| previous by date | index | next by date |
| previous in topic | topic list | next in topic |
----- Original Message -----From: korgpolyex800Sent: Monday, July 02, 2007 11:27 AMSubject: [korgpolyex] Another little updateGooday Poly fans,
I managed to spend another few hours pouring over the EX800 assembler
code and discovered a fundamental error in my interpretation of one
variable. That single mistake had then led me to make multiple errors
in the interpretation of other parts of the code which then led on to
further errors and so on.
While decoding the MIDI code that carries out a program change, I
incorrectly interpreted a check against a variable as being the
program change enable disable flag (parameter 87). But it turns out
that the variable was in fact the flag that determines whether the
EX800 display is in program or parameter set mode. The ramifications
of that little mistake spilled into my interpretation of a great deal
of the code that handles all of the keypad and buttons and display of
information etc.
So after correcting that mistake, I've been able to make a great deal
of progress interpreting the keypad, buttons and display code.
Also, I've been looking at the multiplexing part of the code where the
various sound control chips are controlled. I've slowly been picking
away at it and it's been tough going because this part of the software
has the least connection to external inputs (such as the MIDI or
keyboard input). To put it another way, the code is very mathematical
in nature using many memory variables the use of which is still
unknown. For example, one part of the multiplexing code is set up just
to provide a toggling mechanism that allows the display to flash two
horizontal bars when doing a patch copy. Who would have thought that
inside of all of the EG's and VCF controls etc there would be a little
section of code just set up for flashing those silly horizontal bars
when doing a patch copy. And another section of the code handles
recalculating the VCF cutoff according to the trigger parameter (para
46) after a new key is hit.
The most interesting question on my mind right at the moment is: What
is the mathematics used to operate the EG's? If I can get a good
answer to that question soon then I am hoping to be able to answer the
next really important question which is: Does the CPU have enough
spare cycles left to operate a fourth EG? Because if it does have
enough CPU cycles left for that then we'll be able to use that fourth
EG to control maybe noise level, resonance, or even MG depth. And that
would be a great new feature.
I wish I had more to report but that's about it for now. This email is
more of a note to let you know that I am still plugging away at this
dang thing.
Mike H.