[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ft-devel] Rendering artifacts
From: |
malc |
Subject: |
Re: [ft-devel] Rendering artifacts |
Date: |
Wed, 8 Sep 2010 18:20:13 +0400 (MSD) |
User-agent: |
Alpine 2.00 (LNX 1167 2008-08-23) |
On Mon, 26 Jul 2010, Werner LEMBERG wrote:
>
> > Not sure whether this is a bug, known limitation or something else
> > entirely,
>
> It's something completely different...
>
> > but here goes:
> >
> > ftview 14 LiberationSerif-Regular.ttf [1]
> >
> > Toggling anti-aliasing ('a' key) makes it obvious that it
> > significantly degrades the quality of some glyphs (the leg of 'k'
> > the slashes in rationals become almost invisible)
>
> Looking into the font's `gasp' table you can see that the font expects
> anti-aliasing starting with a font size of 18ppem. In the range
> 9-17ppem, you have to use B/W rendering. ftview ignores the `gasp'
> table completely ? it's just a diagnosing tool, not a real user
> application, thus it allows switching between AA and B/W rendering for
> comparison purposes. However, `fontconfig' or similar libraries
> *must* respect this table, otherwise you get bad rendering results.
>
> In other words: Your bug report is invalid :-)
>
FWIW other fonts are also affected at certain point sizes
Liberation Mono (Regulr) diagonals making the 'v' are very thin,
Liberation Sans Italic the leg of 'y' is completely invisible,
I have stumbled upon [1] and it looks as if Times New Roman's 'z'
suffers from the same problem and the author gives a clue as to why.
The way i understand it - Windows's rasterizer (with or without ClearType)
processes the bytecode (or output of the BCI) differently so that the
problem never manifests itself.
--
mailto:address@hidden
- Re: [ft-devel] Rendering artifacts,
malc <=