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

[Echoes in the Attic]

Going by that test, Zebra's XMF filter exhibits zero-delay-feedback behaviour!

Actually the question becomes, just what filters out there aren't 0df now? I don't htink I have any! Horray, all my synths just got even better!

You probably have tons. They just use some other methods to get the independence properties.

Ya I was joking.

[toitoi]

Le Lotus Bleu Strikes Back

[izonin]

When researching for an adequate filter for my Stargate project, I ported Vadims filter from Reaktor to SM. It didn't sound better than the stock one. The non-linearities are far more important for the character and the quality.

Doesn't make sense. The whole point about 0DFb filters is that they do a better job of reproducing nonlinear (=feedback) behaviour.

My point is that Diva's filters in non-zero (draft) mode sound a lot better than Vadim's zero-delay-feedback one. So 0DFb is no guarantee for good sound.

[awol9000]

SNORE............

[braj]

I second the dongle as being the reason I avoid XILS synths.

[vata44]

+1 ditch your dongle and I will stand in line.

[Lotuzia]

Le Lotus Bleu Strikes Back

Finally someone did get it

Else :

There are already some non dongled Xils synths ( Xils3 LE and PKBII Advanced player .)Be assured there will be more in the future.

These tests ( there are more tests in the advanced page, as there's more detailed information ) are not only for Xils synths, the interesting thing is that you can do them with all your synths, including your real analog ones. I did them with my 7 HW analog synths, and all did exhibit the same exact behaviour. Then some soft synths were tested, but well, you can now try them and do the same with your synth(s) That's all we provided, along with some explanations, only for those who like to read some ( as a poster stated, you dont need to perform these tests, there are infomercial blurbs )


Ltz

[toitoi]

Is it the right guess? I like Tintin world a lot.

[Lotuzia]

Is it the right guess? I like Tintin world a lot.

Yes it is, I'm a fan too, and Lotuzia was the name of my 1rst Sorceress in Diablo II hardcore mode a long time ago, she never died, though Blizzard must have erased it by now, so I thought it might be a good idea to offer her some after life I like magical names

[Richard_Synapse]

This is NOT true in general. It is true for certain specific topologies implemented in certain specific ways, but there are tons of examples where this is NOT the case (ie the analog originals are NOT well-behaved in such a sense). Further, many analog designs that have significantly non-linearities can further have dependencies between input signal levels vs. frequency and/or resonance.

Yep exactly. It pretty much reduces to a few well-behaved designs such as the Moog, and even that isn't 100% constant-Q. Besides constant-Q digital filters exist since ages. The interesting part is when the filters become nonlinear, such as in Diva. The 0df linear case seems somewhat overrated to me, after I experimented with that for a while. In Dune 2 we'll definitely employ nonlinear designs instead.

Richard

[Bronto Scorpio]

In Dune 2 we'll definitely employ nonlinear designs instead.

Nice
Can't wait to hear the results

Cheers
Dennis

[pdxindy]

I second the dongle as being the reason I avoid XILS synths.

Me too... I quite like them sonically... but no dongles for me!

[Kriminal]

blah blah blah...

Here is a simple test that you can make to figure out if you favorite soft synthesizer has a zero-delay-feedback filter...

blah blah blah

just ask the developer, much less hassle

[Kriminal]

I second the dongle as being the reason I avoid XILS synths.

yes, thats one of the reasons

[xavier]

This is NOT true in general. It is true for certain specific topologies implemented in certain specific ways, but there are tons of examples where this is NOT the case (ie the analog originals are NOT well-behaved in such a sense). Further, many analog designs that have significantly non-linearities can further have dependencies between input signal levels vs. frequency and/or resonance.

Yep exactly. It pretty much reduces to a few well-behaved designs such as the Moog, and even that isn't 100% constant-Q. Besides constant-Q digital filters exist since ages. The interesting part is when the filters become nonlinear, such as in Diva. The 0df linear case seems somewhat overrated to me, after I experimented with that for a while. In Dune 2 we'll definitely employ nonlinear designs instead.

Richard

Hi,

I will explain a little deeper because it seems that something was misunderstood.

_ Take a non constant-Q hardware topology. Reproduce it with the needed delay(s) in the digital world. The links between the Q and the frequency you will get with this digital filter won't have nothing to do with the relation between the Q and the frequency of the hardware model.

_ Take a constant-Q digital filter which doesn't self-oscillate. A simple test can show that even the Q is independent of the frequency, the spectrum, due to the digital delay, is warped (here is an other consequence of the delay which is managed by the 2-poles XILS 3 filter).
But we are not talking here such a filters, but self-oscillating filters.

What was stated is that a self-oscillating digital filter which does need a non-linearity in its internal part (otherwise it would output infinite numbers) must compute the input and the output at the same time.
In the digital world, a delay is needed for this. Either you manage it and it doesn't change the filter characteristics, either you don't and the result won't be the same than the analogue model (what ever its topology).

Now this non-linearity can be done in very different way, giving very different sound and results. But this doesn't change the mathematics rules.

I hope this is a little more clear (despite my poor English)

Best regards
Xavier