Author: Dragos Gabudeanu
Date: 21:31:36 11/25/00
Go up one level in this thread
On November 26, 2000 at 00:23:22, John Merlino wrote: >On November 25, 2000 at 23:50:18, Luis E. Alvarado wrote: > >>I just bought CM8K and I was surpriced that the rating of some personalities >>changed with respect to CM7K. Do you know if the CM8K personalities ratimg was >>afected by the survey that was conducted with USCF chess players when they were >>asked to play some of the personalities. Does anybody know how the survey data >>was used.... Luis > >Yes, that is precisely what the test with the USCF players was for. A great >amount of human vs. computer games were played with five "anchor point" >personalities, and it was pretty much discovered that most of the personalities >(especially those between 2000 and about 1300) were rated WAY too high, some by >as much as 200 points or more! > >Therefore, what you are probably seeing is the personalities that had their >ratings lowered, although a few (not many) personality's ratings were actually >raised. > >jm On my k6-2 500 with 256MB CM7k rates itself 2641 US. But CM8K on the same machine rates itself 2818. Is that because the newer King version is thought to be much stronger, or was the rating inflated? How is a rating like that calculated? Dragos
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.