On Sat, Feb 09, 2013 at 08:02:48PM +0100, Jürgen Rühle wrote: > Sandro Santilli wrote: > > Was this patch superceeded by your second patch ? > > If not, are there tests (existing or added by yourself) > > showing this patch implements the correct (or closer to > > the correct) behavior ? > > This first patch was just imeant as a debugging aid since I > was unsure whether the patch was in the right direction. > > Now after messing around with the testsuite I think it is, > though it probably is not the best solution. > > I've attached a minor patch to the testsuite. It seems that > drawing is now pretty correct. The remaining problem looks > like the color boundary is exactly on a pixel boundary in > the reference player but slightly displaced with gnash. > Visually no difference can be seen. > > Thanks for pointing me to the testsuite!
Thank you, your patch was pushed as d44294a4cf83e9ce1fd1d26e625910d3ba592dbe For single pixel error it could be either a matrix resolution issue or pixel hinting (but I think we have a known difference in matrix resolution) --strk; _______________________________________________ Gnash-dev mailing list Gnash-dev@gnu.org https://lists.gnu.org/mailman/listinfo/gnash-dev
