Closed sgolovan closed 6 years ago
Thanks for the detailed suggestions. I thought I’d written some code to improve this behaviour already, but I must have been imagining it!
What should more realistically happen is that if the three sizes are
sf@size = 5 tf@size = 7 f@size = 10
then the transitions happen at 6 and 8.5 (i.e., the average between each nominal size).
I definitely can’t promise a quick update on this but it’s now on my radar. Thanks again.
When
unicode-math
loads new math font (for size 10 main font for example) it uses the following size features:This means that if a font has mathscript variant which looks differently than the main font then starting from
\small
size and smaller text and math will look differently. As an example, STIX2 Math looks suboptimal for\small
or\footnotesize
text sizes.If I use the following size features:
then small text/math become better.
May be it'd be better to switch MathScript and MathScriptScript on starting with smaller sizes than it's now? Is it really a good idea that 9.99 size font uses MathScript style?
The two files attached to show the issue: fsize1.pdf fsize2.pdf
Note the inconsistent size for zeros in fsize1.pdf for small size and brlow. fsize2.pdf exhibits any notable differences only for scriptsize and below, which is fine, I think.
The code is the following (fsize1.tex):
and fsize2.tex: