I ran into a problem today when trying to set a Paragraph style border thickess equal to my Rehearsal mark enclosure thickness. At first I wanted to put this in a reply to @dspreadbury 's comment in this topic which also talked about points vs. spaces, but I thought this might be a bit too far off topic for that one.
I created a Paragraph style for section headings with enclosures and I want the border thickness to be equal to the enclosure line thickness for Rehearsal marks. This proved to be a lot more complicated that I could imagine. Rehearsal mark enclosure thickness (like just about any other line thickess setting I can find in Dorico) is measured in spaces, even when setting your preferred unit of measurement to points in the preferences. For Paragraph styles however, border thickness has to be input in points and the box is limited to 1 decimal. This means that even after doing the manual calculations to convert your Rehearsal mark line thickness (default value 1/8 spaces) to points based on your rastral size, you still can’t input the exact value because more likely than not you’ll end up with more than one decimal.
So unless I’m missing something, it’s currently impossible to set a paragraph style border thickness exactly equal to Rehearsal mark enclosure thickness. Is there a reason why Paragraph style line thickness has to be input in points, as opposed to pretty much all other line thickness settings? In any case I think having to ability to input this in spaces would not only be more consistent but would also eliminate a whole bunch of manual calculating and workarounds (think of all the extra steps you have to take due to the fact that score and part layouts likely have different rastral sizes… not to mention the scenario when you have a few parts with different rastral sizes than others.).
At the same time, there are certainly some inconsistencies here. Paragraph styles let me specify a different font size for score and parts, but the border, padding, indent, etc. remain fixed, which is less than desirable.
Well, if you want, you can set a border on the rehearsal mark paragraph style. But if you’re setting it in Engraving Options, then you’re thinking of it as an object related to the score and the system.
Please note that I’m not saying that all of this is completely consistent (and in fact, I said just above that it’s not!), or that I agree with all of it. I’m just laying out why I think there’s a certain logic to what was done.
Actually your first comment about the relation to font size gave me an idea and I did some experimenting: if I’m not mistaken, when font size is set to Staff-relative, the Border attributes (thickness, padding) also appear to be staff relative:
This is with a rastral size 8 to emphasize the differences. The left text box has a font size 12 Staff-relative and some random values for padding. The big box has the exact same parameters, but font size is set to Absolute.
If I’m correct, this means you don’t need multiple paragraph styles for different rastral sizes. That at least eliminates a lot of extra work but I’ll have to shut down that OCD voice deep inside me that keeps pointing out that the enclosure thicknesses of text vs. rehearsal marks are never exactly the same.
Interesting – I tried that when I responded earlier, because I thought it would be a neat trick, but it didn’t work that way for me. And it still doesn’t! How are you getting those results?
Edit: Oh, wait – you’re doing something different than I thought. Yes, if you have a staff-relative font size, then the point sizes for border etc. are also staff-relative. I thought that was a given. But if you have a paragraph style with different (staff-relative) font sizes for score and parts, the border size remains the same between the two.
Ah yes you’re right. It does make sense that Border attributes are Staff-relative when font size is set to Staff-relative, but it’s not explicitly mentioned in the dialogue.
This is something I hadn’t even noticed myself! But that’s indeed inconvenient. Incidentally, that would also be solved if border attributes could be entered in spaces rather than points.