24 Mar
2010
24 Mar
'10
6:22 p.m.
From: David Olofson <david(a)olofson.net>
These issues seem orthogonal to me. Addressing
individual notes is just a
matter of providing some more information. You could think of it as MIDI
using
note pitch as an "implicit" note/voice ID. NoteOff uses pitch to
"address"
notes - and so does Poly Pressure, BTW!
Not exactly note-pitch. That's a common simplification/myth.
MIDI uses 'key-number'. E.g. key number 12 is *usually* tuned to C0, but is
easily re-tuned to C1, two keys can be tuned to the same pitch yet still be
addressed independently.
It's a common shortcut to say MIDI-key-number 'is the pitch', it's
actually
an index into a table of pitches. Synths can switch that tuning table to
handle other scales.
A MIDI note-on causes a synth to allocate a physical voice. That physical
voice is temporarily mapped to that MIDI-key-number so that subsequent note
control is directed to that voice. The mapping is temporary. Once the note
is done the mapping is erased. Playing the same key later will likely
allocate a different physical voice.
The MIDI-key-number is therefore an 'ID' mapping a control-source to a
physical-voice.
Anyway, what I do in that aforementioned prototyping
thing is pretty much
what
was once discussed for the XAP plugin API; I'm using explicit "virtual
voice
IDs", rather than (ab)using pitch or some other control values to keep
track of notes.
I agree that addressing notes unambiguously regardless of pitch (or any
other arbitrary property) is the ideal. I wish more sequencers were not
locked into a narrow 'western pop music' mode of operation.
But many MIDI alternatives have been proposed without looking deeply
enough to realise that MIDI already supports very flexible note control.
MIDI's significant flaw is it's grossly outdated 7-bit resolution, the
underlying voice model is sound.
Virtual voices are used by the "sender" to
define and
address contexts, whereas the actual management of physical voices is done
on the receiving end.
You have re-invented MIDI with different nomenclature ;-).
Best Regards,
Jeff McClintock
24 Mar
24 Mar
10:03 p.m.
New subject: automation on Linux (modular approach)
On Wednesday 24 March 2010, at 22.21.53, "Jeff McClintock"
<jef(a)synthedit.com>
wrote:
True. My point was really just that pitch is somehow "hardwired" to the key
number on the receiver end, rather than explicitly specified by some CC or
similar.
(Well, there is the SysEx extension you mentioned, but that's not helping much
unless you have the bandwidth to spare, and equipment that supports it - but
that goes for the existing and proposed alternatives as well.)
[...]
Precisely. Or, I just solved the problem in the most straightforward way I
could think of. Design and implementation done in a few hours. (Ok; it's a
pretty high level language, but still. Maybe I can score a bonus point for
having designed and implemented the language too? ;-)
I could add SysEx key-based control to the MIDI parser, of course - but I'm
still wondering where I'd get such data from anyway! Perpetuating the
catch-22, that is; sorry about that. :-)
So... How about a library for sending and parsing MIDI with these SysEx
extensions? I'm thinking along the lines of providing a nice API, but using an
existing standard protocol. The idea would be to use ALSA and/or JACK (would
need an extension for SysEx, or does it support that these days?) for
transport layer, to avoid reinventing that too.
Later, when the API has proved to work, and stuff can be wired, one might add
more efficient protocols and transport layers for special cases, such as
between JACK clients, or between plugin hosts and plugins. (LV2 extension?)
Or, one starts with that, just looking at the feature set of MIDI + SysEx, and
add MIDI gateways later, if/when needed. Depends on how widespread support for
these extensions is, I guess.
Just thinking out loud here...
--
//David Olofson - Developer, Artist, Open Source Advocate
.--- Games, examples, libraries, scripting, sound, music, graphics ---.
| http://olofson.net http://kobodeluxe.com http://audiality.org |
| http://eel.olofson.net http://zeespace.net http://reologica.se |
'---------------------------------------------------------------------'
From: David
Olofson <david(a)olofson.net>
These issues seem orthogonal to me. Addressing individual notes is just a
matter of providing some more information. You could think of it as MIDI
using
note pitch as an "implicit" note/voice ID. NoteOff uses pitch to
"address" notes - and so does Poly Pressure, BTW!
Not exactly note-pitch. That's a common simplification/myth.
MIDI uses 'key-number'. E.g. key number 12 is *usually* tuned to C0, but
is easily re-tuned to C1, two keys can be tuned to the same pitch yet
still be addressed independently.
It's a common shortcut to say MIDI-key-number 'is the pitch', it's
actually an index into a table of pitches. Synths can switch that tuning
table to handle other scales. Virtual voices
are used by the "sender" to define and
address contexts, whereas the actual management of physical voices is
done on the receiving end.
You have re-invented MIDI with different nomenclature ;-). 
25 Mar
25 Mar
8:55 a.m.
New subject: automation on Linux (modular approach)
Jeff McClintock wrote:
Indeed, even low cost sound modules support different tempered scales. A
good example for the mapping are percussion modules. For percussion it's
more common to have non-western tempered scales.
From: David
Olofson <david(a)olofson.net>
These issues seem orthogonal to me. Addressing
individual notes is just a
matter of providing some more information. You could think of it as MIDI
using
note pitch as an "implicit" note/voice ID. NoteOff uses pitch to
"address"
notes - and so does Poly Pressure, BTW!
Not exactly note-pitch. That's a common simplification/myth.
MIDI uses 'key-number'. E.g. key number 12 is *usually* tuned to C0, but is
easily re-tuned to C1, two keys can be tuned to the same pitch yet still be
addressed independently.
It's a common shortcut to say MIDI-key-number 'is the pitch', it's
actually
an index into a table of pitches. Synths can switch that tuning table to
handle other scales.
A MIDI note-on causes a synth to allocate a physical voice. That physical
voice is temporarily mapped to that MIDI-key-number so that subsequent note
control is directed to that voice. The mapping is temporary. Once the note
is done the mapping is erased. Playing the same key later will likely
allocate a different physical voice.
The MIDI-key-number is therefore an 'ID' mapping a control-source to a
physical-voice.
Anyway, what I do in that aforementioned
prototyping thing is pretty much
what
was once discussed for the XAP plugin API; I'm using explicit "virtual
voice
IDs", rather than (ab)using pitch or some other control values to keep
track of notes.
I agree that addressing notes unambiguously regardless of pitch (or any
other arbitrary property) is the ideal. I wish more sequencers were not
locked into a narrow 'western pop music' mode of operation.
But many MIDI alternatives have been proposed without looking deeply
enough to realise that MIDI already supports very flexible note control.
MIDI's significant flaw is it's grossly outdated 7-bit resolution, the
underlying voice model is sound.
Virtual voices are used by the "sender"
to define and
address contexts, whereas the actual management of physical voices is done
on the receiving end.
You have re-invented MIDI with different nomenclature ;-).
Best Regards,
Jeff McClintock
5353
days inactive
5354
days old
linux-audio-dev@lists.linuxaudio.org
2 comments
3 participants
participants (3)
-
David Olofson -
Jeff McClintock -
Ralf Mardorf