ca 4 pole filter , fake or not

[gentleclockdivider]

Just a simple question and a n honoust ansewer
I suspect that the resonant peak in the 4 pole filter is just an added sine wave (tuned to filter freq .) and not the result of a feedbacking filter .
The resonant peak is just too narrow and it is located slightly above the cut-off frequency , I've had a lot of synths but have never actually seens this kind of peak before .
Sure is some some actual real resonance going on ( see screenshot ) , but that narrow peak looks verry strange


And here I've done the same in reaktor ( without actual real resonance ) .


I am pretty sure the r(extra narrow )esonant peak is faked , I can get there 99% close with 20 year old reaktor primary modules , behaving under the exact same circumstances .
My theory is that the 2pole is using real resonance , the 4pole is 2 pole ins series ( so also a bit of real resoance ) but automatically engages a sine osc .
I am almost sure about it , the filter just behaves too weird and I can get it to replicate it exactly with standard primary modules
I would really like a answer from cherry audio

[cherryaudio Greg]

Hi,

I don't know the details of the inner workings of the filter, but I passed your comments along to the software developers.

Greg

[gentleclockdivider]

Any news on this ?
Here are two files , both files contain a cherry audio and reaktor ensemble audio file , with stock primary modules using a sine wave for fake resonance
Thy sound verrry close , to close to consider the CA2600 be analogue modelled at all , given the fact I used 20 year old primary modules and fake resonance

https://app.box.com/s/o5q0etf24dl9acrgvb1fhv9feldno2mv
https://app.box.com/s/ksu4a78fg6oi9axd6tx4frx5j7twvqs5

[gentleclockdivider]

bump

[dergundel]

Hello gentleclockdivider,
good finding!
I can confirm this strange behaviour of the filter. I compared high resonance settings with Arturia’s ARP2600 V3 and WayOutWare’s TimewARP 2600 just feeding a simple sawtooth into the filter. Sweeping the filter you typically get a reinforcement of the partials. Both Arturia’s emulation and TimewARP do show this behaviour, while CA2600 doesn’t, there is just some kind of beating when the presumedly added sine wave and the gentle resonance peak is passing the partials.
Hope we hear from cherry audio soon.

[gentleclockdivider]

Could you post a screenshot of the spectrum analyzer of both timewarp/arturia , so we can compare
-resonance 80 ( just make sure there is an obvious spike )
-freq 4Khz (non kbd tracked ) ,4 pole filter
-osc 3 set to 16' , disable kbd

Thanks

[dergundel]

Hi, I tried my best. Both synths behave a bit different, but you can clearly see, that there is a considerable resonance peak (at 90% with Arturia and about 70% with TimewARP) but it is building gradually around the cutoff frequency, getting steeper and steeper as the resonance is increasing. There is no singular spike whatsoever.
Hope this helps.
Best.

[gentleclockdivider]

Thanks
I will use your images for reference and point them to this thread



https://www.kvraudio.com/forum/viewtopi ... 2#p7932022

[gentleclockdivider]

Still waiting for one of the developers to chime in , why so quiet ?

[Cherry Dan]

Thanks for your interest in our products. I can assure you that we are as interested in getting these details right as you are in analyzing what we've built, and I commend you for your dedication to this issue.

There was indeed a small math issue in the CA2600's filter code that caused the weird resonance peak that you observed. Sonically it was correct but the filter was not behaving as it should. We've just published CA2600 Build 86, which will automatically update the next time you run CA2600 (unless you've turned off automatic updates). Build 86 features improved filter code that correctly self-oscillates.

Please let us know how Build 86 works for you!

Thanks,
- Dan @ Cherry Audio