>> Beg to differ, too. A `web font' is essentially a font suitable
>> for a computer display with a (quite) low resolution. In
>> particular, such a font contains hints to intentionally distort the
>> glyph shapes, making them render fine on such devices.
>>
>> Lilypond fonts and its other graphical objects like slurs don't do
>> this. It looks extremely ugly if you get `crystal clear' rendering
>> of a display font like Arial (or probably even an embedded bitmap
>> for the given PPEM value), while the rest of the score appears
>> fuzzy.
>
> That means that we probably should be focusing about how we can get
> reasonable hinting for Feta.
Hmm. It would mean that we get good rendering results only at some
discrete values, namely where stafflines have (a) a thickness of an
integer multiple of a pixel and (b) where the distances between the
five staff lines are an integer multiple of a pixel also. Sub-pixel
rendering doesn't help a lot for staff lines, so this is quite a
strong physical limitation IMHO.
> Tablets and/or electronic ink are becoming increasingly important
> players in music.
Fortunately, the resolution of hand-held devices increases rapidly.
> Now Feta has the advantage of following traditional music type
> shapes, and those are designed for low resolution anyway as stamping
> with a type does not really allow for integrated hairline shapes:
> fine lines have to be etched separately.
Yes, at the very beginning of my work on the Feta glyphs I took some
measures to ensure good rendering results at low resolutions. It
would probably be easiest to add embedded bi-level and gray-level
bitmap strikes to the feta font. On the other hand I really doubt
that it is worth the trouble.
Werner
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user
