The one parameter that I struggle to get a grip on is the tone shift one. I’ve edited this parameter for voices at their lower part range, under which I’ve learned how to handle the aforementioned parameter, in that a higher value (more or less in tandem with a likewise elevated tension value) intensifies the tone; (sudden) lower values more often than not will contribute toward a falsetto.
The other day, though, I began work on voices at the higher ends of their range, and tone shift no longer works the way I would think. At times, really high values gives almost the inverse effect from what I’ve described above, i.e., it lends the voice a softer timbre.
I read somewhere an explanation from @dcuny about this parameter, explaining that it borrows tones from lower and higher ranges, depending on the value it’s set at. I would appreciate a bit of a further explanation of this.
I believe the “borrowing tones from higher and lower ranges” idea is accurate, though I don’t think we’ve ever had a precise description from Dreamtonics so it’s a bit uncertain exactly how it works.
To elaborate on that idea, note that Tone Shift is measured in cents. The rendered output will have a tone as if the note were x cents different from the actual note position, where x is the value of the Tone Shift parameter, without modifying the pitch of the note.
This is an example that pushes that to an extreme situation, note that I’ve clicked the “x2” option to expand the parameter range to -800 to +800 cents.
In this case the pitches of the notes are C4, G#4, E5 (800 cents apart for this example, since that’s the limit of the parameter range), however the tone is calculated as if all three notes were G#4.
The result is a more consistent tone across larger pitch changes. It could of course also be used to change the tone over time even when there aren’t significant pitch changes.
