Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CM9K Personalities - How Accurate Are Their Ratings?

Author: John Merlino

Date: 20:08:46 02/06/03

Go up one level in this thread


On February 06, 2003 at 21:15:08, Ralph Patriquin wrote:

>On February 06, 2003 at 15:25:09, John Merlino wrote:
>
>>On February 06, 2003 at 14:23:24, Ralph Patriquin wrote:
>>
>>>On February 06, 2003 at 07:52:34, Harpo wrote:
>>>
>>>>After a 15 year hiatus, I'm planning to return to tournament play. In
>>>>preparation, I've been playing 30/75 SD60 games against a handful of CM9K's
>>>>personalities. I recall reading somewhere that these were guaged with USCF-rated
>>>>players, but I would appreciate any opinions from OTB players as to how accurate
>>>>these ratings "feel." I'm only curious about the personalities rated from
>>>>2100-2500.
>>>>
>>>>Thanks in advance.
>>>
>>>Harpo,
>>>
>>>As John Merlino mentioned, some of the personalities exhibit rather bizarre
>>>playing styles that cause them to make moves that in my opinion even a
>>>relatively weak human player would not . I think you might find it will hurt
>>>your play by training with them.
>>>
>>>I prefer to use the Chessmaster and other Grandmaster personalities but simply
>>>weaken them by adjusting their Strength of Play setting. You might want to start
>>>out at 50% and adjust from there in 10% increments up or down until you get a
>>>challenging game.
>>>
>>>In this way you get a more balanced game with different playing styles, but
>>>without the crazy play exhibited by some of the other personalities. You will
>>>also play against better and varied opening books. The only downside is you
>>>won't know the true rating of these opponents. Maybe John can explain how
>>>Strength of Play works and a possible rating estimate from it, as I have no idea
>>>as to its internal workings.
>>>
>>>Ralph
>>
>>Simply adjusting the Strength of Play setting is a good idea, but this setting
>>is not exactly intuitive. The way Johan set it up this variable, it uses an
>>inverse cubic function. So, going from 100 to 90 (or even 85) will result in
>>VERY little reduction of strength. Once you get below 80, the strength starts to
>>fall off more dramatically. So, looking at other personalities will help in this
>>regard as well.
>>
>>jm
>
>
>Inverse cubic, wow! Well I guess he had his reasons. Although it would have been
>nice if it was linear, so that simple assumptions could be made, like:
>
>"If Chessmaster is 2600 at 100%, then it's approximately 1300 at 50%."
>
>John do you know anything about how the play is actually weakened? Does it use
>second best move, third best move etc. (probably a little too simplistic) or
>something else?
>
>Ralph

I agree, from a user standpoint, that a linear scale would be much more
intuitive. And, honestly, I have no idea how the Strength of Play setting works.
Sorry....

jm



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.