Performance feedback was Re: adsr and lfo abolish

[Thomas Hudson]

One interesting approach I thought of this weekend would
be to use CSound or another tool to analyze the level
and spectral response of a live signal. Then use these
responses for control over a VCA and VCF.

I've owned a GR50 guitar synth since they first came
out. After the initial fun of controlling pads and
such (especially mixed w/ straight signal to mask
bad tracking), you start realizing how much of
your playing isn't translated. And I'm not sure
how much could be translated to MIDI, or any other
predictable medium.

Imagine transcribing a Hendrix solo in a MIDI sequencer.
There are just not enough parameters on a synth.
And no clear way to assign them to any performance
input. I know of guitarists "marking out a room" with 
tape on the floor. Standing in various locations and
holding the guitar at different angles gives
different feedback. 

So it seems we need level sensors in performance devices,
along with motion sensors. Does a saxaphone make a different
sound depending on whether the player is leaning forward
or backward. It must, because those sax players are always
rocking back and forth :-)
 
It seems that what we want is "almost repeatability."
If we could introduce some level of chaos that was
pushed towards comformity by parameters of our
playing style, would it feel more like a traditional
instrument? After all, every attempt I've made at
playing sax proved I didn't have the ability to
push the parameters away from total chaos.

There is more to playing than pitch, velocity,
and aftertouch.

Thomas