attack simulation problem

Aug 27, 2012 at 2:59pm

attack simulation problem

Hello to everyone, I had a problem that I can’t understand. I was following sone examples on a book and I’ve done this patch

– Pasted Max Patch, click to expand. –

As written, I can’t understand the difference between the two processes : how a gain could “limit” the feedback?


Aug 27, 2012 at 4:19pm

I’m not 100% sure I understand the question, but if you are asking about a reciprocating delayline, with control over the feedback amount, then this is how to do it (using your [tapin~]/[tapout~] pair):

– Pasted Max Patch, click to expand. –

If the feedback signal is summed with the input, then that signal must be scaled to < 1. Otherwise the system will reverberate infinitely, and even explode if you exceed a scalar of 1.


Aug 27, 2012 at 4:50pm

And here is a patch that is almost certainly not the correct pedagogical answer…but it’s fun. It also has the nice property of being (almost?) impossible to blow up.

– Pasted Max Patch, click to expand. –
Aug 28, 2012 at 6:14am

it can “limit” the number of noticeable delays, because there is a limit where the
human ear – especially under the condition that the original of the delayed audio
(or other audio) still plays – wont no longer hear them.

if gain in a feedback loop is set to * 0.5, the gain of the second loop is at * 0.25
and the 8th is already at * 0.0078125, which is no longer hearable even when
nothing else plays and your stereo is at full power.

so using * 0.5 will … feel like … 4-5 taps.

if gain is set to * 0.1, the second delay is at * 0.1 and the third one is already
at 0.01 – which means -40 db/A compared to the current full power.
you wont be able to hear that. you will hear only one tap.

otoh, using a gain of 0.99999 or less will make 100% sure that your feedback
loop will never ever become continioulsy raising.



You must be logged in to reply to this topic.