Author: Bruce Moreland
Date: 11:36:12 07/20/98
Go up one level in this thread
On July 20, 1998 at 14:30:47, Dan Newman wrote: >I think I've seen this priming-the-pump effect--though perhaps not as >pronounced as a 4% change in node rate. Is it due to branch prediction >being pre-set by the first run or some sort of alignment variation? I don't know. 4% seems like a lot, but I've seen several percent at least, discounting any disk I/O which would be cached on later runs, and could of course result in even more difference. I'm not a hardware guy. >One of the biggest effects I see is that I get a much larger node rate >the longer I run. If I run a test for 10 s I might get 170 knps. If >I run for 40 s 200 knps. It does level off of course. I'm using a >P6/200, so I've attributed this effect to improving branch prediction >as the code runs. Any ideas? Maybe it has something to do with the deeper search having characterstics that differ from the shallower search. bruce
This page took 0.19 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.