u-he @ Superbooth 25 - Zebra 3, Uhbik 2, Cen2rion, CVilization

[Shiek927]

Oh, that's more than fair! Focus on the essentials for now and then you can tap into the community, former Zebra2 users and, naturally, Hans Zimmer, to see where'd they like Zebra3 to organically go next.

[Howard]

What about preset variations?


Base preset and expandable tree of variations.

Not really worth it. Any benefit of organizing presets into "variation" folders is offset by the hassle of having to dig through them—instead, just naming presets like "Preset 1 – Alternative Macros" is simpler and more efficient.

[Urs]

What about preset variations?


Base preset and expandable tree of variations.

That concept is new to me. It seems like something that is even more work intensive and thus even harder to convince sound designers of, or requires us to put three or four people on the task to create variations for.

I suppose with a significantly simpler synth this could be done programatically, but not for Zebra 3. It won't be possible for developers to reliably distinguish what makes a presets CPU intensive, what makes it play sequences, let alone what makes one set of standard macros useful and another useless.

[Urs]

Oh, that's more than fair! Focus on the essentials for now and then you can tap into the community, former Zebra2 users and, naturally, Hans Zimmer, to see where'd they like Zebra3 to organically go next.

That's what we're doing.

Zebra 3 initially comes without a sequencer, nor an arpeggiator. We need presets and example cases based on the performance oriented features to see how any of this integrates with onboard sequencing.

[edited for clarity]

[Fannon]

XYs are a complicated topic. We spent more years on setting up XYs for Zebra2 and Hive preset banks than it took to make those banks. That is despite both to some degree having automated setup.

Many sound designers reject setting up macros. The time/price per preset goes up significantly, and often someone else (a poor coworker) has to go through the draining act of setting them up.

Hence, for now, there will be no XYs or macro controls.

We have an idea that is more digestible to sound designers and that gets us half way there. We will expand the MIDI A/B controllers to A/B/C/D, and add on-UI-control over them. We will require each factory preset (or bank in our shop) to be set up with useful modulation on all 4 of these. So if we ever bring back like 8 macros or 4 XYs, half of them will already be set up to mirror MIDI A/B/C/D.

Thanks for sharing your thoughts on it, I can understand those reasons.

Well, I'm also not sure if I would really need 4 XY pads, but having at least 4 macro knobs I would see as quite essential. It's just so convenient to modulate several values with one knob in a controlled, pre-defined way.

For a macro knob it would need to be clear if its unipolar or bipolar (like for XY) - if you want to keep XY pads open to the future, maybe it should be clear that macro knobs are meant to be bipolar from start? With 4 bipolar macros, I feel you can already do so much. Having 8 macros + 2 MIDI A/B felt more than necessary to me, and I can see how it's bothersome to do this well for each patch.

So if MIDI control gets A/B/C/D and they are also the macros (with labels and well placed in UI), it feels like a good simplification and starting point.

[Miralis]

Now that the whole product lineup is almost fully upgraded, I guess some new exciting software will be announced next year.

Out of curiosity, are there still plans for a Cat emulation (or even a Voyetra)? I also remember that there was a poll a few years ago around autumn to gather ideas/feedback. I wonder if any suggestion stuck.

[Urs]

I think we'll be revisiting a few of the things we always wanted to do.

We have a well of technologies that we have developed over time which have not yet been implemented in any of our products, or just as stripped down versions, such as the transient shaping module in MFM2's input stage.

The Cat is still a huge favourite of mine. I'm merely for a monophonic version though.

[Hanz Meyzer]

So, was video about Zebra 3 taken at Superbooth 2025? If so, where to find it?

[Urs]

So, was video about Zebra 3 taken at Superbooth 2025? If so, where to find it?

It hasn't appeared yet. Pretty bugged out... he seems to be posting one video a day, let's hope it still comes.

Other than that, our booth was packed like never before. It was a constant presentation of Z3 and Uhbik. I guess there wasn't really much of a chance for the usual suspects to take a video, except for those who arranged for appointments.

[Chalisque]

This has turned into a bit of a ramble, but anyway...

I get the thing about macros: preset designers being obligated to map them if they're in the synth. And sequencer and arpeggiator: the place for all those is in some kind of universal synth host (vs every synth and its dog having its own take on the arpeggiator and sequencer, each one offering different features, being programmed differently, and having things arranged differently in the ui and so on). But nobody's got that concept right yet, and all too often (e.g. NI or Novation) try to turn it into an exclusive walled garden. The other issue with having a plugin-host-as-a-plugin vs a built-in sequencer/arp, is the limited means for the intermediary host to specify per-note stuff (e.g. consider the arp mod parameters in Z2).

On the other hand, it is very useful to have a knob for, say, filter cutoff (to pick something at random), for which you can set a lower limit, and an upper limit, so that e.g. the full range of the macro knob moves the filter cutoff in the range 2k-5k or whatever. When a single knob controls the entire range of a parameter, most of which lies outside the useful range, you're left with a few pixels of slider range to play with. (Often with Z2, without shift things are too coarse, and with shift they're too fine.)

[eclipse_soundlab]

What about preset variations?


Base preset and expandable tree of variations.

Not really worth it. Any benefit of organizing presets into "variation" folders is offset by the hassle of having to dig through them—instead, just naming presets like "Preset 1 – Alternative Macros" is simpler and more efficient.

In my mind, it works like this:
The base preset is collapsed and, initially, no one needs to dig into the tree. But when someone likes the preset and finds it useful for production - yet needs a thinner, less intrusive version - they can click the "+" sign to expand the tree and get a ready-to-use preset tailored for that.

It's more of a bonus feature for producers and composers who are building their own library or sound palette - not something meant to show off in the featured factory bank.

Doing dark ambient now - I have examples where two presets and their variations are taking half the preset browser screen. Because theres no chords with one preset each note gets own preset voicing.

And setting up macros to transform the preset is too much work (and creates more convolution in further edits) - I'd rather just tweak the preset directly for higher registers or different dynamic scenarios to make it work. (and save it as variation)

[pierb]

And many users never even touch them.

Probably the majority.

I will add XY mapping for custom sound design projects if the composer requires it but I stopped doing it as a default in my soundsets.

[Funkybot's Evil Twin]

I’ve been thinking about the X/Y system, and honestly, I wonder if just 1 or 2 X/Y pads might be more practical in the long run. Four feels like a lot, especially in real-world use.

I remember the public tagging project—it was brutal! You’d have a simple patch with one oscillator, no filter, a basic envelope, no effects… and still have to come up with four meaningful X/Y assignments. The X/Y complexity vs patch complexity ratio could get way out of whack!

In many cases, finding four good, musically useful X/Y mappings took more time than creating the patch itself. And that didn't guarantee anyone else would find them musically useful!

Then from the user side, when browsing presets, if you see unused X/Y's, there’s this weird sense of, “Huh…why did they not bother finishing the patch?” Which just leads to this weird disappointment.

Another thing: a lot of people got excited about mapping X/Ys to external controllers or building OSC templates. Especially when first getting Zebra2. But really, how many of us actually stuck with those setups and used them consistently?

Personally, I’d advocate for a single X/Y pad—or maybe even a different macro system altogether. Just my two cents, but I was never a huge fan of the 4-pad X/Y approach in Zebra 2.

[pierb]

Zebra 3 comes without a sequencer, nor an arpeggiator.

That's disappointing. As a sound designer, getting a better sequencer was one of my most expected features for Zebra 3. I was hoping to see something like what's available in Serum 2, Avenger, or the latest version of Nexus.

Hopefully you'll reconsider and at least add what's already available in Hive or Repro.

[Shiek927]

Zebra 3 comes without a sequencer, nor an arpeggiator.

That's disappointing. As a sound designer, getting a better sequencer was one of my most expected features for Zebra 3. I was hoping to see something like what's available in Serum 2, Avenger, or the latest version of Nexus.

Hopefully you'll reconsider and at least add what's already available in Hive or Repro.

It'll get one, just not in the 1.0 version. Same with the XY pads. If every conceivable feature were available at launch, Zebra3 would never come out. Plus, they want to see the community response and where the audience wants to go.