Author: pavel
Date: 18:54:20 11/01/00
Go up one level in this thread
On November 01, 2000 at 17:42:26, Peter Skinner wrote: >>unless the results (or similar results) are confirmed by several others, who >>have messed with 4.32e too, I wuold'nt even drop my pants ;) >> >>Pavel > >These are personal games for my own benefit. There have been testers of Rebel >Tiger II that posted games, that on similiar hardware I could not reproduce. > >This does not mean my results are false, nor would their's be, it could just be >one person had better results than others. I am not saying that your results are false, but it is too complicated. How often do you see one version of a program beating a differant version of the same program, and yet not get enough coverage? it's unussual, and needs confirmation from more than one person.... Pavel > >Perfect example was some of the Rebel Tiger 12.0e users on ICC. Most were >running on equal Mhz value computers, yet most times there was large gaps in >rating, even though they were all beating but on the same Crafty clones, and >other computers. I remember LeTiger(C) was leaps and bounds ahead of the other >Tiger clones, and none could reproduce it's results. > >The only thing I changed my my matches were the hash table settings. All boths >ran with default settings, and default books.
This page took 0 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.