You are probably already using Zero Delay Feedback filters, so let your customers know!

[Acid Mitch]

I just hope Urs and others will stay away from this crap..

I'm guessing you haven't actually read the thread.

[andy-cytomic]

Dude you are not telling anything new here. So what?

Good, I'm glad you already realise this.

I am far from being developer but reading amazing threads from developers (such as you), Urs, Valhalla, tony tony chopper, Relab etc. we already did get it - simply building ZDF filter won't be automatically "suppa analog".

It takes far more then that, fine tuning, fine algos, and guess what, anyone reading anything about ZDF already know that.

As i see it you are trying to discredit other people work. That's how i see it and i don't like it.

edit: so anyway what is your real point - at first you are starting a thread where you are calling devs to call their product already ZDF, then you are saying that this isn't much at all. Like i said it started to smell right in the first post..

and before conspiracy theory chime in - i am not vadim, don't know him i only heard for him by reading documents. I am not fan. And i don't have anything against anyone and i like Cytomic work. I just don't like this "knowledge" measuring shitty thing which stink like disrespectful covert agenda..

Please don't kill the messenger, I'm just trying to highlight that perhaps Zero Delay Feedback filters isn't a particularly useful term to use in marketing and since it is likely that everyone already uses them in one form or another. At least the first few people replying to the thread could see the joke, come on people lighten up!

[pinkpanther]

andy-cytomic diss to Z1202 should get him banned. One thing is questioning a work or product and another is going personal. He lost all credibility and respect doing that. Sad. Some people need to relax his sphincter more often.

[toothnclaw]

So which developers were using Vadim's approach before he made it public?

[Kriminal]

andy-cytomic diss to Z1202 should get him banned. One thing is questioning a work or product and another is going personal. He lost all credibility and respect doing that. Sad. Some people need to relax his sphincter more often.

The whole thing is a scam. Look past it and move in. Im sure you have better things to worry about.

[Z1202]

Antti is a personal friend of mine, and I pointed out to him the other day that you had published his analysis in your "Art of VA design" book but not cited his prior art. You've also done this for the diode ladder. Antti was not impressed.

May I assume that this happened in the following way: "hey, see, Vadim used your work here and didn't credit it"? While I agree, that it would have been in principle appropriate to credit the pioneering art (but that's a long other discussion, then all pioneering art, probably starting with the RC filter and bilinear transform should be credited, and it would take me enormous time to find and collect all those references) there is a big difference between not crediting the pioneering art and not crediting the works which you have used as your sources (which you quite smartly trying to hide to make your attacks have much more weight). I should say that I didn't use Antti's results at all. In fact, when seeing his article many years ago, I just noticed: okay, he is using Ebers-Moll and then Euler, and didn't bother to read any further, because it is straightforward from that point on. So I must thank you not only for running public campaigns against me, but also for specifically alienating people to me behind my back. It's a pity that Antti didn't contact me personally on that, to hear the other side's point of view.

[Z1202]

Im still waiting to hear a sound that shows off zdff in action....

Emperors new clothes...

http://www.native-instruments.com/filea ... _knobs.wav
http://www.native-instruments.com/filea ... _booth.wav

Although, here is much more work under the hood than mere TPT (or ZDF). So, 95% of the credit goes to the actual creators of that synth.

Edit: oh, and so much as credit goes, the tracks are by Max Zagler, one of the creators of that synth.

[Kriminal]

Im still waiting to hear a sound that shows off zdff in action....

Emperors new clothes...

http://www.native-instruments.com/filea ... _knobs.wav
http://www.native-instruments.com/filea ... _booth.wav

Although, here is much more work under the hood than mere TPT (or ZDF). So, 95% of the credit goes to the actual creators of that synth.

Edit: oh, and so much as credit goes, the tracks are by Max Zagler, one of the creators of that synth.

Thanks, will listen tonight.

[xh3rv]

The whole thing is a scam.

I might disagree, one could have two Zero Delay Feedback filters that are vastly different creatures, but none would be unicorns or jackelopes

Unless we're talking DIVA which is IMHO the unicorn of VAs.

[Urs]

I just hope Urs and others will stay away from this crap..

How could I?

I strongly disagree with Andy. I neither find pointing out a delayless implementation of a feedback loop useless or misleading, nor do I find it hasn't a substantial contribution to analogue sound, where it applies.

In fact I don't find it very clever to bring up the Direct Form filters as a funny example. As I posted elsewhere, there's some irony when a filter whose implementation necessarily requires delays in feedback path(s) is able to exactly perform like a model of a filter with a delayless feedback path - but is otherwise useless at the task in question. Ha ha, have we laughed.

I will however credit Vadim for pioneering zero delay feedback filters (among other aspects) in a way that I was able to understand. Without his examples, we wouldn't have gotten that far.

[andy-cytomic]

I just hope Urs and others will stay away from this crap..

How could I?

I strongly disagree with Andy. I neither find pointing out a delayless implementation of a feedback loop useless or misleading, nor do I find it hasn't a substantial contribution to analogue sound, where it applies.

In fact I don't find it very clever to bring up the Direct Form filters as a funny example. As I posted elsewhere, there's some irony when a filter whose implementation necessarily requires delays in feedback path(s) is able to exactly perform like a model of a filter with a delayless feedback path - but is otherwise useless at the task in question. Ha ha, have we laughed.

I will however credit Vadim for pioneering zero delay feedback filters (among other aspects) in a way that I was able to understand. Without his examples, we wouldn't have gotten that far.

Hi Urs,

You are confusing the implementation details with the the equations being solved. A DF1 biquad does indeed solve the delay free feedback path in a linear time invariant way. All filters need memory to operate, and in digital this is a delay. If you note in a directly integrated SVF the output depends on previous outputs, this is called an IIR filter.

Perhaps thinking of an explicit filter being turned into an implicit one is useful for you, but it is not a useful distinction in general, and as you can see leads to all sorts of confusion and ambiguity over the intended meaning. The useful mathematical term is called "implicit integration", since this covers both the active (eg one pole low pass with opamp and feedback and capacitor) and passive cases (eg one pole low pass with resistor and capacitor) perfectly:

http://en.wikipedia.org/wiki/Explicit_a ... it_methods

"Explicit methods calculate the state of a system at a later time from the state of the system at the current time, while implicit methods find a solution by solving an equation involving both the current state of the system and the later one."

Even then you need to specify you integrate the circuit equations directly and don't operate through the laplace space, so probably the most accurate definition would be "direct implicit integration of a circuit". If you are fond of acronyms then you could easily make one from this, I'm not fond of acronyms.

My job is done here, Zero Delay Feedback has been shown to be a misleading and so not very useful term. It's a shame people don't have a sense of humour about it, perhaps it's cutting too close to the bone and is disappointing for you to realise the truth.

I'm glad Vadim could show you a simplified version of linear MNA and how it can be used to solve circuits accurately, I think his publication is a useful summary of existing methods applied in a well thought out method. Just like Antti's paper applies existing methods to produce a useful simplified expression to approximate the nonlinearities of the moog ladder filter, so to does Vadim list some useful solutions to specific linear circuits, this is worth publishing and helpful. Antti does not pretend he invented the prior art he uses to come to these solutions, and neither should Vadim.

[Z1202]

Antti does not pretend he invented the prior art he uses to come to these solutions, and neither should Vadim.

Andy, I'm always under the impression that you're trying to put the words in my mouth which I didn't say and then argue against those claims which I made in your imagination.

Please show me a particular place where I pretended to have invented the prior art. I never claimed to have invented bilinear transform or zero delay feedback resolution. In fact, in my paper I have credited the only mention of ZDF I was able to find at that time. So if there is any fault of mine, then it's only that I didn't manage to find the truely pioneering references. But in the meantime I have been made aware of other ZDF works and also publicly admitted their priority.

As for the TPT method , you still didn't show anyone using a combined version of BLT integration and ZDF as a method of VA model design prior to my publication. So, if you excuse me, I will continue to count myself as the inventor of the method, until you (or anyone else) shows the opposite.

As for being equivalent to the trapezoidal integration I in fact several times publicly admitted the mathematical equivalence of the two. And I definitely never claimed to have invented the trapezoidal integration approach. Also, I publicly admitted Serafini doing exactly this quite a few years ago.

As for your claim that you never argued about the TPT method, I think you're trying to be too smart here. The only thing I ever claimed to invent is the TPT method, which currently is being referred to by others as the ZDF.

[Urs]

The useful mathematical term is called "implicit integration"

I'm curious to see to what effect you're able to create a successful marketing campaign around this term.

Nevertheless, you solve circuit models for non-linearities, we solve for feedback. The mathematical result is the same, but the way to get there is different. Ours however can be optimised for execution speed by adding a unit delay, without change to the actual set of equations. With your method you might have to rearrange equations.

[Z1202]

Actually, I have just thought, but maybe it's all due to the fact than Andy doesn't like working with block diagrams. In that case, the TPT method obviously has no value over the trapezoidal integration of ODEs for him personally whatsoever, and can be seen only merely as another expression of the already known method. This still doesn't excuse the way he behaves, IMHO.

BTW, TPT was not designed as another interpretation of the trapezoidal integration, but rather as a topology-preserving extension of the bilinear transform (you can see that the original article doesn't even mention the trapezoidal integration, IIRC). So, Andy, your hints at stealing other people's work are not justified.

[mdsp]

hi

I've been following all these discussions for some time, and I'm no expert and don't pretend to understand all the details and subtleties of each approach, so I prefered to stay silent until now.

But what I've missed most when reading about andy's or vadim's work (and thanks both for taking the time to publish) was the few comparisons of their methods against the state of the art. Probably because lack of time and knowledge about prior work. I guess that if you had to make a more formal conference or journal publications you'd get the same comments from reviewers and that we'd all benefit from having more perspective. Like RBJ's AES article showing how all the different biquad computations methods where equivalent.

Here's to my knowledge a widely accessible presentation of how the wave digital filters formalism + bilinear transform (invented by Fettweiss) can be used to simulate analog circuits:
https://ccrma.stanford.edu/~jos/WaveDig ... lters.html

Note the section about delay free loops:
https://ccrma.stanford.edu/~jos/WaveDig ... Loops.html

And how the state variable change to wave variables solves the implicit delay-free loops (in the linear case):
https://ccrma.stanford.edu/~jos/WaveDig ... ables.html

Also note the section warning about the potential drawback in the non-linear and non LTI case:
https://ccrma.stanford.edu/~jos/WaveDig ... lters.html

Because the frequency axis is warped, we cannot expect nonlinear or time-varying systems to behave in discrete time as they did in continuous time:
Modulation sidebands land at wrong frequencies. Harmonic relationships are destroyed harmonic distortion becomes inharmonic.

The main difference I can see with andy's and vadim's approach is that you both try to preserve the original state variables while the WDF converts that to forward and backward travelling waves. The closest match I could find that compares both is David Yeh' article "Simulating guitar distortion circuits using wave digital and nonlinear state-space formulations" that applied both the K-method and the WDF to the same circuits.

So here's a list of publications I could find that are related to the problem(s) being adressed:

Tutorial on Wave Digital Filters, David Yeh
https://ccrma.stanford.edu/~dtyeh/paper ... torial.pdf

Digital Implementation of Musical Distortion Circuits by Analysis and Simulation. D. T. Yeh, Ph. D. dissertation, Stanford University, June 2009
https://ccrma.stanford.edu/~dtyeh/paper ... esided.pdf

Simulating guitar distortion circuits using wave digital and nonlinear state-space formulations. (slides) ( sound examples) D.T. Yeh, and J.O. Smith, Proc. of the Int. Conf. on Digital Audio Effects (DAFx-08), Espoo, Finland, Sept. 1—4, 2008, pp. 19 — 26.
https://ccrma.stanford.edu/~dtyeh/paper ... fx_sim.pdf

PRESERVING THE STRUCTURE OF THE MOOG VCF IN THE DIGITAL DOMAIN, Federico Fontana
http://quod.lib.umich.edu/cgi/p/pod/dod ... 2.2007.062

F. Avanzini, F. Fontana, and D. Rocchesso. Efficient computation of nonlinear filter networks with delay-free loops and applications to physically- based sound models. In Proc. of "The Fourth In- ternational Workshop on Multidimensional Systems, 2005 (NDS 2005), pages 110-115, Wuppertal, Ger- many, July 2005. IEEE.

A. Sarti and G. De Poli, "Toward nonlinear wave digital
filters," IEEE Trans. Signal Process., vol. 47, pp. 1654-1668,
June 1999

F. Fontana. Computation of linear filter networks containing delay-free loops, with an application to the waveguide mesh. IEEE Trans. on Speech and Audio Processing, 11(6):774-782, Nov. 2003.

J. Szczupak and S. K. Mitra. Detection, location,
and removal of delay-free loops in digital filter con-
figurations. IEEE Trans. on Acoustics, Speech and
Signal Processing, ASSP-23(6):558-562, 1975.

A. Fettweis, "Wave digital filters: Theory and practice,"
Proc. IEEE, vol. 74, no. 2, pp. 270 - 327, Feb. 1986.

A. Sarti and G. De Poli, "Toward nonlinear wave digital
filters," IEEE Trans. Signal Process., vol. 47, pp. 1654-1668,
June 1999

S. Petrausch and R. Rabenstein, "Wave digital filters with
multiple nonlinearities," in XII European Sig. Proc. Conf.
(EUSIPCO), Vienna, Austria, Sept. 2004, vol. 1, pp. 77-80.

K. Meerkötter and R. Scholz, "Digital simulation of nonlin- ear circuits by wave digital filter principles," in Proc. IEEE Int. Symp. on Circuits and Systems, May 1989, vol. 1, pp. 720-723.


I don't know the Control system litterature very well, but each time I hear "state-space" I immediately think "Control Systems". I'm sure there's more to find there about non-linear, non-LTI, continuous, discrete-time and mixed state-space formalisms.