freenode/#clim - IRC Chatlog
Search
4:03:39
slyrus
Ok, I like the idea of using XRENDER, and I think the frame-buffer backend is maybe a useful tool/experiment, but I don't think it should replace the traditional approach. Seems like it would negatively impact HW acceleration.
4:05:28
beach
We (the collective "we", not the royal "we") also have some documentation work to do if we want to establish the use of the XRENDER extension. The documentation of the X11 protocol extensions for XRENDER is totally incomprehensible, at least to me. And that is after several attempts to read and understand it.
4:08:16
beach
Presumably, the same person who wrote the documentation of the protocol extension also wrote the C xlib code to use it.
4:09:06
beach
But CLX goes to the protocol directly (which is a good thing in my book), and gilberth defined the CLX layer differently from the way the C xlib layer is defined.
5:44:00
beach
Sounds plausible. I don't think we can do that in general since it requires sharing memory between the Common Lisp system and those extensions.
9:39:25
jackdaniel
http://stackoverflow.com/questions/2025282/what-is-the-difference-between-px-dp-dip-and-sp-on-android (explanation what densiti-independent pixel is)
14:13:11
jackdaniel
slyrus: "may be rendered moot by the bezier-clx-draw branch" could you rephrase? I don't understand this sentence
14:14:38
slyrus
i have a branch that uses native CLX stuff to draw the bezier designs, instead of relying on the mcclim-raster-image stuff. both approaches are interesting, but the new one should be the default IMO.