Page 1 of 2

Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 09:39
by benwiggy
I've been using the Sebastian Figured Bass font, which is available here: https://github.com/fkretlow/sebastian

However, when I use it in Finale, the numbers disappear from the screen when I zoom in to more than 300%..! I've discovered that the reason for this is because the characters have zero width, which is odd.

I've tried using TypeTool to set the width margins and reference point of each character, but when I do, the symbols all move to the left a bit. I would have to adjust all the chord suffixes libraries in my existing documents.

As there's quite a lot of font knowledge here, I was wondering if anyone might know a way of fixing the font without having to alter the documents?

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 10:15
by OCTO
Maybe it is problem with Finale, have you tried using the font in other older versions?
Also, does it print, or just screen zoom problem?

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 10:20
by benwiggy
It's simply a display issue at > 300%. They re-appear with a Redraw, but won't appear as you scroll around. The document prints fine.
MM have confirmed to me that it's Finale's problem with displaying characters that have zero width.

I don't mind changing my existing documents, but if there's an easier way to fix the font without having to do that, that would be better.

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 11:45
by Knut
Apparently, zero width glyphs has been known to cause problems in a lot of software. Particularly, this is the case with browsers, but I'm sure other software like Finale is not immune either. Unfortunately, I don't think there is a remedy for this problem on the font side of things, except modifying the font, substituting zero width glyphs for negative kerning pairs, which of course is a lot of work.

There was some discussions on compatibility issues with zero width glyphs in the pre-release stages of SMuFL, which finally lead to discouraging their use in the standard. As a result, SMuFl uses zero width glyphs only in exceptional cases, instead relying on kerning and software development to achieve the desired spacing of relevant characters.

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 12:22
by benwiggy
Knut wrote:Unfortunately, I don't think there is a remedy for this problem on the font side of things, except modifying the font, substituting zero width glyphs for negative kerning pairs, which of course is a lot of work.
Well, I only use 1-9 and the augmented numbers 4+, 6\ etc, so it might be a possibility. So after adjusting the margins, I need to "counteract" this by setting a negative value for kerning pairs? (The numbers tend to be used individually.) I'll have more of a play.

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 12:42
by Knut
benwiggy wrote:
Knut wrote:Unfortunately, I don't think there is a remedy for this problem on the font side of things, except modifying the font, substituting zero width glyphs for negative kerning pairs, which of course is a lot of work.
Well, I only use 1-9 and the augmented numbers 4+, 6\ etc, so it might be a possibility. So after adjusting the margins, I need to "counteract" this by setting a negative value for kerning pairs?
Yes, exactly. The benefit to this is that you can set separate values for each pair of numbers, which, unless your numerals already have tabular spacing, can give you more satisfactory vertical centering. If the numbers have tabular spacing, however, you would of course use the same negative value for all pairs.

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 14:11
by benwiggy
What constitutes a pair, exactly? Do I have to do it for every combination of two numbers?

I think it would probably be quicker to adjust the chord suffix libraries.

Re: Problems with Sebastian-FigBass

Posted: 03 Dec 2016, 15:53
by Knut
benwiggy wrote:What constitutes a pair, exactly? Do I have to do it for every combination of two numbers?

I think it would probably be quicker to adjust the chord suffix libraries.
In principle, yes, but most font editors allow you to set up kerning classes, which applies the same amount of kerning to any glyph in a particular class, so the amount of work would depend on the number of variables needed.

I'm almost certain tweaking the chord suffixes is at least just as quick, unless you know your way around the font editor very well. However, tweaking the font may be the better long term solution.

Re: Problems with Sebastian-FigBass

Posted: 04 Dec 2016, 07:30
by benwiggy
Well, I've created a "clean" version of the font, fixing all the errors that Fontforge flagged. I didn't make any kerning alterations, and I'll adjust my existing documents. For reasons that need not bore you, it's not that much of a job to adjust the chord suffixes.

Thanks for the advice. There aren't many fonts out there with all the figured bass symbols, and I really like this one.

I suppose I could give the font a new name, and keep old and new versions on my system, but that's probably not necessary, as I have PDF copies of all my files, and don't go back to Finale until I need to change something.

Re: Problems with Sebastian-FigBass

Posted: 04 Dec 2016, 12:04
by Knut
benwiggy wrote:Thanks for the advice. There aren't many fonts out there with all the figured bass symbols, and I really like this one.
I took a look at the Sebastian figured bass font. It contains a number of ligated +_numeral and numeral_+ glyphs which is not found in the current version of SMuFL. Do you know if these can be found in the literature, or are they of the font designer's own creation?

Here's the figured bass symbols currently featured in SMuFL:
https://w3c.github.io/smufl/gitbook/tab ... -bass.html
https://w3c.github.io/smufl/gitbook/tab ... ement.html

As you can see, the only ligatures featured are 2+, 4+, 5+ and +7.