January 29, 2017
I recently bought the books and have lots of fun so far! But I think I spotted a mistake or maybe I just don't understand what's going on...
I tried to come up with the difference equation for the fractional delay. Let x(n) be the input to the delay and y(n) the output (no wet/dry mix here, it's fully wet). We want the input signal to exit the output after D+d samples where D is the integer part and d the fractional part. So we have to delay the input x(n) and the fedback output signal feedback*y(n) by D+d samples using D unit delays and linear interpolation with d.
The difference equation I obtained is a bit lengthy but it contains interpolation between input samples and output samples. Taking a look at your code on pages 238 and 239 it seems the difference equation to your implementation is y(n) = x(n-D) + feedback * (d*y(n-D) + (1-d)*y(n-D-1)).
So only the feedback path contains the interpolation. That means that the input sample is put out after D samples but the fedback output sample after D+d samples and this is not correct!
Am I on the wrong track here? I hope we can get this clear!
Most Users Ever Online: 36
Currently Browsing this Page:
Guest Posters: 1
Newest Members:Merril Bradshaw, BillPlunkett, Laxmi123, Pajczur, michaelwayneharwood, RickM, rainbow wind, Tekkerue, CRay, clc
Moderators: W Pirkle: 185
Administrators: Tom: 66, JD Young: 80, Will Pirkle: 0, W Pirkle: 185