posted 04 March 2005 02:37            

sorgenkind.kym

some questions regarding the attached sound:
- why is it that the scale doesn't influence the feedback parameter at
all? How can you accomplish exactly that?
- Are the AmpFollower and the SoundToGlobalController the right
sounds to translate the scale of the input into a control signal that
can be used in the feedback field of the delay?
- How do you know before how to scale the generated control signal
appropriately (for the feedback parameter), so that you make optimal
use of the control area (0...1), without having to rely on 'trial and
error'?

So what I need is a smoothed (reduction to a kind of ‘envelope’) and appropriately scaled input amplitude for the feedback of the delay.

Another issue: is it possible to use the ‘nextChaotic’ expression with a threshold? That is, if the threshold is exceeded and outputs a ‘1’, the next chaotic value is generated.
Thanks.

posted 04 March 2005 12:16            

I think that the output of the AmpFollower is very small so that the amount of feedback is quite small. I think you may be overlooking that the Delay will output a delayed version of the signal even when feedback is zero.

--- Are the AmpFollower and the SoundToGlobalController the right
sounds to translate the scale of the input into a control signal that
can be used in the feedback field of the delay?

The SoundToGlobalController is not necessary, but it is helpful so that you can see the output of the AmpFollower. Since the AmpFollower outputs the RMS squared of the signal, it will have very small output levels. You can get a more accurate measurement of the amplitude by taking the square root of AmpFollower output (add "sqrt" after where you have pasted the AmpFollower).

The level may still be very small, so you may want to add a gain or compressor in the signal path.

--- How do you know before how to scale the generated control signal
appropriately (for the feedback parameter), so that you make optimal
use of the control area (0...1), without having to rely on 'trial and
error'?

There really isn't a very good way to know the levels in advance when doing live performance, since too many things can change the level of the signal: distance from microphone, pre-amp settings, the moods of the performers, etc.

If you add a gain or compressor to the signal, then you can change the signal level only (and keep your feedback settings constant). The compressor is even better since that can reduce the variation in signal level that the AmpFollower sees if the audio input level changes.

---Another issue: is it possible to use the ‘nextChaotic’ expression with a threshold? That is, if the threshold is exceeded and outputs a ‘1’, the next chaotic value is generated.

Yes, you could use:

posted 06 March 2005 02:31            

That is what I did. But the strange thing was that the gain didn't seem
to have any influence. What did have an influence were very large settings of the feedback. But you want to limit that to control signal
ambitus. Hmmm, I'll try again, probably a little mistake somewhere, it
should work better with a gain sound.