Zebra 3 Public Beta (final beta)

[Echoes in the Attic]

The presets do not follow my taste. But I am sure the architecture allows me, to come up with presets that sound like real instruments. Just needs time to learn alot.

There will be a lot of those as well, I presume. It definitely is possible to do a lot of realistic sounding instruments. But I don't think it was the focus with the preview set of the factory library.

That piano sound in your video intro was pretty impressive.

[kraster]

Zebra 3 is not what I'd consider a "limited" synth in any shape or form.

…

It's also not really true that most modern synths are somehow limitless. Outside of modular environments Halion, Falcon, Msoundfactory and Phase Plant 99% of synths don't have limitless modules.

Zebra sits more with synths like Current, Serum 2, Rapid and Pigments. ie. A more limited but extremely versatile feature set.

You say it yourself: „..outside of modular environments…“

Zebra is marketed as a semi-modular synth. And no, it does not sit more with synths like Serum 2, Current …
It does more sit with synths like MSoundFactory, Phase Plant …

And compared to how unlimited you can use modules there, the „Numbered limited modules approach“ is a bummer.

Modular just means that the signal flow isn't set in stone like it is in fixed architecture synths. It doesn't mean infinite modules. Besides Zebra is Semi modular which implies that it has some characteristics of a modular environment but lacks others.

And yes it absolutely does sit with Current etc. because like those synths Zebra has always had a limited module paradigm.

Zebra is very good at expanding on something within its modules. Like the 16 part unison on each oscillator if you want big and wide. Or its enormously variable and impeccably modeled filters. Or it's amazingly versatile MSEGs or the mapping functions.

Again, rather than get bummed out by what it supposedly lacks delve into its enormous capabilities. It's a phenomenal piece of software.

[Urs]

The presets do not follow my taste. But I am sure the architecture allows me, to come up with presets that sound like real instruments. Just needs time to learn alot.

There will be a lot of those as well, I presume. It definitely is possible to do a lot of realistic sounding instruments. But I don't think it was the focus with the preview set of the factory library.

That piano sound in your video intro was pretty impressive.

Just don't play it outside the lower mid range...

And note, it's pretty hot by gain, and it's totally not a finished or polished preset...

Jazz Piano 4.zip

[zerocrossing]

God forbid we should expect an intuitive UI.

"Intuitive" is subjective. I found the Mono-Poly-Legato switch in the first fifteen seconds.

Maybe it would be better if everyone spent some days instead of just a few hours with Zebra 3 before passing judgement on the UI. The more time you spend with a UI the more intuitive it gets.

Is it 100% perfect for 100% of the people? No, of course not there isn't one single UI that passes that test.

I'm finding it easy to navigate but of course some of the deeper features are going to require time on task and that's fine with me....that's half the fun of it.

Of course some things are going to need to be learned, but Zebra has always had some issues in this respect. Even modulation used to be confusing. I've been using synthesizers since the dawn of the Juno 106, and working with spline based graphic editors since the mid 90s, and I find the wave shape editors to look pretty confusing. Hell, I couldn't even find a way to load a new wavetable. Even the way to morph through waves is called "curve morph," when the expected name would be "wave morph," like pretty much every similar type of software. Even "Shape Morph" would make more sense.

As a UI designer, I'm always pretty sensitive to this type of things. Sometimes a better idea should be discarded because there's already an established way to do things. For instance, I was tasked to find a way to replace the icon for "NPC has a quest" for you icon. I'd used the ubiquitous "!," which I agree makes little sense, but everyone who plays RPGs knows it. When we play tested alternative symbols, players were confused. What did we end up using?

https://www.charmedparticle.com/portfol ... ct-title-3

[zerocrossing]

Second critic is: I would appreciate to have a full modulation matrix and drag&drop approach, instead having all the fields und the parameter knobs. It's very limitating, it's not needed anymore and makes the synth look crowded. What was a good idea in 2007 doesn't mean it's still needed in 2025.

It is odd to have two very different modulation paradigms. To a new user, it seems like a lot of things are just not mod targets, when they are, but only in the mod matrix. I'd also appreciate a "focus" screen, where all modulation can be seen as a list, like Omnisphere does it.

[Urs]

Even the way to morph through waves is called "curve morph," when the expected name would be "wave morph," like pretty much every similar type of software. Even "Shape Morph" would make more sense.

We're pretty clear about the terminologies:

A Curve is a single complete waveform, spectrum or envelope, made up of stitched together Bezier Splines.

A Curve Set is the complete set of up to 16/8 of each

A Shape is something like a sine, square, triangle, blip whatsoever that can be drawn into a Curve. The Shape based editor offers a bunch of Shapes to be drawn into the Curve, and a Curve can consist of multiple Shapes, like you can draw 7 blips into a single Curve.

Morphing happens between Curves, regardless of what shapes they contain, hence "Curve Morph". The result of morphing is a Curve again, to be interpreted in whatever way, consisting of whatever shape or form.

A Waveform in this context is a Curve that's used in the oscillator, when interpreted as such.

A Spectrum in this context is a Curve that's used in the oscillator, when interpreted as such.

An Envelope in this context is a Curve that's used in the MSEG.

We use "Wavetable" in the actual sense of the word, as a waveform that is stored in memory as a set of 2048 samples. Those samples are not directly editable, but we render Curves into them.

Likewise, and Additive Spectrum is a set of frequencies and volumes of up to 1024 sine oscillators.

It is not difficult, and IMHO not even all that convoluted. But then, I've spent the past four or so years with this...

[zerocrossing]

And it's annoying that they are limited to 4.

If "King Crimson is not a band, it is a way of doing things", then maybe "Zebra is not a synthesiser, it is a way of doing synthesis".

The whole concept and workflow of Zebra is and has always been about the economics of connecting as few modules as needed for as much outcome as possible.

It is one middle ground between simple fixed architecture synths like Hive, Diva and Serum and full blown modular environments like VCV Rack or Reaktor.

It is not the only concept that tries to achieve this, there is e.g. Phase Plant, MPowerSynth and probably a few more that I don't recall. However, these concepts are not the same as Zebra, since they either sacrifice the depth that Zebra offers, or they sacrifice the workflow for whatever infinity they may have.

In my experience, if you can't make a sound with the modules in Zebra and its 4 filters, you're probably looking for a different concept. Maybe you're looking for a layer based approach such as Dune 3.

Speaking of which, Zebra is the opposite of a layer based approach. This has advantages and drawbacks. The advantage is, you can probably build more complex and/or refined patches quicker in Zebra than in a layered synth. The disadvantage is, if you need gazillions of the same kind of modules, you lose all the advantages. And then you need a different synth.

So, with all the comparisons about Diva, Serum, Dune, Phase Plant, Modular Synths and so on and so on, I think what most of the time is missed is that if you pick just one aspect, you might miss the bigger picture.

Besides, more than 4 filters per voice will bring CPUs into serious trouble

It's a semi modular. No one plays an Arp 2600 plugin and expects unlimited modules.

I'd much rather have a guardrail of "limited," (which is still pretty damn high) modules that sound great. Sure, MSoundFactory has the ability to add a lot more modules, but when you start using it, you will notice that you run out of CPU pretty quickly if you set the oversampling high enough to get the quality that's the default in Zebra 3. Same thing is true with modular plugins like VCV Rack. It's all fun and games until you start using polyphony and up the oversampling. I get that's what you want sometimes, but then Zebra isn't the correct tool for that job.

[Urs]

It is odd to have two very different modulation paradigms.

Pretty much every synthesizer since the Minimoog has a dual modulation paradigm

More obvious maybe in the Prophet 5, which has direct Modulation Sources such as Env->Filter, but also an early ModMatrix with a direct and a wheel controlled path, which are both neither visually assigned to source not to target.

We have always offered on-target direct modulation slots for common modulations. In some synths we offer a ModMatrix, which by principle is indirect. It is not a surprise, and not at all odd.

[zerocrossing]

Even the way to morph through waves is called "curve morph," when the expected name would be "wave morph," like pretty much every similar type of software. Even "Shape Morph" would make more sense.

We're pretty clear about the terminologies:

A Curve is a single complete waveform, spectrum or envelope, made up of stitched together Bezier Splines.

A Curve Set is the complete set of up to 16/8 of each

A Shape is something like a sine, square, triangle, blip whatsoever that can be drawn into a Curve. The Shape based editor offers a bunch of Shapes to be drawn into the Curve, and a Curve can consist of multiple Shapes, like you can draw 7 blips into a single Curve.

Morphing happens between Curves, regardless of what shapes they contain, hence "Curve Morph". The result of morphing is a Curve again, to be interpreted in whatever way, consisting of whatever shape or form.

A Waveform in this context is a Curve that's used in the oscillator, when interpreted as such.

A Spectrum in this context is a Curve that's used in the oscillator, when interpreted as such.

An Envelope in this context is a Curve that's used in the MSEG.

We use "Wavetable" in the actual sense of the word, as a waveform that is stored in memory as a set of 2048 samples. Those samples are not directly editable, but we render Curves into them.

Likewise, and Additive Spectrum is a set of frequencies and volumes of up to 1024 sine oscillators.

It is not difficult, and IMHO not even all that convoluted. But then, I've spent the past four or so years with this...

It's not difficult to understand. It just bucks convention. Is the red octagon stop sign the best thing for a stop sign? Maybe a hand shape would work better, but it would be confusing because people are so used to the octagon. Not that it's so confusing, but I had to stop for a few seconds when I wanted to modulate the waveform, which is ultimately what's happening, and think about it. I can't remember the last time I had to think about what control was the one that modulated the waveform.

[Teksonik]

Or, if you refuse to compromise,

Why should anyone want to (or have to) compromise? That's a defeatist attitude.

they force you to figure out a way to get done what you want to get done.

No, they force you to get something else done instead of what you wanted done.

I mean look what musicians did with early synths that didn't have near the functionality we have today. Greg Hawkes of The Cars wasn't creative? He played some of the most iconic synth riffs to this day.

Those limitations were mandatory and here's the salient point: because that's all they had at the time.

Are you saying that Greg Hawkes couldn't make more iconic synth riffs if he had the options we have today back in those days?

We have so much power at our fingertips these days why limit ourselves? I'll make an example of Zebra 3. I'm a huge fan of multiple built in Arpeggiators in synths. I love using them in different directions at different speeds triggering different Oscs, etc (and before someone even starts typing, no you can't do that with the Arp in your Daw). I've made countless such patches in other synths and use a lot of them in compositions.

At this point in time Zebra 3 does not have even a single Arp (no, the Mappers are not the same). So I am forced to compromise by not being able to make those patches that I love in Z3. Do you see what I'm saying?

What I'd love to see is 4 Arp Modules in Z3 that could be placed on each Column instead of having an Arp "Mode". That would certainly follow the semi-modular paradigm of Z3. Is that going to happen any time in the future? Who knows, it's out of my control.

Zebra 3 is a great synth and we can love it for what it is today but please no one try to dumb it down by pushing back against features that will expand the range of sounds it can make. That's a self defeating attitude.

[Urs]

It's not difficult to understand. It just bucks convention. Is the red octagon stop sign the best thing for a stop sign? Maybe a hand shape would work better, but it would be confusing because people are so used to the octagon. Not that it's so confusing, but I had to stop for a few seconds when I wanted to modulate the waveform, which is ultimately what's happening, and think about it. I can't remember the last time I had to think about what control was the one that modulated the waveform.

It just would be inconsistent if we said "Morph Waveform" in the one editor, when the same function is "Morph Envelope" in the other. And dawg only knows what else we'll do with the editors in the coming years. Then we have to say "look, here's a tutorial, but whenever we talk about morphing a waveform you need to imagine it say morphing whateverelse".

So the challenge we're posing is to learn that the tools are the same, regardless of what the underlying Curve represents. Plus, sometimes in the same editor it can represent completely different things. Hence we stick to the structural terminology, as the semantics can change (oh hey, terms from my own studies in product design)

[Teksonik]

Sometimes a better idea should be discarded because there's already an established way to do things.

Then my question would be do you advocate for a major rewrite of a plugin that's been in development for who knows how many years? How many man hours of work would that create? I don't have any idea I'm just asking. I'm not sure what you want discarded and added. Maybe it's a couple of lines of code or a couple of months of coding. Either way I don't oppose anything that makes Z3 better.

Keeping in mind the hundreds if not thousands of patches created in Zebra 2 and those already for Z3 by countless sound designers (not to mention Z2 being used in several movie themes) and we see proof that Zebra is very usable in it's current state.

Again is it 100% comfortable for 100% of the people? Nope, nothing is...

Don't get me wrong, I have a laundry list of things I 'd like to see added. Some them are features of convenience and some are features of function which add to the range of sounds that can be made. I'll give a specific example of a feature of convenience.

I've been at it since 4am with Z3 and one thing that keeps seeming like a good idea would be for a way to have the contents of each column shown in order in the left hand panel and not showing any modules that aren't in that stack.

Z3 121125-2.png

Notice how Osc 4 is selected and shows on top (I have that option chosen in Rack View) but none of the other modules in that stack are shown without scrolling. I've tried the other two rack view options and they don't seem to accomplish the task unless I'm missing something.

There is room at the top of each column for a "Focus" button where if you click on it only the modules in the column that has focus would show in the pane of the left in their natural order. Sometimes I want to edit the modules in a stack one after the other in order without having to click back and forth.

Does that make sense to anyone else or am I trippin? It's not something that adds more features or accomplishes a task that can't already be done, it's just something that makes things more....convenient. Again I don't know if that would take hours or months to accomplish.

Anyway I'm going back in to spend a few more hours with Z3........

[coroknight]

It's 2025 and after 11 years of development Zebra3 is finally in public beta and it's still not good enough

Users need unlimited LFOs, MSEGs and envelopes, with deep feature sets each, cutting edge DSP, effortless workflow and all for $99 new or a free upgrade from version 2 that they bought 15 years ago

Yes it’s not good enough. Even in Serum we now have 10 fully featured MSEGs which also function as LFOs.

Zebra 3 might have “extra” LFOs tucked into various places but… why? It’s so much simpler to have a list of modulators in one place that can be used everywhere.

[exmatproton]

I'm a huge fan of multiple built in Arpeggiators in synths. I love using them in different directions at different speeds triggering different Oscs, etc (and before someone even starts typing, no you can't do that with the Arp in your Daw). I've made countless such patches in other synths and use a lot of them in compositions.

At this point in time Zebra 3 does not have even a single Arp (no, the Mappers are not the same). So I am forced to compromise by not being able to make those patches that I love in Z3. Do you see what I'm saying?

Or use different synths that do EXACTLY what YOU want, or try a different approach with a synth that makes you think about how to get a specific result. Play with those different types of gate triggers, MSEG's and ModMappers..they are awesome.

I for one love the limitations. Makes me think harder to get to a certain result, with often surprises along the way. Which i find cool. Otherwise it is just waiting on a AI tool with one text input field for a prompt and a 'run' button.

Just use Serum or something else if you think Z3 isn't enough for you..quite easy choice right?

[KBSoundSmith]

Sometimes a better idea should be discarded because there's already an established way to do things.

Then my question would be do you advocate for a major rewrite of a plugin that's been in development for who knows how many years? How many man hours of work would that create? I don't have any idea I'm just asking. I'm not sure what you want discarded and added. Maybe it's a couple of lines of code or a couple of months of coding. Either way I don't oppose anything that makes Z3 better.

Keeping in mind the hundreds if not thousands of patches created in Zebra 2 and those already for Z3 by countless sound designers (not to mention Z2 being used in several movie themes) and we see proof that Zebra is very usable in it's current state.

Again is it 100% comfortable for 100% of the people? Nope, nothing is...

Don't get me wrong, I have a laundry list of things I 'd like to see added. Some them are features of convenience and some are features of function which add to the range of sounds that can be made. I'll give a specific example of a feature of convenience.

I've been at it since 4am with Z3 and one thing that keeps seeming like a good idea would be for a way to have the contents of each column shown in order in the left hand panel and not showing any modules that aren't in that stack.


Z3 121125-2.png


Notice how Osc 4 is selected and shows on top (I have that option chosen in Rack View) but none of the other modules in that stack are shown without scrolling. I've tried the other two rack view options and they don't seem to accomplish the task unless I'm missing something.

There is room at the top of each column for a "Focus" button where if you click on it only the modules in the column that has focus would show in the pane of the left in their natural order. Sometimes I want to edit the modules in a stack one after the other in order without having to click back and forth.

Does that make sense to anyone else or am I trippin? It's not something that adds more features or accomplishes a task that can't already be done, it's just something that makes things more....convenient. Again I don't know if that would take hours or months to accomplish.

Anyway I'm going back in to spend a few more hours with Z3........

Try the Group function -- select the modules you want to display, and a blue/white + will appear just above the modules. Click on that and it will create a Group (which you can name with a double click). Now when you click on the Group Header you've just created, only the modules in that Group will display on the left.