I'm attempting to emulate something in PD that calls for a 4 pole
Butterworth low pass filter with a cutoff frequency of 10 Hz. I have
very little idea what that is. But I'm wondering if it's possible to
construct something that would do the same thing in PD. If so, anyone
have any ideas on how to make something like that with PD's existing
objects?
-- Pall Thayer palli@pallit.lhi.is http://www.this.is/pallit
On Thu, 26 Oct 2006, Pall Thayer wrote:
I'm attempting to emulate something in PD that calls for a 4 pole Butterworth low pass filter with a cutoff frequency of 10 Hz. I have very little idea what that is. But I'm wondering if it's possible to construct something that would do the same thing in PD. If so, anyone have any ideas on how to make something like that with PD's existing objects?
Had a look at [cpole~] yet?
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Hi Mathieu,
I did take a look at cpole~ and it made no sense to me. I'm trying to
avoid having to learn too much :-)
On 26.10.2006, at 14:07, Mathieu Bouchard wrote:
On Thu, 26 Oct 2006, Pall Thayer wrote:
I'm attempting to emulate something in PD that calls for a 4 pole
Butterworth low pass filter with a cutoff frequency of 10 Hz. I
have very little idea what that is. But I'm wondering if it's
possible to construct something that would do the same thing in
PD. If so, anyone have any ideas on how to make something like
that with PD's existing objects?Had a look at [cpole~] yet?
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
-- Pall Thayer palli@pallit.lhi.is http://www.this.is/pallit
On Thu, 26 Oct 2006, Pall Thayer wrote:
I did take a look at cpole~ and it made no sense to me. I'm trying to avoid having to learn too much :-)
Trying to avoid to learn too much? Then just use four [lop~] in a row and hope that no-one will notice the difference.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Yeah, I thought about that but does one then set each of the lop~'s
for 10Hz or try to spread the 10 between them like:
[lop~ 2.5]*4 or [lop~ 10]*4
?
On 26.10.2006, at 14:46, Mathieu Bouchard wrote:
On Thu, 26 Oct 2006, Pall Thayer wrote:
I did take a look at cpole~ and it made no sense to me. I'm trying
to avoid having to learn too much :-)Trying to avoid to learn too much? Then just use four [lop~] in a
row and hope that no-one will notice the difference._ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
-- Pall Thayer palli@pallit.lhi.is http://www.this.is/pallit
On Thu, 26 Oct 2006, Pall Thayer wrote:
Yeah, I thought about that but does one then set each of the lop~'s for 10Hz or try to spread the 10 between them like: [lop~ 2.5]*4 or [lop~ 10]*4
You want to cut at 10 Hz so you put four of them at 10 Hz.
Each linear filter has a fixed gain for each frequency. The log-log spectrum of one filter drops suddenly around 10 Hz with a straight slope. With several filters in series, the log-log spectra add up, so the slope becomes steeper.
I suppose that you've seen log-log graphs before? for a function y=f(x), they have log(y) and log(x) axes, instead of y and x axes.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Not 100% sure, but I think all filters in the stack should have the same cutoff frequency:
input | [lop~ 10] | [lop~ 10] | [lop~ 10] | [lop~ 10] | output
To be sure, I would personaly use the iemlib filters though the above model "sounds" quite good
Pall Thayer palli@pallit.lhi.is:
Yeah, I thought about that but does one then set each of the lop~'s for 10Hz or try to spread the 10 between them like:
[lop~ 2.5]*4 or [lop~ 10]*4
?
I'm attempting to emulate something in PD that calls for a 4 pole Butterworth low pass filter with a cutoff frequency of 10 Hz. I have very little idea what that is. But I'm wondering if it's possible to construct something that would do the same thing in PD. If so, anyone have any ideas on how to make something like that with PD's existing objects?
Try this filter designer to get coefficients to use in biquad~ objects: http://www-users.cs.york.ac.uk/~fisher/mkfilter/trad.html
Here's some code too, which might be more clear: http://www.musicdsp.org/showone.php?id=227
For 4-pole you'll need to use more than one biquad, I think. I'm not sure how, but you can split it over multiple biquads.
Start with 2-pole for a single biquad, to make things easy.
Steve
Pall Thayer wrote:
I'm attempting to emulate something in PD that calls for a 4 pole Butterworth low pass filter with a cutoff frequency of 10 Hz. I have very little idea what that is. But I'm wondering if it's possible to construct something that would do the same thing in PD. If so, anyone have any ideas on how to make something like that with PD's existing objects?
either use [biquad~] (and your favourite DSP book ;-)) or the ready-made butterworth filters from iemlib.
mfg-asdr IOhannes
Ah yes. My favourite DSP book. That would be um... the PD list!
I'll take a look at the iemlib stuff, thanks.
On 26.10.2006, at 15:13, IOhannes m zmölnig wrote:
Pall Thayer wrote:
I'm attempting to emulate something in PD that calls for a 4 pole
Butterworth low pass filter with a cutoff frequency of 10 Hz. I
have very little idea what that is. But I'm wondering if it's
possible to construct something that would do the same thing in
PD. If so, anyone have any ideas on how to make something like
that with PD's existing objects?either use [biquad~] (and your favourite DSP book ;-)) or the ready- made butterworth filters from iemlib.
mfg-asdr IOhannes
-- Pall Thayer palli@pallit.lhi.is http://www.this.is/pallit
-
IOhannes m zmölnig -
julien breval -
Mathieu Bouchard -
Pall Thayer -
Stephen Sinclair