[PD] Noise in bp-filter

[Claude Heiland-Allen]

Hi,

I would do it by having multiple filters in parallel, one for each 
overtone, and mix between their outputs, instead of changing a single 
filter's parameters (which is problematic).

This would use more CPU power, but [bp~] is pretty cheap.

It would also make possible mixing overtones in different proportions, 
which may or may not be useful.


Claude

On 18/02/2023 18:42, William Brent wrote:
> Did you smooth the frequency changes with a [line~] object or 
> something else before sending to [vcf~]'s 2nd inlet? You can do the 
> same to make gradual transitions for Q changes, but since the Q inlet 
> doesn't take a signal you'll have to use [line] instead.
>
> On Sat, Feb 18, 2023 at 8:19 AM Daniel Martin via Pd-list 
> <pd-list at lists.iem.at> wrote:
>
>     As you say, the problem is that I jump from a frequency to the
>     next non-continuosly. Is there a way to do that without
>     interrupting the signal and without noise?
>     El viernes, 17 de febrero de 2023, 23:52:22 GMT+1, Peter P.
>     <peterparker at fastmail.com> escribió:
>
>
>     Daniel, can you try the [vcf~] object instead of [bp~]? The noise you
>     hear most likely happens when you change the filter frequency
>     non-continuously.
>
>     * Daniel Martin via Pd-list <pd-list at lists.iem.at> [2023-02-17 20:15]:
>     > Dear PD-Users,                         I'm having a problem with
>     a patch that filters overtones from a sample. I use for that a bp
>     filter. The center values for frecuency, Q and output volume are
>     changed according to which overtone one wishes to hear. It is so
>     configured, that one can select between overtones 3 to 17,
>     pressing keys d, r, f, t, g, y, h, u, j, i, k, o, l, p and ñ (I
>     have a spanish keyboard). The problem is when I want to change
>     between overtones: one hears a noise. I solved that partially by
>     reducing the output volume of the bp filter to 0, than changing
>     the filter parameters values and then raising up the volume again
>     to 1. That works only with adjacent overtones (for ex. overtone 3
>     to 4 or 3 to 5), but with larger jumps (for ex. 3 to 15), the
>     noise comes again, the larger the jump, the stronger the noise.
>

-- 
https://mathr.co.uk