DIVA: "Unison" vs. "Stack"

[Nokenoku]

Hi

I am using a lot of DIVA lately!
And now I'm actually getting to the limit of my processor (i5 4670K @ 4.4GHz).

One thing I noticed is, that "Stack" in DIVA with for example 3 voices seems to actually triple the amount of needed ressources.
Earlier I though, "Stack" would just be "Unison". But is it maybe, that not only the OSCs get multiplied but also the whole signal flow including filters?

In that case, it would be cool to have a "Unison" mode as an alternative, since it would save quite a big amount of processing power for some patches, and in a lot of cases would not even sound that much different.

[Urs]

Yes, "Stack" is a true form of unison with whole voices multiplied. That's important because just doubling oscillators and running thorugh the same (mono) filters would neither allow for stereo spreading nor sound the same, due to the intermodulation distortion.

This is why we added the Digital Osc. Because it has a Supersaw model that does the more cpu friendly oscillator-unison-thing.

We've tried to add wave multipliers to create multiple "detuned voices" out of the oscillators, but the result isn't anyhwere as good as one would imagine.

Cheers,

- Urs

[Nokenoku]

Yes, "Stack" is a true form of unison with whole voices multiplied. That's important because just doubling oscillators and running thorugh the same (mono) filters would neither allow for stereo spreading nor sound the same, due to the intermodulation distortion.

Ah, that stereo-thing makes sense.
You could get that though with just two filter-streams if needed.

The distortion part makes sense as well ... but in some cases it might even sound good or won't make a big difference.

I get, that the stack mode (as it is) is the more puristic approach.
Still I think an optional "only multiply OSCs" mode would be great for certain patches to free up some ressources.

Greetings