Hi! Nice day to everyone:

I´m want to know about one effect named "convolve".
I listen the effect a lot of times in works created in MAX/MSP and I want to reproduce it in Pd.

I try to find an block diagram to understand the action and elements involved, but I haven´t lucky. I found a lot of mathemathical expressions, but was very complex.

My question is:

The effect "convolve" have other name?
Sombody have an block diagram to explain the action?
"Vocoder" it´s the same than "convolve"?

I hope you can help me. Thank´s to everyone.

->
convolve ->
->

=

-> rfft ->
* -> irfft ->
-> rfft ->

where * is complex multiplication (a+bi) * (c+di) = (a*c-b*d)+(a*d+b*c)i

http://en.wikipedia.org/wiki/Convolution

"convolve" is the operation of "convolution", which in the time domain is roughly equivalent to two signals getting into each other's face. It's pretty laborious to calculate, but in the frequency domain it is performed by a simple multiplication (as in ClaudiusMaximus' post)

Ah, didn't see the wikipedia link.

The easy way to do it is to download and install fftease library

There are a number of handy objects here if you want to avoid the confusion of fft
http://www.sarc.qub.ac.uk/~elyon/LyonSoftware/MaxMSP/FFTease/

Am I allowed to cheat?

S

ps. for universal binaries of this lib, just do a search on this forum and theres alink to download the zip

Hi!!! Thank you to everyone:

I´m very new in this stuff of electronic music and I must learn a lot of theory about the acustic phenomenon (it´s the correct way to say it?).

Thank you again. Have a nice day.

@boonier said:

There are a number of handy objects here if you want to avoid the confusion of fft
http://www.sarc.qub.ac.uk/~elyon/LyonSoftware/MaxMSP/FFTease/

nice stuff thanks!
anyway i did not already manage to performe what i want
looks loke a holly grail i'll not reach that soon

@ClaudiusMaximus said:

->
convolve ->
->

=

-> rfft ->
* -> irfft ->
-> rfft ->

where * is complex multiplication (a+bi) * (c+di) = (a*c-b*d)+(a*d+b*c)i

http://en.wikipedia.org/wiki/Convolution

so if i understand right, programing:

in1-> rfft -> A in2-> rfft ->C
-> Bi ->Di

and then (A*C-B*D)-> irfft -> out
(A*D+B*C)->

should make a convolution of in1 and in2 ??

Hi,

I was just wondering if you managed to get convolution reverb working using this method? My maths knowledge is next to none, so it would be great if you could explain how to achieve this effect with FFTease? (I've installed it, examples sound great, but how do I make it work as a convolution reverb...?) I just need a processor to take a clean sample and an impulse response and perform convolution on these two.

By the way, this is my first post and I've been using Pure Data for about two weeks! I'm simply loving it!

Many thanks!

Vytis

check the Help browser for example I08.pvoc.reverb

I think this is a good starting point...good luck!

anyone knows which window (hann, hamming, rectangular, gauss, etc) is the best if you want a convolution with a wide range of formants?

Hann is not so good in real time, I noticed, because the sound generated has just one formant, and the result is very similar to a resonant filter...

if anyone has a clue, please share it!

There's an external in PD-Extended that does partitioned convolution. You can find it in the Help Browser under:

5.reference/bsaylor/partconv~-help.pd

In that help the author demonstrates a simple reverb.

thanks, saturno.