u-he

Urs · Post by **Urs** » Fri Apr 15, 2016 8:35 am

urosh wrote:I guess this is really difficult, if not most difficult VCF to accurately emulate.

Thanks for your analysis! I haven't even begun to math this out.

Thing is, the Polivoks filter sounds wildly crazy and each sounds different. I'm fairly sure that one could get away with a bunch of shortcuts, if only one finds the right approach.

urosh · Post by **urosh** » Sat Apr 16, 2016 4:13 pm

Urs wrote: Thanks for your analysis!

Well, I wasn't much of a help.

davidka · Post by **davidka** » Tue Apr 19, 2016 2:23 pm

EvilDragon wrote:
knowix wrote:Was Cytomic modelling the Polyvoks filter, or the Wasp? I can't remember.
WASP.

--> First hit on YT

urosh · Post by **urosh** » Wed Apr 20, 2016 9:43 pm

Regarding that bug I mentioned earlier (VCA mode getting stuck to Env even when Gate is selected on front panel), I've tested on different machine, in 32b and 64b Reaper, both 32b and 64b VST2 and VST3 version, and result is the same. I save project with VCA in Gate mode, on reopening it's Env and it's not possible to set it to Gate. Now, does anyone else have this problem except pdxindy? And is there any chance that this might get addressed? Thing is, RePro Alpha very much ruined for me everything else ITB for analog-like bass duties (at least), and I have snaky suspicion that filter model I like will not end up in final version, so I would be more than happy with more usable alpha version.

Urs · Post by **Urs** » Wed Apr 20, 2016 9:57 pm

urosh wrote:I would be more than happy with more usable alpha version.

Hehe, sorry... we've moved the codebase quite a bit since the alpha was built and we didn't expect the need to fork it. The gate option isn't even in it anymore since the dedicated VCA envelope took over. So the best bet would be the final version I'm afraid...

I'm not sure which filter model you voted for... I'm not even sure if the final version will be exactly the same, because I have since discovered a few more aspects that I'd like to see modeled.

urosh · Post by **urosh** » Wed Apr 20, 2016 10:25 pm

Urs wrote:So the best bet would be the final version I'm afraid...

Then pretty please with sugar on top, sneak me in on beta testing program. Currently, when I reopen project I manually recreate settings to new instance and then copy midi+envs from saved one, and I sample it. I might as well get Pro One.

Urs wrote:I'm not sure which filter model you voted for...

I've voted 4, but since I've moved to 1, it sometimes has subtle but definite edge on plucky low cutoff reso bass patches. I thought it might be that more high end is fooling me, but did not have luck with EQing others to have quite the same punch/whip.

****
On unrelated topic, I made simplified model of voltage amplification stage (mentioned earlier in discussion about Polivox filter), with just 3 exp functions. And I got convergence problems in simulator. Go figure, I thought convergence problems for simple topologies were myth.

david.beholder · Post by **david.beholder** » Thu Apr 21, 2016 12:51 am

urosh wrote: Then pretty please with sugar on top, sneak me in on beta testing program. Currently, when I reopen project I manually recreate settings to new instance and then copy midi+envs from saved one, and I sample it. I might as well get Pro One.

He he I want to get into beta testing too.

aaron aardvark · Post by **aaron aardvark** » Thu Apr 21, 2016 4:28 am

I would also be interested in a beta testing program since I have owned a real Pro-One since 1981.

AnX · Post by **AnX** » Thu Apr 21, 2016 4:59 am

Uhe seems to do public betas these days, so you already are on the team, kinda.

Urs · Post by **Urs** » Thu Apr 21, 2016 7:18 am

It's not even in beta testing yet

david.beholder · Post by **david.beholder** » Thu Apr 21, 2016 8:24 am

Urs wrote:It's not even in beta testing yet

May be all 3 of us could join alpha program?

urosh · Post by **urosh** » Sun Apr 24, 2016 8:55 pm

Ok, I've found another issue, this might be real issue for U-he team. On particular patch I get occasionally loud pop/click (attached screenshots from editor). This happens with models 1 and 4. What is really really weird is that (when it happens, which is not every time) it happens on the same place on same note in sequence. Just that particular note in sequence, and it always happens on edge of first cycle of saw wave after note start.
Not much of an issue for me, but it is quite noticeable, and I certainly would mind if RePro started to click loudly in the middle of live performance.
(Urs, if you need it I'll send whole project and renders via email)

u-he-william · Post by **u-he-william** » Mon May 09, 2016 10:11 am

urosh wrote:Ok, I've found another issue, this might be real issue for U-he team. On particular patch I get occasionally loud pop/click (attached screenshots from editor). This happens with models 1 and 4. What is really really weird is that (when it happens, which is not every time) it happens on the same place on same note in sequence. Just that particular note in sequence, and it always happens on edge of first cycle of saw wave after note start.
Not much of an issue for me, but it is quite noticeable, and I certainly would mind if RePro started to click loudly in the middle of live performance.
(Urs, if you need it I'll send whole project and renders via email)

Hi,
a late response:
The clicks and pops you got are a side effect of the way filters 1 and 4 are implemented. No need to worry, it will not be in the full release.

Cheers

louderr · Post by **louderr** » Wed May 11, 2016 9:48 pm

EvilDragon wrote:
urosh wrote:
Urs wrote: Polivoks filter is 12 dB LP/BP and the most f**ked up filter ever put in hardware. I'd be happy to model it, but I currently dont know how.
What's the problem, too computationally expensive or not possible to solve with current approach you guys use in Diva/RePro?
Polivoks filter is extremely unpredictable and has lots of non-linearities in it. It's so "bad" that some people say that there's in fact no such thing as "Polivoks filter", there's just an idea of how it's supposed to work (and it rarely works as intended).

Schwa OLGA plugin has a filter based on such design, AFAIK. It's probably not as minutely modelled as filters in Diva, but it sure screams like a pig being slaughtered!

Interesting comment. Yeah Olga is a good unit and fills in a lot of gaps in the virtual synth world. I thought it was all about the overdrive so this is useful to know from an educational standpoint. PS I've been following your posts, O Evil One, and find them on target each time. Thx

louderr · Post by **louderr** » Thu May 12, 2016 3:33 am

Funkybot's Evil Twin wrote:
Urs wrote:I would need an audio example to see what they mean. My guess is, due to the principle-induced difference in filter frequency, one filter method often has more "resonance locking" artifacts than another at a given oscillator frequency, or it pronounces some noise a bit more than another. That said, I don't think that any of the filters differ much in their aliasing behaviour, and that's also not the point of this.

In general, yes there is a tad of aliasing, there always is, "aliasing free" is a myth, but it doesn't make much of a difference for the result. The final version (which goes to 50+ kHz) can be tweaked for less aliasing, if need be.
I don't think it's aliasing but 1 has a TON of artifacts at max resonance and a fairly low cutoff compared to the others. I think that's what people have been calling aliasing. The difference in the amount of noise and behavior is obvious in moscom_electronic's post here: http://www.kvraudio.com/forum/viewtopic ... 0#p6431179

I'm pretty sure that's why some people think 1 sounds terrible, while others kind of dig the instability of that filter.

YES

u-he

KVR Audio

[ANN] Repro-1