Updates: Status: Fixed
Comment #1 on issue 450 by alexandre.bergel: Non embedded edges culling in Mondrian http://code.google.com/p/moose-technology/issues/detail?id=450
In the version 2.30 of Mondrian
Updates: Labels: Milestone-4.3
Comment #2 on issue 450 by tudor.girba: Non embedded edges culling in Mondrian http://code.google.com/p/moose-technology/issues/detail?id=450
Sounds great. What is the impact on speed for a large visualization?
Comment #3 on issue 450 by alexandre.bergel: Non embedded edges culling in Mondrian http://code.google.com/p/moose-technology/issues/detail?id=450
I did some benchmark, but I am not sure on how close they are to the reality. I generated the blueprint of the AST classes. Doing 100 repaints of take 9214 ms without the improvement. It takes 6045 with.
To obtain the canvas, I put a "self haltOnce" in MORoot>>displayOn:
Comment #4 on issue 450 by tudor.gi...@gmail.com: Non embedded edges culling in Mondrian http://code.google.com/p/moose-technology/issues/detail?id=450
Nice
Comment #5 on issue 450 by alexandr...@gmail.com: Non embedded edges culling in Mondrian http://code.google.com/p/moose-technology/issues/detail?id=450
it works. But it can be optimized I think. For example, I check if the bounds given by the current window frame intersect the rectangle bounds of the edge. This is clearly not optimal since we should check if the line crosses the window frame.
-
moose-technology＠googlecode.com