Zebra 3 feature suggestions

[dermage]

Frankly, I'm fed up and I WON'T be buying Zebra3!

...I'll be getting it for free (as a Z2+Dark Zebra owner). Which is generous. So I can wait. I guess I'm not fed up. Not about this anyway.

I made this decision a year ago (when buying DZ during their soundset sale )
I bet all the people asking for Z3 not even half explored the possibilities of DZ (if ever)
I'm sure Z3 will be groundbreaking/game changer/whatever, but I'd rater wait 6 months-year longer than have a subpar/unfinished 3.0 (Massive X anyone?).
Anyway It looks like (as Urs said) they'll have some product releases in between, which should satisfy the U-HE needs.

Z3 is their most ambitious project yet and it has to live up to the work it took.

[Urs]

I also wish to reiterate this:

Shortly after Zebra 2 came out, it became a running gag to answer feature requests which were beyond its scope with "maybe in Zebra 3 then". It was clear to everyone that this wouldn't happen anytime soon as Zebra2 was pretty new. However, a few years later we got more and more support requests asking "shall I buy Z2 now or wait for Z3?" even though there was absolutely nothing going on. People started posting "I'll hold off until ..."

So one fine day we started some brainstorming over beers in our favourite hangout. We figured out what Zebra 3 could be, we did some drawings and feature lists. That was a few months after Diva was updated ("40% less CPU") and in a time where we thought we'd have the freedom to start something new (before RE, before VST3 issues, before AAX, before Mavericks I think, before Win8). It was also the time when The Dark Zebra was gonna come out.

It was very clear to us that no matter what, it would be months if not years until we'd have something to show. But we still had to address the issue that, apparently, more and more people were holding off on Zebra2, and dwindling sales. So this is when we devised the plan laid out in the infamous "Zebra3 info" thread: We assured people that Zebra2 was still the way to go, that it'll have a future and that buying it was making the upgrade even cheaper than holding off. We also added that owners of TDZ would get a free upgrade.

At this time we were estimating two years maybe, but we hadn't counted on the Rack Extension issue, operating systems upgrades, new plug-in formats and all these things that a) required us to do tedious maintenance upgrades and b) required us to release new products (mainly Bazille, Presswerk and Hive) which in turn also required further TLC for themselves.

Then also, people at u-he came and went. Staff became bigger, the company moved into bigger and bigger spaces.

I spent much of my time managing, training and supporting rather than developing. Hive, if you remember, was "Urs' weekend project". Almost all of the code I wrote was written on weekends, because the weeks were busy with other stuff.

However, every single release we did since 2011 contains technologies we created with Zebra 3 in mind: MPE and NKS stuff, Hive's wavetable playback, the technology behind filters in Bazille, Hive and Repro, Repro's voice management, Hive's improved ModMatrix, the new preset browser, CoCo's bucket-brigade-style delays (which we'll use for Z3's comb filters), and so on and so on.

All of these things had to happen "on the side", wrapped up into other projects. Even the drum synthesis modules we have (which weren't realistic enough to compete with sample based systems) will find their way into Zebra3, even if they probably won't be used in their own product.

At the beginning of everything we do, I ask myself: How does that fit into Z3? - Simply because we won't do it if it doesn't.

- U

[AnalogGuy1]

If we're blue-skying, how about integrating a patch-finder tool that lets a user feed it an audio snippet of the sustain-portion of a given sound and it returns the 10 closest installed patches, leaving it to the user to choose one and then adjust the ADSR envelope and tweak parameters?

Maybe with a "quick" option that just looks for a harmonic match (minimizing the error in the frequency domain), a "medium" option that adds some time-domain modulation information (same as before but minimizing error for several different, time-overlapped windows a-la STFT), and a "comprehensive" option (that tries matching for several different points in the xy modulators' dimensional space). (Since there is such a freakishly huge sound library for Zebra, maybe further fine-tune with pure time-domain analysis like a matched filter - with several thousand patches, we can afford to be picky).

Would require a little additional space in each patch to save pre-computed harmonic information, but perhaps better than wondering if a sound is "warm" or "thin".

The ultimate would be a machine-intelligence matrix trained to match a given sound to an installed patch, but we're not there yet - there's no way to pre-compute the training when every user has a unique patch library. It could, however, be trained against the factory default, which would give folks a pretty good starting off point...

[Urs]

Our attempts training a neuronal network to automatically tag presets have not been overly successful

[pdxindy]

However, every single release we did since 2011 contains technologies we created with Zebra 3 in mind: MPE and NKS stuff, Hive's wavetable playback, the technology behind filters in Bazille, Hive and Repro, Repro's voice management, Hive's improved ModMatrix, the new preset browser, CoCo's bucket-brigade-style delays (which we'll use for Z3's comb filters), and so on and so on.

All of these things had to happen "on the side", wrapped up into other projects. Even the drum synthesis modules we have (which weren't realistic enough to compete with sample based systems) will find their way into Zebra3, even if they probably won't be used in their own product.

At the beginning of everything we do, I ask myself: How does that fit into Z3? - Simply because we won't do it if it doesn't.

- U

That all sounds amazing!!

So when is Zebra 3 gonna be released?

[pgwalsh]

This is a great thread. I'm a newbie to Zebra 2, so I may be a bit naive of the capabilities, but I hope Zebra 3 incorporates more physical modelling. I had a Korg Z1 and it was one of the most fun synths to program. Between that and my Ensnoiq TS-10 I could create a wide variety of sounds.

I hesitated for years on purchasing Zebra 2 because of FUD, the modular concept is intimidating.

[lunardigs]

You know, I read these threads and find myself laughing out-loud often. Especially when reading belabored responses from Urs ... Yup, I must be a dork.

[digitalboytn]

That all sounds amazing!!

So when is Zebra 3 gonna be released?

Tommorow at 4pm sharp

[neverbeeninariot]

That all sounds amazing!!

So when is Zebra 3 gonna be released?

Tommorow at 4pm sharp

Timezone?

[Shiek927]

Urs' explanation about every major product being a tease to the eventual Z3 is pretty much what I've always known about it and they should all be causes for excitement. Zebra3 is going to be the new flagship of the company that's just as tonally flexible as Z2 if not much more, so it makes sense that it's going to have a little bit of everything i.e. a little bit of all the products released since 2011.

One question I did have is if Z3 was going to have any sampling and granular type of synthesis! I have a Roland V-Synth XT hardware synth that can do incredible kinds of manipulation. Not only is the sampling very natural sounding, even if it's just one note, but you can individually adjust the time aspect, pitch and formants, for some really unique and wild experimentation. It's very easy to create beautiful and lush textures, sampling real world sounds on one PCM oscillator and layering it with an analog oscillator. Is there any intention to do anything like this? cheers!

[Funkybot's Evil Twin]

Urs had previously said he's not a fan of samples in synths (paraphrasing) so I'd expect no.

[Stollmeister]

That all sounds amazing!!

So when is Zebra 3 gonna be released?

Tommorow at 4pm sharp

Timezone?

Gallifrey

[Shiek927]

Urs had previously said he's not a fan of samples in synths (paraphrasing) so I'd expect no.

Aww!! that would open up a world of possibilities, but I suppose some limits are a good thing. Would still love to see some sort of granular ability since it would open things up!

[exmatproton]

Zebra 3 feature suggestions : RELEASE TIME!

[pdxindy]

Urs had previously said he's not a fan of samples in synths (paraphrasing) so I'd expect no.

Aww!! that would open up a world of possibilities, but I suppose some limits are a good thing. Would still love to see some sort of granular ability since it would open things up!

Zebra 2 already has a vast sonic landscape...