Good for you PeWe,

These days It seems we are all software(media)programers in one way
or another. I have lost friends over this, but I won't pirate.

Karl



--- In xpantastic@..., PeWe <ha-pewe@...> wrote:

Congrats ! That?s exactly the right way, but don?t ask me where he

is.

There are some guys in the Notator.org group which do a online

tutorial

for Notator, they went this way and got a go AFAIR


envia94 schrieb:

Hi,

I wonder if somebody would like to contact Jacques Isler and as

his

permission. In any
case, I think that all patent rights etc. get old in 20 years,

but

this had to me verified.

Tiitu.

Hi Again,

I finally have a little time to be more clear on the LFO help post.

I know the M12/Expander is one the few synths to allow a mod source
to modulate itself. I use this all the time when I want really long
envelopes. Now, I understand this is common knowledge among the
Modular Synth Community. However in theory the LFO modulating it's
own rate should make the LFO rate speed up and slow down right?
But what seems to happen is the rate will increase a small amount and
stay there(or decrease if mod is negative). I like to experiment with
LFO triggered envolopes. When I try to get these rates close to a
Beats Per Minute number I find It's somewhere in between what I need.
By modulating the LO rate by the same LFO I can usually adjust out
any error.
I don't know what led me to this discovery but It works on my M12. I
can't really say about the Xpander.

Karl

--- In xpantastic@..., Karl Schmeer <shire03@...> wrote:

Hi Tony,
I'm sorry, the rate of the LFO by the output of the same LFO.

eg.. LFO1 RATE by LFO1 .

Karl

Tony Cappellini <cappy2112@...> wrote:
Karl

You can almost modulate any source by itself. (in the XP/M12)

Which parameter in the LFO are you modulating by the same LFO?

On Tue, Aug 12, 2008 at 2:51 AM, Karl <shire03@...> wrote:

Hello All
Has any one noticed that you can adjust the XPANDER/M12 LFO speed

in

between the standard 0-63 settings by using recursive modulation.

Just

modulate the LFO with Itself. Adjusting the modulation amount will
change the rate in small increments. I don't know if this works

on the

M1K though.

Karl

开云体育

Congrats ! That?s exactly the right way, but don?t ask me where he is.

There are some guys in the Notator.org group which do a online tutorial for Notator, they went this way and got a go AFAIR


envia94 schrieb:

开云体育

Hi Karl !

Yes, if you sample your gear,- see the instrument you sample as "the oscillator- the sound source" and the sampler as the instrument you use later.
A high quality sampler like Kontakt 3 p.ex., has a lot of modulators like envelopes, lfos, filters, step sequencer, VCA as also FX including amp models and so on.
So, best is to sample all the patches you use the same way since decades ( there are for sure some leads and pads as also basses) as the are and w/ multisampling techniques. Decide for a compromize of count, lenght of samples and memory usage.
With a hi end sampler ( I have Halion and Kontakt ) which is running on a computer, you don?t have to care on memory at all.
Listen in which zone/range of a keyboard the selected patch you want to sample sounds best and how you used it yourself on the original instrument. Sample that range only but w/ as much velocity steps as possible and possibly all the keys of that range separately.
This is what Samplerobot automatically does and it?s doing more !!!
Now you have your patch sampled as it is p.ex. ...

Next,- edit exactly this patch. Remove all the filter modulations and eventually almost everything comparable what your sampler can do w/ the resulting samples later,- but keep everything whats related to the oscillators like fixed frequency and/or PWM modulation amounts, detune of osc 1+2, FM/lag settings p.ex. or ring modulations if it is important for the basic sound. Now sample the resulting sound at full velocity, full open filter but no resonance (eventually several times w/ different filter modes) at max level., now you get samples the way you can use the modifiers of your sampler to a full potential later.

If your sampler does portamento, don?t sample it. If it does PWM to raw wave samples, don?t sample PWM.
Instead multisample the oscillator section w/ open filter but w/ only 1 Osc. switched on,- do this the 2nd time with the other Osc. switched on and the 1st off, both w/ separate Pulse Widths settings but no modulation. Do this w/ the OSCs not detuned against each other. Better do a layer of both of these multisamples in your sampler later and detune th layers against each other in the sampler.

Controllers and Midi settings of the synth to sample you can ignore,- Samplerobot allows you to tell the program what the sampler shall do later to your samples w/ midi controllers, vibrato included.

It?s a bit of a learning process and time consuming, but you can get very intersting results which sound not like the original for sure but sometimes also better and you can use many sounds of your beloved vintage gear in modern DAWs without having the dinos connected all the time, audio and midi wise, and without dealing w/ probs of midi and external gear on your DAW.

Also it saves the live of your old machines because they aren?t always running.

The biggest advantage of modern DAW usage isn?t virtual instruments, - it?s advanced sampling, automation and total recall. Recording audio to disk is nothing else than sampling.

Karl schrieb:

Hi All,

Somewhere around here I have all the patch names for the M1K factory
patches. I know this is an Xpander/M12 Group. But I thought I would ask
anyway. Does anyone want a copy of these.

Karl

Hi,

As you may know, there has been quite recently some discussion about
the possibilities for re-manufacturing Notitake Itron VFD displays
for Oberheim Xpander and Matrix-12. We therefore have arranged a poll
in the Xpantastic forum.

Until now, 20% of the members have answered that they would probably
like to order altogether 110 spare VFD displays.

Please, we need Your vote too!

To vote, please visit the following web page:


Note: Please do not reply to this message. Poll votes are
not collected via email. To vote, you must go to the Yahoo! Groups
web site listed above.

Best regards,
Tiitu.
List owner
xpantastic · Xpantastic!

Hi,

I wonder if somebody would like to contact Jacques Isler and as his permission. In any
case, I think that all patent rights etc. get old in 20 years, but this had to me verified.

Tiitu

--- In xpantastic@..., PeWe <ha-pewe@...> wrote:

There?s so much to think about for a good cloning of the
characteristics of a synth.

Hello there PeWe,
Yes, I kept wondering who Karl Schrieb was Ha Ha, And why he was
putting his name on my emails.

Well as I mentioned I don't take the M12 out anymore. I just don't
want anything to happen to it. So my original intent was. I would
take my favorite patches and sample them in and play them out of my
EMU Ultra.
Heres a problom I never expected:
When you sample a sound, usually you sample a high note. Then the
sampler divides this note down to give you your lower ranges. Well
there is significant high frequency noise in a M12 sound. The
switching power supply/ high voltage display multiplexing are two
sources I identified. Well these are usually inaudible but when the
sounds are divided down so is this noise and now it's part of the
audible signal. I know I could multisample this but a lot of my
patches were ambiant shifting kinds of things and use too much
memory. You probobly know this is also a problem with LFO rates also.
By the time you switch everything off. You are just sampling a pure
wavform.
So I scratched this Idea. What is cool to do is sample in some M12
base tones and you get some incredible low freq stuff. And there is a
neat transfer algorithim which can create some new sounds.

Karl

Hi,

just my 2 cents before someone dives deeply into Xpander's SysEx
implementation and pull out his hairs...I'm currently doing it for the
single patch part (and only for that).
I did a check of a the spec for each parameters, and it has some
tricky bugs. I've written a command line software that dumps the
contain of a single patch sysex to show what'is in it. It's working
fine,but some tests have yet to be done (comparing what's on my PC and
what my XP shows is really a boring task...) I'll upload it in the
next days with the C/C++ source code and the updated MIDI Spec.

/Greg.





--- In xpantastic@..., PeWe <ha-pewe@...> wrote:

>>>

Tony Cappellini schrieb:

If you already have this, then why not just use it to send the program
to the XPander, then save the XPander Bank as sysex and send it to us?

no time !

It??s illegal if would spread contents of this disk out over the

web.

Atari hasn't been in business for a long time and is not making any
money from any software sales.
I"m sure the author is no longer selling the program nor

supporting it.

Since you know him you can talk to him.

Atari isn??t the author and I don??t know the author, he??s not

mentioned

in the manual.
The person I mentioned was the distributor.

Some is free available some not. P.ex. Steinis and C-Labs software

isn??t

freely available unless it??s cracks.

The manual of the program is 4 shitty pages in german and doesn??t

cover

the functions of the program properly and there??s no list of the

patches

at all.
Typing 1000 patchnames by reading ??em from a small Atari SM124

monitor,

for me, is much work,- too much.

I wanted to help by pointing on that exsisting program in the hope
someone wants to have it and do a search at ebay or wherever possible.

Much software from the computers of the 80's is now freely available
to run in software emulators.



As I've already said- adding the patche names to a bank of Xpnder
patches is a piece of cake.
Someone just needs to come up with the 1000 patch names- unless they
are already supplied in the manual for that Atari software.

If you think this is a lot of work, it's not. This would be the
EASIEST part of it.We just need the patch names in a text file.
They can be added to banks of patches with a simple program.

开云体育

>>>

Tony Cappellini schrieb:

If you already have this, then why not just use it to send the program
to the XPander, then save the XPander Bank as sysex and send it to us?

no time !

>
> It?s illegal if would spread contents of this disk out over the web.
Atari hasn't been in business for a long time and is not making any
money from any software sales.
I"m sure the author is no longer selling the program nor supporting it.
Since you know him you can talk to him.

Atari isn?t the author and I don?t know the author, he?s not mentioned in the manual.
The person I mentioned was the distributor.

Some is free available some not. P.ex. Steinis and C-Labs software isn?t freely available unless it?s cracks.

The manual of the program is 4 shitty pages in german and doesn?t cover the functions of the program properly and there?s no list of the patches at all.
Typing 1000 patchnames by reading ?em from a small Atari SM124 monitor, for me, is much work,- too much.

I wanted to help by pointing on that exsisting program in the hope someone wants to have it and do a search at ebay or wherever possible.

Much software from the computers of the 80's is now freely available
to run in software emulators.



As I've already said- adding the patche names to a bank of Xpnder
patches is a piece of cake.
Someone just needs to come up with the 1000 patch names- unless they
are already supplied in the manual for that Atari software.

If you think this is a lot of work, it's not. This would be the
EASIEST part of it.We just need the patch names in a text file.
They can be added to banks of patches with a simple program.

Karl

You can almost modulate any source by itself. (in the XP/M12)

Which parameter in the LFO are you modulating by the same LFO?

开云体育

>>>

Tony Cappellini schrieb:

> The most important difference is, M1K deals different w/ waveforms and the
> waveshaping parameter decides which kind of waveform the oscillator creates
> in the end,- this is what a Xpander and Matrix 12 cannot do.

Why?

I think the pulse width of all the waves is affected in the M1K, - am I wrong ???

> As mentioned before in another post, - I did this programming,-
> Xpander>Matrix and vice versa manually for some patches and IMO, this was
> one of my most time consuming and boring programming experiences fo all the
> time because of all the comparison.

Why ? XPander single patches will go directly into an M12 via sysex.

Missunderstanding: I meant Xpander to Matrix 1000

Buy a sustain pedal ... :-)))))))))))

I dont even have a sustain pedal ;-)

That means, the patches on this disk are licenced to the user who buyed the
program.
I myself buyed it and I hope I?m not the only one.
Definitely, I didn?t found any info of this program in the web last nite,
but I cannot imagine no one has it except me because I buyed it in a regular
music shop in the past and it was not cheap... (DM 249.- was ~ USD 750.- in
1988).

If you already have this, then why not just use it to send the program
to the XPander, then save the XPander Bank as sysex and send it to us?

It?s illegal if would spread contents of this disk out over the web.

Atari hasn't been in business for a long time and is not making any
money from any software sales.
I"m sure the author is no longer selling the program nor supporting it.
Since you know him you can talk to him.

Much software from the computers of the 80's is now freely available
to run in software emulators.

I have no interest to be liable for anything/something in that direction

Nobody is going to pursue any legla actions for such old property.

Beside this,- it would be a very time consuming work to reload all the banks
into an Xpander and save ?em in single patch sysex format by reading the
patchnames in the launched program on the Atari and to rename ?em one by one
on a PC.

As I've already said- adding the patche names to a bank of Xpnder
patches is a piece of cake.
Someone just needs to come up with the 1000 patch names- unless they
are already supplied in the manual for that Atari software.

If you think this is a lot of work, it's not. This would be the
EASIEST part of it.We just need the patch names in a text file.
They can be added to banks of patches with a simple program.

Finally, I think it?s a better idea to buy a used Matrix1000 in addition and
if a Xpander exsits already...
200 bucks,- how much time someone can invest for 200 bucks coding a program
or transfering files and rename ?em ?

I havent seen an M1000 go for sless than $350
Even if 1 person has both sysnt together ( I used to have all 3 at one
point:M6Rm M1k, Matrix 1000) even doing 10 patches by habd is time
consuming..
Once that person figures out which params map directly and which
params dont it can be automated.

All is possible in the end, but what a complex work and what for ?

It will take a long time, probably more than anyone is willing to wait.
I have little time to work on projects like this- and I am already
working on one for the XPander/M12.

So this might not be a prob because the Xpander/Matrix12 has more and in
theory, all what?s more in these machines would be unused for Matrix 1K
patches,- but ...
I only refer to the modulators available in the modulation matriss of a M1K
!
So, the 1st thing what?s different is the mod-matiss.

It is a problem somewhat, because the extra parameters inthe XP
patches have to bse set to something.
Zero is *probably* a safe bet, unless a 0 causes the sound to change.

The most important difference is, M1K deals different w/ waveforms and the
waveshaping parameter decides which kind of waveform the oscillator creates
in the end,- this is what a Xpander and Matrix 12 cannot do.

Why?

Boith synths have the same basic wavehsapes, the XP/M12 have noise in
VCO2 as well.
These can be selected in the patch.
Now- the DCOs on the M1k wont sound the same as the VCOs in the
XP/M12, even for the same wavsehape & frequency settings, but there's
nothing
we can do about that (not easily, anyway)

As mentioned before in another post, - I did this programming,-
Xpander>Matrix and vice versa manually for some patches and IMO, this was
one of my most time consuming and boring programming experiences fo all the
time because of all the comparison.

Why ? XPander single patches will go directly into an M12 via sysex.

I doubt this can be done by a simple command line proggi, - it must be a
more "inelligent" program which exactly knows the differences in the voices
and so on, not forget to mention, the enduser has to "tell" the program what
he wants then.

a configuration file could be read to make the decisions that a gui
otherwise would provide.

To 1.)
Even if parameters map directly, they don?t scale the same,- speeds of LFOs,
envelopes, ramps are different p.ex.

All your questions are right

In the end the result is a different sounding patch you have to edit again
anyway and by using your ears and taste.

This will have to be done no matter what method is used.
The synths are just too different for the patches to be ported and
expect to sound the same.
Myabe some basic boring patches might, but both synths have matrix
modulation. The more complex patches are likely to sound too
different.

2.)
Most factory patches, in the Xpander and/or the M1K, don?t have the simplest
controllers implemented which you need all day, - this is p.ex. sustain
pedal (both) and pitchbend-ranges on your demand (Xpander),- you?d have to
program a lot of usefull stuff anyway yourself.
Sometimes, the controller setups only demonstrate what the machine is able
to do and sometimes it makes no musical sense in daily work.

I dont even have a sustain pedal ;-)

When you say cmd line do you mean a DOS based running under say XP or
would I have to get a linux shell running in my PC?

The program would run in any Microsoft OS in the CMD shell (what you
called DOS-based), and also Linux & the Mac OS.

I've know of a program written this way to translate patches from a
first generation synth to a second gnereation synth, of which the
two synths have NOTHING in common other than the manufacturer. The
patche formats are 100% completely different..

I like your Idea of just filtering and replacing the appropriate
M1K code with M12 code. I am not clear on the Expander data format

Greg has alreayd uploaded the XPander patch format, as a PDF,and it's
beautifully formatted compaired to the original.

but I have taken a look at my M1000 manual. The sysex patch file is
pretty straight forward. Everything hinges on the data position in
the file.

Position, range, ( 0-63, -127) and sign to be more precise.

As PeWe ponts out it would certainly be easier to go from the M1K to
the Xpander than the other way around.

I agree

Looking at the M1K sysex file :The MOD sources/destinations all Have
a number assigned to them, I suspect it's the same on the Xpander/M12.

Yes, download the news sysex spec that was uploaded by Greg.

But wait, there are all those hardwired mods on the M1K.

And on the XP/M12 too

These may not be the same on both machines

This creates a problem with the Xpanders 20 routing max.

One of the many questions that must be answered before a program could
be written to assist.

BTW: If you find the posted M1K sysex file lacking I have the manual
and my wife has a scanner that will create a PDF file from the scan.

I've found a PDF of the M1k manula. No need to scan it.
Thanks

开云体育

Hi !

Well, you see there?s the "schrieb" again in this mail. Now I know it?s being created by Thunderbird, german version, if I recieve a mail.

Yes, I know exactly the intention of all the trial ?n error? keeping the old gear alive because it sounds good and sometimes great and because of we all used it so long and therefore are very familiar w/ it.
It?s the old "never change a running system" rule, especially if it runs well and made great music, or not so great but commercially successful music for decades if treated by the creative people,- whatever they did,- sounddesign or composing, arranging or all.
The risk of sales is to run into a never ending learning curve to use new equipent which probably is outdatet already if you buy it and dies after the warranty becomes invalid.

Actually, cloning the dinos is best done by sampling technology and support of complex cloning software.
Samplerobot is such a program and Mimik, used in the Open Labs Neko and Miko machines, is a derivate of Samplerobot.



Even these programs do a lot automatically and over nite, you have to prepare your dinos well for getting the best results as also you need some advanced knowledge to setup such a program properly.
All this is time consuming and steals lifetime, except you are a person who prefers to sit in front of a computer day and nite for cloning gear instead of? doing other things which also don?t have to be music related ... :-)

I?m also pretty sure it?s possible to create great virtual instruments, also 1:1 clones, assumed there?s enough calculation power available and great sounding audio interface hardware w/ no latency issues as also the graphics GUI won?t steal all the cpu and shares RAM too much, which is both the typical showstopper today.

Sonicprojects OPX has knobs in the GUI to switch animation of keyboard and levers to off because it introduces clicks and pops w/ some systems and hosts if animation is activated.

Cloning a Xpander or Matrix 12 in software w/ a excellent sound is a target of many coders and I talked w/ some of these because it?s my interest to get such a clone.
All said, the quantity of modulation connections is a potentially problem as also is the menue based user interface, which isn?t up to date anymore. But if you change this, you never get a Xpander clone.
Additionally, the hardware machines are multitimbral in multi patch mode,- means, there are 6 or 12 synthesizers w/ a complex voice architecture running simultaneously and depending on the loaded patches in the multi-patch slots, the cpu consumption would vary or sometimes dramatically rise.

Now think of several instances of all this in a host,- you?d need a very powerful computer which you don?t find actually in a laptop form or in machines of Muse Research and Open Labs p.ex.,- and other software has to run simultaneously on such a machine too.

If life changes and you have to sell your gear because of whatever it is circumstances,- the gear is gone IMO.
"No software replaces a good piece of hardware", - rules !
In such a case, you have to go other creative ways and that?s possible ever unless you resignate.

If you have the time, adjust all your patches you need carefully w/ the hardware and use Samplerobot to clone the patches.
You?ll be surprised how many of the actual sampling libraries are been done w/ this program and are big sellers.
Have in mind to edit the patches such way, you can use your software sampleplayers edit functions, midi controller features and synthesis features to modify the sound later in possibly any way you like and did w/ the real instrument.
Sample pure waveforms at full level and w/ different filter cutoff settings, envelopes to zero attack, no env-amount on filter, full sustain level and a long release time. Sample different pulse width and do longer samples w/ PWM which you can loop later such way you get longer periods of PWM modulation in a sampled waveform.
There?s so much to think about for a good cloning of the characteristics of a synth.

In any way, you?ll need so much time that in the end you make no deal if you sell a instrument like the Xpander, unless you make the decision to do a cut and accept the Xpander (and/or all the other vintage gear) will be history for you.


Karl schrieb:

--- In xpantastic@..., PeWe <ha-pewe@...> wrote:

Finally, I think it?s a better idea to buy a used Matrix1000 in
addition and if a Xpander exsits already...
200 bucks,- how much time someone can invest for 200 bucks coding a
program or transfering files and rename ?em ?

All is possible in the end, but what a complex work and what for ?

Hi PeWe,

Your right. I mostly worry about how to clone the dinosaur, not if we
should or not.

Karl