Re: [ft-devel] a flag making FT_Open_Face() to use fallback family name

[mpsuzuki]

Dear Werner,

On Sat, 28 Aug 2010 09:06:47 +0200 (CEST)
Werner LEMBERG <address@hidden> wrote:
>> Although the question "which is better result, hinted or unhinted?"
>> should be evaluated since FreeType 2.4.x (because of fundamental
>> change of default hinter) again, I'm afraid it cannot make everybody
>> agreed to use hinted result.  Some people may say: "I don't know the
>> complex story in TrueType rasterizer, but I prefer unhinted result.
>> I don't want to enable hinter by default".
>
>I have no opinion on this, but providing the ability to use the
>autohinter exclusively might have some benefits, so I don't object to
>improvements here.

In fact, in poppler list, JustFillBug posted an idea
the user configurable file listing the fonts to be hinted
(or unhinted) forcibly:
  http://lists.freedesktop.org/archives/poppler/2010-August/006291.html
However, I think, it is difficult for FreeType2 to have
an interface to manage such file, and it is the task of
higher layer libraries/applications.

However, in this case (nameless MingLiU embedded in PDF),
the target font is difficult to identify. So David's
proposal is very attractive.

>Just curious: Is the result of the `tricky' operations on a font
>really resolution independent, this is, are the subglyphs first
>shifted to default positions, then moved again to fit the selected
>ppem value?

I think the tricky operation is almost independent with
the resolution, I will make some pictures with different
resolution / point sizes.

BTW, while collecting the fonts using such tricky hints
(to make a list of table checksum), I found yet-another
Kaishu font showing heavily broken shape:
  https://www.codeblog.org/blog/mpsuzuki/images/20100827_0.png

I have to register this :-)

Regards,
mpsuzuki

P.S.
Tomorrow my office will have scheduled black out, and next
my post will be lated.