What did people do with time-varying filters prior to 2014?

[soundmodel]

This paper:

http://dafx14.fau.de/papers/dafx14_aaro ... s_for_.pdf

has a good overview on time-varying musical filters, but it was published in 2014. However, plenty of time-varying filters have been implemented before that.

I wonder what these filters prior to 2014 used as references? Or are they possibly individual "hacks" regarding e.g. stabilization of DF-II?

[soundmodel]

Okay, maybe nevermind. The references for stabilizing DF-II in that paper are, e.g.:

M. Gold and B. Rader, “Effects of parameter quantization
on the poles of a digital filter,” Proceedings of the IEEE, vol.
55, no. 5, pp. 688–689, May 1967.

[4] R. Rabenstein and R. Czarnach, “Stability of recursive
time-varying digital filters by state vector transformation,”
Signal Processing, vol. 8, pp. 75–92, 1985.

Then it's odd to see the time-varying case still be treated as somehow new though.

[whyterabbyt]

Given that the paper refers to prior work at least as far back as 1985, why are you claiming that the paper is treating it as new? Abstract says the paper is a novel proof of the SVF, nothing more, certainly not treating anything other than that proof as new.

Strawman vibes here.

[soundmodel]

Well but the question stands. What did they do prior to the SVF then?

If you read the paper more carefully, then it makes it seem like other attempts must be somehow "wrong", because the SVF fits the problem so naturally. E.g. I think transient minimization sounds like a dumb idea even on paper.

Based on the paper they would have looked at DF-IIs that "happen to be nice", coupled forms or then using transient minimization.

[whyterabbyt]

Well but the question stands. What did they do prior to the SVF then?

Maybe that stuff in the earlier papers you referenced. Maybe something else, given that, say, Antti's paper on the Moog filter was presented at DAFX a whole decade earlier.

You do get that academic research papers arent the entirety of real-world research and development, dont you?

[soundmodel]

You do get that academic research papers arent the entirety of real-world research and development, dont you?

Which is why this question is asked here.

[soundmodel]

This paper claims to review the history, but also says it does not mention everything.

Variable Digital Filters
Georgi Stoyanov and Masayuki Kawamata
https://citeseerx.ist.psu.edu/document? ... edd683549e

Possibly, the chapter "2. Basic Theory" could be quite revealing already(?)

[whyterabbyt]

You do get that academic research papers arent the entirety of real-world research and development, dont you?

Which is why this question is asked here.

the only question i see you asking is 'how did people do this thing before a point in time where Im claiming that someone else said it was new even through they didnt'

[whyterabbyt]

This paper claims to review the history, but also says it does not mention everything.

yes, because the person who wrote it clearly didnt assume they knew everything of that history.

[soundmodel]

This small paper claims that SVF is not known because of poor documentation:

Digital State-Variable Filters
Fons ADRIAENSEN
http://kokkinizita.linuxaudio.org/papers/digsvfilt.pdf

It references a 2020 paper though, so maybe the writer is not informed.

---

To answer viewtopic.php?p=8872347#p8872347 below:

Based on the superior advantages of the SVF presented in the first post's paper, people should be driven to use the SVF, if they'd be aware of the results from that paper or elsewhere.

[BertKoor]

maybe the writer is not informed.

How ironic...

[soundmodel]

This could be a somewhat significant paper (just a guess based on the year 2005 and the content):

High-Order Digital Parametric Equalizer Design
Sophocles J. Orfanidis
http://www.ece.rutgers.edu/~orfanidi/ece348/hpeq.pdf

Here

https://www.ece.rutgers.edu/~orfanidi/hpeq/

it says that the filters are in Matlab's Filter Design Toolbox.

The Filter Design Toolbox only seems to implement these and RBJ's biquads. https://se.mathworks.com/help/audio/ug/ ... ation.html

I wonder if it's reasonable to expect that Matlab has chosen the most prominent references? Or i.e. it would then point to the idea that other plug-ins are likely using Orfanidis' paper to some extent? In some other domains the implementations that Matlab has are usually a good representation of the "state of the art".

[gaggle of hermits]

This small paper claims that SVF is not known because of poor documentation:

Digital State-Variable Filters
Fons ADRIAENSEN
http://kokkinizita.linuxaudio.org/papers/digsvfilt.pdf

It references a 2020 paper though, so maybe the writer is not informed.

no it doesn't:

"So one may wonder why it is not used more often. The main reason seems to poor documentation."

not used more often /= not known

why lie about this?

[soundmodel]

Yet another paper (from 2010):

Accurate Discretization of Analog Audio Filters
with Application to Parametric Equalizer Design
Simo Särkkä, Member, IEEE and Antti Huovilainen
https://users.aalto.fi/~ssarkka/pub/eq-article.pdf

It's not about musical filters though. It writes:

The disadvantage of the methods is that it introduces
a slight delay to the signal (order of tens of samples), which
limits its applicability to control engineering applications.

However, the references section of the 2010 paper looks like it could a very good historical reference for time-varying filters.

[Urs]

Then it's odd to see the time-varying case still be treated as somehow new though.

The time-varying case is not treated as new. The first sentence in the paper reads:

"A variety of methods are available for implementing time-varying digital filters for musical applications."

So, if a variety of methods are available, then certainly they predate the paper and thus are not treated as new.

(I for sure remember implementing time-varying filters for musical applications in the previous Millennium)