The long DIVA thread

[Urs]

I really like the softer sound of analogue...Diva is very accurate and I think you are right - Diva does sound more direct and it still sounds much more modern to me than real analogue. However this I think is good for people (a lot of people) who make modern music styles...I heard some ABs of Diva with a MOPHO on Gearslutz and I think Diva sounded better in the contaxt it was in (modern electro sound)- but it was still less analogue because of this 'direct' and accurate aspect.

They compared Diva to a Mopho?!? Why would they do that? It's perfectly clear that Diva hasn't got a dual oscillator with Subosc... Get your Minis out guys...

For someone who loves more old school sound - I'd love it to have this option on Diva. However - rather than a static frequency response - could it be more dynamic softening for the high freq? Static hig freq cuts sound muffled and dull to me - whereas dynamic/saturated highs sound musical and sweet...

Agreed. We'll see what's possible before release, otherwise we'll add it later.

As far as I can tell both Howie and Clemens are into the idea, more so than I am. So we'll eventually get there.

[Echoes in the Attic]

Would it possible to simply lift the Arp right out of Zebra and smack it down on a page in DIVA?

It's not that easy. We have not yet touched the multicore things, and we have yet to see how it affects future options for architectural changes. We're relatively sure that we can not have a ModMatrix like in Zebra or TyrellN6 as is, which is why I added the Adder/Multiply section. I'm also not sure about sequencers and arpeggiators that add a level of complexity to the MIDI Note stream, because we need to pre-schedule the whole timing of voices to ahve them render on different cpu cores.

I'm afraid that, if we did add that extra complexity, we will run into trouble with the multicore support. And as of now, I'd prefer multicore and playability over seq/arp. An arpeggiator or sequencer (or both) may surely be an option for a later update, and we'll work on this stuff next year when concentrating on Berlin Modular.

I figured there would be a bunch of technical reasons why it wouldn't be as straight forward as it sounds.

By the way, what't he deal with multicore support? Is there going to be a standalone version? I thought plug-ins cpu allocation was decided by the DAW. For example Ableton Live allocates each track to a different core. Is that not the case with other DAWs?

[SWAN808]

They compared Diva to a Mopho?!? Why would they do that? It's perfectly clear that Diva hasn't got a dual oscillator with Subosc... Get your Minis out guys...

haha I know what you mean. Must be confusing after you slavishly model a specific synth. However I suppose they are more comparable in real life price and availability. Ultimately the decision this market might think about...so it is relevant I think. And you can argue that comparing to ANY real analogue is relevant because of the general shared qualities of analogue sound...even the chip / digital based MOPHO...

For someone who loves more old school sound - I'd love it to have this option on Diva. However - rather than a static frequency response - could it be more dynamic softening for the high freq? Static hig freq cuts sound muffled and dull to me - whereas dynamic/saturated highs sound musical and sweet...
Agreed. We'll see what's possible before release, otherwise we'll add it later.

As far as I can tell both Howie and Clemens are into the idea, more so than I am. So we'll eventually get there.

Thats great - I really respect the extent you have applied to this emulation. It seems like the devil is in the detail so I think this aspect would be really worthwhile.

[MFXxx]

Would it possible to simply lift the Arp right out of Zebra and smack it down on a page in DIVA?

It's not that easy. We have not yet touched the multicore things, and we have yet to see how it affects future options for architectural changes. We're relatively sure that we can not have a ModMatrix like in Zebra or TyrellN6 as is, which is why I added the Adder/Multiply section. I'm also not sure about sequencers and arpeggiators that add a level of complexity to the MIDI Note stream, because we need to pre-schedule the whole timing of voices to ahve them render on different cpu cores.

I'm afraid that, if we did add that extra complexity, we will run into trouble with the multicore support. And as of now, I'd prefer multicore and playability over seq/arp. An arpeggiator or sequencer (or both) may surely be an option for a later update, and we'll work on this stuff next year when concentrating on Berlin Modular.

I figured there would be a bunch of technical reasons why it wouldn't be as straight forward as it sounds.

By the way, what't he deal with multicore support? Is there going to be a standalone version? I thought plug-ins cpu allocation was decided by the DAW. For example Ableton Live allocates each track to a different core. Is that not the case with other DAWs?

I have a child like mind. What are the possiblities and would there be any point in creating a module that would allow zebra, diva and ace to integrate to make one mega synth. Asking this question reminds me why I never became an enginneer and designed a car. Regardless the question stands.

[Funkasizer]

We have a noiseless version of the famous synth chorus, the ensemble modelled after a digital emulation of *that* ensemble (I don't have a Polysix for reference), the other chorus is taken from a legendary digital multifx box, but tweaked to my liking.

Wow. Now i know why that ensemble sounds great. I love the Ensemble FX from the Korg emulation. I never played a real one so this is my only reference.

I'll go and try to recreate my favorite Polysix patches with Diva now...
Will be difficult to replicate the SSM filter with Diva but i'll try.

I won't insist that the ensemble is perfect though, I just looked at the behaviour of the other one and made some wild guesses. There's 6 taps and 3 LFOs, and the relation of the LFO speeds may be slightly off, whereas I think the depths are pretty correct.

Regarding SSM, we have an Opera 6 which sounds nothing like Korg's Polysix emulation. That's why we haven't even bothered With any OpAmp ladder until we can get hold of either a Polysix or (better) an Rev 2 Prophet 5.

To my ears the ensemble effect has too much presence and feels as if the audio is being sucked into a vacuum hose. Therefore I prefer the ensemble emulation in the Korg Legacy bundle, and also the Dimension D emulation by EMS. The combination of the latter with the Diva synth engine brings me in pure heaven (typically, I use the Dim D in stereo mode, setting 1, 40-50% dry/wet). Hope this doesn't sound too negative, just trying to give some feedback. The Diva synth engine is just wonderful! Congratulations, Urs, with this fantastic achievement!! Cheers, Jack

[Bronto Scorpio]

We have a noiseless version of the famous synth chorus, the ensemble modelled after a digital emulation of *that* ensemble (I don't have a Polysix for reference), the other chorus is taken from a legendary digital multifx box, but tweaked to my liking.

Wow. Now i know why that ensemble sounds great. I love the Ensemble FX from the Korg emulation. I never played a real one so this is my only reference.

I'll go and try to recreate my favorite Polysix patches with Diva now...
Will be difficult to replicate the SSM filter with Diva but i'll try.

I won't insist that the ensemble is perfect though, I just looked at the behaviour of the other one and made some wild guesses. There's 6 taps and 3 LFOs, and the relation of the LFO speeds may be slightly off, whereas I think the depths are pretty correct.

Regarding SSM, we have an Opera 6 which sounds nothing like Korg's Polysix emulation. That's why we haven't even bothered With any OpAmp ladder until we can get hold of either a Polysix or (better) an Rev 2 Prophet 5.

To my ears the ensemble effect has too much presence and feels as if the audio is being sucked into a vacuum hose. Therefore I prefer the ensemble emulation in the Korg Legacy bundle, and also the Dimension D emulation by EMS. The combination of the latter with the Diva synth engine brings me in pure heaven (typically, I use the Dim D in stereo mode, setting 1, 40-50% dry/wet). Hope this doesn't sound too negative, just trying to give some feedback. The Diva synth engine is just wonderful! Congratulations, Urs, with this fantastic achievement!! Cheers, Jack

There will be a Dry/Wet/ controlfor the chorus in the final version iirc.

BTW: I can't wait for ValhallaÜberMod! The combination with DIVA will be awesome

Cheers
Dennis

[Darrenaudioguy]

Hi folks,

I have noticed that, at least in Reaper, Diva gives a PDC value identical to the audio hardware buffer size. So if I set my RME card to 1024, Reaper reports Diva is using 1024 in the PDC monitor. If I then change down to 256 on the RME, Diva follows suit again in the Reaper PDC monitor.

Is this expected behavior? I have tried a few other VSTi and they do not produce any value in the PDC monitor.


TIA

Darren.

[Ingonator]

We have a noiseless version of the famous synth chorus, the ensemble modelled after a digital emulation of *that* ensemble (I don't have a Polysix for reference), the other chorus is taken from a legendary digital multifx box, but tweaked to my liking.

Wow. Now i know why that ensemble sounds great. I love the Ensemble FX from the Korg emulation. I never played a real one so this is my only reference.

I'll go and try to recreate my favorite Polysix patches with Diva now...
Will be difficult to replicate the SSM filter with Diva but i'll try.

I won't insist that the ensemble is perfect though, I just looked at the behaviour of the other one and made some wild guesses. There's 6 taps and 3 LFOs, and the relation of the LFO speeds may be slightly off, whereas I think the depths are pretty correct.

Regarding SSM, we have an Opera 6 which sounds nothing like Korg's Polysix emulation. That's why we haven't even bothered With any OpAmp ladder until we can get hold of either a Polysix or (better) an Rev 2 Prophet 5.

To my ears the ensemble effect has too much presence and feels as if the audio is being sucked into a vacuum hose. Therefore I prefer the ensemble emulation in the Korg Legacy bundle, and also the Dimension D emulation by EMS. The combination of the latter with the Diva synth engine brings me in pure heaven (typically, I use the Dim D in stereo mode, setting 1, 40-50% dry/wet). Hope this doesn't sound too negative, just trying to give some feedback. The Diva synth engine is just wonderful! Congratulations, Urs, with this fantastic achievement!! Cheers, Jack

My favorite Chorus/Ensemble FXs so far are from Korg Polysix, Synthmaster 2.5, PPG Wave 3.V, Xils Synthix and the Schwa Oligarc Chorus but the Chorus/Ensemble FXs in Diva are very nice too.

Some of the "broad" sound of the Polysix emulation is from the unison Spread and not from the Ensemble. This should be possible to be replicated in Diva with a second OSC and/or the Voice panning. Or with a Unison in a later version...


Ingo

[justin3am]

I think I could crank something out tonight.

Well, I thought I'd have a video for you guys this morning but something seems to have gone wrong in encoding. The video quality is extremely poor and the colors are all messed up, so I've got it set to private at the moment. The sound quality isn't bad, but it was a bit of a rushed job so I think I'll just wait for the weekend when I can really put a good effort into making a quality video. I guess it's not a big deal anyway, since the offending video has been taken down.

[Urs]

By the way, what't he deal with multicore support? Is there going to be a standalone version? I thought plug-ins cpu allocation was decided by the DAW. For example Ableton Live allocates each track to a different core. Is that not the case with other DAWs?

Well, DAWs can only distribute a whole plugin among cores, but it's possible that a plugin distributes its voices among them, with a little overhead. This can ease things, expecially when some cores stay idle (that's even the case with Logic often, and Logic afaiaa is renouned for its multicore goodness (apart from a few bugs)).

Thus we'll try to do it, as an option. I'm confident it'll help.

[pdxindy]

Would it possible to simply lift the Arp right out of Zebra and smack it down on a page in DIVA?

It's not that easy. We have not yet touched the multicore things, and we have yet to see how it affects future options for architectural changes. We're relatively sure that we can not have a ModMatrix like in Zebra or TyrellN6 as is, which is why I added the Adder/Multiply section. I'm also not sure about sequencers and arpeggiators that add a level of complexity to the MIDI Note stream, because we need to pre-schedule the whole timing of voices to ahve them render on different cpu cores.

I'm afraid that, if we did add that extra complexity, we will run into trouble with the multicore support. And as of now, I'd prefer multicore and playability over seq/arp. An arpeggiator or sequencer (or both) may surely be an option for a later update, and we'll work on this stuff next year when concentrating on Berlin Modular.

Easy answer... make Zebra output midi (on the VST)

[EvilDragon]

Easy answer... make Zebra output midi (on the VST)

As I suggested a few pages ago.

[pdxindy]

Easy answer... make Zebra output midi (on the VST)

As I suggested a few pages ago.

Oh yeah?? Well I suggested it a few pages before that!!

[EvilDragon]

Then we're even!

[mystran]

Finally tried it. Don't like the envelopes (tried all 3 versions, somehow they refuse to do what I want, especially with fast settings) and it would be nice to be able to drive the transistor ladder and SVF at least another 6dB higher (or maybe up to 20dB; you can barely get any distortion out of them). I kinda like the OTA clean mode (the rough mode might improve with more drive). The Sallen-Key is the opposite to the rest and sounds best to my ears with oscillators around 10-15% (can't imagine ever wanting over 25% for any purpose whatsoever).

The oscillators are not bad, but I don't like the character of the drift. The 3xOsc sounds like it would sound nicer if the filters had stronger low freq response (as is it sounds weak, but has the most interesting character otherwise). In general the low/low-mid range sounds boringly flat.

One non-sound related thing I don't like: while dragging knob pressing/releasing shift has no effect unless you stop dragging and start dragging again with the new shift state. When doing large adjustments (where the final value needs accuracy) this means one needs to first do the approximate adjustment, then move mouse back on top of the knob (since it's usually not there anymore), then redrag with shift for fine tuning.

Overall I'm kinda disappointed. It doesn't sound bad. It actually sounds kinda nice, but I guess I was expecting more character or something. For that much CPU it just sounds far too flat and clean and basically far too nice. Not my cup of tea I guess.