All about the Roland Jupiter-series

-----Original Message-----
From: Jason 'Fish' Proctor [mailto:jproctor@...]
Sent: Wednesday, January 10, 2001 7:09 PM
To: media@...; Analogue Heaven
Subject: Re: [AH] MKS-80 w/PC-1600x ??


Sounds like the 1600 is pushing stuff out too quick for the MKS to
deal with it. If the 1600 has a delay parameter, try tweaking that.

>I'm using a PC-1600x to program an MKS-80. I noticed that when I move the
>faders all the way up, or all the way down, the parameter displayed on the
>PC-1600x goes all the way up or down, but the parameter on the MKS-80 does
>not. If I repeatively move the fader up a bit then all the way back down,
>I get an assortment of small values on the MKS-80, eg. 1, 3, 4, 7, 2, 3 . .
>. but not zero. In the same way, if I keep moving it up, I get an
>assortment of high numbers, but never 100. Any ideas??

-----Original Message-----
From: N. Kent [mailto:ndkent@...]
Sent: Wednesday, January 10, 2001 10:16 PM
To: analogue heaven; media@...
Subject: Re: [AH] MKS-80 w/PC-1600x ??

> Sounds like the 1600 is pushing stuff out too quick for the MKS to
> deal with it. If the 1600 has a delay parameter, try tweaking that.

I don't believe it has a delay but I may be wrong, also just moving the
slider slower should achieve somthing like that. A delay parameter, if
there was one would seem kind of counterproductive for any kind of live
tweaking, would it not. Sending one single parameter sysex string, even
very quickly shouldn't choke anything. Sure you could space the 7 or
whatever number of bytes the MKS uses out via a delay if there is one,
but thats probably not the problem.

>
> >I'm using a PC-1600x to program an MKS-80. I noticed that when I move

the

> >faders all the way up, or all the way down, the parameter displayed on

the

> >PC-1600x goes all the way up or down, but the parameter on the MKS-80

does

> >not. If I repeatively move the fader up a bit then all the way back

down,

> >I get an assortment of small values on the MKS-80, eg. 1, 3, 4, 7, 2, 3 .

.

> >. but not zero. In the same way, if I keep moving it up, I get an
> >assortment of high numbers, but never 100. Any ideas??
>

For what its worth I have the original PC-1600 and an MKS-80. The
original factory patches had only a patch of some popular MKS-80
parameters not all of them so I fleshed it out to all the parameters in
3 patches.

But speaking of data overload, make sure you don't have some kind of
midi loop or software thru going on. For example just test your MKS with
the Peavy and I guess the keyboard but keep out any interfaces or
computers, see if it works. Theres that switch on the back for the real
programmer, but I don't think that would make it function wrongly
instead of just not function.

Then I'd reload the PC-1600 patch assuming it was the factory one. Maybe
it got scrambled a bit or got updated wrong. You can program the sliders
to output a range other than 1-100 or whatever, some patches need it,
you can do 0-1 if thats what you need. For some parameters on the MKS
there are only values like for 0-7 or whatever. Is the parameter in
question really supposed to go to 100? Some don't. Your Peavy will
should read small number values for some things like waveshapes or
octave detunings.

So *none* of the parameters work corectly with the MKS or just a
specific few while others work correctly? A reload or patch edit of the
slider range on the improperly working ones should fix it if its a
matter of just certain parameters.

As a last resort I guess check out the output of the peavy with a
realtime midi viewing utility. If you use a computer there should be a
share or freeware one out. I find one essential when doing PC-1600
patches to see if whats coming out of the PC-1600 is what I expect it to
be. You shouldn't see anything coming out except short sysex strings
just when you nudge the slider. If stuff is coming out when the slider
isn't moving then you have something in your rig generating unneeded
midi when it shouldn't or something is broken.

nick kent

-----Original Message-----
From: Cap'n F.M. Bleep [mailto:bleep@...]
Sent: Wednesday, January 10, 2001 10:22 PM
To: Sweet Analogue Heaven
Subject: Re: [AH] MKS-80 w/PC-1600x ??

On Wed, 10 Jan 2001, N. Kent wrote:

> > Sounds like the 1600 is pushing stuff out too quick for the MKS to
> > deal with it. If the 1600 has a delay parameter, try tweaking that.
>
> I don't believe it has a delay but I may be wrong, also just moving the
> slider slower should achieve somthing like that. A delay parameter, if

there *is* a delay parameter on there somewhere, but i can't recall where
at the moment (the manual is downloadable from www.peavey.com if you don't
have it). however, the problem might be that the pc1600 is sending 127
increments to the mks instead of the 100 it wants. you can edit the max
and min values per fader, so setting the max to 100 (and the min to 0 or
1, whatever the mks wants) should work...

hope that helps!

bleep.
out.

-----Original Message-----
From: Rob [mailto:cyborgzero@...]
Sent: Thursday, January 11, 2001 12:04 AM
To: analogue heaven
Subject: Re: [AH] MKS-80 w/PC-1600x ??


Er, no, bit/word delay is *NOT* the same as just "moving the slider slower".
Nice try though! ;))

Okay, what you have here sounds like either

1) You need to set the parameters to generate checksum (CS in your string)
OR 2) You need to set a proper offset and/or range for the parameter.

This comes down to really going over the values in the sysex implementation
and checking to see if what happens is what is "supposed" to happen.. I had
some parameters in the same piece of gear that require checksum, require
offset, range or all three!! And the sysex chart doesn't neccesarily give
you all the info you need all the time.. Some sysex charts are missing info.

----- Original Message -----
From: "N. Kent" <ndkent@...>
To: "analogue heaven" <analogue@...>; <media@...>
Sent: Wednesday, January 10, 2001 4:16 PM
Subject: Re: [AH] MKS-80 w/PC-1600x ??


> > Sounds like the 1600 is pushing stuff out too quick for the MKS to
> > deal with it. If the 1600 has a delay parameter, try tweaking that.
>
> I don't believe it has a delay but I may be wrong, also just moving the
> slider slower should achieve somthing like that. A delay parameter, if
> there was one would seem kind of counterproductive for any kind of live
> tweaking, would it not. Sending one single parameter sysex string, even
> very quickly shouldn't choke anything. Sure you could space the 7 or
> whatever number of bytes the MKS uses out via a delay if there is one,
> but thats probably not the problem.
>
<snip>

-----Original Message-----
From: N. Kent [mailto:ndkent@...]
Sent: Thursday, January 11, 2001 5:08 AM
To: analogue@...
Subject: Re: [AH] MKS-80 w/PC-1600x ??

> On Wed, 10 Jan 2001, N. Kent wrote:
>
> > > Sounds like the 1600 is pushing stuff out too quick for the MKS to
> > > deal with it. If the 1600 has a delay parameter, try tweaking that.
> >
> > I don't believe it has a delay but I may be wrong, also just moving the
> > slider slower should achieve somthing like that. A delay parameter, if
>
> there *is* a delay parameter on there somewhere, but i can't recall where
> at the moment (the manual is downloadable from www.peavey.com if you don't
> have it).

I think someone would have encountered the problem by now since its a
reasonably popular combo. And as I said, while not impossible, its
unlikely its going to choke on a slowly moving slider, its not a whole
dump of a patch bank, a single slider shouldn't be choking it unless you
go crazy and adding delay I doubt is the answer. But who knows, the
person in question has a 1600x, I have a 1600


however, the problem might be that the pc1600 is sending 127

> increments to the mks instead of the 100 it wants. you can edit the max
> and min values per fader, so setting the max to 100 (and the min to 0 or
> 1, whatever the mks wants) should work...
>
> hope that helps!
>
> bleep.
> out.

Not quite, I've done that plenty of times by accident while programming
all the other parameters . If you slide past 100 to say 120 the MKS just
stays at 100, it doesn't loop around or bug out... but the advice of
checking out the range on any parameter that doesn't work is well worth
checking out, who knows maybe somehow it got set to 0-7 instead of 0-100.

> Er, no, bit/word delay is *NOT* the same as just "moving the slider

slower".

> Nice try though! ;))

no I did not say it was the same, I said "something like" because in
this case with about 7 bytes or whatever a strings length, is the
chances of just one of those coming out and making the unit malfunction
seems unlikely. Sorry if you misintrerpreted me but I was mentioning
that as almost proof that it wasn't an issue of spacing the MIDI stream.
(while of course it might if many sliders are moving at once and
certainly for bulk dumps that don't work out)

>1) You need to set the parameters to generate checksum (CS in your string)

Huh?? The MKS-80 doesn't utilize checksum for SYSEX control of
parameters. Say you generate it then what are you going use it for???
Throwing in extra bytes is a sure fire way to not get SYSEX to work.

> 2) You need to set a proper offset and/or range for the parameter.

there is no offset on MKS parameters. We've mentioned the range which we
all think is good advice to check out

> This comes down to really going over the values in the sysex

implementation

> and checking to see if what happens is what is "supposed" to happen.. I

had

> some parameters in the same piece of gear that require checksum, require
> offset, range or all three!! And the sysex chart doesn't neccesarily give
> you all the info you need all the time.. Some sysex charts are missing

info.

mmmm... you are guessing here much more than I am. The Sysex on the MKS
80 is very straight forward, no checksums and Peavy does give a sample
bank of about a third of the MKS-80 parameters when you buy the unit.
(Unless this changed for the 1600x) Its somewhat time consuming but
pretty troublefree to flesh out all of the others from the ones you get,
you don't necessarily need charts, nothing tricky is going on on
Roland's side of things. As mentioned, I did it and everything worked
you just need a plan for what slider you want doing what ans some
patience.

nick kent
(not the person with the PC-1600x problem)

-----Original Message-----
From: media@... [mailto:media@...]
Sent: Friday, January 12, 2001 6:17 PM
To: ndkent@...
Cc: analogue@...
Subject: Re: [AH] MKS-80 w/PC-1600x ??



Thanks to everyone for their help.

I'm using a PC-1600x, a Rev. 1 MKS-80, a Studio 5LX, and running OMS 2.3.8
and Cubase Audio XT 3.0.7 PPC on an 8100/110.

>For what its worth I have the original PC-1600 and an MKS-80. The
>original factory patches had only a patch of some popular MKS-80
>parameters not all of them so I fleshed it out to all the parameters in
>3 patches.

Then I think I might be using your patches!! I downloaded them from Dan
Nigrin's website.

Does anyone know where I can get the original MKS-80 patch for the PC-1600??

>But speaking of data overload, make sure you don't have some kind of
>midi loop or software thru going on. For example just test your MKS with
>the Peavy and I guess the keyboard but keep out any interfaces or
>computers, see if it works. Theres that switch on the back for the real
>programmer, but I don't think that would make it function wrongly
>instead of just not function.

Yes, thank you, that solved the problem. When I connected the PC-1600x
directly to the MKS-80 everything worked fine. "STUPID @$%#&!! OPCODE!!"
Yet, I swore to soon. The Studio 5LX wasn't the problem. As it turns out,
Cubase is the problem. It's clearly buggy. It can record parameter values
from 0 to 100, it can play back parameter values from 0 to 100, but it
can't thru parameter values from 0 to 100 (or play them while recording).
This doesn't come as much as a shock as one would think. Until version
3.0.7 Cubase couldn't even record SysEx through OMS.

"STUPID %$#@&*!! STEINBERG!!!"

-----Original Message-----
From: N. Kent [mailto:ndkent@...]
Sent: Friday, January 12, 2001 8:15 PM
To: media@...; analogue@...
Subject: Re: [AH] MKS-80 w/PC-1600x ??


media@... wrote:

>
> Thanks to everyone for their help.
>
> I'm using a PC-1600x, a Rev. 1 MKS-80, a Studio 5LX, and running OMS 2.3.8
> and Cubase Audio XT 3.0.7 PPC on an 8100/110.
>
> >For what its worth I have the original PC-1600 and an MKS-80. The
> >original factory patches had only a patch of some popular MKS-80
> >parameters not all of them so I fleshed it out to all the parameters in
> >3 patches.
>
> Then I think I might be using your patches!! I downloaded them from Dan
> Nigrin's website.

I think they were already up there done by someone else when I
discovered that site, anyway
the original 1600 came with some of the parameters as an example patch
as I've mentioned so that almost shouts to do the rest

> Does anyone know where I can get the original MKS-80 patch for the

PC-1600??

I'd look for the whole factory bank that shipped with it, its got like
1/6th of the DX7, part of a proteus,etc. though if what you have works
its a waste of time since Peavy's bank isn't anywhere near a complete
set of MKS parameters

>
> >But speaking of data overload, make sure you don't have some kind of
> >midi loop or software thru going on. For example just test your MKS with
> >the Peavy and I guess the keyboard but keep out any interfaces or
> >computers, see if it works. Theres that switch on the back for the real
> >programmer, but I don't think that would make it function wrongly
> >instead of just not function.
>
> Yes, thank you, that solved the problem. When I connected the PC-1600x
> directly to the MKS-80 everything worked fine. "STUPID @$%#&!! OPCODE!!"
> Yet, I swore to soon. The Studio 5LX wasn't the problem.

Actually I have Studio 5LX now (and a Studio 4 before it when I was
messing with the Peavy which shares most oif the 5's concepts). The
thing is the manuals are kind of thick and get into all kinds of obscure
features not how to get something simple running. This might be your
problem, I know I've run into other people on the net who like I did
essentially thought the same---

The matrix grid with ins and outs of all the connections on your
interface and the names of all the gear you registered in OMS that comes
up in OMS on your computer **looks like something you are supposed to
connect**. Like you seem to be encouraged to connect your masterkeyboard
to say your modules. But actually if you are using a software sequencer
on your computer you should connect nothing on that Matrix. Maybe save
nothing as patch #0 on your Studio 5 with nothing routed in OMS after
you've set up all your synths with OMS (so your sequencer knows whats
where).

Your sequencer should be doing the patching like from your master
keyboard to the modules when you are using the sequencer not you
manually patching with the OMS matrix interface. You'd only use that
matrix interface in OMS for routing MIDI when *not* using a software
sequencer. There are probably some exceptions but you shouldn't really
use that matrix at all with a software sequencer... but seeing it so
prominant when flipping through the manual while scratching your head
about how to set all this stuff up sure encouraged me think I was
supposed to patch my synths together with it. (well I do with some
Studio 5 patches I made for playing without booting the computer, but
when I'm sequencing I stick to using an empty patch matrix interface
that I have set up as patch #0 on the Studio 5)

As it turns out,

> Cubase is the problem. It's clearly buggy. It can record parameter values
> from 0 to 100, it can play back parameter values from 0 to 100, but it
> can't thru parameter values from 0 to 100 (or play them while recording).
> This doesn't come as much as a shock as one would think. Until version
> 3.0.7 Cubase couldn't even record SysEx through OMS.
>
> "STUPID %$#@&*!! STEINBERG!!!"

I never personally found cubase feature rich enough to follow (but I
like their VST and ASIO standards). So I don't know the answers. Be
aware that a lot of apps default to filtering out SYSEX though it can be
turned back on. I don't know if OMS defaults but you can make your
interface filter it away via OMS something you don't want to do in this
case. Anyway with something like a Studio 5 LX and a PC1600 with its own
thru jack and an MKS there are a couple ways to cable everything so you
might want come up with a better cable path. While maybe not the
optimum, maybe you can have your master keyboard out to the in on the
Peavy then the Peavy thru to the Studio 5. The Peavy has a Merge. This
way Cubase has to see your playing and your SYSEX all coming in on the
same port. Then connect the MKS on some other Studio 5 port.

The MKS I believe echos the SYSEX back out it's MIDI out so you want to
be very wary if you have the midi out on the MKS going to back into the
MIDI interface. You need it connected for say a software patch editor.
If you are aware that the MKS is spitting all the SYSEX back out the out
then you can probably manage it. The thing is your software sequencer
might default to resending the SYSEX being echoed by the MKS mistaking
it as another controller keyboard which can be big trouble (it forms
loop) so you have to set things up right.

I don't know whats going on with cubase but I'd think it would just
record the sysex or not or choke, not interpret it to filter out values
burried in a Sysex string without a lot of special code. We aren't
talking about CC#s where this sort of filtering is commonplace. The
parameter values for the MKS contained in the SYSEX are within a string
of several bytes which changes from company to company, I don't think
its realistic that Steinberg is saying hey lets interpret this whole
sysex string unique to the MKS just to drop the higher values but leave
the rest alone so it still makes the MKS respond, something more messed
up is going on. (or maybe you are looking at hex numbers when you think
you are seeing decimal numbers)

nick kent

-----Original Message-----
From: media@... [mailto:media@...]
Sent: Friday, January 12, 2001 10:20 PM
To: ndkent@...
Cc: analogue@...
Subject: Re: [AH] MKS-80 w/PC-1600x ??

>> I'm using a PC-1600x, a Rev. 1 MKS-80, a Studio 5LX, and running OMS
>> 2.3.8 and Cubase Audio XT 3.0.7 PPC on an 8100/110.

>> Yes, thank you, that solved the problem. When I connected the PC-1600x
>> directly to the MKS-80 everything worked fine. "STUPID @$%#&!! OPCODE!!"
>> Yet, I swore to soon. The Studio 5LX wasn't the problem.
>
>Actually I have Studio 5LX now (and a Studio 4 before it when I was
>messing with the Peavy which shares most oif the 5's concepts). The
>thing is the manuals are kind of thick and get into all kinds of obscure
>features not how to get something simple running. This might be your
>problem, I know I've run into other people on the net who like I did
>essentially thought the same---

The Studio 5LX works fine without Cubase, using a patch that sends the
output of the PC-1600x to the MKS-80. However, with Cubase and a blank
patch the data doesn't thru properly.

>> As it turns out, Cubase is the problem. It's clearly buggy. It can
>> record parameter values from 0 to 100, it can play back parameter values
>> from 0 to 100, but it can't thru parameter values from 0 to 100 (or play
>> them while recording). This doesn't come as much as a shock as one would
>> think. Until version 3.0.7 Cubase couldn't even record SysEx through
>> OMS.
>>
>> "STUPID %$#@&*!! STEINBERG!!!"

> The MKS I believe echos the SYSEX back out it's MIDI out so you want to
>be very wary if you have the midi out on the MKS going to back into the
>MIDI interface.

In mode III (SysEx mode), my MKS-80 seems to echo everything thru its MIDI
out -- notes and all. Occasionally it spits a string that screws up my
PCM-80. So I have mine disconnected.

>I don't know whats going on with cubase but I'd think it would just
>record the sysex or not or choke, not interpret it to filter out values
>burried in a Sysex string without a lot of special code.

You would think so, but apparently that isn't the case. Cubase seems to
record and playback SysEx without problems, but it can't thru the same
exact data. I assume that as far as the Studio 5LX "sees" the data coming
from the Macintosh, it would not distinguish between recorded data being
played back by Cubase and data being passed thru Cubase. So Cubase must be
at fault.

>We aren't talking about CC#s where this sort of filtering is commonplace.
>The parameter values for the MKS contained in the SYSEX are within a string
>of several bytes which changes from company to company, I don't think
>its realistic that Steinberg is saying hey lets interpret this whole
>sysex string unique to the MKS just to drop the higher values but leave
>the rest alone so it still makes the MKS respond, something more messed
>up is going on.

I don't think its specifically intentional, although I agree something
messed up is going on!! It's only intentional in that Steinberg makes more
of an effort to market new features than fix existing products. Nor do I
think Cubase is interpreting the data. It seems to me that it's simply
dropping strings, and somehow always dropping the very last string
(representing 0 or 100).

>(or maybe you are looking at hex numbers when you think
>you are seeing decimal numbers)

That's not it -- I'm looking at the MKS-80's LCD display.

Thanks again for your help.