|
From: Frank C. <dev...@ch...> - 2004-09-28 23:48:34
|
On 28-Sep-04, at 6:13 PM, James W. Walker wrote: > Frank Condello <dev...@ch...> wrote: > >> James: Has there been any progress with your new sorting algorithms? >> I would very much like to see your fast centroid method checked in. > > I wasn't sure that we had a consensus that it was the right thing to > do, since in at least one case the centroid method looks worse than > the current method. Which case was that? I can't seem to render anything properly with the current code. Triangles pop unpredictably at the slightness change in the viewport... From my perspective, your centroid code is the lesser of two evils :) > Also, you proposed ideas of accumulating transparent triangles in > chunks, rather than one big heap, where different chunks can have > different sorting methods. I liked the idea, but haven't found the > time/energy to actually implement it. Well, if you check in some new sorting code I might be persuaded give you a hand with that part. I think for the time being it would be nice to fix or at least improve upon the following transparency issues: <http://sf.net/tracker/index.php? func=detail&aid=989156&group_id=45158&atid=442052> <http://sf.net/tracker/index.php? func=detail&aid=967773&group_id=45158&atid=442052> <http://sf.net/tracker/index.php? func=detail&aid=973022&group_id=45158&atid=442052> <http://sf.net/tracker/index.php? func=detail&aid=967958&group_id=45158&atid=442052> Then we can work on speeding things up - though at this point your code appears to be faster than the current method anyway. Frank. |