DIVA - A question on the BITE filters

[moscom_electronics]

First, I am new to the forum, and I like this place a lot a lot of nice pieces of info can be found here, and I really appreciate this.

I like the sound of the MS20 filter a lot, and I think DIVA is close to it. I just find the behavior of the version 1 BITE filter to be odd. Is the high pass really a high pass? it sounds like the filter tries to cut the frequencies but does not manage to do that I remember that something was stated regarding this filter by Urs during beta test, just wondering if this still needs to be improved.

Thanks a lot!

[moscom_electronics]

As a side remark, version 2 of the BITE filters seem to work fine, or at least have the proper behavior.

Also another thing that I have noted (because I have a real ms 20 at home ) is that the ring mod of the ECO osc behaves very differently from the MS 20. I understand completely that this behavior may vary from one MS 20 to the other. I am just wondering if this difference comes from limitations in the current possibility in modeling or if it is just dependent on the MS 20 used to model.. or anything else?

Thanks!

P.S. : in all the testing I took care of the peculiarities of DIVA's behavior regarding the osc levels with the BITE filters.

[whyterabbyt]

I like the sound of the MS20 filter a lot, and I think DIVA is close to it. I just find the behavior of the version 1 BITE filter to be odd. Is the high pass really a high pass? it sounds like the filter tries to cut the frequencies but does not manage to do that

The MS20's highpass filter was a 6db/Octave cutoff, not 12db/Octave, according to Tim Stinchcombe and others.

[moscom_electronics]

Thanks for your reply!

Well, it should still cut frequencies and the difference in behavior between version 1 and version 2 in DIVA is pretty obvious.

Moreover, my MS 20 is a korg35-chip version (which if I am correct is version 1), and the cut in frequency is clear and obvious.

I do not look for the exact same behavior between the MS 20 and DIVA (though I am pretty it would be seen as highly desirable by many members here). I am just wondering if it is normal that the version 1 high pass bite filter does not seem to cut that much of the frequencies...

[Urs]

We somehow screwed up on the first revision (hehehe, "as confirmed on Gearslutz"), also on the lowpass. It's somewhat too tame. Rev 2 is spot on though. The highpass is however not really a highpass, even in the hardware. There's always some bleed through on the low end. That's because the signal is injected into the resonance path, and then kind of lowpassed from there.

Or so.

[moscom_electronics]

Thanks Urs for the reply

indeed the rev2 is really (really, really... to the point where I may sell my MS;)) close to the real thing. I will keep using this one. Do you plan to update the rev 1 filter in the future?

and also,the ring mod sounds different compared to the hardware. Any hints on that?

[Urs]

Thanks Urs for the reply

indeed the rev2 is really (really, really... to the point where I may sell my MS;)) close to the real thing. I will keep using this one. Do you plan to update the rev 1 filter in the future?

I'm not sure. Andy from Cytomic might do the ultimate emulation and even he's been struggling. This seems to be one hell of s difficult issue (google "stiff problem")

and also,the ring mod sounds different compared to the hardware. Any hints on that?

I think we used the way the Legacy collection did it, using the actually selected waveform from VCO1 instead of just the square wave. I can check it up after the Musikmesse.

(There's a hidden switch for this that we forgot to put on the gui - space for future updates)

[moscom_electronics]

Nice, thanks a lot!

[Sendy]

Thanks Urs for the reply

indeed the rev2 is really (really, really... to the point where I may sell my MS;)) close to the real thing. I will keep using this one. Do you plan to update the rev 1 filter in the future?

I'm not sure. Andy from Cytomic might do the ultimate emulation and even he's been struggling. This seems to be one hell of s difficult issue (google "stiff problem")

and also,the ring mod sounds different compared to the hardware. Any hints on that?

I think we used the way the Legacy collection did it, using the actually selected waveform from VCO1 instead of just the square wave. I can check it up after the Musikmesse.

(There's a hidden switch for this that we forgot to put on the gui - space for future updates)

I'm fairly sure that at least on my copy of the Legacy MS-20, the Manual states that ringmod takes place between square and the currently selected waveform, but in reality it's always between the square and variable pulse. Personally it's nice to have the option to shop around with different values, if it doesn't mess up the historically correct setup (whatever that may be!)

[Sendy]

(google "stiff problem")

I wouldn't. Just think of the spam you're setting yourself up for.

[moscom_electronics]

I'm fairly sure that at least on my copy of the Legacy MS-20, the Manual states that ringmod takes place between square and the currently selected waveform, but in reality it's always between the square and variable pulse. Personally it's nice to have the option to shop around with different values, if it doesn't mess up the historically correct setup (whatever that may be!)

I also like this option a lot anyway, DIVA sounds great per se, so it is not really a problem if it does not match an old dinosaur