Hardsync with non-zero start

[torsig1967]

I read about hardsync in at the Modular zone at www.clavia.se and the fact that the syncpoint occurs on zero crossings.

I wonder if it's possible to simulate the more "analogue" style of syncing at the transient, perhaps by shifting the phase. I'm not sure of exactly how this should be done correctly though.

Anyone tried this?

[Falk]

One way is to invert the signal you want to sync to (for the saw wave) to make some other oscillator sync at the transient.

Another is to use a LevAdd module to shift the zero crossing point to somewhere else in phase so to speak.

---

SyncAtTransient.PNG
Description:	Syncing at transient of saw wave
Filesize:	23.7 KB
Viewed:	5825 Time(s)

---

[torsig1967]

Thanks for the suggestion though it doesn't seem to work. I can hear no sonic difference with or without the inverter.

I'm not sure which module is LevelAdd but I tried to add offset in different ways and it seems to me there's no sonic difference at all doing this.

[ian-s]

[davep]

I believe Torsig is not trying to change the relation of phase sync between the master & slave osc, he is trying to change the point in the waveform at which the sync slave osc resets, is that correct? The NM resets to zero crossing, which is in the middle of the ramp on a sawtooth wave, but real analog sync resets to the LOWEST point of the wave, at the start of the ramp's rise, and I think THAT is what you are trying to achieve, correct?

These two sync responses do indeed sound very different, and to my ears, the real analog behavior is much smoother and more musical, especially with sawtooth waves. Unforunately, I've never had any luck figuring out how to achieve this on the NM and as far as I know nobody else has either. There were a few threads on this topic over the years on the mailing list but we never found a solution.

But on the topic of phase modulation with sync, this can be done a few ways. One of the easiest is to use a sawtooth as the sync master and conbine this with an LFO or other control signal in a mixer before patching it to the SYNC input of the slave osc. The LFO will slowly 'bias' the sawtooth possitive & negative, which moves the point at which it crosses zero up & down the length of the ramp, which changes the phase relation of the sync'd osc.

You won't hear this effect unless you then combine the audio waves fromt eh master & sync'd oscs. You can use one sync master with multiple LFOs + mixers to modulate the phase of multiple sync'd oscs to get a huge buzzy detuned multi-osc sound that also has hard sync sweeps.

[torsig1967]

Thanks - a lot of interesting ideas here.

And yes - davep - it was the classic kind of hardsync I was interested in. The Modular sync hasn't got the bite I'm after with the zero-crossing sync scheme.

Well I guess there's no solution then.

[ian-s]

What about using a sawtooth frequency doubler (level compare, mixer and level shift) on the slave? The reset point would then be on the 'bottom' of the sawtooth.
I suspect that there are other reasons why the sync sound is not to your liking.
Not all analog synths had good sync sounds by the way, big difference between ARP/CAT/Roland/SCI.

[Tusker]

I haven't patched on the Nord in a long time, but I remember posting this in one of the communities some time ago. Dunno if it works. Hope it does ...

OK, if found an even better solution to the sync stepping. Simply send the synced oscillator and the syncing oscillator through a ring modulator. It cures the stepping, when you sweep the sync pitch.

Why? Well, it's a little complex, but you remember my saying that Nord code triggers waveforms when they cross the x axis rather than when they begin (like the top of a descending saw wave). When you use sync, the pitch attempts to lock to where it crosses the X-axis rather than the point where the syncing oscillator retriggers the wave. The syncing oscillator disruption of the pitch causes these zero-crossings to be jittery in the synced oscillator. When you run the two waves through the ring modulator, you are multiplying them together. Zero multiplied by anything is zero, so due to the zero value in the syncing wave, you get a jitter-free synced oscillator, if you run them through a ring modulator. Whoopee!

I can imagine you saying, doesn't the ring modulator alter the tone? It can, depending on the waveform you select for a syncing oscillator. Generally I believe a sine wave will give you an unaltered but jitter free signal out of the ring modulator. It's like multiplying by 1. Soon as I get some web space, I'll put up some sample files. This is too cool. It's also usable on any fixed architecture synth, where you have a ring modulator to spare, I assume. Hope this is still interesting. I kind of dived off the deep end there.

I tried the trick out with standard oscillators, the FM oscillator, and the sine bank, and it works. This is the first time, I got a trick straight out of the synth and not from the community, so I'm kind of pumped about further exploration. This synth is truly instructive.