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

[EvilDragon]

[kmonkey]

Like i said right from the beginning. This thread stink a lot and prove that even a well known gentleman can play di** measuring ridiculous game..

[kmonkey]

Well, I for one feel crucified by people who have neither seen our code nor our method.

I don't believe it was our marketing campaign that started the hype about zero delay feedback filters with users, it was Diva's sound. Which even after two years is testament to our approach - where a unit delay in the feedback path is the only difference in the lowest quality mode, albeit an audible one. QED.

We users don't want to know (it's a stretch i know but). You are doing amazing plugins without any crappy buzzwords - your plugins have attitude, sound, and your userbase is telling all what is really important. Really it's that simple.


Just stick to what you and your team is doing best - we love you - not because we are gay (no pun intended) but because you actually deliver stuff in your products - end of story

(did i mentioned amazing technical support haha..)

- this thread is really top trolling thread of the 2013..

[kmonkey]

need to highlight the truth.

I guess you never saw this:

[kmonkey]

People are stupid. I'm a moron, you're a moron, that's how things are. We're all basically shit-flinging monkeys. Imagine how we look to some super-advanced race of robots or whatever. No surprise they always want to massacre us in movies.

No you are wrong. In these movies they want to massacre us because these same morons you are describing are responsible for making that same movie you are referring to

There are other circles of people, not all are (necessary) morons..

this thread is even more dramatic than when Jessie Spano got hooked on diet pills

i was just thinking, i hope there's a lot of people on kvr today telling me who the bad person is, because i can never figure that shit out for myself

Its me

Krim..now that was funny...

[aciddose]

There are other circles of people, not all are (necessary) morons..

Sounds like something a moron might say

You know what I mean though, relative to what? We're all morons when you take into account what we still don't know. My point wasn't whether any particular person is stupid, my point was to say that we're all going through a learning process and before attacking the knowledge some people have it's important to consider whether it was their choice to be limited to that knowledge, or whether they simply need to take a few more steps forward.

[toothnclaw]

Dear non-technical reader, please allow me to give you some commentary which I hope will allow you to understand this scenario entirely.

In 1996...

Thanks Dave! A very helpful history lesson. Sums up the thread nicely.

[Ichad.c]

- this thread is really top trolling thread of the 2013..

I actually think 0df is the most trolled topic in this sub-forum, and the trolling usually starts with a certain semantic 'genius'. Kind of stupid really, because of all the threads - there are only a very few that are actually on-topic i.e. that will be useful for new devs. And learning and teaching is what this sub-forum is about, *sigh*.

[FastTriggerFish]

You're missing the point too. There's no need to do any maths. Any 2nd order biquad in any form has up to two unit delays in its feedback path(s) - no matter what analogue response it perfectly models. Claiming otherwise is purely ridiculous.

Urs, please get Clements on this, you are out of your depth here.

That type of response is exactly what I was talking about earlier.
You don't seem to have got the hint from more earlier comments so I'll spell it out for you:
this really makes you sound like an insecure d*ck. And you do this all the time !
Except in this case you managed to give this response to a very well known, respected and admired developer, which will make you look even worse than usual in the eye of many people.

I'll just give you a tip for next time: if you think someone made a mistake or said something wrong, don't assume that's because they're less smart or less competent than you. Just assume they've made a mistake which everyone does occasionally, and explain your point. And maybe, just maybe try imagining that you might be wrong too or have misunderstood... self doubt is a critical part of scientific thinking after all.
That might help you communicate your ideas in a decent manner and leave the passive agressive insults out of it.

[andy-cytomic]

You're missing the point too. There's no need to do any maths. Any 2nd order biquad in any form has up to two unit delays in its feedback path(s) - no matter what analogue response it perfectly models. Claiming otherwise is purely ridiculous.

Actually, this is perfect Urs. I am so glad you posted it, since this is a part of the confusion I hadn't even brought up yet, so thanks very much for swapping team on this If the implementation of a digital filter that requires previous state fed back with delay to solve it disqualifies it from being a "zero delay feedback filter" then you have just disqualified all IIR filters, all of your own, all DF1 biquads everything. So now, by your very own definition no such thing as a "zero delay feedback filter" exists.

[andy-cytomic]

The RBJ DF1s add no such delay, they use a trapezoidal integration via the bi-linear transform which doesn't add a delay, it solves for the filter implicitly, and in doing so it just doesn't preserve the original topology

So if "implicit integration" does not necessarily maintain the topology then I just had an

It means that "implicit integration" isn't useful as a buzzword either, while, e.g. "TPT" is

This is correct. Implicit integration does is not sufficient to describe maintaining the topology of a filter. Here is the definition for explicit and implicit:

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

I certainly hope that "implicit integration" never becomes a buzzword. Solving systems of non-linear equations with implicit methods is certainly useful, but it does not describe the full extent of what is going on either. Preserving the topology of a circuit while solving the systems of non-linear implicit equations is almost there, but how about people just judge the quality of the filter by listening to the results?

Urs your work is excellent, I love your products and you are a friend. I don't have a beef with you, only the terminology. You continually didn't get what I was saying, didn't take it on board for whatever reason, which is why I was asking for Clements to comment instead as I no matter how many times I said things in different ways it just wasn't getting through. I hope that has now changed and you can see the ambiguity and confusion the terminology is capable of causing, even amongst developers using these methods.

PS: calling something "topology preserving" is useful, but please remember that the chamberlin svf is topology preserving, and it is an explicit method.

[Urs]

You're missing the point too. There's no need to do any maths. Any 2nd order biquad in any form has up to two unit delays in its feedback path(s) - no matter what analogue response it perfectly models. Claiming otherwise is purely ridiculous.

Actually, this is perfect Urs. I am so glad you posted it, since this is a part of the confusion I hadn't even brought up yet, so thanks very much for swapping team on this If the implementation of a digital filter that requires previous state fed back with delay to solve it disqualifies it from being a "zero delay feedback filter" then you have just disqualified all IIR filters, all of your own, all DF1 biquads everything. So now, by your very own definition no such thing as a "zero delay feedback filter" exists.

I love your logic. Well played.

Your style however, and the attitude displayed... not my cup of beef.

Even if you're right in the RBJ case (I'll lay out my doubts elsewhere), you miss to address a central point. Pretty much what Dave said.

[xh3rv]

If the implementation of a digital filter that requires previous state fed back with delay to solve it disqualifies it from being a "zero delay feedback filter"

Is it completely correct to say previous state is fed back here? It's a discretely localized sample of state. Computationally it's very easy to consider previous output as state. But mathematically I wonder if the methods of (insert semantic) filters are continuously localized state - this is trivially true with discrete LTI, but harder with non-LTI. So a sample of previous output isn't exactly in the same domain as state.

[Urs]

Here's my point: As soon as you modulate coefficients in a biquad filter, you break it. Then the change of coefficients propagates through the "state" with unit delays. A ZDF filter doesn't suffer form that problem. Because any state variabale at any time is calculated with a set of coefficients that's corresponds to that point of time. Sorry I can't express this in any more useful terms.

(I guess I'm grasping at straws here, but even if I can't point my finger on it, I'm very sure that it's the delays that f**k biquads up for things. Maybe if one did a reverse z-transform or some such thing, a change in coefficients would appear as a form of delay in the analog counterpart. Or some such thing. I'm being silly here. Wodka helped)

[andy-cytomic]

Specifically, Stilson and Smith introduced their delay into the "feedback path". That term "feedback path" really only means anything if you're used to looking at schematics and can make sense of the circuitry; it's being used in a specific technical sense referring to the Moog ladder filter.

Nonetheless, Urs' version based on Vadim's book had *zero* additional *delay* in the *feedback* path of the Moog ladder *filter*.

zero delay feedback filter. A phrase is born.

Hi Dave, thanks for the chipping in

I wanted to add that I believe that Vadim's technical summary also uses the term, for example section 3.10 is called "Zero-delay feedback".

This is also not specific to the Moog ladder filter structure, feedback loops appear in lots of circuits, not just filters.

Another problem here is that it's not just the resonance feedback path that contains "difficult" parts to model, the moog ladder, in its most basic non-linear form that Antti presented, has 5 non-linear implicit parts to it, 4 in the filter core and 1 in the resonance path. The four in the core aren't "feedback" in the classic sense, since there is no wire shown on the schematic that takes the output and "feeds it back" to the input, but they are equally as hard to solve as a non-linear circuit that does have feedback.

In fact there are loads of non-linear filters that are difficult to solve, and the "zero delay feedback" part, as in the main resonance feedback path, is actually the easiest bit to solve.