Is this a bug? The first rehearsal mark gets lowered (to the point of colliding with chord symbols) when placed over a double bar line. Am I missing something in the engraving rules?
I don’t think it has anything to do with the double barline. Instead, it’s moving closer to the staff based on the available space to the left of the chord symbol.
You could increase the minimum distance from the staff in Engraving Options–Rehearsal Marks–Vertical Position.
It’s nothing to do with it being the first mark. It’s to do with it being the letter “A”, and the width of that letter. The letter “V” in Academico Bold is also slightly narrower than the other letters.
Switch the font style to Italic and you get this:
It seems like there must be something else in play here too though, as C is more narrow than A and on the same system. I is certainly more narrow. I wonder if there is another factor influencing the positioning too.
Maybe it has something to do with the kerning and letter spacing built into the font? The characters that Dorico is perceiving as more narrow have sloped sides, even more so in the italic. Characters with sloped sides often require more kerning pairs to make the spacing correct. I’m not sure how Dorico calculates the padding between the character and enclosure, but perhaps it is incorporating some sort of kerning value into the calculation? Just a guess.
I can set any of the Rehearsal Marks’ indexes to 1 (making it an A), or 22 (making it a V), and it’ll jump down to meet the baseline of the chord symbol, because it thinks it has space. I can also use the Note Spacing tool to adjust the spacing to the right of any of the barlines, and (depending on which direction I go) I can force any of the Rehearsal Marks to jump up or down.
I don’t understand how font kerning works, but I do know that looking at just a column of letters, as FredGUnn has done, doesn’t tell the whole story.
If I change the Engraving Options for Minimum Width and Left and Right padding as follows:
I get this:
I repeat: this is nothing to do with “the first repeat mark”.
Yes I think that’s been made clear! This thread has been very illuminating and I appreciate everyone chiming in. I’m wondering if it will be less work to leave the engraving options as they are and just make adjustments manually.
Completely agree it has nothing to do with first. In your example directly above with 1/2 space padding, the width of the C rehearsal mark looks to be more narrow than A. I’d have to overlap them to be sure, but this is perfectly logical as the character width of the letter C isn’t as wide as A in Academico bold. It does raise the question though - why does Dorico think the C will cause a collision in your first example, but not the A? Especially if Dorico is reading the boundary of the character C as more narrow than A as shown in your example.
A. When I said “repeat mark” obviously I meant “rehearsal mark”. Sorry!
B. Re: kerning and font width, I’m not qualified to answer the question. Hopefully somebody more knowledgeable (Daniel?) will chip in at some point.
I’m not sure there’s a lot more I can add here. The rehearsal mark has room to squeeze in next to the chord symbol when the double barline is shown because that effectively moves the rehearsal mark a tiny bit to the left in relation to the chord symbol, making room for it to appear there. Rehearsal marks have a minimum width that will be used unless the width of the letter(s)/number(s) in the enclosure plus the specified padding to the left and the right exceeds that minimum width. Everything is working as I would expect.
In the OP’s post I think that’s pretty clear. The double bar gives it a bit more room and Dorico thinks that’s enough to avoid a collision. I’m more interested in leo’s first example actually. In that example, why does Dorico think the A rehearsal mark takes up less room than the C? The character C has a more narrow character width than A. If rehearsal marks have a minimum width that is used unless there is an exception that exceeds it, the C rehearsal letter width has to be less than or equal to the A enclosure width, correct? It can’t possibly be greater than A, although it certainly could be the same if they both meet this minimum width criteria, unless there is some other kerning value coming into play here. So if the C enclosure width is less than or equal to A, why does Dorico think there is room for A, but not C?
(Yes, I realize I could have adjusted all the rehearsal letters in less time it took to type this, just trying to understand why this is the result by default.)
Try typing ‘xCx” and “xAx” in Academico Bold. You’ll see that the serifs on the Xs touch (or almost touch) the serifs on the A, whereas there’s much more space around the C.
That’s apparently a ligature in Academico, I guess, as they are definitely touching for me.
xAx is still wider than xCx, so I’m still not sure why Dorico thinks the C is wider than the A. My hunch (and this may be wildly off base) is that since characters like A and V have a lot more kerning pairs due to their angled slope, Dorico may be calculating the width based off a non-existent kerning pair value, as it is obviously only one character. Anyway, it’s not a huge deal, but it does seem a bit inconsistent.
No, it’s not a ligature. The characters have zero or even negative side bearings, or there are kerning pairs defined for these combinations, so they slightly overlap.
I don’t think this is about kerning at all. Dorico calculates the box padding around the glyph’s metrics (spacing) rather than around the actual contour. The letter C has usually larger side bearings than A or V, hence the difference. The vertical font metrics have an impact, too. You can see that the characters are not vertically centered in the box, as they would be if Dorico was padding the contours. (This can be remedied by setting different padding values in Engraving Options, though.)
Strange! It looks like I was jumping to conclusions.
In Academico Bold the letter A is considerably wider than C, spacing included. There must be something else at play here.
Dorico uses the tight bounding rectangle around the string to determine the width and the height, but it also does render strings like “xAx” as a single run of text, so advance widths and kerning pairs etc. will all come into play in rendering that text.
… and that implies that the “black ink” of some character glyphs will extend slightly outside of the bounding rectangle. That is a normal feature of font design, right back to the days of metal type.
You can’t really make sense of all this except either by looking at the metrics in the font itself, or by having enough experience to make an informed guess about what they are. Just “looking at the letters as rendered” isn’t enough information.
That’s what I did! Both the contours and the metrics suggest that A should be wider than C. A bit of a mystery… though admittedly of little importance.
Florian, I know you understand in detail how fonts “work”, but possibly some the other posters in this thread don’t - hence my comment.