?

>>>

Am 20.04.2012 00:37, schrieb Bakis Sirros:

?

hi list,

let's say you have two Xpanders midied together (midi out from one to midi in of the second).

does the Xpander support a 'midi overflow' mode, so you could play two Xpanders as one 12-voice machine?

NO !!!

thanks in advance for your help!

There?s no help possible except a firmware/OS hack or a ext. hardware device which doesn?t exist AFAIK.

The Xpander MIDI Out set to "MIDI ECHO" just echoes what comes in at it?s MIDI Input.

A 2nd Xpander receiving MIDI info from the 1st Xpander?s MIDI Out (in MIDI Echo Mode) receives the SAME notes, controller data and sysex.
That way, it will be 2 MIDI layered 6-voice Xpanders in general,- not discussing the common zoning features and combinations w/ 2 XPanders linked via MIDI here.

To make 2 6-voice Xpanders a 12-voice Xpander, in theory there would be 2 ways:

MIDI spill-over or recognition and MIDI filtering of odd and even MIDI note numbers.
Some 19" MIDI tone generators use the MIDI note numbers.

Recognition of odd and even MIDI note numbers would work on the same MIDI channel in single-path mode and on the same MIDI channels in multi-patch mode, the zones set up exactly the same.
That would be a functionality, we?d wish to have in the Xpander itself.

To be precise, Xpander #1 plays the even MIDI note numbers and Xpander #2 the odd MIDI note numbers.

IMO, it would be the more interesting method w/ 2 Xpanders because we also deal w/ the voice panning here which is not w/ a Matrix-1000.

If deviding odd and even MIDI note numbers would be a task for a masterkeyboard, the masterkeyboards features would be different,- p.ex. transmission of odd MIDI note numbers on MIDI channel X and even MIDI note numbers on MIDI channel Y (1 MIDI Out port),- and Xpander #1 set to MIDI Ch. X while Xpander #2 being set to MIDI channel Y, - the Xpanders daisy chained via MIDI Out Echo > MIDI In.

Or,- transmission of odd and even MIDI note numbers on 2 MIDI Out ports of the masterkeyboard (if it has 2 MIDI Out ports) but same MIDI channel,- each Xpander connected to one of these ports.

Or split of MIDI note numbers across 2 MIDI channels and 2 MIDI ports.

All is thinkable and would work.

Unfortunately, there?s no masterkeyboard out there I know, splitting odd and even MIDI note numbers and transmitting these on different MIDI channels or different MIDI ports.
I hoped a Oberheim XK can but it doesn?t and none of my other keyboards does.
Also my MIDITEMP PMM88E can?t.

If anyone knows a hardware MIDI processor doing that,- let me know ...

I?d call it "Odd-Even MIDI Note Splitter" (MIDI Ch. or MIDI Port Split)

If someone has the skills and thinks about building such little box,- splitting note numbers into MIDI channels on 1 MIDI port would be the most compact and cost effective solution.

I wonder why it doesn?t exist made by Midisolutions,- they have the skills to realize that w/ ease.
Possibly it could be realized w/ this box,- just the processing function is missing:


They do modifications of their products on demand as also customized products.


If that is done,- you need a keyboard providing power over MIDI to their boxes,- in the FAQ is a long list which products don?t and I doubt it?s complete.

best regards,???? :-)

Bakis.

ditto

PeWe

Sorry for being so late.? I had the same issue with my Matrix-12.? In the end, I had to replace one of the CEM chips.
In the M-12 they are socketed, so it was not too difficult to troubleshoot by swapping the chips between the bad voice and known working voices.

The bad news is the the CEM chips are hard to find, but they do turn up from time to time.

I wasn't very clear about my decompiler and what it does, etc, so I'll
explain.

Example here:

[original 6502 code]
[decompiled code]

The whole idea is to get rid of all register assignments, and also
eliminate the use of the stack. Also, it deals with CPU flags.

It does this by having 3 kinds of variable:

*LLocal (a register used as a loop counter)
*BLocal (branch local; a register or stack is assigned in an If or
If/Else block)
*FLocal (a register or stack is the return value from a function)

Any Global variable, which in most CPUs is just RAM, is not filtered out.

Loops are found, as are any amount of nested If/Else's, and Goto's are
eliminated. (except for jumps from inside a loop).

The point of my decompiler, then, is not to output code that would
compile in a C compiler, but to make assembly code much easier to read -
and much faster to understand. Then you modify the assembly code to do
what you want.

As an example, I've decompiled a game and I could see what the function
(which draws the game screen) did, and how it worked, very quickly.

Also, it can output annotated assembly language, and it can output
addresses before each line of code, so you can load the decompiled code
into a disassembler, and it shows you the code alongside the code in memory.

As for the 6809, I'll look into getting a disassembler for it and get to
writing a module.

Cheers,

Jeremy.

?

Hi Tobbe,

The Europa upgrade is for your Roland Jupiter 6:

As for your strange (but cool) tuning sounds, I'd call that a feature rather than a bug. ;) I don't know why it started doing that, but it does raise the issue of OS versions matching/requiring specific circuit board revisions. Perhaps you have a particularly early/late board version? Other than that, some slight corruption of the EPROM containing the firmware is the only other possibility I can think of offhand.

Cheers,

John :^P

On Tuesday, 3 April 2012 at 15:47, Tobbe Bergman wrote:

>
>
> Sorry for intruding but I was wondering what the Europa Upgrade was ?
>
> Ever since I upgrade to the latest software ( Which I Found here ) I have been having strange sounds when I am tuning the VCF's
>
>
>
> Cheers / Tobbe
>

> 2012/4/3 John Pallister <john@... (mailto:john@...)>

> >
> > Hi Tony,
> >
> > Nobody's saying it would be easy to develop a new version of the Xpander OS, but it would certainly be much easier if there was a version of the source code to start with.
> >
> > Is your friend looking to form a small team to develop a new Xpander OS as a commercial product, like the Europa upgrade (which had a hardware and software component, and therefore would be a much bigger deal than a straight OS update)? If not, could he be persuaded to donate the source code to the Xpander user community (i.e. this mailing list)?
> >
> > Your allusions to the Xpander source code are very tantalising. There are people who would be interested in working with the source code if it became available; what exactly are your friend's criteria for releasing it?
> >
> > Cheers,
> >
> > John :^P
> >
> > On Tuesday, 3 April 2012 at 15:07, Tony Cappellini wrote:
> >
> > >
> > > > > You've mentioned your friend before, although I can't find the reference. By >>"committed to the cause", do you mean "prepared to spend money"?
> > >
> > > A project like this is easily going to take 1 person more than a year, even if
> > > they know 6809 assembly very well.
> > >
> > > For those who know the Jupiter-6, the Europa upgrade took something like 3-5 years before it was finished. Two people worked on the project initially, but I believe it was eventually finished by 1 guy.
> > >
> > > Finding the right people to see this project to its completion isn't going to be easy.
>
>

Sorry for intruding but I was wondering what the Europa Upgrade was ?

Ever since I upgrade to the latest software ( Which I Found here ) I have been having strange sounds when I am tuning the VCF's



Cheers / Tobbe

2012/4/3 John Pallister <john@... (mailto:john@...)>

Hi Tony,

Nobody's saying it would be easy to develop a new version of the Xpander OS, but it would certainly be much easier if there was a version of the source code to start with.

Is your friend looking to form a small team to develop a new Xpander OS as a commercial product, like the Europa upgrade (which had a hardware and software component, and therefore would be a much bigger deal than a straight OS update)? If not, could he be persuaded to donate the source code to the Xpander user community (i.e. this mailing list)?

Your allusions to the Xpander source code are very tantalising. There are people who would be interested in working with the source code if it became available; what exactly are your friend's criteria for releasing it?

Cheers,

John :^P

On Tuesday, 3 April 2012 at 15:07, Tony Cappellini wrote:

You've mentioned your friend before, although I can't find the reference. By >>"committed to the cause", do you mean "prepared to spend money"?

A project like this is easily going to take 1 person more than a year, even if
they know 6809 assembly very well.

For those who know the Jupiter-6, the Europa upgrade took something like 3-5 years before it was finished. Two people worked on the project initially, but I believe it was eventually finished by 1 guy.

Finding the right people to see this project to its completion isn't going to be easy.

?

Hi Tony,

Nobody's saying it would be easy to develop a new version of the Xpander OS, but it would certainly be much easier if there was a version of the source code to start with.

Is your friend looking to form a small team to develop a new Xpander OS as a commercial product, like the Europa upgrade (which had a hardware and software component, and therefore would be a much bigger deal than a straight OS update)? If not, could he be persuaded to donate the source code to the Xpander user community (i.e. this mailing list)?

Your allusions to the Xpander source code are very tantalising. There are people who would be interested in working with the source code if it became available; what exactly are your friend's criteria for releasing it?

Cheers,

John :^P

On Tuesday, 3 April 2012 at 15:07, Tony Cappellini wrote:

>
> > > You've mentioned your friend before, although I can't find the reference. By >>"committed to the cause", do you mean "prepared to spend money"?
>
> A project like this is easily going to take 1 person more than a year, even if
> they know 6809 assembly very well.
>
> For those who know the Jupiter-6, the Europa upgrade took something like 3-5 years before it was finished. Two people worked on the project initially, but I believe it was eventually finished by 1 guy.
>
> Finding the right people to see this project to its completion isn't going to be easy.

@Jeremy,

If Tony's friend could be persuaded to divulge the source code, we wouldn't need to decompile it; it has to be hand-coded assembly language, so turning it into C code is unlikely to help, I would've thought… I suppose you'd get "structured" assembly language, but then you'd have to feed any modifications back through a C compiler, which would be unlikely to produce machine code of the same density. But I'm just guessing here, so I'd be interested to hear your thoughts.

The decompiler is just for figuring out how code works, rather than
recompilable C.

Yes, it is 'structured' with loops, and no registers, etc.

The whole point is to see the program at a higher level, but with far
less effort than doing it by hand. Say, 2 weeks instead of 2 months.

Cheers,

Jeremy.