Re: [Nautilus-list] Xft Anti-aliasing, Libart antialiasing?
- From: Moses Lei <mlei mtmis com>
- To: Keith Packard <keithp keithp com>
- Cc: Alex Larsson <alexl redhat com>, Nautilus list <nautilus-list lists eazel com>, Keith Packard <keithp suse com>, Raph Levien <raph acm org>
- Subject: Re: [Nautilus-list] Xft Anti-aliasing, Libart antialiasing?
- Date: Thu, 13 Dec 2001 22:15:20 -0500 (EST)
> I have a hypothesis that text labels work better when drawn as accurately
> as possible, while text passages work better when drawn as clearly as
> possible.
>
> We use labels to scan the screen for information, and our eyes are looking
> for a particular word shape -- the slightly defocused appearance is not
> objectionable while inaccuracies in letterform and wordform make it harder
> to locate.
>
> Long passages involve accurate focusing on the text; the "fuzzy" appearance
> of pure AA text makes our eyes struggle to sharpen the glyphs. Inaccuracies
> in letterform and wordform are less objectionable because there is a lot of
> additional context used in performing word recognition.
That is an interesting hypothesis that makes sense rationally. But from
personal experience, I work for long periods of time on writing papers and
essays (such is the life of a student) using OpenOffice, which renders
fonts unhinted, and I don't experience much, if any, more eye strain than
the rest of my interface, which is generally hinted aliased text. In
contrast, however, hinted antialiased text seems to strain my eyes very
quickly. This is on a 1280x1024 17" CRT.
Different people have different ways of perceiving things, and maybe we
don't all focus our eyes the same way. So I guess an option would be the
best way to go.
Glad to see that 1) I'm not crazy, and 2) something has been learned from
this discussion.
Moses
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]