On Fri, Jun 15, 2007 at 05:47:25PM +0200, Udo Giacomozzi wrote: > Hello strk, > > Wednesday, June 13, 2007, 5:24:00 PM, you wrote: > s> I dropped support for textured glyphs from device fonts. > > Is this discussion *only* related to device fonts? Please note the PP > also uses the OS font rendering for device fonts. Their placement is > less precise compared to embedded glyphs. So perhaps we can go the > same way for Gnash?
Yes, only to device fonts. We weren't using textured glyphs already, for a long time. I dropped support for them when introducing progressive loading (play while loading) due to the hard-to-understnad and manage implementation of the software renderer found (and still being) in fontlib.cpp. That drop long ago introduced what is know as the infamous "text quality regression". Since AGG was doing a good job with vectors that bug became a not-so-important one. I guess precision of placement is a separate issue, due to the possibly different metrics found in systems and different fonts being used (font name to actual font filenames matching) > s> BTW, the current vectorial device font doesn't seem to work at all with > OGL, left/right fill > s> problem most likely as I saw a similar thing with drawing api. > > I want to point out that shapes of *embedded* glyphs need to be > rendered with different rules (compared to other shapes). I guess > device fonts even more different.. Uh ? I think I used the same rules used for embedded glyhps. The only difference being the way we create the paths (SWF-defined for embedded, gnash-defined for OS fonts) which is where I belive OGL fails (it's known to fail with some of our drawing API tests too). --strk; _______________________________________________ Gnash-dev mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnash-dev
