Resynthesizing a preset to wavetable shifts the phase of resynthesized wavetables in relation to the original wavetables. The amount of shift increases as the sample rate increases. It seems like both the sampling rate and the oversampling coefficient affect the shift. F.ex:
48k os 2x ~8.31
48k os off ~7.7437
So manually compensating for that takes a lot of trial and error to get the exact numbers down.
Windows, Vital 1.5.5, standalone and VST3 tested.
yea I’ve noticed funny things with wavetable phase shift, running everything on Mac, so it’s a universal thing… not testing rn, but is this related to the re-synthesising being a 4 second snapshot of audio from the output of the patch? so playing it back has to be a 4 second LFO ramp… (Probably not relative tbh) I noticed a lot of funky wavetable phase oddities with re-synthesising, I’m run things at 48k in Ableton, so that might make sense why it’s particularly noticeable if the shift is that much as you’ve seen at that sample rate.
Yup. That’s fixable in the waveform editor with a phase shift modifier. Takes an extra step though. Or then set the delta to the osc before resynthesizing. Would be pretty nice if it would work properly oob though.
Dunno what it’s about, could be just some minor error in the math too. If you or someone else has the time it would do a service to everyone to get the exact numbers down and posted here.
It seems like the oversampling setting affects this as well, so I removed the numbers from the original post.
that’s interesting will test asap, I resynthesised a complex patch earlier while making a cello sound for a sample pack i’m working on, but as it wasn’t a basic waveform, I wasn’t easily able to track the phase shift - going to see how things behave soon and report back