Is this possible?
Attachment 6922 not found.
Please also notice the meta connector for the lfo position…
I guess a manual connection to the internal “lfo” would not be possible since the waveform always needs to be 0 in sum for one cycle?
But a selection of waveforms like triangle, square, saw maybe?
A selection of several basic waveforms meanwhile actually should be standard for all modulation devices. Same for the LFO-phase setting. I love the idea of a meta-connector to trigger the modulator from outside!!
+1
But I think its not possible, since the waveform always has to be perfectly bi-polar? the chorus on one side pitches up the sound only because it pitches down on the other side, so no endless fill of buffer.
EDIT: Well, the connector could dc offset any custom lfo waveform by it’s offset in sum to 0… 
No?
A selection of several basic waveforms meanwhile actually should be standard for all modulation devices. Same for the LFO-phase setting.
+1
But I think its not possible, since the waveform always has to be perfectly bi-polar? the chorus on one side pitches up the sound only because it pitches down on the other side, so no endless fill of buffer.
EDIT: Well, the connector could dc offset any custom lfo waveform by it’s offset in sum to 0…
+1
Well, what could happen in worst case, when the device runs out of buffer? Technically not much, I’d guess. Except: no or chopped modulated output. I’d consider that an intentional effect. Sound rape incoming, beyond chorus/flanger modulations known so far. To me sounds very interesting.
Have you guys already tried setting depth = 0 to bypass the standard lfo and then automating the delay parameter with some custom lfo or formula device? I haven’t, but the idea seems tempting to me right now…I’ll try tonight, maybe the parameter is smoothed enough to already let it be useful as a custom curve chorus.
Other than that I say: haha, you sadly don’t seem to know how the usual digi chorus works. The lfo will just shift the delay line read position relative to the current “now” position. Not an absolute detune like a pitch shifter would. So the pitch shift/detuning comes by the relative movement/speed of the read pointer compared to the actual sample rate. Differential math to the rescue! As long as the lfo won’t try to blow the delay line by delivering too high absolute delay values (or even future aka negative values), it can’t cause any serious wreckage to the plugin. But this could always be contained in some manner. The lfo curve could in theory be anything - a dc offset would just mean the same as the “delay” slider does, and the curvature of the lfo would just influence the “curvature” of the envelope of the pitch changes. Discontinuities (saw or square lfo…) would mean glitches aka “clicks”. Extreme modulations could cause issues like clicks, aliasing and such, and would need advanced smoothing, summing and antialiasing techniques (like for fast “audio rate” kind of modulations). I also guess the chorus is not using sine wave, but some optimised lfo curve for pleasant detuning.
OopslFly, dunno if I understand you correctly or you do me…? 
^_^Saw and square waves of course interpolated or done thru FFT (so kind of sinoid saw / sqr) or whatever. DC offset for getting always 0 as sum result of a custom wave, so no buffer overflow. So there would be kind of event required “custom lfo wave changed” in the lfo device, so the custom wave could be aligned to 0 in pre-calculation and not realtime (have to be summed once after each change).
Also variants in between sine and triangle could be interesting in sound maybe? I tried melda’s mmmmmmmmmmmmmmmultimultichorus, there you can setup a custom wave. But it doesn’t not seem to work correctly / as expected. Also IMHO that chorus doesn’t sound like a chorus, more like a flanger?
Have you guys already tried setting depth = 0 to bypass the standard lfo and then automating the delay parameter with some custom lfo or formula device? I haven’t, but the idea seems tempting to me right now…I’ll try tonight, maybe the parameter is smoothed enough to already let it be useful as a custom curve chorus.
Only gives nasty zipper noises, a pity. Also would that be more simple time offset / delay?