Request for implementation of optical spacing
See image:
Related issue: #18941: Cross staff beaming causes messed up layout of chords with seconds.
This did't use to happen some months ago. Compare with the images here:
#14448: Cross staff beams are always horizontal
| Attachment | Size |
|---|---|
| shots.png | 36.14 KB |
| Untitled.mscz | 2.12 KB |
Comments
Also, look at the increased separation between the last two notes of the 3rd bar.
I won't quibble the spacing, but now the stem and beam positioning is messed up:
#3 should be fixed in f262e9c107.
The spacing looks wrong but the note heads are at the right position.
Optically the note stems should be spaced more equally. This is currently not implemented and postponed for version > 2.0. The note head position can be tweaked manually of course.
I cannot reproduce the wrong beam length not covering the last note stem. Maybe its fixed...
Agreed, it looks fine now (both the stem and beam length) except for the spacing, and I understand that issue and why it won't be addressed for 2.0.
This exact same issue was brought up again at https://musescore.org/en/node/110681. This issue dates from the 2.0 development period, when there were other priorities, but with 3.0 now beginning development with a very strong focus on improving score layout, now seems like a good time to revive the topic of optical spacing.
It's a good topic, but as with some other "subjective" areas, is greatly complicated one you take other staves into account. Any adjustment made to optimize spacing on one staff is going to either throw all others staves *out* of optimal spacing, or else will sacrifice alignment of notes. Gould touches on optical spacing on p. 41 but makes no reference to how to solve this problem. Would be interesting to see what LilyPond does,
Some insight as to LilyPond: http://lilypondblog.org/2015/12/optical-spacing-the-swan-part-2/#more-4…
Great, thanks for that article! It demonstrates the issue nicely and shows how LilyPond currently deals with it, and shows that even LilyPond requires currently requires a manual override to work reasonably in multistaff situations, and that there are other complications as well that they don't expect to be able to handle any time soon.
I'm still inclined to suggest we at least attempt to do this for systems with only a single visible staff. Well, really, *measures* with only a single visible staff, since there might be cutaway staves only partially visible for a given system.
Might as well make the title more general, and call this what it is - a feature request. As the other referenced examples show, it goes beyond cross staff notation or beamed notes, but applies to lots of different situations.