Thank you, John! Is it just coincidence that the number's 128 -- an exponent of 2? You'll find I'm incurably curious -- but I try a question in google first whenever feasible
No, no coinicdence. The MIDI 'instrument number' is held in binary in the last seven bits of a 'data byte' within a MIDI message.
As you say "powers of two", using binary we can count from 0 to 127 in 7 bits.
The structure of MIDI messages is described within the MIDI 1.0 detailed specification under the heading "Data Structure".
(NOT easy reading when you're new to this stuff!)
If you go to "rodgersinstruments.com" you will find in the 'support' and then either 'downloads' or 'owners manuals' sections, pdf's of various modern instruments.
In some (all? I had a quick browsw through a few of them) of those guides there is information about MIDI implementation for the organ concerned.
I suspect they are all very similar and probably any one will act as an initial guide.
They will tell you which channel of the 16 (another power of two - 4 bits) is used for each manual and for the pedals ... I suspect.
If you can let us know the Rodgers organ model number you're using, that might help.
Found those specs in the site -- great. I've been working off a list I found online in which the numbering is strangely off by 1! Found this page -- >gulp<
and have emailed Rodgers to ask about this MIDI problem -- as sneaky preliminary to asking about our model's MIDI specs. Googled that and searched the organ's manual PDF -- no luck.
I would suggest you don't try creating one from scratch, but use an already existing .ins file (proper name "instrument definition file"
and edit it to match your needs.
But we need the detailed Rodgers MIDI information from the manual.
Event List view for this organ shows every minutest change in a track as a line of type but isn't, alas, editable other than deleting.
It may be easier to use a very simple sequencer program, like Sekaiju (a freebie), to edit the MIDI files. SONAR is perhaps too complex for this.
I always use a very old (circa '99) sequencer for this sort of work, even though I own Sonar 7 PE and the latest Cakewalk by Bandlab.
Lastly, about the numbering being off by one. Yes, quite right.
Back in '83 when the MIDI spec was first published, people weren't as computer literate as they now are.
So Program numbers were described as being from 1 to 128. A programmer knew that they are encoded as 0 to 127 in the data byte within the MIDI message concerned.
Likewise, MIDI channels are described as being from 1 to 16, but encoded as 0 to 15.