Zebra 3 - final beta Rev 21625 April 8th

[johnnyberrygoode]

I do not know, but maybe if the letters ADSR were placed in the center of the regulators, it would look a little better.

Oh yes. Thanks for spotting! Fixed!

I understand that you are fixing problems in zebra, so I will send you everything that I have discovered.
1. I think that if the "target circle" in the lfo module is moved to the left, as it is done in other modulation modules, it will give more symmetry.
2. Based on the logic of the grid formation in the filter matrix, it seemed to me that this line is not needed.

I understand that maybe some of the things I'm talking about may look insignificant, but since I'll be seeing the interface of this synthesizer for a very long time, I'd like to fix all the visual issues.
In the utility module, in signal mode, the line seems to be a little long, and in meter mode, a slight gluing between the lines is visible.

[Urs]

The x:1 ratio in the FMO changes the carrier frequency even if the FM index is set to 0. Is this behavior intentional?

Initially, when using an FM module, you hear the carrier (a pure sine wave). If you rotate the ratio knob to the left, you change the pitch of the carrier wave, and if you rotate it to the right, you change the pitch of the modulator.

That is correct.

Here's why we do this: Because tuning the modulator below the carrier in pretty much all FM Synths I have tried brings the perceived fundamental *below* the carrier frequency. For the classic Modulator-at-a-ratio-of-0.5 patch, one always has to tune the whole thing up one octave, for some ratios one can't even bring them up enough. Not so in Zebra 3. With FM applied, one can finally compare different ratios without losing perceived fundamental pitch.

(Yes, this is intentionally different from what people are used to, and I'm sure it'll inspire a whole new set of FM sounds. Just be aware that if an FMO is used without any FM, Carrier pitch may change for rations on the left half of the knob)

[loctune]

The x:1 ratio in the FMO changes the carrier frequency even if the FM index is set to 0. Is this behavior intentional?

Initially, when using an FM module, you hear the carrier (a pure sine wave). If you rotate the ratio knob to the left, you change the pitch of the carrier wave, and if you rotate it to the right, you change the pitch of the modulator.

That is correct.

Here's why we do this: Because tuning the modulator below the carrier in pretty much all FM Synths I have tried brings the perceived fundamental *below* the carrier frequency. For the classic Modulator-at-a-ratio-of-0.5 patch, one always has to tune the whole thing up one octave, for some ratios one can't even bring them up enough. Not so in Zebra 3. With FM applied, one can finally compare different ratios without losing perceived fundamental pitch.

(Yes, this is intentionally different from what people are used to, and I'm sure it'll inspire a whole new set of FM sounds. Just be aware that if an FMO is used without any FM, Carrier pitch may change for rations on the left half of the knob)

That perfectly answers my question! Thanks.

I do have a few more question regarding FMO: in this FMO we can have up to 3 sound sources: the input, the internal modulator, and the carrier. So the signal path becomes a bit complicated here...

If input set to dry, it seems to pass through the FMO no matter if the routing is set to carrier or internal modulator, so internal modulator only modulates the carrier and the modulated carrier signal mixes with the input signal?

And it seems like it's possible to have both the input and the internal modulator modulate the carrier together. In this case (input DX7->carrier and FM index>0), which modulation takes place first, or the input and internal modulator both modulate the carrier at the same time?

[trevdog]

some obvious things missing:
- no OTT / multiband comp
- MSEGs being used up on simple curve morph instead of just not limiting us with MSEGs
- range becomes numeric values instead of telling us what frequency we're modulating to on a filter.. -41 means nothing. it should say 132hz for example.
- no notch filter in filter section forcing me to use up an eq
- wavetables being truncated

[trevdog]

also how do i insert a comp into the matrix? or can it only be inserted as post FX?

[Urs]

-41 means nothing. it should say 132hz for example.

Well, the "how much" of the modulation depends on the base frequency of the filter. So when in the modulation it says "-1000 Hz" and then you turn the cutoff a little bit, the modulation is not "- 1000 Hz" anymore but something different. It would still be doable to display a meaningful value, but only if there is exactly one modulation (and never more). As soon as there are more than one, it becomes quite impossible to do so because the values would interact. But in our modulation system, there can be dozens of modulators on one target. Therefor, we keep the modulation depths as % of the parameter range, instead of "making things up for sake of meaningful units".

This is btw. why I have been reluctant to show "Hz" or other translated units in my products before. IMHO one can either scale knobs nicely in a musical way, or display meaningful units, but doing both results in inconsistencies. That is why we also have a mode that shows both the pure, nicely scaled parameter going linearly from 0 to 100 or so, and the translated unit (which loses its meaning as soon as modulation or interaction with other parameters comes in).

So we offer three settings in our preferences:

- "scaled only" for natural units where possible
- "scaled & raw" for both
- "raw only" for linear parameters

Most linear parameters are not without meaning either, in Filter Cutoff it is semitones btw..

(I just see that I dislike "scaled" and "raw" as names for these options... will bring it up with the team)

[ashone]

There has still been no response about any option to switch the color scheme back to the original version. The current one is too ugly and completely unacceptable.

[Urs]

also how do i insert a comp into the matrix? or can it only be inserted as post FX?

Compressors live in FX space.

[dzjin]

+1 on a light mode like Zebralette3 please!

[Urs]

There has still been no response about any option to switch the color scheme back to the original version.

We're probably going to provide a few more themes over the years, as we did with Zebra 2. We do not have any concrete plans, drafts or designs, so I can't make any statements as to what they're going to look like.

[ElSonador]

I tried out a little more yesterday. The preset was original built with the first Beta. I set Zebra to init again and tried to reconstruct the same arrangement. Maybe this is it.
I learned, that my old presets sound quite different anyway.

[Urs]

+1 on a light mode like Zebralette3 please!

Sure, a lighter colour scheme is probably among those we'll eventually add down the road.

[trevdog]

-41 means nothing. it should say 132hz for example.

Well, the "how much" of the modulation depends on the base frequency of the filter. So when in the modulation it says "-1000 Hz" and then you turn the cutoff a little bit, the modulation is not "- 1000 Hz" anymore but something different. It would still be doable to display a meaningful value, but only if there is exactly one modulation (and never more). As soon as there are more than one, it becomes quite impossible to do so because the values would interact. But in our modulation system, there can be dozens of modulators on one target. Therefor, we keep the modulation depths as % of the parameter range, instead of "making things up for sake of meaningful units".

This is btw. why I have been reluctant to show "Hz" or other translated units in my products before. IMHO one can either scale knobs nicely in a musical way, or display meaningful units, but doing both results in inconsistencies. That is why we also have a mode that shows both the pure, nicely scaled parameter going linearly from 0 to 100 or so, and the translated unit (which loses its meaning as soon as modulation or interaction with other parameters comes in).

So we offer three settings in our preferences:

- "scaled only" for natural units where possible
- "scaled & raw" for both
- "raw only" for linear parameters

Most linear parameters are not without meaning either, in Filter Cutoff it is semitones btw..

(I just see that I dislike "scaled" and "raw" as names for these options... will bring it up with the team)

ok but you could still show for just that one modulator amount even if we just keep in mind that the target is moving due to many other modulators. :] plz, because when setting many modulators, we would not be guessing, each one we could set more accurately.

however let me say this... Zebra 3 is still really cool though, even with the quirks!

[stippenstoh]

Props to Urs for how well he deals with entitled and/or rude commenters.

[misterandy]

Props to Urs for how well he deals with entitled and/or rude commenters.

Quite the stoic!