per string on a guitar/violin) then dedicating multiple MIDI
channels is the only way to go. Remember, not everyone plays music
using a mouse ;-)
_______________________________________________
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.