Re[2]: [Gnash-dev] Fast, high-quality, platform independent software renderer for Gnash

[Udo Giacomozzi]

Hello strk,

Thursday, September 7, 2006, 4:51:11 PM, you wrote:
s> I don't think there is any new backend structure coming.
s> The best would be getting adjustments and documentation of the
s> current structure from people actually working with it (you and
s> the cairo guys).

I agree that's probably the right way. When implementing the shapes
filling the result looked awful because a simple rectangle is
converted to 20 triangles or so and because of anti-aliasing you can
see the borders.

We'll try to patch Gnash and implement a new architecture so that the
original shapes are passed to the backend class which then has a
fallback mechanism (tesselator) available for OpenGL and similar.

Don't know yet how much modification of existing code this will
require but it seems not so trivial.

As soon as the new backend architecture has been implemented I expect
a working AGG backend relatively soon.

Udo