Damn, they're some impressive performance gains.
-
-
-
Yep - initial version was good but a bit of bitmasking/shifting makes it shine! To be handled with care as it only support integers on a given range (eg max shift)
End of conversation
New conversation -
-
-
very interesting. my impression was that you can get a decent improvement using radix vs heapsort, with some trade-off with the number of depth bins related to the base you sort in. taking out some of my clunk helps... but doing the sort in memory is a big plus.

-
I already used that array trick on a voxel demo - for byte values, this is indeed faster than for i/table[i]=0
End of conversation
New conversation -
-
-
Nice work!
Thanks. Twitter will use this info to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.