Hi there,

I'm trying to implement the self modulating fm scheme explained at section 4 of this article : http://hci.rwth-aachen.de/materials/publications/poepel2005b.pdf.

Here is a very simple patch that tries to use [fexpr~], but doesn't seems to work like described in the article. Don't see why.

Could someone help me removing the mud I have in the eyes ?

Thank you,

Nau

http://www.pdpatchrepo.info/hurleur/self_fm_with_fexpr.zip

Yop,

thank you again sebfumaster. I understand your patch excepted for the upsampled part. I can't understand why you do that.

Would you explain that to me ?

Thank you.

Nau

just for the reasons described here http://crca.ucsd.edu/~msp/techniques/latest/book-html/node114.html

Thank you sebfumaster, your answer coudn't be more efficient !

Nau

Yop,

sorry, but when I read

If the high-frequency content of the input signal does turn out to give unacceptable distortion products, in general it is more effective to increase the sample rate than the number of points of interpolation. For periods greater than 4 samples, doubling the period (by doubling the sample rate, for example) decreases distortion by about 24 dB.

, I understand that it is talking it is "real" sample-rate doubling, not upsampling using linear interpolation at the inputs of the subpatch, as you do.

Sorry, I'm not at high level in dsp, but is there a way to evaluate the benefits of such an upsampling method ? I mean with real signals, statistically ?

Thanx,

Nau

hmmm... I'm not too hot on dsp either, you might be right in that linear interpolation won't do too much here. the reason I did it was mainly this:

If we run our delay at a sample rate of 88200 Hertz instead of the standard 44100, we will get only about 1 dB of roll-off all the way up to 20000 Hertz.

at the bottom of the page.

as a side note, it would be cool if pd had a cubic upsampling scheme for inlet~ and outlet~

I've been thinking about this and I think a better way for this upsampling would probably just be to put an appropriate oversampling lowpass (FIR?) filter after a sample-and-hold type inlet~ but I don't have the dsp knowledge to design such a thing.
http://yehar.com/blog/wp-content/uploads/2009/08/deip-original.pdf

Thank you, I'll have look at it.

oo, just found this too https://ccrma.stanford.edu/~jos/Interpolation/