> May be do need calculation of AABBs? If do it, leak don't observed. > But without AABBs leak have been and for 16k spheres. > Maybe too many interactions to create (nbSpheres^2) in this case? > Yes, that was the cause. I have some graphs now comparing SAP and QuickSort: http://yade.wikia.com/wiki/Colliders_performace . QuickSort is always faster and SAP sucks big time on the initial sort (3 hours for 128k spheres!!). Could you also run the test (warning: it takes like 6 hours on a 6 way machine!!)? Maybe QuickSort's speed depends on the packing density; the one I use is the initial packing generated by TriaxialTest.
Vaclav _______________________________________________ Mailing list: https://launchpad.net/~yade-dev Post to : [email protected] Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp _______________________________________________ yade-dev mailing list [email protected] https://lists.berlios.de/mailman/listinfo/yade-dev
