[cairo] anti-aliasing quality in glitz/OpenGL vs. Cairo native rasterization?

Carl Worth cworth at east.isi.edu
Fri Jun 11 09:48:25 PDT 2004


On Fri, 11 Jun 2004 00:43:30 +0200, Leon Woestenberg wrote:
> I may start a thesis project exploiting offloading parts of a rasterizer
> into logic (as GPL VHDL code) and Cairo may be an interesting candidate
> there.

I might be quite interested in this. And I'd be glad to help out
regardless.

Do you have a target platform in mind already?

In a former life, I've done a bunch of FPGA and hardware synthesis work,
but I haven't been back in that land since I've been playing with
graphics for the past while. Though there are already a couple of recent
developments that may push me back into the land of programmable
hardware soon.

For cairo itself, we haven't done any profiling yet, which would be a
necessary first step, of course. But at the bottom of the stack, the
fundamental geometric operator is to rasterize a trapezoid then
composited (IN) with a mask, and composite (any Porter/Duff operator)
with the destination.

Our current software for that is in libpixman, but we've designed a new
specification that we haven't implemented yet.

-Carl




More information about the cairo mailing list