Hello list,
I am trying my hand at some basic FFT stuff. I am designing a cross synthesis patch (see attached) where the amplitude from one signal is combined with the phase of another signal and fed into [rifft~].
The problem is that the polar to cartesian conversion using [poltocar~] sounds really bad. I worked out the trig formulas for what is going on, and found that using [expr~] to handle to polar to cartesian conversion sounds much better.
Although I am happy with the [expr~] results, I am worried that I am not using [poltocar~] correctly, or understanding how to use it.The help files are very vague when it comes to the FFT objects in pd, so I was wondering if someone here could enlighten me as to what I'm doing wrong.
Thanks and have an awesome day!
-Reed
if you're using pd extended 0.43, there's a bug that just ruins everything
2015-07-21 15:11 GMT-03:00 Reed Perkins reedperkins32@gmail.com:
Hello list,
I am trying my hand at some basic FFT stuff. I am designing a cross synthesis patch (see attached) where the amplitude from one signal is combined with the phase of another signal and fed into [rifft~].
The problem is that the polar to cartesian conversion using [poltocar~] sounds really bad. I worked out the trig formulas for what is going on, and found that using [expr~] to handle to polar to cartesian conversion sounds much better.
Although I am happy with the [expr~] results, I am worried that I am not using [poltocar~] correctly, or understanding how to use it.The help files are very vague when it comes to the FFT objects in pd, so I was wondering if someone here could enlighten me as to what I'm doing wrong.
Thanks and have an awesome day!
-Reed
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
it's been corrected in the new cyclone library, that you can download with the deken plugin, and it was working just fine with pd extended 0.42, so it's a pd extended 0.43 thing
I was actually the one who raised it up an issue with both poltocar~ and cartopol~, that the signs were inverted. It actually worked fine for spectral processing, even though it wasn't correct. But then only one of the objects were corrected in 0.43 and everything was ruined :) it was better before
2015-07-21 15:35 GMT-03:00 Alexandre Torres Porres porres@gmail.com:
if you're using pd extended 0.43, there's a bug that just ruins everything
2015-07-21 15:11 GMT-03:00 Reed Perkins reedperkins32@gmail.com:
Hello list,
I am trying my hand at some basic FFT stuff. I am designing a cross synthesis patch (see attached) where the amplitude from one signal is combined with the phase of another signal and fed into [rifft~].
The problem is that the polar to cartesian conversion using [poltocar~] sounds really bad. I worked out the trig formulas for what is going on, and found that using [expr~] to handle to polar to cartesian conversion sounds much better.
Although I am happy with the [expr~] results, I am worried that I am not using [poltocar~] correctly, or understanding how to use it.The help files are very vague when it comes to the FFT objects in pd, so I was wondering if someone here could enlighten me as to what I'm doing wrong.
Thanks and have an awesome day!
-Reed
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Thanks I was indeed using 0.43. Also, I didn't know what deken was. So cool! Thanks for your help.
-Reed
On Tue, Jul 21, 2015 at 12:39 PM, Alexandre Torres Porres porres@gmail.com wrote:
it's been corrected in the new cyclone library, that you can download with the deken plugin, and it was working just fine with pd extended 0.42, so it's a pd extended 0.43 thing
I was actually the one who raised it up an issue with both poltocar~ and cartopol~, that the signs were inverted. It actually worked fine for spectral processing, even though it wasn't correct. But then only one of the objects were corrected in 0.43 and everything was ruined :) it was better before
2015-07-21 15:35 GMT-03:00 Alexandre Torres Porres porres@gmail.com:
if you're using pd extended 0.43, there's a bug that just ruins everything
2015-07-21 15:11 GMT-03:00 Reed Perkins reedperkins32@gmail.com:
Hello list,
I am trying my hand at some basic FFT stuff. I am designing a cross synthesis patch (see attached) where the amplitude from one signal is combined with the phase of another signal and fed into [rifft~].
The problem is that the polar to cartesian conversion using [poltocar~] sounds really bad. I worked out the trig formulas for what is going on, and found that using [expr~] to handle to polar to cartesian conversion sounds much better.
Although I am happy with the [expr~] results, I am worried that I am not using [poltocar~] correctly, or understanding how to use it.The help files are very vague when it comes to the FFT objects in pd, so I was wondering if someone here could enlighten me as to what I'm doing wrong.
Thanks and have an awesome day!
-Reed
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Alexandre Torres Porres -
Reed Perkins