[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: fontname codes for variants (was: fontname code for scaled variant)
Concernant « fontname codes for variants (was: fontname code for scaled variant) », Melissa O'Neill écrit : «
» On this issue too, I'm left wonderinging if there is any standardization
» for names. If not, I'd like to suggest we reserve 0 and 1 for this
» purpose and use some other letter for scaled variants.
these one are usefull. BTW, with expert & MM fonts what do you prefer:
optically scaled superiors or the expert ones? :-)
» (i.e. xheight & capheight) matching. In a line of 12 point Minion or
» Kepler, 12 point Myriad looks oversized and out of place, whereas 11
» point Myriad fits well. Thus, it is useful to have an 11/12 scaled Myriad
» available so that when TeX asks for 12 point Myriad, it's really getting
» 11 point.
»
ok now i understand what you want & get yet more puzzled: are you
using latex & the nfss?
If you simply want to systematically scale linearly one font family in
a document, here is what i do: create a new FD file.
For instance i use here every day Univers with
Utopia [my Utopia _is_ scaled down to the x-height of CM, to be usable
with latex's default layout, btw...] : for it i have a normal
t1pun.fd, plus a customized one, called t1punut (univers for utopia),
that reads:
\DeclareFontShape{OT1}{punut}{m}{n}{
<-> s * [.85] punr8t
}{}
[etc.]
This makes your `variant' useless (which seems normal to me because,
as any software is able to scale linearly the metrics, there is no
qualitative difference between pmr8t at 10bp or at 11/12bp!
BTW, having two families may be handy, because you're usually happy
to have a sans serif font at its normal size (with its larger
x-height) in headings where they look visually appealing (especially
to anglo-saxons with fraktur reminiscence), and an other version,
scaled down to the x-height of the serif face used, for font
variations in the running text.
******************************
Having systematically variant names for encoding design decision is
probably impossible, not even wishable?
I'm thinking of my UtopiaText/LucidaNewMath+CMmath/Univers/CMtt layout.
When you're doing that you necessarily scale appropriately some fonts,
but not the other etc. I for instance align in the math the caps to
Utopia caps, & the lowercase to Utopia lowercases, so i refer twice ti
Lucida, with 2 scales, and so on... that can't be standardized!
Thierry