View Single Post
Old 03-20-2016, 05:09 PM   #80
memyselfandus
Human being with feelings
 
memyselfandus's Avatar
 
Join Date: Oct 2008
Posts: 1,598
Default

Quote:
Originally Posted by ijijn View Post
Thanks youyourselfandyouguys! I'll try to digest your recent suggestions shortly. For a start, I imagine that for microtonality, which is a fascinating area and one I should explore more fully, additional accidentals paired with text meta-events, as with everything else so far, would be a solid foundation to build on.

Hmmm, as for feature losses, I can't think of any, but I'm certainly keen to hear if there's something I'm missing. The intention is for it to be an elegant superset of features around what exists already.

From a display angle, it's largely an exercise in hiding/combining staves to keep things sane. If we start with the assumption that we have access to 16 channels with 16 voices per channel on a single track, say, then this gives us 256 voices up our sleeves. We probably don't want all 256 (empty) voices to show up all the time, each on its own staff: that would be very silly.

Perhaps by default the notation editor could show "voice group 1" (usually voices 1 & 2) of channel 1, and then when we add notes and change their channel/voice numbers, more staves could spring up to accommodate. This way it's always very tidy while keeping everything accessible.

This architecture would require the grouping of voices into staves (defaulting to pairs, I'd imagine, but an override later would be useful) and staves into instruments: one staff for predominantly melodic instruments, but two or more for keyboard instruments, harp, etc. (using a brace, and what lovely braces they are after that recent update) or with a bracket/other connecting strategy for a related group of instruments, or ossia, or similar. Then clef events and sneaky non-global key signatures (and maybe even time signatures, that would be amazing to see at some stage) could be a per-staff thing, which reinforces the way notation is typically used.

Some of you may be wondering about the interaction of voices with things like CCs. Being a light traveller, I always generate my own CCs on the fly after routing voices to buses, so it doesn't concern me directly, but I suppose having per-voice CCs (via those text meta-events again, why mess with perfection?) would be another helpful step to shape each voice as you like, as well as providing a suitable link between visible notation and audible result when it comes to interpreting the notation in meaningful ways.

Finally, if you don't want to use voices at all, each channel still has the per-staff grunt it needs. Ditto for one line per track. It just scales up or down as needed.
I love that you are here ijijn and I love reading your awesome posts great ideas!
memyselfandus is offline   Reply With Quote