W3C home > Mailing lists > Public > www-svg@w3.org > December 2010

Re: Scaling sub-pixel fonts

From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
Date: Mon, 6 Dec 2010 10:55:36 +0100
To: www-svg@w3.org
Message-Id: <201012061055.37527.Dr.O.Hoffmann@gmx.de>
Jeff, why do you call it a sub-pixel font, the
font-size for presentation should be around
7.5px, whatever is available in this range for
the viewer ;o)

I think, I provided related accuracy tests
with and without animation for SVG tiny 1.2
(currently they are not in the SVG 1.1 SE test
suite I think, but most of my test apply for SVG 1.1
as well), typically it is more indicated to ensure, that 
the allowed number range is supported, respectively
that something like additive animations are 
accurate within the allowed number range.
Typically this is more done with (basic) shapes,
not necessarily with text. However with these 
and related tests in my own test suite it becomes
obvious, that several viewers have accuracy
problem or do not manage even the much more
restricted number range of SVG tiny. Opera displays
circles in some situations as irregular polygons - 
these are known bugs (at least there is a bug report
in the BTS related to this issue for a long time)
therefore no surprise, that  there happens even more 
nonsense with 'unversioned' SVG documents including 
notation with unit identifiers an a much larger number 
range. It is known as well, that typically browsers have
problems with unit identifiers (therefore nowadays the
CSS-WG tries to corrupt the meaning of unit identifiers
completely to suppress discussion about such common bugs) - 
however px on screens should be typically no problem ;o)

Olaf
Received on Monday, 6 December 2010 09:56:13 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:47 GMT