< Back to IRCAM Forum

Summed curves (perlin-esque)

Hello all,

I’m interested in utilising graphic representation (in a variety of ways as musical data) through BPFs and BPF-Libs. I’m seeking to create a final summed curve from an initial input (drawn or via x,y list - line or spline) curve, plus the addition of self-similar, but modified, curves with double the frequency (occurrences) that will ultimately produce a summed curve (basically a form of 1d Perlin noise).

After the initial curve, subsequent curves should double in double the frequency, halve in amplitude. As an added treat it would be wonderful to be able to adjust separate amplitude for each frequency (‘persistence’ in Perlin terms, specifying the amplitude’s increase or decrease).

See the attached to show what I’m after:

There’s probably a very simple solution utilising scaling and mathtools especially given OM’s powerful suite of BPF/BPF-libs (possibly BPF-INTERPOLX too?) but I’m just drawing a blank at formalising the problem in OM terms. ANY assistance would be really appreciated.

Thanks in advance, and good wishes to all,

Dear Krhes,

Could it be something like this(?):

(It’s too bad you didn’t attach your original bpf)

However, what are you describing as a perlin-esque effect (cf. sorry in french:
Bruit de Perlin — Wikipédia) is more like you should use the om chaos library.
(Here the old documentation:)
PW-Chaos.pdf (284.5 KB)

It’s some kind of a brownian noise, instead of a exact summed curves, no?

Anyhow here is the patch. (made it fast, so maybe some issues…)
Patch 6.omp (92.3 KB)

Hope this helps


Thank you so much Karim,
I’ll look through this over the next day or two in detail and I’m sure it will be of great use (why I didn’t think of looking at OM-Chaos I don’t know - ugh!). I really appreciate your help (shouldn’t you have been awarded a legion d’honneur for all your OM work by now?!?). Already, using different input curves I’ve had some interesting (to me at least!) shapes…!

1 Like