AW: Keyboard/Midi Frontend (RE: PolyModular Standards)

Haible Juergen Juergen.Haible at nbgm.siemens.de
Fri May 7 13:44:14 CEST 1999


Hi Ingo,

thanks for your answer.

	>I have a Roland A-80 masterkeyboard. The A-90 has polyphonic
aftertouch
	>too. The keys on the Roland have to be pressed really hard to get
	>aftertouch. I already replaced the sensors (before I did that it
was
	>nearly impossible to get aftertouch). In the schemo there's a
resistor
	>with different values for the A-80 and the A-50 (the
	>61-nonweighted-key-version). I changed that resistor to the A-50
value
	>too, which made it a bit easier to get aftertouch, but it's still
hard.
	>I'll have to look into that again.

Reminds me of the infamous JX-8P aftertouch. I remember a Keyboards review
talking about Sumo Ringers having done the beta test ...

What kind of sensor elements did they use ? Are they expensive ? Can they be
mounted under other keyboards, too ? 

	>> I'm a little sceptical about the
	>> Midi slowing down with polyphonic AT, however. Is there a
solution
	>> for that ?
	>
	>Never tried to reach the limits, with normal playing it's ok, I
think.

I never tried with polyphonic AT, so I can just speculate.
One of my favorite playing techiques with "normal" AT is holding a bass note
with the left hand, and using this note to change the AT while I'm
playing fast runs with "light" fingers (no AT force) with the right hand.
The
downside is that with this method, there *is* a conflict between note on
information and AT changes all the time (which wouldn't be the case if
AT is applied after a new note). 

	>My user interface is indeed not very user-friendly. You enter the
	>parameter number (in hexadecimal) with a 16-button-keyboard, two
digits
	>of the display show that number then, two other digits show the
value.
	>Then you can change the value with up/down buttons. There are two
	>buttons for up and down each, one for small and one for large steps
(for
	>instance, for pitch offsets its octave and semitone steps). But it
is
	>very flexible this way. I can add new parameters and change value
ranges
	>without having to change the UI hardware.

This is what I feared. 8-(
Nothing against you, though - *everybody* seems to build that way. Guess I'm
part of a dying breed. I want switches for everything, and no display at
all.
A feature that's hidden behind a menue is a feature that will not be used.
Sorry to say it so directly, and I know I'm a bit extreme in that point.
Strictly IMO comment, that is.
I don't want to force my own preferences onto all of you. I have found an
interesting alternative which will work for myself, if I'm the only one:
I can retrofit my CS-50 with a few of these DIN jacks, bringing out the
V/Hz CVs and the gates, probably even the VCO outputs, and bringing
in external signals at the end of the signal chain. This should work for me,
so don't kill the project out of regard for my extreme point of view.
(No this wouldn't give me all these keyboard modes either, but it would
be a cheap and space-effective solution, so it is tempting.)

	>Which functions do you consider important? 

I'd say all the switched functions, but no continuous functions.
Things like velocity on the filter is something that can be done
in the analogue / modular domain. Sometimes you want it to
affect the cutoff directly, sometimes you want it to scale the
envelope that modulates the VCF, sometimes you just want
it to modulate the Sustain voltage or Decay rate of the VCF-EG.
No need to do this in the processor. Concentrate on what the processor
is good at (and good at interfacing, i.e. simple MUXes that scan 
a set of switches) - that's what my opinion is.
Switches for *everything*. Want to turn 3 voices off because you're
going to re-tune one voice ? Flick 3 switches. You wouldn't want
to search the menues for that kind of operation, really.
(The *worst* that can be done is what we find in later Doepfer
midi interfaces (and my Kawai CA600 piano isn't better here):
Want to switch something? Switch to "learn" mode and feed
in some Midi code. This is the horror for me. As a naive beginner
I thought it was a great way to set the reference note, and Midi
channel, with just hitting one key on a connected keyboard.
After going thru this "easy" procedure a few times, I wondered
why they asume you'd want to use a different midi channel
and reference note with every power-up. Strange, my keyboard
didn't grow more keys in the bass range over night, and I didn't change
my Midi channel setup on workdays and the weekend either.
Sorry for another UI rant, but I just can't stress enough how important
a good UI is.)

This is not, entirely not, meant to be offensive or against your
great offer to use your previously built interface, Ingo. If I were you, 
I wouldn't go into pains of doing a redesign unless a lot of people really
honor you for that work, so don't listen too much to my talk ! (;->)

Regards,

JH.




More information about the Synth-diy mailing list