Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Human rating differential compared to Computer vs. computer

Author: KarinsDad

Date: 13:16:32 01/29/99

Go up one level in this thread


On January 29, 1999 at 15:57:43, allan johnson wrote:

> In Australia it is not unusual for players rated 1400 to defeat players of 1900
>in the Australian championships in January.Players of 1500 strength have even
>beaten opponents rated 2000.There is generally a logical reason for this
>occuring though.What happens is that ratings are adjusted annually here and so
>up and coming stars are often playing well above their rating when they meet an
>opponent of allegedly greater ability.So your arguement that a 500 rating
>difference suggests a guarenteed win does not work here.

True. If you have a cruder system where a larger percentage of players whose
strength does not correspond to their ratings, then of course you will have more
"exceptions to the rule". I would think a problem like this would influence your
chess federation to modify the frequency of their rating adjustments.

> As for chess programmes playing one another I have tested the Mephisto Risc
>against Rebel 9 on many,many occasions and the Risc has yet to secure a draw.

I would be curious to know what the "approximate ratings" of these two are (and
only you know what type of system you run Rebel 9 on).

KarinsDad



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.