Issue 4323 - On screen display of fonts regressed from version 641 to 641C/D
Summary: On screen display of fonts regressed from version 641 to 641C/D
Status: CLOSED DUPLICATE of issue 2792
Alias: None
Product: General
Classification: Code
Component: ui (show other issues)
Version: 641
Hardware: PC Linux, all
: P3 Trivial (vote)
Target Milestone: ---
Assignee: stefan.baltzer
QA Contact: issues@framework
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-04-29 23:45 UTC by Unknown
Modified: 2003-03-11 17:59 UTC (History)
2 users (show)

See Also:
Issue Type: DEFECT
Latest Confirmation in: ---
Developer Difficulty: ---


Attachments
screen dump using OO641 (17.83 KB, image/jpeg)
2002-04-29 23:47 UTC, Unknown
no flags Details
screen dump using OO641D (16.45 KB, image/jpeg)
2002-04-29 23:47 UTC, Unknown
no flags Details

Note You need to log in before you can comment on or make changes to this issue.
Description Unknown 2002-04-29 23:45:46 UTC
(This question/issue has been posted to the discuss mailing list and also added
to issue 2999, but I was told that this should be entered as a new issue.  I'm
not convinced, but here it is anyways.)

I've recently upgraded to OO641D after having used 00641 for the past several
months.  Note that OO641C exhibits the same 'problem.'  All use of OO has been
on a Redhat Linux (v7.2) box, i386 architecture.  Output display is an SGI
1600SW panel display driven by a Number9 video card.

Other info includes:

XFree86 version 4.1.0
Fonts being served through xfs using the following catalog definition:
   catalog = /usr/X11R6/lib/X11/fonts/misc:unscaled,
	/usr/X11R6/lib/X11/fonts/75dpi:unscaled,
	/usr/X11R6/lib/X11/fonts/100dpi:unscaled,
	/usr/X11R6/lib/X11/fonts/misc,
	/usr/X11R6/lib/X11/fonts/Type1,
	/usr/X11R6/lib/X11/fonts/Speedo,
	/usr/X11R6/lib/X11/fonts/cyrillic,
	/usr/X11R6/lib/X11/fonts/CID,
	/usr/X11R6/lib/X11/fonts/local,
	/usr/share/fonts/default/TrueType,
	/usr/share/fonts/default/Type1,
	/usr/share/AbiSuite/fonts,
	/usr/X11R6/lib/X11/fonts/latin2/Type1,
	/usr/share/fonts/mathml/cmpsfont,
	/usr/share/fonts/mathml/MathT1


This may have been asked before.  My question is what changed in the manner in
which fonts are rendered, and/or anti-aliased and/or hinted on the screen?  The
display quality using OO641 is superior to what I'm observing w/ OO641D.  The
difference is especially noticable when editing documents using the standard
Microsoft fonts (for Office compatibility) available through Microsoft's
download pages.  Documents are being displayed 100% zoom setting.  Increasing
zoom level to 125% at least allows me to discern the difference between plain
and bold text.  Under normal zoom level (100%), this is practically impossible.

Attached are two examples to illustrate what I'm seeing.  Example test is Times
Roman New at 12pt.  Note that in the sample for OO641D it's practically
impossible to tell the difference from plain and bold font faces (at least to my
old eyes)

It would also be nice to return to 641's private font list rather than simply
providing access to all fonts served by your local font server.  This would
avoid displaying unusable CMR fonts (required for MathML under mozilla) in the
font drop down menus.
Comment 1 Unknown 2002-04-29 23:47:20 UTC
Created attachment 1509 [details]
screen dump using OO641
Comment 2 Unknown 2002-04-29 23:47:51 UTC
Created attachment 1510 [details]
screen dump using OO641D
Comment 3 thorsten.martens 2002-04-30 14:38:54 UTC
Hi, seems to be more a font-problem than a framework-problem to me.
Please have a look. Thanks !
Comment 4 behrends 2002-05-05 20:20:07 UTC
This appears to be an incidental problem of the upgrade to
freetype-2.0.5. Specifically, vcl/source/glyphs/gcach_ftyp.cxx has
various uses of #if (FTVERSION <= 204) and #if (FTVERSION < 205) that
normally disables autohinting in conjunction with antialiasing. But
since the code hasn't been updated with the new version of freetype,
v. 2.0.5 has autohinting enabled for antialiased fonts, which leads to
the poor display of antialiased fonts.

(This is a conjecture at this point, but can be corroborated easily by
rotating text by 0.1 degrees in sdraw, which also turns autohinting
off and leads to an improved display.)
Comment 5 t8m 2002-05-20 08:52:04 UTC
See also issue 2792
Comment 6 Joost Andrae 2002-11-28 16:45:31 UTC
Joost: duplicate

*** This issue has been marked as a duplicate of 2792 ***
Comment 7 michael.bemmer 2003-03-11 17:30:51 UTC
As mentioned on the qa dev list on March 5th I will close all resolved duplicate
issues. Please see this posting for details. First step in IssueZilla is
unfortunately to set them to verified.
Comment 8 michael.bemmer 2003-03-11 17:59:24 UTC
As mentioned on the qa dev list on March 5th I will close all resolved duplicate
issues. Please see this posting for details.