in graphics.library fast region functions. YXBanded region algorithms like used by X11 (or aros thanks to Fabio) might be a must, as they (although often resulting in regions having more cliprects than really need to describe a shape) are *much* faster than normal region algorithms.
Yes, that is exactly on point: Graphics library is 68k, and as such the region code is emulated. Furthermore, there is a certain overhead when changing between native code and emulation, making these calls more expensive since layers is PPC. The result is that this takes much longer than it should (Thomas Richter rewrite layers for OS 4)
Quote:
Intuition shouldn't actually have much to do with opaque speed problems at all!? Or maybe you just mean delayed-app-refresh/lagging-refresh symptoms
Indeed. The refresh in the input handler means that it hampers reaction speed. We're experimenting with this, but like I said, one of the major problems is that graphics is still emulated. Especially the fact that there are lots os small functions being emulated incurs a big overhead.
One point that shouldn't be overlooked is that OS 4 isn't finished yet. If it where, then we would already be selling it.
_________________ Seriously, if you want to contact me do not bother sending me a PM here. Write me a mail