W dniu 28 września 2010 02:16 użytkownik Andrea Canciani <[email protected]> napisał: > On Tue, Sep 28, 2010 at 1:27 AM, Krzysztof Kosiński <[email protected]> > wrote: >> W dniu 27 września 2010 23:42 użytkownik Andrea Canciani >> <[email protected]> napisał: >>>> 1. Floating point coordinates in Pixman. >>> Do you mean in the API? Or for internal computations? >> >> Both are required to avoid the rendering corruptions seen with large >> gradients. > Not exactly true. They can be handy, but they are not *required*. > Please note that ensuring full precision with floating point is usually > harder than with fixed point. > Also, see wip/linear2. It makes linear gradient be as exact as possible > (more than it would be possible using doubles).
I am referring to a different kind of corruption than the one seen in the gradient bugs I've filed. I'll try to file a separate bug for that tommorow. In the existing gradient precision bug, the gradient is misaligned with its stops. In the case I'm referring to, which was discussed on the list but not filed as a bug, radial and linear gradients fail to render or become scrambled when any of their coordinates exceed 32767 pixels, even if only a portion of them is visible. Regards, Krzysztof _______________________________________________ Pixman mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/pixman
