freetype
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Freetype] Comments on quality


From: David Turner
Subject: Re: [Freetype] Comments on quality
Date: Tue, 08 Jan 2002 16:52:53 +0100

Hi Michael,

> 
> Actually, there is nothing wrong with the MS core fonts. They are typically
> of much higher quality than most TT fonts, and this is true for Vincent
> Connare's Comic Sans Serif in particular. Also, they are most certainly
> hinted for grayscale usage. Note however that this fonts contains a 'gasp'
> table that says that it was designed to be rendered b/w below 12pt/16px.
> Showing the font with grayscales below this size would be a serious mistake,
> as you can see in the sample picture. The 'gasp' table is there for a
> reasons....
> 
> So, FreeType and the font itself are working just fine. You could argue that
> ftview should be smarter, but I think that would be beyond the scope of such
> a tool. However, a "real" application should use the 'gasp' table (I am a
> bit biased though, as I invented the table back in '96 ;-).
>

Actually, the problem is that using the "gasp" table has been patented by
Microsoft (I don't recall the Patent number but I guarantee you that I have
read and analyzed it several times). There is no doubt that "ftview" could
use it (it is actually already loaded by FreeTyp, mind you), I just don't
want to enter another mined land :-)
 

But there is hope nonetheless. Since the advent of ClearType, Microsoft
seems to have started re-hinting their core fonts in order to avoid
rendering problems similar to the ones described by Artur. I don't know
if this will "fix" "Comic Sans" for anti-aliased rendering at small pixel
sizes, but I hope it will :o)


> Writing an auto-hinter is a very complex task. I think that Freetype 2.0.5
> (I have not looked at 2.0.6 yet) does more than a decent job. We should all
> be grateful that there are people that invest so much time and effort in
> this project, and we should not expect the impossible.
>
Actually, I'm pretty certain that it's possible to do much better. The
problem is that I lack the time to study this further because it will
require lots of experimentation, and a couple strokes of genius (and
believe me, these don't come too often ;-)

I'm also getting a bit bored about this whole hinting topic.. it's time I
switch to a few more interesting things. I mean, what a _niche_ segment of
Computer Science !! ;o)


Regards,

- David



reply via email to

[Prev in Thread] Current Thread [Next in Thread]