2 Oct
2009
2 Oct
'09
7:06 a.m.
Hey all,
I've read and read and re-read the README of the latest release of aeolus
that includes midi control of the stops and I still don't understand how to
specifically program the 'bytes' on midi CC 98 or so..
Is this like sending raw midi sysex data or what?
Thanks,
Andrew Coughlan.
2 Oct
2 Oct
8:39 a.m.
New subject: So what's the deal with controlling the aeolus organ stops via midi
On Fri, Oct 02, 2009 at 10:06:08AM +0100, Andrew C wrote:
I've read and read and re-read the README of the
latest release of aeolus
that includes midi control of the stops and I still don't understand how to
specifically program the 'bytes' on midi CC 98 or so..
Is this like sending raw midi sysex data or what?
No, it's just using controller 98. The data in a control message
is a single byte, that value is exlained in the readme.
You typically need two controller messages: one that
selects a groupt of stops (corresponding to the layout
in the GUI) and the mode: on, off, toggle, and the
second message selects a stop in the group. You don't
need to repeat the first if it's identical to the
previous one.
--
FA
Io lo dico sempre: l'Italia è troppo stretta e lunga.
7:12 p.m.
New subject: So what's the deal with controlling the aeolus organ stops via midi
Fons,
Thanks a lot for your informative answer, but I'm still slightly confused.
:(
I've set my a button on my controller to midi CC 98 and then set it to
toggle between a value of 00 and 2, so it should toggle the second stop of
the manual on and off?
I really feel I have missed something vitally important here.....
The README says that the value used is a bit 6 set... Is it even possible to
get standard hardware midi controllers to send that kind of info?
Also, thanks a million for writing the software!
Andrew.
On Fri, Oct 2, 2009 at 11:37 AM, Fons Adriaensen <fons(a)kokkinizita.net>wrote;wrote:
On Fri, Oct 02, 2009 at 10:06:08AM +0100, Andrew C
wrote:
I've read and read and re-read the README of
the latest release of aeolus
that includes midi control of the stops and I still don't understand how
to
specifically program the 'bytes' on midi
CC 98 or so..
Is this like sending raw midi sysex data or what?
No, it's just using controller 98. The data in a control message
is a single byte, that value is exlained in the readme.
You typically need two controller messages: one that
selects a groupt of stops (corresponding to the layout
in the GUI) and the mode: on, off, toggle, and the
second message selects a stop in the group. You don't
need to repeat the first if it's identical to the
previous one.
--
FA
Io lo dico sempre: l'Italia è troppo stretta e lunga.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
4 Oct
4 Oct
11:34 a.m.
New subject: So what's the deal with controlling the aeolus organ stops via midi
On Friday 02 of October 2009 23:12:16 Andrew C wrote:
Thanks a lot for your informative answer, but I'm
still slightly
confused. :(
I've set my a button on my controller to midi CC 98 and then set it
to toggle between a value of 00 and 2, so it should toggle the second
stop of the manual on and off?
I really feel I have missed something vitally important here.....
The README says that the value used is a bit 6 set... Is it even
possible to get standard hardware midi controllers to send that kind
of info?
Unfortunately it's not that simple. Most hardware MIDI controllers will
let you send continuous controller values or toggle between two values.
But to control the stops in Aeolus you basically need to be able to
manipulate individual bits of the controller values...
Some time ago I wrote a small script that might be just what you need.
It uses mididings (http://das.nasophon.de/mididings/) to convert normal
MIDI controllers (one CC for each stop) to Aeolus' CC 98. You'd just
need to insert it between your MIDI controller and Aeolus.
The script is here:
http://das.nasophon.de/mididings/examples/aeolus.py
Dominic
8:21 p.m.
New subject: So what's the deal with controlling the aeolus organ stops via midi
On Sun, Oct 04, 2009 at 03:33:46PM +0200, Dominic Sacré wrote:
Unfortunately it's not that simple. Most hardware
MIDI controllers will
let you send continuous controller values or toggle between two values.
But to control the stops in Aeolus you basically need to be able to
manipulate individual bits of the controller values...
Some time ago I wrote a small script that might be just what you need.
It uses mididings (http://das.nasophon.de/mididings/) to convert normal
MIDI controllers (one CC for each stop) to Aeolus' CC 98. You'd just
need to insert it between your MIDI controller and Aeolus.
The basic problem here is that MIDI doesn't have anything
close to what is required. On a pipe organ you can switch
each stop on/off independently of all others. MIDI is not
designed to do that, its 'preset' model is something quite
different. The standard Aeolus setup has around 55 stop
buttons, and that amounts to more than 2^55 ~ 3e16 =
30000000000000000 combinations or 'presets'.
The easiest way to control Aeolus from MIDI is to create
a number of presets in Aeolus (it can have up to 1024)
and then use the standard MIDI messages (bank/preset) to
select one of them.
If you really want to control individual stops that is still
possible using normal controllers. You need (for the standard
instrument) 4 buttons to select one of the four groups, and
16 buttons to select a stop within a group. If you organise
things this way then each button just needs to send a single
controller message.
The essential thing to understand is that switching a stop
on/off is a two-step process. First you select a group and
and action (on/off/toggle), then you select a stop within
that group. The first (group/action) message is stored, you
only need to send it when it changes. This is really no more
complicated than the concept of preset banks.
Apart from all this, mididings is the perfect tool to
manipulate all things midi. Using it you can map Aeolus'
control messages to whatever you want.
Ciao,
--
FA
Io lo dico sempre: l'Italia è troppo stretta e lunga.
5 Oct
5 Oct
3 p.m.
New subject: [LAU] So what's the deal with controlling the aeolus organ stops via midi
On Monday, October 5, 2009, Fons Adriaensen wrote:
On Sun, Oct 04, 2009 at 03:33:46PM +0200, Dominic
Sacré wrote:
That is the main purpose of the MIDI System Exclusive messages, precisely.
Almost all manufacturers of hardware MIDI synthesizers provide sysex messages
in order to store and retrieve sound configurations (programs) or modify
single parameters of the programs. The format of the messages is free, and as
long as you need.
Here is a general description, if you need it:
http://home.roadrunner.com/~jgglatt/tech/midispec/sysex.htm
The MMA requires that you use a registered manufacturer ID, but only for
commercial products. There is a special ID = 0x7D that is intended for
educational or development use only, and should never appear in a commercial
design.
Regards,
Pedro
Unfortunately it's not that simple. Most
hardware MIDI controllers will
let you send continuous controller values or toggle between two values.
But to control the stops in Aeolus you basically need to be able to
manipulate individual bits of the controller values...
Some time ago I wrote a small script that might be just what you need.
It uses mididings (http://das.nasophon.de/mididings/) to convert normal
MIDI controllers (one CC for each stop) to Aeolus' CC 98. You'd just
need to insert it between your MIDI controller and Aeolus.
The basic problem here is that MIDI doesn't have anything
close to what is required.
3:21 p.m.
New subject: So what's the deal with controlling the aeolus organ?stops via midi
On Mon, Oct 05, 2009 at 07:00:40PM +0200, Pedro Lopez-Cabanillas wrote:
The MMA requires that you use a registered
manufacturer ID, but only for
commercial products. There is a special ID = 0x7D that is intended for
educational or development use only, and should never appear in a commercial
design.
Where it is silently assumed that 'educational' and 'development'
implies 'not distributed', or at least 'never used together with
any other app using the same ID'.
If two or more open source programs use 0x7D and they happen to
see the same MIDI stream then one of them will be screwed.
Ciao,
--
FA
Io lo dico sempre: l'Italia è troppo stretta e lunga.
6:47 p.m.
New subject: So what's the deal with controlling the aeolus organ?stops via midi
Completely agree that SYSEX is where this kind of functionality should reside.
This use of 0x7d is a bit antiquated, no? The reassignment of 0x00 to indicate 3 byte
SYSEX ID means there is a bit more flexibility in the system. Currently the following
are assigned:
0x00 0x00 0xXX American group
0x00 0x01 0xXX American group
0x00 0x20 0xXX European group
0x4X Japanese group (with holes)
0x5X Japanese group (with holes)
0x00 0x04 Japanese group
Getting a registration requires it be paid for, pretty ludicrous for what purports to be
an open standard. I would suggest that Open Source developers should simply take
one of the unassigned values for its own first digit, agree between themselves who
get the next two digits for their apps and SYSEX then be sent with 3 byte ID.
SLab hijacked 0x53 about 10 years ago for this purpose and bolted on a three more
digits for personal identification and there is little reason why this should not be done
again, this time in line with the current MMA SYSEX ID defs as above? If open source
goes and squats on 0x7d then everybody can go and haggle over which app gets the
next two digits.
There is probably going to be some complaints that SYSEX require an extra two
bytes and how that is going to degrade system performance at 31.25KHz, which
probably needs a pre-canned response as it is a non-issue.
This discussion is perhaps better aimed at LAD rather than LAU: LAU justs wants it
to work, LAD can horsetrade on the details.
It is possible that the MMA will take an issue with this since it does not actually fit
within the new SYSEX ID specs (0x7d has not been assigned in the new definition)
but I am pretty sure that if open source applications restrict themselves to their own
set of second and third digits then the MMA will not be so daft as to assign that
number to any manufacturer. For the apps that eventually go commercial then
it is their job, as a part of the commercialisation, to formally request their own ID
from MMA and then to recognise both of them if desired - the MMA assigned one
thanks to paying $$$ for the number, and the squatted ID for compatibility purposes.
Now, to tie this back into the original subject: this does not really help with assigning
MIDI controllers back to app controls as now the surface has to be configured to
generate more complex SYSEX messages, neither easy nor even possible with some of
them, but the argument of applications getting screwed by using the same number
is actually very easy to avoid, admittedly as long as it is done unanimously. Perhaps
linuxaudio.org might want to pick up the gauntlet of assigning the open sourced
digits?
Regards, nick
"we have to make sure the old choice [Windows] doesn't disappear”.
Jim Wong, president of IT products, Acer
Date: Mon, 5 Oct 2009 19:19:17 +0200
From: fons(a)kokkinizita.net
To: linux-audio-user(a)lists.linuxaudio.org
Subject: Re: [LAU] So what's the deal with controlling the aeolus organ?stops via
midi
On Mon, Oct 05, 2009 at 07:00:40PM +0200, Pedro Lopez-Cabanillas wrote:
"we have to make sure the old choice [Windows] doesn't disappear”.
Jim Wong, president of IT products, Acer
The MMA requires that you use a registered
manufacturer ID, but only for
commercial products. There is a special ID = 0x7D that is intended for
educational or development use only, and should never appear in a commercial
design.
Where it is silently assumed that 'educational' and 'development'
implies 'not distributed', or at least 'never used together with
any other app using the same ID'.
If two or more open source programs use 0x7D and they happen to
see the same MIDI stream then one of them will be screwed.
Ciao,
--
FA Date: Mon, 5 Oct 2009 19:19:17 +0200
From: fons(a)kokkinizita.net
To: linux-audio-user(a)lists.linuxaudio.org
Subject: Re: [LAU] So what's the deal with controlling the aeolus organ?stops via
midi
On Mon, Oct 05, 2009 at 07:00:40PM +0200, Pedro Lopez-Cabanillas wrote:
>
> Io lo dico sempre: l'Italia è troppo stretta e lunga.
>
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user(a)lists.linuxaudio.org
> http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
_________________________________________________________________
Windows Live: Make it easier for your friends to see what you’re up to on Facebook.
http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/so…
The MMA requires that you use a registered
manufacturer ID, but only for
commercial products. There is a special ID = 0x7D that is intended for
educational or development use only, and should never appear in a commercial
design.
Where it is silently assumed that 'educational' and 'development'
implies 'not distributed', or at least 'never used together with
any other app using the same ID'.
If two or more open source programs use 0x7D and they happen to
see the same MIDI stream then one of them will be screwed.
Ciao,
--
FA
6 Oct
6 Oct
6:02 p.m.
New subject: [LAU] So what's the deal with controlling the aeolus organ?stops via midi
On Monday, October 5, 2009, Nick Copeland wrote:
Getting a registration requires it be paid for, pretty
ludicrous for what
purports to be an open standard. I would suggest that Open Source
developers should simply take one of the unassigned values for its own
first digit, agree between themselves who get the next two digits for their
apps and SYSEX then be sent with 3 byte ID.
SLab hijacked 0x53 about 10 years ago for this purpose and bolted on a
three more digits for personal identification and there is little reason
why this should not be done again, this time in line with the current MMA
SYSEX ID defs as above? If open source goes and squats on 0x7d then
everybody can go and haggle over which app gets the next two digits.
The process for the sysex manufacturer ID registration is not only expensive
($200 per year) but also quite ridiculous. It is almost an invitation to
sysexquatting. Have you seen the "Recently Assigned Manufacturer ID Numbers"
page? http://www.midi.org/techspecs/manid.php
There are a lot of companies with a status of suspended, relinquished, renewal
pending ... In red, no less, to add embarrassment to the crisis?
Now, to tie this back into the original subject: this
does not really help
with assigning MIDI controllers back to app controls as now the surface has
to be configured to generate more complex SYSEX messages, neither easy nor
even possible with some of them
I agree. It doesn't help in many situations. However, sysex messages could be
used to store patches inside of MIDI files, attached to the music using the
patches. This can be done as well in Aelous with the current controller 98
messages. In either way, it won't be "easy peasy lemon squeezy".
Regards,
Pedro
5:34 p.m.
New subject: [LAU] So what's the deal with controlling the aeolus organ?stops via midi
On Monday, October 5, 2009, Fons Adriaensen wrote:
On Mon, Oct 05, 2009 at 07:00:40PM +0200, Pedro
Lopez-Cabanillas wrote:
The only devices or programs that would be screwed are those bad or poorly
designed/written.
In addition to the manufacturer ID, there should be enough additional bytes to
uncertainly identify a particular model among others using the same
manufacturer ID. The device (soft in this case) receiving a message should be
able to distinguish between legitimate and spurious messages using the model
ID bytes, checksums and/or some other mechanism.
Regards,
Pedro
The MMA requires that you use a registered
manufacturer ID, but only for
commercial products. There is a special ID = 0x7D that is intended for
educational or development use only, and should never appear in a
commercial design.
Where it is silently assumed that 'educational' and 'development'
implies 'not distributed', or at least 'never used together with
any other app using the same ID'.
If two or more open source programs use 0x7D and they happen to
see the same MIDI stream then one of them will be screwed.
11 a.m.
New subject: So what's the deal with controlling the aeolus organ stops via midi
Hey Fons,
Yeah, I'd say for my needs, I can just create a few presets and just cycle
through them as needed!
Thanks again!
Andrew.
On Sun, Oct 4, 2009 at 11:18 PM, Fons Adriaensen <fons(a)kokkinizita.net>wrote;wrote:
On Sun, Oct 04, 2009 at 03:33:46PM +0200, Dominic
Sacré wrote:
Unfortunately it's not that simple. Most
hardware MIDI controllers will
let you send continuous controller values or toggle between two values.
But to control the stops in Aeolus you basically need to be able to
manipulate individual bits of the controller values...
Some time ago I wrote a small script that might be just what you need.
It uses mididings (http://das.nasophon.de/mididings/) to convert normal
MIDI controllers (one CC for each stop) to Aeolus' CC 98. You'd just
need to insert it between your MIDI controller and Aeolus.
The basic problem here is that MIDI doesn't have anything
close to what is required. On a pipe organ you can switch
each stop on/off independently of all others. MIDI is not
designed to do that, its 'preset' model is something quite
different. The standard Aeolus setup has around 55 stop
buttons, and that amounts to more than 2^55 ~ 3e16 =
30000000000000000 combinations or 'presets'.
The easiest way to control Aeolus from MIDI is to create
a number of presets in Aeolus (it can have up to 1024)
and then use the standard MIDI messages (bank/preset) to
select one of them.
If you really want to control individual stops that is still
possible using normal controllers. You need (for the standard
instrument) 4 buttons to select one of the four groups, and
16 buttons to select a stop within a group. If you organise
things this way then each button just needs to send a single
controller message.
The essential thing to understand is that switching a stop
on/off is a two-step process. First you select a group and
and action (on/off/toggle), then you select a stop within
that group. The first (group/action) message is stored, you
only need to send it when it changes. This is really no more
complicated than the concept of preset banks.
Apart from all this, mididings is the perfect tool to
manipulate all things midi. Using it you can map Aeolus'
control messages to whatever you want.
Ciao,
--
FA
Io lo dico sempre: l'Italia è troppo stretta e lunga.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
5570
days inactive
5574
days old
linux-audio-user@lists.linuxaudio.org
10 comments
5 participants
participants (5)
-
Andrew C -
Dominic Sacré -
Fons Adriaensen -
Nick Copeland -
Pedro Lopez-Cabanillas