← Back to team overview

yade-dev team mailing list archive

Re: perfomance benchmark

 

But scaling O(NlogN) is not so bad.. And perfomance can be significantly optimized now by playing with verletDist, I think. But it need some tries.
You are between two influences; if verletDist is too big, then many potential interactions will exist, which slows down the InteractionLoop (many Ig2 calls). OTOH if verletDist is too small, the collider will have to run more often.
But what about

 40% of collider time is spent in the "find(id1,id2)" ???

Because, if more verletDist => more interactions per body => yet more time for collider?!
Big proportion of find(...) is not surprising, because it has to fetch different parts of interaction data from RAM every time; but I would guess that is part O(N), unlike the sort algorithm.




Follow ups

References