Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: LCTII test vs SSDF results

Author: Peter Kappler

Date: 12:25:00 09/27/99

Go up one level in this thread


On September 27, 1999 at 12:55:25, Joshua Lee wrote:

>SSDF might as well start over, hire some masters or GM's to play some of the top
>programs ,(and i don't mean 2 games or whatever the small number of human vs
>computer games they originally had played) but since they are going to cheese
>and  not do this the best thing is to take LCTII results and subtract 200points.
>For example Hiarcs rates itself 2450 on my K6-2 300, it gets 2620 on the LCTII
>test. It sounds pretty close....


The LCT2 rating formula was specifically designed to match the results in the
SSDF rating list.

I think it's obvious that estimating playing strength via test suites is
inferior to playing hundreds of test games.

That being said, I think people pay way too much attention to the SSDF ratings.

--Peter



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.