Question of Hardsync of Zebra2

[Urs]

Wow, that explains a lot. I thought I was just being a dumbass (that's not necessarily ruled out) trying to figure out why synch didn't behave the way I thought it should. I figured everything would be slaved to 1. Forgot all about the wavetable/grane flimflammery.

Thanks for the discussion!

Aha! Urs, seems that i'm not the only one~

I guess Howie will make a note for the next Zebra manual

[SJ_Digriz]

Wow, that explains a lot. I thought I was just being a dumbass (that's not necessarily ruled out) trying to figure out why synch didn't behave the way I thought it should. I figured everything would be slaved to 1. Forgot all about the wavetable/grane flimflammery.

Thanks for the discussion!

Aha! Urs, seems that i'm not the only one~

I guess Howie will make a note for the next Zebra manual

To be fare, I'm not sure I ever looked in the manual to figure out sync. So, the blame isn't completely on the manual. I just ASSUMED I knew how sync works. Bad on me.

Obviously a complete explenation of how it works would be great in the manual. But, us users need to be smart enough to go read the damn thing. (NOTE: I have read significant portions of the Zebra and Diva manual. But, sometimes you take prior knowledge for granted)

[linearhit]

I believe Zebra hasn't slave oscs. I think it just squeezes waveform and adds new copies of squeezed waveform until cycle is not full. Then it just renders result to actual waveform that is playing.

I have another question. What is the order: first sync, and then OscFX, or vice versa? And PWM is calculated before sync?

just have a test, you're quite right.
attached is a sine wave with sync=0,14 & 27.

[AnalogGuy1]

@linearhit: do you mind sharing what scope VSTe you used to post those waveforms?

[Gonga]

Each osc in Zebra has its own internal sync slave osc (it seems). Move the Sync knob and you'll hear your classic sync sound.

Little known fact - this parameter can run negative via modulation and instead the waveform will expand as the last part of the waveform is discarded out the right-hand side of the buffer/memory/whatsitcalled, giving you PWM on squareoid waves, and a kind of 'negative sync' if the wave has a lot of complexity to it.

Wow. Kewl, must-try!

[Gonga]

"Sync: Offset for the oscillator-internal hard-sync effect. This classic 'analogue' sync adds a lot of upper harmonics, and is probably why it was often used to imitate screaming guitar sounds back in the shoulder-padded and hairsprayed 1980s..."

And by certain stubborn modern-day luddites

And now for something completely different (de-rail):

I always wondered whether it could be possible to hard-sync a synced oscillator to a third oscillator, so that you had a dual-layer hard-sync. Then you could set up two controllers to control both synced parameters. It might be a bit like running a filter through another filter, where you can get great vocal formant type sounds, except, of course, much better because, after all, it's SYNC!

Could Bazille do this?

[trance_lucent]

I always wondered whether it could be possible to hard-sync a synced oscillator to a third oscillator, so that you had a dual-layer hard-sync. Then you could set up two controllers to control both synced parameters. It might be a bit like running a filter through another filter, where you can get great vocal formant type sounds, except, of course, much better because, after all, it's SYNC!

Could Bazille do this?

Zebra can (to some extent). You can use SyncMojo and usual hardsync. Or you even can to assemble quad sync: use sync waveform (there are some geomorph wavetables with sync waveforms), then use syncmojo twice in two fx slots and then usual hardsync.

[trance_lucent]

To illustrate the quadsync (also featuring quad detune osc mode for stereo):
https://dl.dropbox.com/u/57065234/quadsync.mp3

[Sendy]

Serial or recursive synchronization is very good for creating vocaloid sounds, in fact, there's a synthesis technique called VOSIM based entirely on interlocking synchs to create speech.

Speaking of Bazille, you can also do double-sync right off the bat with that. Use one of the resonant 1-3 waveforms modulated, and apply fractal resonance modulation. You'll end up with a synched sine wave (with a tapered smoothing window applied to the waveform to get rid of the rattles normally associated with sync on smooth waveforms), which then becomes the basis for a second sync process, which will also be tapered to avoid frizz. The "tri/saw/max" option selects the window used.

The effect is a lot like double formant filtering, only it's in the time domain, and one time-based process messes with the other one, scattering the phase of the original waveform unpredictably. I like to imagine what might result from continuing the process indefinitely, a truely fractal waveform

[Sendy]

Here's a really cool VOSIM implimentation demo done on the Nord G2 modular. Apparently all done with sync.



PROOF that sync > filters

[Gonga]

Sendy, you have given me another new quest!

Thanks to trance_lucent also!

[linearhit]

@linearhit: do you mind sharing what scope VSTe you used to post those waveforms?

it's bluecat

[hollo]

Sendy, you have given me another new quest!

Thanks to trance_lucent also!

+1 Thanks for you 2! It's incredible what is happening with those sync sounds in the video!

[trance_lucent]

Well, I don't familiar close with VOSIM technique, but a quick supposition: may be it is great to use "ripples" OscFX in Zebra for that?

[Urs]

Well, I don't familiar close with VOSIM technique, but a quick supposition: may be it is great to use "ripples" OscFX in Zebra for that?

Yeah, the video looks much like the windowed phase distortion resonance, and actually much like Bazille.

Ripples is cool, but starts at the wrong phase (I wasn't aware of the exact PD technique back then). I do think I've coded a "Ripples2" OscFX, but haven't added it yet. That's the one to go for. Then.