Used to be in separate static RAM (really fast), chips, but the algo's they used were the same one's they use today. Thankfully, the cache's today are larger. Sure that helps. How much this helps each sorting algorithm, I just couldn't say. The pivot in Quicksort is (hopefully) held in a register while in it's inner loop, but the larger a sub-group it can fit in a cache, the better it will perform.
Branch prediction has noticeably improved in the cpu, also. Yes, in the old days you used to be able to turn off the cache in the BIOS. And that was always a fine prank to play on a buddy you wanted to annoy. All of a sudden his machine ran like it was stuck in molassis. Due to pre-load in the HD controller, disk fragmentation isn't the problem it used to be. Still a problem, just not such a big one. I'd very much like to see a test of these algo's on modern PC's, compared "fairly" (if that's even possible), side by side. Adak --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to algogeeks@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/algogeeks -~----------~----~----~----~------~----~------~--~---
