What is the easiest resonant lowpass filter to point students toward?

bpf~ is bandpass. As far as I can see, so is bob~. (C'mon... any dance musicians will want a Moog-style lowpass filter...)

As it stands right now, I can tell the students to add the mmb abstractions into their search paths and use [filtercoeff.mmb] --> [biquadm.mmb~] and that works great. And I don't mind to teach them about coefficients (it's a synthesis theory class -- I won't ask them to do the math themselves, but I do want them to be aware that coefficients exist and more or less what they mean).

But I can't shake the feeling that there must be an easier way.

hjh

I've put together mmb's [filtercoeff.mmb~] and [biquad.mmb~] abstractions in one hand abstraction with two versions. One is [multiFilter_abs~] and the other is [omniFilter_abs~]. You can get them here https://github.com/alexdrymonitis/filter_abstractions
Also, I've elimited the use of [expr~] which is used quite a bit in mmb's abstractions, as I think it's rather CPU hungry.

Ah, OK, that looks quite useful.

If it sounds occasionally like I'm a bit cross with some of these questions, it's because of things like... this afternoon, I planned to largely finish my class slides for the filters lecture (brand-new class, so a lot of the material is new). Instead, I lost about 45 minutes trying to make a demo of biquad -> SuperCollider's SOS (second-order filter section) and then another two hours playing with various filter types in PD. So we're at the time when normal people get to go home from the office, and I haven't even started on what I was supposed to be working on.

I get the feeling sometimes that Pd is a bit stubborn about the "vanilla" concept, at the expense of fundamental usability. I'm at pains to imagine a convincing reason why a modern audio programming environment should not have a 2nd-order resonant low pass filter built into the basic distribution. MSP has lores~. SC has RLPF. Pd has... "wish you luck finding an extension for it" (where the saving grace is the user community on this forum).

Sorry. Just, one wasted afternoon later, I'm in a somewhat less than celebratory mood. I very much appreciate the simpler abstractions. Thanks!

hjh

I think bob~ is actually a lowpass filter... at least, it sounds like it at low frequencies and you can still hear the fundamental when you use it at higher frequencies generally.. it is based on the moog ladder filter and that filter is lowpass afaik.. the moog bandpass module in the original was an add-on that combined the high-pass and lowpass filters I think.

This screenshot of a spectrogram recorded by alternating bob~ with a biquad low pass filter shows:

• bob~ attenuates both low and high frequencies somewhat.
• The biquad attenuates high frequencies significantly, but not low (the stronger band in the middle is resonance from Q).

bob~ doesn't sound to me like low pass, nor does its observed frequency response look like low pass.

hjh

@ddw_music here https://github.com/pure-data/pure-data/tree/master/extra/bob~ miller refers to the "cutoff frequency" which would be an odd term to use for a bandpass filter.. if you look at bob~ without any resonance it definitely behaves like a a lowpass.. I think that as the resonance goes up the pass-band might get attenuated? most discussions of the moog ladder filter refer to the lowpass filter (rather than the High-pass one) and since this certainly is not a high-pass filter I would say that in all likelihood it's an emulation of the lowpass species.
If you want to go for instructional clarity I agree that maybe other filters look more like "lowpass" filters in spectrograms tho (but bob~ still sounds like a lowpass to me imho)

probably the right outlet of vcf~

Fair enough about passband attenuation -- sure, it's legit for a filter either to boost the resonance or to attenuate elsewhere.

in all likelihood it's an emulation of the lowpass species.

Rant time.

See, but this gets to my broader point: Why do you have to guess what kind of filter it is?

Why does the help patch not simply say explicitly that it's a lowpass filter?

This shouldn't even be a conversation. Just add one word to the help patch, and then the help answers rather than raises questions.

Same problem with [vcf~]'s help patch (re: "probably the right outlet of vcf~"):

Outlets:
0 signal: - the filtered signal (real part).
1 signal: - the filtered signal (imaginary part).

Now let's suppose you're a digital media artist with some degree of technical sophistication, but not quite enough mathematics to understand Julius Smith's digital filters book.

How, actually, is one supposed to guess, based on the information in the help patch, that real = bandpass and imaginary = lowpass? Oh right. I'm gonna spend a couple of months to grok Hilbert transforms until it finally dawns on me. Now that is user-friendly documentation.

If in fact it really is the right outlet of vcf~ (and it appears to be), then I shouldn't have to open the thread on the forum at all. I should be able to find the answer in the help. (I did look at the help patch before asking, but plainly the help is not written for PD's actual user base.)

FWIW I have a similar complaint with SuperCollider's documentation -- quite often, details are either missing or stated in terms that are not plainly relevant to the user's most likely question.

hjh

I wouldn't consider any outlet of vcf~ as a proper resonant lowpass filter. More about vcf~ here: https://github.com/porres/Live-Electronic-Music-Tutorial/blob/master/Examples/Vol.2/Part.09-Filters-Reverb-Karplus/31-Filters(Advanced)/3.Z-plane/4.[vcf~].pd

If you want you can also use biquad~ but you need to know how to generate the coefficients. For externals, I can suggest [else/bicoeff] to generate the coefficients of [else/lowpass~] for a lowpass filter.

cheers

For now I'm going with @alexandros's filter_abstractions.

I wouldn't consider any outlet of vcf~ as a proper resonant lowpass filter...

I guess that would make sense, then, why the help patch doesn't claim that it's lowpass -- it's funny, [vcf~] help says it's "like [bp~]" and refers to center frequency, but it seems to attenuate high frequencies more sharply than low, so I'm not sure if it really is bandpass, classically, or not.

Oh well... [omniFilter_abs~ lowpass] works a treat, and it's painless to add it to a patch. Thanks!

hjh

I actually meant to show this example where you can have a look at the frequency response of [vcf~] : https://github.com/porres/Live-Electronic-Music-Tutorial/blob/master/Examples/Vol.2/Part.05-Filters(Basic)/22-Filters.types/2.Filter.Types/2.Resonant/2.Bandpass/2.[vcf~].pd

why the help patch doesn't claim that it's lowpass

cause miller seems to consider it a lowpass, but I wouldn't consider it a proper one. You can see in that example that you have a constant zero dB, which is not how lowpass filters usually work.

it's "like [bp~]" and refers to center frequency, but it seems to attenuate high frequencies
more sharply than low, so I'm not sure if it really is bandpass, classically, or not.

yeah, it does attenuate more the high frequencies, the previous example to the one I posted above shows that too. I wouldn't consider it a proper bandpass as well.

Both [bp~] and [vcf~] are just pretty cheap filters. You'd need a biquad (with 2 poles and 2 zeros) to make a proper bandpass and a lowpass.