- From: Thomas E Deweese <thomas.deweese@kodak.com>
- Date: Thu, 21 Nov 2002 07:57:00 -0500
- To: Tobias Reif <tobiasreif@pinkjuice.com>
- Cc: www-svg@w3.org
>>>>> "TR" == Tobias Reif <tobiasreif@pinkjuice.com> writes: TR> Thomas E Deweese wrote: >> Given your interest in high quality rendering [...] [...] the >> anti-aliasing was not very high quality. TR> [...] >> Well, SVG can be used in a large number of places and this is >> certainly one of them. You should take a look at Batik it is >> currently the most conformant SVG renderer available (unfortunately >> for you it does not do font hinting at all). But it is certainly >> capable of taking the place of a tool like GhostScript (who's font >> rendering is generally much worse than Batik's - IMHO). TR> Two issues I found with Squiggles rendering quality: TR> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14673 This is because we do no font hinting, so when we turn off anti-aliasing in response to optimize-speed or geometricPrecision, the text looks bad. For text to be readable at 'small' font sizes (<18pt) you either need bitmap fonts, font-hinting, or anti-aliasing (and anti-aliasing is really only good down to ~12pt). TR> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=12013 The rendering of the feather is different and I agree that the Adobe rendering is more pleasing here, however I suspect that the real problem is with the SVG, the Apache feather uses line-to's rather than curve-to's for most of the feather segments - so I don't think it is batik that is introducing the faciting, the Java2D anti-aliasing functions may show the faceting more but I think the source had it in it to start with (there is also little we can do about this sort of problem as we simply pass the vector data on to the Java rendering engine that does most of the work). You also seem to have caught it at a really bad moment :) If I zoomed in or out a little bit most of the uglyness goes away. Once again I didn't mean to claim that Batik was perfect - just good enough to do real work.
Received on Thursday, 21 November 2002 07:57:06 UTC