Grumbling Feathers

Im creating a patch for adaptive tuning in Max Msp and am translating a PD patch into Max Msp. That's all fine but im not sure how to go about calculating roughness for every sine pair in a given spectrum. 

I need to take the outgoing FFT frequencies compile them into a list and then perform the calculation for roughness on every pair of frequencies within the spectrum. Im guessing I need to implement a loop function using count or uzi but im not 100% sure quite how. The aim is to have a list of all the ratios between every sine tone pair in the spectrum with their corresponding roughness values.

Im creating a patch for adaptive tuning in Max Msp and am translating a PD patch into Max Msp. That's all fine but im not sure how to go about calculating roughness for every sine pair in a given spectrum. 
    I need to take the outgoing FFT frequencies compile them into a list and then perform the calculation for roughness on every pair of frequencies within the spectrum. Im guessing I need to implement a loop function using count or uzi but im not 100% sure quite how. The aim is to have a list of all the ratios between every sine tone pair in the spectrum with their corresponding roughness values.

Many many thanks in advance


Hi all,
Im creating a patch for adaptive tuning in Max Msp and am translating a PD patch into Max Msp. That's all fine but im not sure how t

dissonance-curves-in-max-msp

You could do it the way you say, but it would be more efficient to do that as a matrix operation (a genuine matrix multiply-and-add)-- I'll see if I can post an example later.

You could do it the way you say, but it would be more efficient  to do that as a matrix operation (a genuine matrix multiply-and-add)-- I'll see if I can post an example later.

Thanks Floating Point that's intriguing is that using the matrix object?

Yes, use the matrix object, which would contain a table of frequency differences . Sorry, haven't gotten around to doing an example for you, but instead I found some C++ code by Nick Collins which I've adapted to javascript. It's after work by Bill Sethares. It's a naive, incomplete adaptation, but I might see if I can do a gen~ implementation using his entire methodology if you're interested. Have a look at the attached and see if it's of use.

Dissonance_test.zip

Yes, use the matrix object, which would contain a table of frequency differences . Sorry, haven't gotten around to doing  an example for you, but instead I found some C++ code by Nick Collins which I've adapted to javascript. It's after work by Bill Sethares. It's a naive, incomplete adaptation, but I might see if I can do a gen~ implementation using his entire methodology if you're interested.    Have a look at the attached and see if it's of use.


OK, here's a more complete, functioning example-- hope it is of use.

Dissonance measure.zip

OK, here's a more complete, functioning example-- hope it is of use.


Nice patch FP! There is also a package dedicated to this kind of analysis: zsa.descriptiors. In particular you can look at zsa.flatness~

Hi Michelez, oh yes I'm aware of the zsa descriptors, have used them in the past. But my patch is to do with calculating the 

 of a particular spectral profile using an objective measure of "dissonance" as outlined in this paper:

http://sethares.engr.wisc.edu/paperspdf/cmj98.pdf

Hi Michelez, oh yes I'm aware of the zsa descriptors, have used them in the past. But my patch is to do with calculating the dissonance of a particular spectral profile using an objective measure of "dissonance" as outlined in this paper:

http://sethares.engr.wisc.edu/paperspdf/cmj98.pdf

oh ok, I will take a look at the paper... thanks for sharing!

Dissonance Curves in Max Msp