u-he rePro in the works

[Urs]

it's a matter of five days to create an algorithm thats suspiciously similar sounding.

Maybe take 10 days, and make it sound identical?

Send me your QRS and I'll give it a shot!

(part fo why we did the Protoverb experiment is, we don't have a QRS and I don't really want to cough up the dough for a Yardstick based on a hunch)

[Urs]

Also: just because something takes X amount of cycles, it doesn't necessarily follow that the ONLY way to get that sound is to use X amount of cycles. For all I know, the Bricasti is using 30 seconds of time varying direct convolution (i.e. dense "velvet noise" that is frequency + amplitude shaped and constantly randomizing), and therefore requires computing several thousand delay taps per output sample. Which would make for an amazing reverb, but there might be WAY CHEAPER ways of getting a similar sound.

Yep.

The Bricasti feels much like cleverly layered reverbs. The theory that convolution is used might be true if it's true that Casey had a computer run some algorithm (genetic?) for a few years. Maybe what came out of it wasn't an algorithm per se, maybe it was a set of tens or hundreds of impulse responses that layer nicely.

However, visually and audibly the Bricasti still has that Lexicon-thing going, which is that stereo ping pong effect one gets with low diffusion settings (in the Lexicon style ones). Interestingly - and IIRC - the taps spread out instead of closing in on each other, which looks a lot like what one could do by increasing the number of delay-allpass sections in the Lexicon/Alesis/Schroeder concept. As typical reverb decays shouldn't exceed, say, a few seconds until decaying below oblivion, I'm convinced that a traditional Schroeder approach will get one close to the Bricasti beyond human perception (there!).

Also, I haven't yet come up with tests to see if the assumed layers in the Bricasti interact in any way. If they don't, splitting the algorithm onto multiple cores is trivial.

[Urs]

So how about that rePro? Seems like it will be cool. I'm glad U-he's taking on stuff like this. Wonder what the pricing will be? I know I'm not the first to say it, but I'd really like to see a U-he Prophet 5.

Rev 3 of the Prophet 5 is probably closest to the Pro One because it uses the same Curtis chips. However, I've looked at the schematics and they're not exactly the same. There's a pesky little capacitor that might make the filter sound different. Furthermore I often hear that Rev 2 sounds better, and it uses a whole different set of chips... as much as I'd love the idea, as long as we don't have the original hardware we won't consider it.

Question for Urs: where's the modulation section and amp ADSR?

The screenshot is for the "research ware" version and even for this isn't all that final. We'll have a knob for Osc2->Cutoff because it'll help evaluating the filter quality.

If we do a full emulation, it'll have the full feature set, but also a few tweaks necessary to make it a good experience. For instance, it would be awkward to program the sequencer as is, i.e. without a rest button.

That said, at the moment I wouldn't even rule out a poly version, but this wouldn't exactly be like any polyphonic Prophet. The concept of the final product is also a bit of work in progress because we might be ambitious enough to extend it somewhat. After all, we posted this because we wanted to secure "RePro" as a name for a plug-in because of its versatile features for puns and wordplays on vintage synthesizers.

The pricing will depend on the extent of what I described in the previous paragraph.

[Urs]

I would also like to follow up on the debate of whether or not it makes sense to continue doing emulations of analogue synths.

If we look at our sales numbers, Diva is still the strongest seller - neither Hive nor Bazille could change that in the long run. Zebra generates the most revenue, but also because of ZebraHZ with its Diva filters. If we look at feature requests in general, the request for more analogue models in Diva or Zebra is probably the number one request, way ahead of drum machines, samplers or what have you.

If we were a purely marketing driven company, from the data we have and including a reflection of our strengths and reputation, we would focus on emulations of analogue synthesizers. We have two defining factors that drive sales: Our endorsement for Zebra by Hans Zimmer and our reputation for Diva as the reference for analogue sounding softsynths. Both are co-dependent, because Diva came about due to Hans' wishes for better filter quality in Zebra. He sent us his Minimoog, we used it as reference, created Diva and eventually built those filters into his special edition Zebra - which we're allowed to market with his name. In effect, building analogue sounding filters is what we should do most, were we purely driven by marketing agendas. (also in line with some stuff Hans' and I talked about during dinner a few weeks ago)

However, I can't cope with the idea of being driven solely by marketing forces. I'm merely motivated by development challenges and my motivation has a relatively short life span. For instance it'll be difficult for me to get back into the Protoverb business right now - even though a commercial version would create some nice income. It'll have to wait until some day I have some sleepless nights pondering about feedback delay networks. Before Protoverb it was preset tagging and related stuff. But at the moment it's filter stuff. I couldn't care less about anything else. There's a short window of concentration, maybe a few weeks, and I have to use it (which includes teaching my employees so they can work on these things when my window is done). This is how I work best, and if it coincides with whatever a marketing department would say, even better!

[glokraw]

Successful marketing of products consumers want or need,
funds further technical progress and perfection.
Hums 'Let The Circle Be Unbroken' on the way to the bank

[fluffy_little_something]

Nothing wrong with emulations, do one of the Rhodes Chroma

[TheNickC]

I don't see u-he's emulation efforts as an either or thing, Zebra and Bazille both have lots of good strictly digital tools -as well as- the analog modeled filters. It's not as if Urs has simply rehashed what's been done in hardware decades ago, these things benefit the straight up digital synthesizers as well. I don't care at all personally about VA being 1:1 identical with real analog, it just needs to sound good. But if researching hardware is how u-he makes good filters they're not wrong for doing so, the results definitely validate the approach. There are also developers who specialize in additive synthesis or physical modelling or whatever other digital techniques you'd like to see refined, it's ok for different niches to exist and for some people to be "stuck in the past" if you want to call it that, the growth of technology is doing just fine.

[valhallasound]

Rev 3 of the Prophet 5 is probably closest to the Pro One because it uses the same Curtis chips. However, I've looked at the schematics and they're not exactly the same. There's a pesky little capacitor that might make the filter sound different. Furthermore I often hear that Rev 2 sounds better, and it uses a whole different set of chips... as much as I'd love the idea, as long as we don't have the original hardware we won't consider it.

I've heard that the Pro-One sounds considerably better than a single voice of the Rev 3 Prophet-5. I owned a Rev 3.3 Prophet 5 in the 1990s, and it didn't stand out as the best sounding synth. A poly Pro-One might be a better emulation goal.

[nevernamed]

I would also like to follow up on the debate of whether or not it makes sense to continue doing emulations of analogue synths.

I think if you like recreating vintage analogue circuits in software it makes sense to do that . It doesn't need rationalising much beyond that. What doesn't make sense is being prescriptive about what's a worthwhile investment of one's time and effort.

And on the Diva hardware thing what might make sense is partnering up with a hardware company (risk management)? I'm not sure who the natural ally might be though. Maybe the Waldorf chaps? That knobby 5 octave Waldorf Q interface looks comprehensive and could be repurposed maybe.

[Funkybot's Evil Twin]

A poly Pro-One might be a better emulation goal.

What he said.

Reading this site got me totally stoked for this release:
http://greatsynthesizers.com/en/test/se ... onophonic/

[nevernamed]

...or maybe a Kickstarter project?

[pdxindy]

And on the Diva hardware thing what might make sense is partnering up with a hardware company (risk management)?

I think it makes sense to forget the idea and keep making cool software synths like they are obviously good at and can be successful doing. That door is wide open, full of creative potential and has little risk and lots of reward and the iron is hot so to speak.

There is already 3-4 years worth of projects wanting to be completed. Down the road, who knows. Keep making great stuff, and some hardware manufacturer will approach u-he.

[Urs]

A poly Pro-One might be a better emulation goal.

What he said.

Reading this site got me totally stoked for this release:
http://greatsynthesizers.com/en/test/se ... onophonic/

Hmmm... that description is a bit over the top. The Pro One isn't exactly the greatest/phattest/cleanest/smoothest sounding synthesizer of all times. But it is very flexible, making it a good workhorse mono synth.

[AnX]

it's a matter of five days to create an algorithm thats suspiciously similar sounding.

Maybe take 10 days, and make it sound identical?

Send me your QRS and I'll give it a shot!

(part fo why we did the Protoverb experiment is, we don't have a QRS and I don't really want to cough up the dough for a Yardstick based on a hunch)

It was very flipant remark...and i have no idea what a QRS is

[EvilDragon]

Quantec Room Simulator. Yardstick hardware reverb box.