How was it possible to do FM synthesis in Zebra to a non sine wave, e.g. pulse, saw or triangle...?

When I put OSC1 above FMO1 I can modulate sine with anything but not even vice versa. I think Urs answered me once that there is even many ways to do this but I don't remember any more

There are some more FM-carrier waveforms in the FMO tab ('Generator'). The different choices on an FMO's panel are also interesting.

Checking a spectrograph might be interesting The different carrier waveforms are all sinusoidal in nature, but they sort of approach an LP-filtered version of saw/square/triangle. They are all some combination of less harmonics or dampened harmonics relative to a saw/square/triangle. They are not listed in order of harmonic complexity, that's sort of audible.

I don't think you can do it, it's one of the requested features, to be able to used non-sine-based waveforms as an FM carrier. As has already been stated, you can choose different waveforms for the FMO, but they're all transformations of a sine wave.

Personally, I hardly ever (never?) get anything useful out of carriers other than sine. I even suggested dropping those extra modes.

You clearly don't produce psytrance

A lot of sounds come from nordlead there where you get really good sharp sounds (that cut clearly trough the mix) by modulating something with more harmonics than a sine.

Howard probably patches a synth to self destruct every time he hears one alias

I don't know Nord Leads but I gather they're bright, dunno how aliasing or filters all play into that. Not that anyone asked*, but a tangent on 'brighter' from Zebra:

On the PatchLib there are plenty of wavesets that would be worth checking out. It's possible to capture some very bright FM-like qualities in the oscillators, many of the synth import sorts of sets have things along those lines. Also, some great squelching is in some of the more VA-oriented sets (check out Sendy's for sure).

For oscillators - try 'crisp' rather than 'soft' in the OSC tab, 'normalize' can help balance amplitudes following changes to waves across the set or from FX or whatever.

For filters the 'allaround' and 'vintage2' filter modes can be a little screamier, or the XMF is always amazing. The 'EQ peaking' mode on a filter might be really nice on top of things here, too - just like an EQ band but it's in the synth grid so it can track keyfollow for each voice.

*[e] Well, this comes up now and then for Zebra - I believe it's fair to say that by design it doesn't default to being super bright in very high frequencies. Possible, but not preferred.

I've noticed that the carrier WILL unashamedly alias in Zebra, when you start with a bright wave and essentially make it brighter, everything starts to bounce off of nyquist and reflect back down. If you start with a sine or similarly dull wave, you can build up a bright wave without aliasing.

The only real purpose of FM on bright waves that I can see, is to produce gritty waveform variations. Since Zebra has so many virtually alias-free ways of making and animating gritty waves, I don't think doing FM on bright waves is really worth it in Zebra at the moment.

In a VA environment where you've only got square, saw, etc, and the FM's going to react in a more analogue way (Diva *cough*), FM on bright waves really comes into it's own.

Nice reply, Sendy - precisely the way I see it.

I've got more than a handful of sounds that use the other FM waveforms in Zebra, and those sounds wouldn't have been heard without them, using any other means.
Meaning, if I switched them to the sine wave, the sound would have never been heard, and would be lost, and would also sound worse.
I feel they are very useful, and any nasty highs that come from them can be filtered or eq'ed out.
Getting them used with a ring mod and shaper, and sideband (for frequency adjustments) can produce some great things.

(just to clarify, not sure I was clear in my post )

Totally agree that it doesn't help to feed something brighter to an FMO - would probably the exact wrong thing to do if the FMO has to be involved. I was thinking skipping FM altogether.

Joke or serious? Now I'm confused!

Serious - I feel like using the FMOs in Zebra2 for a really bright VA tone - I guess I mean potentially dangerously bright - doesn't work nearly as well as using OSCs tweaked for brightness and maybe a little FM character embodied in the waveform, without any FMOs whatsoever. I believe the latter way avoids a lot of nastiness.

I was afraid I might have inadvertently wrote something that looks like a suggestion to feed something brighter into an FMO, though ... just makes the problem brighter, I'd say.

The part about patching synths that do something ugly to self-destruct was a joke though

@xh3rv... So who are you agreeing with? We were talking about non-sine carriers (I thought). BTW: Bright modulators are fine - just remember that the FM amount is crucial (modulate it from e.g. an envelope).

You and Sendy.

I'm not sure exactly how much I disagree with mcnoone (sorry - I think in the right hands 'bad' can always somehow be good, that's not what I would argue). As sort of a best practices, my personal preference for more exotic FM in Zebra would always be sine carriers, complex modulators, and probably the filtered FM mode but not exclusively. My personal preference for exotic FM would be Ace or Bazille, though.

I would disagree strongly, never would want to try it, with using Zebra's FMOs to do a bright Nord Lead sound. I really don't know, but strongly suspect there are details about how FM works in a Nord versus Zebra that are different. As a consequence, any sort of intentional use of enharmonics or aliasing just wouldn't translate between the two - it would be cleaner, closer, and better to use OSCs - they can be pretty bright with more flexibility. Without knowing anything about Nords that's just a hunch.