Skip to content
Browse files

report: Figure

  • Loading branch information...
1 parent df8b736 commit 839324e9f5ce7579d029261eced674c7beeaad56 @thomasintveld thomasintveld committed May 20, 2012
Showing with 1 addition and 1 deletion.
  1. +1 −1 report/spacePartitioning.tex
View
2 report/spacePartitioning.tex
@@ -27,5 +27,5 @@ \subsection{Space Partitioning}
Hence, there will be a significant amount of boxes that are completely empty, and vice versa a significant amount of boxes with a much higher than average number of particles (sites).
Caching in modern CPUs may also play a role, in the sense that large numbers of monomers and boxes imply that the entire working set does not fit in the fastest cache levels on the CPU, potentially giving a bias to smaller boxes (though we can't immediately see a reason for this).
-Lastly, the performance scaling with the ideal number of boxes is compared to the scaling without space partitioning in figure \ref{spacePartVsNoSpacePart}. The quadratic trend of the naive method is clearly worse than the linear scaling when space partitioning is enabled.
+Lastly, the performance scaling with the ideal number of boxes is compared to the scaling without space partitioning in Figure \ref{spacePartVsNoSpacePart}. The quadratic trend of the naive method is clearly worse than the linear scaling when space partitioning is enabled.

0 comments on commit 839324e

Please sign in to comment.
Something went wrong with that request. Please try again.