Zebra 3 - Sound design, tips & tricks

[tq]

We do have this thread for Zebralette, but I think we also need a tips & tricks thread for Zebra. Personally, the multitude of options is still a tiny bit overwhelming for me, so I would love to hear from you how you achieve a specific sound or where you did find interesting ways of doing things.

As a start, I'd be very interested to hear how some of you would go for a more lo-fi, wobbly, fragile sound - something like old dusty analogue that comes from your granny's tapes? (I like what the Unfinished does here, and also Beautiful Void Audio.) I see that some people use LFOs to do minimal pitch variations to thicken the sound and make it more unpredictable, but also noise and the new OSCs might be good tools. Opinions, tips & tricks?

[tq]

Ah well, this thread doesn't fly...

[Chris-S]

I see that some people use LFOs to do minimal pitch variations to thicken the sound and make it more unpredictable, but also noise and the new OSCs might be good tools. Opinions, tips & tricks?

Yes, pitch wobble via LFO is a good start, additionally roll of the highs and lows with EQ, add some distortion/saturation.

[Chris-S]

keyboard splits

A Mapper module can be used to create key splits. Check out the preset "Dr Chow's lownely Digital Band" for example.

[Urs]

keyboard splits

A Mapper module can be used to create key splits. Check out the preset "Dr Chow's lownely Digital Band" for example.

I am experimenting on a one-octave GM drum kit with 12 distinguished instruments... using positive elevation of ModMappers to set volume of instrument layers, negative elevation to modulate stuff in others. Differences in Snares and HiHats are mostly created by different MSEG Curves, which I use one full Mapper for to select per key. The same Mapper defines pitch for Toms.

I might even use one octave to play loops for base and hihats...

[NAD]

As a start, I'd be very interested to hear how some of you would go for a more lo-fi, wobbly, fragile sound - something like old dusty analogue that comes from your granny's tapes?

The most immediate way to do that is by going to the pitches tab and in the calibration section selecting "destabilize" (and obviously cranking the amount way up). Here's a quick preset I made to demonstrate that:

Bored of Canada.zip

I also used some of the new features in the noise module to add some reasonably convincing (at least to my ears) vinyl noises.

(Side note: I would have liked to make the oscillator polyphonic and the noise monophonic so that it doesn't stack up when you play chords, but I couldn't find a way to do that. Any ideas?)

[Urs]

(Side note: I would have liked to make the oscillator polyphonic and the noise monophonic so that it doesn't stack up when you play chords, but I couldn't find a way to do that. Any ideas?)

Interesting... I have to think about that.

[spektralisk]

I was trying to build a dynamically changing, variable-length sequence. I approached this using a Mapper: an LFO increments the Mapper, and I wanted a second trigger, running at a different rate, to reset the Mapper.

I can see there’s a Manual Reset option available. Is this something that can be addressed as a modulation target, so it can be triggered programmatically?

Is there another recommended way to implement this kind of behavior? I was thinking of MSEG but I can't see a way to trigger it with a modulator.

[Urs]

So for now we avoid having modules which are both trigger sources and targets. Mappers and ModMath are the exception, but they are also pretty simple, code wise, as they do not have an internal timing mechanism.

For different length MSEGs I would suggest simply using the Curve Morph, and probably quantize and/or S&H any modulation. But it depends on what you want to do.

[FragileX]

Hello Urs,

I was wondering if it would be possible to implement some "pitch end modes".
Like it is now "always all voices" are bend when pitchbend is used.
In a "sustained only mode" the released notes wouldn't be affected by pitchbend.
And, what I would really love to have, would be a "last played note only mode" where only the last note would be affected by pitchbend.

This would be fantastic for us not yet having a keyboard with poly Aftertouch...and be able to play a bit more expressive.

Thanks for considering this.

By the way - thanks a lot for Zebra 3. I really like the sound and am starting to wrap my head around it !!
Which you all a nice time and fun & success with Zebra 3!

[Tuskerfort]

(Side note: I would have liked to make the oscillator polyphonic and the noise monophonic so that it doesn't stack up when you play chords, but I couldn't find a way to do that. Any ideas?)

Interesting... I have to think about that.

Not to derail a sound design / tips & tricks thread with something nobody asked for , but this exchange nudged a half-baked idea I have been carrying around, and it felt just coherent enough to mention as a concept, not a request.

I have long been intrigued by keyboard-logic-aware modulation sources. That is, modulation driven by what the player is doing musically, not just by velocity, aftertouch, or controller movement.

One example would be a signal representing the number of notes currently sounding. With noise layers, this could theoretically let noise pull back as chord density increases, without the player having to manage it manually.

More broadly, it suggests musically intuitive behavior. Dense rhythmic comping sits lower, single-note lines step forward, purely as a consequence of playing style.

In the same dangerous-ideas bucket, a controller derived from the consonance of currently sounding intervals (octaves, fifths, seconds, etc.) could allow harmonic relationships themselves to influence timbre or level.

Again, this is just a thought experiment, not a feature ask, and very possibly a terrible idea in practice. It felt close enough to the poly vs mono noise discussion to mention, so I will leave it there and get back to happily patching with a silly grin on my face.

[stippenstoh]

Has anyone a trick for this: I randomly pan every new note via a random LFO driving the pan on the main grid. But here's what I would like to achieve as well: the higher the notes, the further to the sides the random pan range should move.

Also, it's a polyphonic sound (so I don't think the main grid pan can work)

Example: when a low and a high key are played simultaneously, the random range of the low key should stay close to the center while the random range of the high note should be further to the sides.


I am mostly worried about low frequencies moving too far from center, so using the Bass Mono Utility would do the job - but maybe there is a more elegant way?

Mapper in 'Key' mode?

[NAD]

Has anyone a trick for this: I randomly pan every new note via a random LFO driving the pan on the main grid. But here's what I would like to achieve as well: the higher the notes, the further to the sides the random pan range should move.

Try this on for size:

Key-Tracked Random Pan.zip

I have it set to output a static random value for each note but you can change that to an LFO in the A slot of Mod Math 1. Go into Mapper 1 to set how much each key gets modulated.

[stippenstoh]

Thanks NAD!
I'll give it a try!

EDIT:
man, you're a genius. Exactly what I tried to do and in such a simplistic way. Thanks so much!

[NAD]

I have long been intrigued by keyboard-logic-aware modulation sources. That is, modulation driven by what the player is doing musically, not just by velocity, aftertouch, or controller movement.

That's a cool idea. It brings to mind advanced Kontakt scripts for multi-articulation sample libraries. It's obviously been done a ton for sample libraries but when I think of it no synths come to mind. It would definitely be a tricky endeavor though....