FYI this is a file that evaluates cost of various ops when optimizing (for i386):
http://repo.or.cz/w/official-gcc.git/blob/trunk:/gcc/config/i386/i386.c It is grouped by processor models; interesting is that (for core2 family) sqrt costs 58, float division 32 (!), float multiplication 5 (for AMD k8 processors, it is 35 for sqrt and 19 for fdiv). (that doesn't mean that you should rewrite * to /, gcc is smart enough to precompute if used multiple times) Cheers, v. _______________________________________________ Mailing list: https://launchpad.net/~yade-dev Post to : yade-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
