New DIVA components

[Ch00rD]

I'm not sure I want any further vintage analogues.

Phew!

Phew indeed - I'm glad I somehow managed to sneak my favorite vintage analogues into Urs' possession. Hopefully he's not spending all his cash on six-packs of batteries nowadays.

It could also be that "this is the actual model carefully and lovingly modelled by u-he" would increase the resale value of such old-timers. If that would turn out to be the case, and (big) if I ever decideIf to sell them, I'll be glad to kick back half the profits to u-he for funding his high-maintenance Diva's & Dino's.

[fmr]

Please enlighten me. What does the Voyetra Octave offer so special? The prices mentioned are ridiculously expensive.

[GruvSyco]

Well, from what I read it's special offer even before vintage was being finicky and prone to not working. At least that's what I gather from a few New Order interviews.

[olikana]

New components would have to offer something substantially different from what is available.
currently diva has
the 2 most sought after 24dbs : Moog and Roland juno/jupiter
the 2 most sought after 12dbs : ms20 and oberheim

it misses maybe a classic Curtis CEM and an SSM to cover all shades of 24 db. but it covers more than enough already.

I think a formant filter would be more useful at this stage.
when i dreamt of building a modular i thought i would end up buying this http://www.doepfer.de/a104.htm
and the Trautonium is definitely one hell of a dinosaur! interesting read.
http://www.doepfer.de/traut/traut_e.htm

[News From The Sky]

The Voyetra 8 has the SSM filter, one of the reasons it's considered one of the best-sounding analog polysynths. The VCOs can output saw, pulse, triangle, and suboctave waves simultaneously in any combination, and there's a separate noise source. It has some fairly powerful modulation capabilities (including independent PWM, linear FM of VCO1, and exponential FM of VCO1, VCO2, VCF, or Q), but an excruciatingly painful UI, another reason why a good emulation would be welcome.

That said, my suggestion wasn't serious (don't worry Urs!). US$6000 is an insane price to pay, though the original retail price was $4600.

I'd be happy to just get a good SSM 4-pole LPF in Diva - the Dual VCO can cover most (though not all) of the same ground.

[EsmileZ]

New components would have to offer something substantially different from what is available.
currently diva has
the 2 most sought after 24dbs : Moog and Roland juno/jupiter
the 2 most sought after 12dbs : ms20 and oberheim

it misses maybe a classic Curtis CEM and an SSM to cover all shades of 24 db. but it covers more than enough already.

I think a formant filter would be more useful at this stage.
when i dreamt of building a modular i thought i would end up buying this http://www.doepfer.de/a104.htm
and the Trautonium is definitely one hell of a dinosaur! interesting read.
http://www.doepfer.de/traut/traut_e.htm

Oberheim xpander with his different 15 filter types

[fmr]

New components would have to offer something substantially different from what is available.
currently diva has
the 2 most sought after 24dbs : Moog and Roland juno/jupiter
the 2 most sought after 12dbs : ms20 and oberheim

it misses maybe a classic Curtis CEM and an SSM to cover all shades of 24 db. but it covers more than enough already.

I think a formant filter would be more useful at this stage.
when i dreamt of building a modular i thought i would end up buying this http://www.doepfer.de/a104.htm
and the Trautonium is definitely one hell of a dinosaur! interesting read.
http://www.doepfer.de/traut/traut_e.htm

Oberheim xpander with his different 15 filter types

THIS

[Ch00rD]

New components would have to offer something substantially different from what is available.
currently diva has
the 2 most sought after 24dbs : Moog and Roland juno/jupiter
the 2 most sought after 12dbs : ms20 and oberheim

That summary makes quite clear what exactly is still missing, right there in the middle: the 1 most sought after 18dbs : TB-303.

[stillshaded]

model Waldorf waves scanning oscilators.


multiply sound possibilities by x100

[EvilDragon]

Or, even better, the crusty PPG Wave 2 wavescanning.

[Sendy]

Or, even better, the crusty PPG Wave 2 wavescanning.

Or Zebra's wave morphing oscillator with an optional "crust" switch which adds alising and makes the switching between waves less smooth.

Actually, I'd LOVE a synth with the option to reduce it's control-rate globally. A lot of the chiptune sounds I like get their computery quality from modulation that's made up of rapid steps instead of smooth changes. If that feature was in the tweak panel it'd add a whole new dimension of character to Diva.

[3*s]

When do we get this JP8000 osc? Don't mean to be impatient but the suspense is killing me.

[News From The Sky]

Or, even better, the crusty PPG Wave 2 wavescanning.

Or Zebra's wave morphing oscillator with an optional "crust" switch which adds alising and makes the switching between waves less smooth.

Actually, I'd LOVE a synth with the option to reduce it's control-rate globally. A lot of the chiptune sounds I like get their computery quality from modulation that's made up of rapid steps instead of smooth changes. If that feature was in the tweak panel it'd add a whole new dimension of character to Diva.

The "quantize" modulation processor in Diva basically does this (though not globally). Perhaps "All" could be added as an available modulation source.

+127 on the wavetable oscillator (with a variable quality parameter) in Diva - the JP8k osc is a step in that direction.

And I'd much rather have an SSM 4-pole and the Xpander/M12 filter before the 303, personally, but I think that race is run.

[Sendy]

Indeed the quantize function is good for stepped modulation, but it has it's own drawbacks. Modulation passed through it is quantized in value as opposed to time, which is a weird distinction to think about and explain. Instead of a constant rhythmic change, the changes will occur depending how quickly the inputted modulation is changing. Like I said, hard to explain

But basically a sample and hold "circuit" is much more fluid.

Another thing I dislike about quantize is that there's a weird asymmetry around zero, which can throw off the timing of "groove based" rhythmic effects.

[Ch00rD]

For 'stepped' modulation you could simply forego internal modulations and use parameter automation from the host for everything, then increase its audio buffer size. That way you decrease the update frequency of all parameter modulations globally. The only thing that may be getting in your way is the built-in parameter smoothing, which results in less-than-perfectly-steppy changes, comparable to a slight portamento effect on the pitch of notes played.