Zebra 3.0 - out now

[Aloysius]

i love zebra.
something about the grip of the modulations makes it so unique and inspiring and always sets it apart from other synths. with the routing/mixer architecture being so consequently open and free, possibilities truly seem to be infinite, no bottom to be reached. with like no other synth sometimes you are half an hour deep into designing a sound and then just one supersmall increment on one parameter opens up a whole new spectrum. otherworldly. one-of-a-kind.

but back on my topic here and to pick up on some other posts in this thread, i am using zebra for nearly ten years now and i too have to say, the one thing i always get stuck on is that the MSEGs are not usable as "free-running" LFOs (synced to host tempo).
I mean, it is right there! in the most awesome way possible! complex edit tools, bezier, grid, etc.....
it breaks my brain. it truly does.
that i cannot simply click and have them running unaffected by any note/gate-triggers seems to me to be such a limitation.

as it strangely seems to be unclear to the majority here, why this functionality would be insanely useful, let me explain.
the idea is (even the need to mention that seems crazy to me), to have parts of the track automation already included in the patch.
so for example, what i and i am sure others as well, find myself doing often is having a little swell in multiple parameter values towards the end of a 4-(or 8 or 16 or 32)bar section, simply to have a little transition to resolve on the new "1" downbeat, where all the modulation cycles start anew again.
top candidates for this are for example env release or of course cutoff but really any parameter is employable for these little bumps at the end of the respective cycle.

techniques (not even really worth that description, its just so basic) like that are absolutely vital in so much of sequenced dance and electronic music for the last 3 decades. just look at all the tools popping up over the last years that offer exactly that functionality: lfotool, shaperbox, infiltrator, modulation processor from dialog audio, gooey snake, duck, tantra, gatekeeper... clearly there seems to be quite the demand for it.
and of course this is only one use-case among many. in fact, only scratching the surface. having these super tight, super flexible, graphically well represented curves as lfos would expand the overall functionality of zebra by quite a lot, without being more overwhelming.

if i understand it correctly, as they are implemented now, only long held notes can really benefit from the usage of long MSEGs, for a pluck or any plain eighth or sixteenth note there is no use of the possibility of having note-on triggered envelopes being multiple bars long.
for long held notes/chords sure, you can draw in beautiful curves over the duration of the note (or even over the duration of consecutive notes, but only if the first one is being held), but not everybody is trying to be Klaus Schulze these days.

It for sure is subjective but i dont need 8 (!) note-on triggered envelopes for my (mostly on the shorter side, sub 1/2 note) sequenced sounds. I'd rather have 8 lfos (or more)! or even better the ability to switch between env/lfo-functionality according to the task at hand.


on a related note, i saw with some sadness that the global lfos were cut out of the zebra 3 release.
so we now have only 4 true cyclical modulation sources.

sure, i can have the mappers "triggered"/modulated by an lfo (or external cyclical source via ABCD) to have them run as stepseqs/lfos, but that means basically sacrificing said lfo/ext source for having that mapper run cyclically, as lfo and mapper then have of course the same tempo division and only a "ramp up"-waveform functions as reliable trigger for a simple unidirectional step sequencer function.
once more, not being able to simply click to have the mappers run as true simple step sequencer loops, is to me such a limitation. with the possibilty to have up to 128 steps a rudimentary automation/arrangement could already be included in the patch! again, it is right there!

is trigger-agnostic cyclical modulation somehow looked down upon in the more scientifically-oriented synth community or frowned upon as too simple or too "new school" (which it isnt!)?
true cyclical MSEG-functionality "not rooted in actual sounddesign needs", "is only for the sake of having it", "not thinking how it can be useful"??? (paraphrased qoutes from this very thread)
gentlemen, please...

for now, i enable these additional cyclical functions with external modulation sources (cubase modulators) via the ABCD-knobs (that awesomely became automateable since the last beta release, thank you!) but the point of sacrificing stands here as well, these knobs could be used for more advanced modulation purposes than simple lfo tasks. still, it is at least a way to get to the functionality and i am glad about that!

of course, all that could be accomplished via MidiCC or host automation but then you have the problem of the parameter being no more accessible via the respective knob on the GUI. much better to have it included internally in the instruments own excellent mod system. also of course it keeps the patch portable across platforms.

for any user saying it is already possible to reach true note-agnostic functionality with the MSEGs, no it is not. the "single"-trigger option doesnt cut it, it is still note-on dependent.
but i ll be happy to learn otherwise. i have a 4/4-bar with say 5 discrete 16th-notes spread across it and want to increase say the env release exponentially over the second half of the bar via an MSEG (i know there are other ways, eg via velocity etc, but that is not the point here). how?

and just to be clear, MSEGs are of course amazing as they are as note-triggered envelopes, this text here is just a pledge to expand their functionality.

so, to sum it up: if the MSEGs and Mappers would be able to run as independent cyclical lfos/stepSeqs (preferably with length up to 64 or even 128 bars!), i would welcome it soo much, it would be heaven for me. HEAVEN!
-----> please Urs, can you make it happen some day?



while i am here all wishlisting away, i cant resist to mention three more items i'd love being implemented:

- having individual outs for the 4 out mixer channels (Main/1/2/3).
would save me so much time, as now i often record the layers individually in multiple passes, having one armed while the others are muted (in the signal mixer). individual outs would enable multitrack recording in one go. (many synths would benefit from such a feature)
having the layers as individual audio tracks is highly valuable for arrangement/mixing purposes, a bit less so when using the 1/2/3 as sends, but even then.

- mute buttons for the out-mixer channels (post-fader/knob).
would be helpful for setting/comparing levels, as due to the complex possibilities of interconneting the channels of the two mixers sometimes muting a lane in the signals mixer doesnt mute enough or mutes too much of the out-mixer channels content and one has to resort to turning the respective level knob to zero, which then has to be set again, creating some sort of levels whack-a-mole. clean separated mutes at the end of the 4 chains would solve this.

- 2 more compressors to then have 4 available, one for each of the 4 mixer channels (the compressor is one of the best ones implemented natively in any synth i know btw, although i liked the beta design a bit more for having the gain reduction more exactly readable)


to round it out, some few possible inconsistencies i came upon (cubase 15/windows10):
- curve mod in OSC/MSEG is no longer accessible via the little grey triangle, was very intuitive that way, now only accessible via the knob.
- modulation source value in the mod matrix has inversed mouse-wheel behaviour, scroll-up reduces value
- secondary modulation source value in the mod matrix defaults to 100 when activated, should be zero as with the primary modulation source


this is my first KVR post, although obviously i follow these subjects for a very long time, it was that much a matter close to my heart that i had to chime in but it also is primarly a true ask for the described functionality, which would improve working with zebra so much for me.

i hope the post is comprehensible and not too convoluted with all the brackets and insertions. as you know, zebra is a super complex creation in itself and one has to try to be exact when writing about it.

thank you and cheers!

The user praises U-He Zebra for its uniquely deep, open-ended modulation system, but argues that its MSEGs and mappers are limited by being note-triggered rather than freely running, tempo-synced modulation sources, which restricts common electronic music workflows. They strongly advocate for adding independent cyclical LFO/step-sequencer functionality (plus a few workflow improvements), saying it would massively expand creative possibilities and make the synth even more powerful.

[Aloysius]

Jesus, get a life.

Ah, “get a life”—such a compact phrase, so deceptively efficient in its delivery, and yet, when one pauses to examine it with even a modest degree of curiosity, it reveals itself to be a remarkably dense and multifaceted proposition, one that gestures toward questions of existence, agency, fulfillment, and the allocation of one’s finite cognitive and temporal resources.

To begin with, the imperative “get” implies acquisition, as though a “life” were a something that can be obtained, misplaced, upgraded, or perhaps even installed like a software patch—raising immediate ontological concerns about whether the speaker believes the recipient currently lacks such a thing, or merely possesses an inferior or misconfigured version of it. This alone opens an entire avenue of inquiry: what constitutes a “life” in the sense being invoked here? Is it a socially sanctioned pattern of behavior? A certain ratio of offline to online activity? A curated set of hobbies deemed acceptable by some invisible cultural committee?

Then there is the matter of implied deficiency. The phrase presumes that the current state of the addressee is somehow insufficient, that their present engagement—perhaps writing an elaborate, deeply considered post about a niche synthesizer modulation architecture—is evidence not of passion or expertise, but of absence. And yet, one could just as plausibly argue the inverse: that the willingness to invest time, thought, and expressive effort into a highly specific domain is, in fact, a hallmark of having a richly textured life, one populated by interests, curiosities, and a desire to refine both tools and understanding.

Furthermore, the directive offers no actionable parameters. If one were to sincerely attempt to comply—if one were to stand up, metaphorically speaking, and say, “Very well, I shall proceed to get a life”—what would the next step be? Should one diversify interests? Reduce depth in favor of breadth? Engage in more socially visible activities? Or is the suggestion less about actual behavioral change and more about a rhetorical shorthand for “your priorities do not align with mine”?

In that sense, “get a life” functions less as advice and more as a conversational compression algorithm, reducing a potentially complex disagreement about value, focus, and enthusiasm into a three-word dismissal. It bypasses engagement entirely, opting instead for a kind of semantic eject button.

And yet, paradoxically, its very brevity invites expansion—because the moment one interrogates it, even lightly, it collapses under the weight of its own vagueness. What seemed like a definitive statement becomes a prompt, a doorway into a broader reflection on how people choose to spend their time, what they consider meaningful, and why deeply caring about something—anything—is so often misread as excess rather than richness.

So, in the end, “get a life” might best be understood not as an instruction, but as an accidental philosophical question: what does it mean to have one in the first place, and who, exactly, gets to decide?

[ptnah]

Jesus, get a life.

my life is music, i thought yours was too, Mr.S.?

[jtsterays]

kvr

[PieBerger]

kvr

Synth go brr!

[bftucker]

Ah, a potential KVR Klassic Kharacter!

Welcome and best wishes.

[ptnah]

Ah, a potential KVR Klassic Kharacter!

Welcome and best wishes.

i dont know what inside jokes you geezers are on about.
i just really want that feature and thought i'd better make a detailed case for it.
what's the problem?

[pdxindy]

i dont know what inside jokes you geezers are on about.
i just really want that feature and thought i'd better make a detailed case for it.
what's the problem?

There's no problem. Don't take it personally, there's always some unfriendly voices on KVR.

All your requests are reasonable wishes. You articulated them well.

[bftucker]

“Tempering is a process that reduces brittleness and relieves internal stresses in hardened steel by heating, increasing toughness and ductility while slightly lowering hardness to prevent cracking”

Don’t lose your temper. We geezers like to tease.

[BBFG#]

There's a support group axiom that states "It's easier to buy a pair of slippers than it is to carpet the world."

They also often list "litigiousness" and "quixotic" as character defects. And the medical experts agree.

Perhaps "getting a life" is simply a constructive nudge to sign out and go take a walk in the real world.

[ultrashall]

Update2: Found the source of error now: I had installed Zebra3.vst3 beta in a subfolder in December, so there were two different versions of the plugin in the VST3 folder. Deleting the old version solved the problem.

However, since the new GUI had been introduced, Cubase crashes (freezes) every time when I try to open Zebra's GUI. Loading the synth into the DAW and playing the default sound - without opening the GUI - is no problem, on the other hand.

ultrashall, Germany

[ThoughtExperiment]

There's a support group axiom that states "It's easier to buy a pair of slippers than it is to carpet the world."

They also often list "litigiousness" and "quixotic" as character defects. And the medical experts agree.

Perhaps "getting a life" is simply a constructive nudge to sign out and go take a walk in the real world.

Perhaps...but to be fair, it wouldn't look that way to a new guy

[Sampleconstruct]

I'm sorry.

[BBFG#]

There's a support group axiom that states "It's easier to buy a pair of slippers than it is to carpet the world."

They also often list "litigiousness" and "quixotic" as character defects. And the medical experts agree.

Perhaps "getting a life" is simply a constructive nudge to sign out and go take a walk in the real world.

Perhaps...but to be fair, it wouldn't look that way to a new guy

Maybe somewhat directed to the verbose seasoned member waxing the philosophical as well.
Realizing I should just keep away from this thread since it has so little to do with its subject line now.

[pdxindy]

i have a 4/4-bar with say 5 discrete 16th-notes spread across it and want to increase say the env release exponentially over the second half of the bar via an MSEG (i know there are other ways, eg via velocity etc, but that is not the point here). how?

Not arguing against your feature request.

But the example above can be done with a Math Module and an LFO.