Offset noteheads (unisons & seconds) between voices not offset by correct amounts
Windows 7, GIT commit: 60e5fe4
1) load attached score, which contains a number of seconds and unisons between voices
Result: the amount by which noteheads are offset appears inconsistent. On closer examination, it's not random.
Any time there is a whole note on the left (voice 1), the note to the right is not offset by enough - as if it is being offset by only the width of a quarter note notehead. Although, notice that in measure 2, when it's two whole notes, they look good together. This may mean that what's really happening is that whole notes are simply being drawn too far to the right.
Here's a screenshot of a relevant passage:
In the second measure shown (measure 4), there should be a slight gap between the C and B, so the C doesn't look like a half note. This is specifically mentioned by Elaine Gould in "Behind Bars" on page 53.
Also, in measures 8 - unison combinations of quarter and half - the stems touch, and Gould also specifically warns against this.
Finally, I would also observe that in the case where it's a quarter note on the left and whole note on the right and there is a second between them (measure 5), there is maybe a bit too *much* space. You don't really need to offset the whole note on the right by as much as you would if it were a unison rather than a second (measure 10). That's more subtle, and maybe not worth special casing unless it's easy.
BTW, it looks basically the same in Bravura; this isn't a font issue per se.
First reported by chenlung via IRC.
| Attachment | Size |
|---|---|
| offset-notes.mscz | 1.64 KB |
| offset-notes.png | 3.83 KB |
Comments
Does it look better like this?
Or like this?
the 2nd looks better to me
Generally, #2 is a bit tighter (less increase in offset), right? I like that better too. Ideally I might still want measures 4-7 even tighter if that can be done without messing up measures 9-12, which could possible be a little looser. Like I said, seconds can be tighter than unisons in absolute X-position terms and still *appear* to be spaced similarly. But if it's more convenient to always use the same X offset, this works as a compromise. One can always fine tune from there if one has particular ideas.
Neither of these addresses measure 8 - there should have space between the stems. That's probably more important than the quibble about measures 9-12.
BTW, I'm basing these comments primarily on Gould p. 53, but as we know, hers is but one opinion. So I'm also comparing to published literature. One thing I see is that quite a few editors like to put extra space between seconds even for measures 1 & 3. This seems unnecessary to me, but it does illustrate the subjective nature of these things. Overall, I'd say Gould's recommended offsets are on the tight side compared to what I see in published literature. So I'd say, offsets should be no tighter than recommended by Gould, but if it's a little looser in a few places, that's fine too.
I just wanted to upload the text and images from page 53 of 'Behind Bars' that elicited the issue (I hope there's no problem):
"Offset the lower part to the right. Vertically align the upper part with a part on another stave:"
"Keep noteheads separate; stems should not touch. Semibreves in particular should not appear to be joined to stems:"
Compare MuseScore in unaltered and altered forms - the latter attempts to match the above.
Also attached are mscz files reflecting these respectively: The unaltered was produced in 1.3 and the other saved in 2.0.
Using MuseScore 2.0 Nightly Build (bb9c7b3) - Mac 10.7.5.
Yes, as I mentioned, that's one of the sources I consulted to inform my comments above. My question is, what about lasconic's version #2 do you feel falls short? Other than quibbling about how *much* separation to have - and as I mentioned, Gould generally leaves less than other publishers - the only actual issue I saw is with the unisons in measure 8, where there is no separation at all. are you seeing something else?
Regarding 2:
For adjacent notes that are mixed quarters and halves, the stems should probably align.
I think I agree with your observations on others.
This was fixed along with #3325: Seconds between voices overlap
Automatically closed -- issue fixed for 2 weeks with no activity.