Anybody who enjoys graphs might like to fiddle with this one..
It demonstrates that a unit-impulse signal input to a filter, outputs the impulse-response..
..a brick-wall filter giving us the sinc function.
Only needs vanilla
Impulse response
Anybody who enjoys graphs might like to fiddle with this one..
It demonstrates that a unit-impulse signal input to a filter, outputs the impulse-response..
..a brick-wall filter giving us the sinc function.
Only needs vanilla
That's a nice illustrative patch, moog1. Such demo's can be equally useful as mathematical formula's, when it comes to understanding dsp.
Your patch shows two responses: one around index zero and one around index N/2. If you take out the window and overlap, only the response around index zero remains. Then center the unit impulse at index N/2, and the response is centered as well. See attached.
Katja
@katjav said:
If you take out the window and overlap, only the response around index zero remains. Then center the unit impulse at index N/2, and the response is centered as well..
Katja
Thankyou .. now it's close to the true theoretical impulse response of the filter
I've been playing around with the patch, as I was trying to get an accurate 'filter kernal' from it..
It certainly gives insight into how it all functions. I've increased the fft block size to 4096, and added a routine to copy to a centered kernel of 128 samples.
Added in a 'bp object' routine for interest as well .. bit more lipstick too
..still vanilla
is there any particular reason why you don't use a bandlimited impulse?
Oops! Looks like something went wrong!