DIVA known issues

[ford442]

here's a small issue - i was in REAPER just now and when i closed the main window of REAPER, the box with 'save, yes no' popped up underneath DIVA's GUI.. i don't think that is normal for VSTs in REAPER..

[Viktor [TUC]]

Has bug reporting now switched to this thread from the beta forum? I might have missed that.
So, before these go unnoticed, I'll post them here too, and sorry, in case they've already been mentioned a million times:
2 bugs (or sort of strange features?) and 2 things for the wishlist:

1. In the modifications tab, the knob next to the lag input/ouput, which I assume is the amount of lag, could use some recalibration. There's little effect up to 99, then at 100 suddenly there's "ultimate" lag. That could be more evenly distributed, I think.

2. When I use a squarewave LFO for stereo panning, I get severe clicks when the sound switches from left to right when playing. Even routing that LFO through "lag" first, then to panning, doesn't remove the clicks completely. Maybe the LFO section could use an "almost" squarewave with rounded edges, a less perfect squarewave?

(now the less important stuff)
3. Also under "modifications", I could use "amount" knobs for modwheel and pressure. Example: Sometimes all inputs for filter modulation are taken, so I add the modwheel and, say, an envelope together (with the adder function), but I don't want the modwheel to open the filter as much as the envelope does. And if the other filter mod input is also not scaled to what I want the modwheel to do, I run out of options.

4. The vibrato could go twice as high, for my taste. Of course I could always use another LFO to directly modulate the pitch, but if they all have different jobs already, I would have to make a sacrifice.

Also, can someone tell me the difference between mod sources "KeyFollow" and "KeyFollow2" ?

Cheers,

Le Cokteur

[Synthbuilder]

First off, this is one ruddy marvellous VSTi.

So these are not exactly a bugs but simply observations:

1. The ladder filter clearly allows itself to be overdriven. You can see that when you stack three sync'd VCOs together at the same pitch and push all three of them into the filter at full volume. You get that lovely soft overdriven sound that ladder filters do. Probably some tanh waveshaping going on there.

However, the DIVA doesn't have enough gain on any individual input to allow the filter to go into serious overdrive. So can we have some more gain to allow any one input to force the filter into overdrive. This would bring it into line with the Minimoog and Micromoog. The latter of which seems to overdrive the filter loads.

2. Maximum filter FM modulation depth is, I think, too low. This is particularly apparent on the ladder filter. I'm a bit of a filter FM fan so perhaps I'm being too fussy.

3. I'm not quite sure what the feedback knob is doing but it's not sounding like the Minimoog's output to input 'trick'. Not that I'm actually too keen on that sound anyway.

4. Waveshaping. There's an apparent pitch shift going on when moving from the sawtooth to the pulse. It's caused I think because the pulse wave is being morphed into the wave from the left hand side of the waveform. This creates an apparent phase shift which when modulated appears as a frequency shift to the ear/brain. It's an interesting sound but I'm not sure I like it. I can't be sure whether the Voyager or Micromoog do this but my memory tells me that they don't. I could be wrong though.

Phase shifts like this are an interesting side effect of generating pulse waves from the falling sawtooth waveform in a real VCO. They can be avoided by using the triangle wave as a reference which generates a symmetrical change in pulse width with respect to the central point of the waveform.

Build 360 Win XP SP3

Tony

[Ingonator]

3. I'm not quite sure what the feedback knob is doing but it's not sounding like the Minimoog's output to input 'trick'. Not that I'm actually too keen on that sound anyway.

The Moog Little Phatty (or Slim Phatty) got a knob for filter overload which is one of my favorite features in my Slim Phatty. Would be nice to have something like this in Diva. I agree that the Overdrive (or the drive in the Rotary speaker FX) does not really do the trick.
In the Waldorf Blofeld for example the filter drive (i'm mostly using the tube drive) makes the difference between the synth sounding "average" and sounding real fat.
A nice external Drive i got is Schwa Oligarc Drive but i still prefer the one n the Little Phatty/Slim Phatty for this.


Ingo

[hakey]

I get some noise when playing polyphonically using LFO "single" to modulate, eg, vibrato, filter etc. - experienced the same problem with Tyrell.

EDIT: Example patch: Noisy LFO

Q6600, 2GB RAM, Edirol UA25, Windows XP, Live 8, Beta 360

[Urs]

First off, this is one ruddy marvellous VSTi.

So these are not exactly a bugs but simply observations:

1. The ladder filter clearly allows itself to be overdriven. You can see that when you stack three sync'd VCOs together at the same pitch and push all three of them into the filter at full volume. You get that lovely soft overdriven sound that ladder filters do. Probably some tanh waveshaping going on there.

However, the DIVA doesn't have enough gain on any individual input to allow the filter to go into serious overdrive. So can we have some more gain to allow any one input to force the filter into overdrive. This would bring it into line with the Minimoog and Micromoog. The latter of which seems to overdrive the filter loads.

Diva's maximum gain with the triple osc and Transistor Ladder is exactly the one in our Minimoog, and the bahaviour is pretty similar. It might be different in others.

The overdrive of e.g. the Sallen-key filter is a few dB higher than in our MS-20s.

Thing is, those filter circuits only operate in a certain range. Exceeding that range may lead to unexpected behaviour

2. Maximum filter FM modulation depth is, I think, too low. This is particularly apparent on the ladder filter. I'm a bit of a filter FM fan so perhaps I'm being too fussy.

This might indeed be possible. We're replacing the generic and expensive filter tuning algorithm with empirical tunings for all sorts of quality settings and sampelrates. In this vein we might be able to extend teh range without sacrificing performance.

3. I'm not quite sure what the feedback knob is doing but it's not sounding like the Minimoog's output to input 'trick'. Not that I'm actually too keen on that sound anyway.

Well, we've worked on this very long time. Our SH-09 and Minimoog behave very similar to what we got. I'll give it another try though.

4. Waveshaping. There's an apparent pitch shift going on when moving from the sawtooth to the pulse.

We're currently using so called miniBLEPs to simulate instant changes in level or angle of the waveform. This method is erratic and will be replaced by a method that is not prone to dc offsets

[Urs]

I get some noise when playing polyphonically using LFO "single" to modulate, eg, vibrato, filter etc. - experienced the same problem with Tyrell.

EDIT: Example patch: Noisy LFO

Q6600, 2GB RAM, Edirol UA25, Windows XP, Live 8, Beta 360

LFO single is broken, but fixed in next beta.

[crackershaun]

When doing fast runs Diva "seems" to miss the occasional note.

I can't do this "on demand" and it appears to be just occasional.

Because I have played the same type of runs with so many other VSTs, and not noticed it, I am pretty sure it is real, but I can't work out what seems to be causing it.

I think it is when notes are extremely closed together and not with full velocity - but this could be a "crimson kipper".

[crackershaun]

When doing fast runs Diva "seems" to miss the occasional note.

I can't do this "on demand" and it appears to be just occasional.

Because I have played the same type of runs with so many other VSTs, and not noticed it, I am pretty sure it is real, but I can't work out what seems to be causing it.

I think it is when notes are extremely closed together and not with full velocity - but this could be a "crimson kipper".

Oops - Already known at atart of thread!! "occasionally missed notes (need reproducible procedure)"

[Urs]

Yeah... we really need a reproducible Logic/Reaper project or something for it... I can't work it out at the moment...

[crackershaun]

Re: "occasionally missed notes (need reproducible procedure)"

Could it be to do with "legato"? I just switch the patch I was using to "mono" and it seemed OK, but missed some notes in "legato".

Cannot do it on demand though - so this could be "by accident"!

[hakey]

Not sure if this is a bug or a design feature - double clicking on the oscillator detune knob returns detune to zero *and* the oscillator range knob to its default position.

It might be more useful if only the detune was altered and range left where it is.

[Urs]

Not sure if this is a bug or a design feature - double clicking on the oscillator detune knob returns detune to zero *and* the oscillator range knob to its default position.

It might be more useful if only the detune was altered and range left where it is.

Fixed in our internal builds... (we're beyond Rev400 now)

[hakey]

In that case, I give up. Seems any bug I discover is already old news...

[Urs]

In that case, I give up. Seems any bug I discover is already old news...

Nono... keep them coming...