sqaure, sync, a stupid theory and Zebra

[Maplesyrup]

a stupid idea came to me while driving today.
a square has a negative and positive phase right?
so in theory, if i take a sqaure wave and sync it to an oscillator an octave higher i should get silence/DC offset...right?
so i tested this completely useless theory of mine with zebra with no success.
i tried it with Poizone2, and it did work.

two questions:
1. what gives? does zebra sync uses some magic voodo that bypasses the laws of math? not that i have any complaints, sync in zebra sounds wonderful
2. incase i do manage to create DC offsets from a zebra OSC, any intresting audio applications i can use it for within a zebra patch?

[Urs]

Hi,

in Zebra you can't sync an oscillator to a higher pitched one - The oscillators have "self-sync" where the slave pitch can only go above the master pitch.

I am thinking of adding a dc offset functionality though... good for physical modelling flutes...

[Howard]

Actually, you can offset the Sync value negatively using a unipolar source e.g. mod wheel. BTW: negative cyclic modulation via MSEG or User-mode LFO gives you a fantastic-sounding PWM!

EDIT: I knocked up a little tutorial about this today...

[Sendy]

Zebra can do negative sync? I had no... freaking... idea! WE NEED A BIGGER SYNC KNOB!

[Howard]

Zebra can do negative sync? I had no... freaking... idea! WE NEED A BIGGER SYNC KNOB!

@Sendy: Any idea how "negative sync" in Zebra actually works i.e. why we get PW? (Urs isn't available at the mo')

[bmrzycki]

Any idea how "negative sync" in Zebra actually works i.e. why we get PW? (Urs isn't available at the mo')

I'm playing with it right now -- not sure exactly what's going on. Two things though:

1. When you turn on OSC sync the waveform flips horizontally. The normal falling saw becomes a rising saw. The up down square becomes a down up square.

2. When performing negative sync on a saw or a triangle wave it too seems to "thin" out the waveform until it's either a spike (saw) or a very very weak saw (triangle).

I'd speculate it has to do with running the sync algorithm backwards and we're possibly removing synced waves instead of adding them?

It's quite interesting nevertheless.

[Sendy]

Negative sync is effectively like 'rejecting' more and more of the end of the waveform, and stretching the remainder to fit the cycle frequency. It's been used to create PWM on synths which don't have PWM but do have sync, and it also lets you do weird stuff like morph a triangle into a sawtooth. If you imagine a triangle chopped in half, and stretched to be at the same frequency and resulting in a saw, you're imagining the process of negative sync.

Incidentally, it can sound really cool on complex waveforms! Also, on a lot of synths, if you push it to extremes, you can hear the oscillator engine break down into sines or jitter noise

EDIT: another interesting fact about sync in Zebra (and other good synths), if you change the phase of the waveform, you can effectively change the sound of the sync effect, as you're altering which part of the wave the process begins at.

[3ee]

Very interesting, thanks for the info!

[Howard]

Very interesting, thanks for the info!

+1 Sendy & Brian

[Maplesyrup]

nice info, so we have 4 ways of achieving PWM now? wonder which one is least CPU intensive, i'll check that later

@ Urs, how would you use a DC offsets to physically model flutes? is there a synth that is known to do it with it? can i possibly make an attempt in Bazzile or in my ACE demo or do they miss certain modules?

[Urs]

@ Urs, how would you use a DC offsets to physically model flutes? is there a synth that is known to do it with it?

If I recall correctly, there's that Reaktor ensemble called "Steampipe" which quite famously sported noise and dc as sources for comb filters. DC into a comb filter with negative feedback creates a squarewave type sound, hence fluteish...

[Sendy]

Incidentally, if you try and set up this PWM modulation using the Mod Matrix, it "isn't 'avin it" and refuses to enter the negative range of sync. This is really weird considering if you assign the modulation in the oscillator section everything's PWMtastic... Bug?