Zero Delay Feedback Filter (How to test if your synth has a )- Xils-Lab White Paper -

[ObsoleteAcc99]

[DELETED]

[ObsoleteAcc99]

[DELETED]

[hakey]

Disinformation thread.

[ObsoleteAcc99]

[DELETED]

[ariston]

there is no hidden intention other than the one you are creating.. and i have read ALL the back stories and topics related to what you are insinuating.. So please STOP trolling an information thread!

Well, someone's lost his pacifier and is a bit cranky, it seems.

FYI: I did stop. You dragged it back up, days later I might add, and obviously felt that your opinion was so desperately needed that it required two posts. How does it feel to troll and obscure an information thread by accusing someone else of trolling? Can you spell "hypocrite"?

You don't see a hidden intention, I do. You can unclench your butt cheeks now, it's only a difference of opinion.

edit: Damn, broke my promise. See what you made me do? Well, back to anger-management classes.

[ObsoleteAcc99]

[DELETED]

[ariston]

I didn't see that it was days ago, in that case i am sorry. I honestly just saw fresh posts and did not think to look at your particular posting date. That is no BS, that is true.

Peace.

I do really feel Laurent was only at least trying to be helpful though..,

Okay, fair enough, no harm done.
I still don't quite see the point of it all, beyond the fact that we've only been talking about 0 delay FB filters ever since you-know-who released you-know-what. It has been remarked that this proposed test is far from conclusive; I tend to agree. By this test, Z3ta2 has 0dfbfs. Does that even mean anything? The only software filters I've heard so far that are even close to you-know-what are the audio demos of the upcoming The Drop.
And, in the end, it's all about the sound. I might be curious as to the tech (just so I can bore my wife with the details), but it's all just numbers, innit? Isn't that why we pay talented devs like Urs and Xavier, so we don't have to worry about it?

[rod_zero]

The point here is very simple:

Does the test can prove with 100% accuracy that a synth use zero delay filters?

If not, why?

Was the test designed to prove synths using one method and discriminate others?

Or, how many ways are there to "pass this test" using other methods?

[hakey]

Does the test can prove with 100% accuracy that a synth use zero delay filters?

No. It's next to useless - many standard, non-0df filters will "pass" the test.

[kv331]

problem with zero-delay filters is they don't solve the aliasing issue. it still requires massive over-sampling of about 8x-16x to get anywhere near where you want to be. by that time the filter is so expensive you might have considered using an analog one instead.

I havent played with ZDFs yet, but if they will still require high oversampling rates (>= x4), then in that case why would I use them? Because if there's oversampling (>=x4), then that'll compansate for the z-1 delay/phase errors to a certain extend, as nyquist frequency is divided by the oversampling factor... Please correct me if I am wrong guys!

Bulent

[kv331]

Does the test can prove with 100% accuracy that a synth use zero delay filters?

No. It's next to useless - many standard, non-0df filters will "pass" the test.

Yes, as long as they are designed correctly.

[hakey]

I havent played with ZDFs yet, but if they will still require high oversampling rates (>= x4), then in that case why would I use them? Because if there's oversampling (>=x4), then that'll compansate for the z-1 delay/phase errors to a certain extend, as nyquist frequency is divided by the oversampling

When you say ZDFs are you thinking about zero delay filters or zero delay *feedback* filters?

From what I understand, in the second case 0-delay makes a considerable difference as the effect of nonlinearities (=feedback) can be more accurately modelled.

[kv331]

When you say ZDFs are you thinking about zero delay filters or zero delay *feedback* filters?

From what I understand, in the second case 0-delay makes a considerable difference as the effect of nonlinearities (=feedback) can be more accurately modelled.

I think (!) meant zero delay feedback filters, where you get rid of z-1 in the feedback loop.

[hakey]

From what I understand, the problem is that in a real electrical circuit any feedback between input and output is instantaneous.

However, a digital model with 1-sample delay feedback uses the output from the last iteration as the input for feedback in the current iteration - ie it fails to model the effect of instantaneous feedback.

In contrast, a 0-DFb filter effectively uses the output from the current iteration as the input for feedback in the current iteration, and so models instantaneous feedback.

[brok landers]

My idea of a good filter, I press a note on the keyboard, I hear the sound, it sounds good to my ears. That's a good filter.

It sounds unpleasent to my ears, its a bad filter.

Do some tests with Diva in draft and divine mode. Might notice something you weren't aware of before. The z-1 draft has this comic/cartoony quality to it, while the zdf divine sounds more audio-realistic.

I am very aware of the difference in sound of the filter in Diva at different quality settings, particuarly the difference between draft - fast/great/divine. No additoinal tests required

My point is that for me personally I really don't care what goes on under the hood, unless its part of a bug solving thing for example if there was say a DC offset issue with a synth I might analyse n try and help the developer get to the root of the problem.

I will also go one step further and say that for some particular types of sounds, draft mode in Diva works can work better than the higher quality modes. And by better I mean in getting a particular kind of sound suited to genre and taste, not authenticity to some old vintage piece of equipment.

One mans awesome bright trance lead is anothers horrible plasticy lifeless sound, and vice versa. To argue that one particular way is the 'correct' way when it comes to something as subjective as sound and music in this instance is a bit silly. Maybe something as obvious as hideous amounts of aliasing sure, but this is different in my view.

For example, BigTones presets were designed in draft and thus to work at their intended filter response ideal in draft mode .But again its a matter of taste, we have the choice, of multiple quality settings, which is a nice touch.

There are times I want a really organic filter sound that sounds like its running through big chunks of hardware, and there are times I want a glassy pure filter sound, or a very resonant sharp filter sound, or a warm but very smooth analog/digital hybrid sound, or a very bright digital filter sound. Depends what mood I'm in, what I'm trying to accomplish, how I feel it'll suit a sound in a particular mix. No right or wrong way here.

Got nothing against marketing, its part of business, and it works, I'm just saying for me personally, I just play, I just listen, if I like the sound of it, thumbs up from me.

this summs it up rather nicely, imo.
also, i don't know, but i assume that the difference between draft and the higher quality modes in diva are not only the zero delay fb-mode, but way more enhancements, so i don't know if the comparsion between draft an the high q-modes fits clothes... i could be wrong on this though...