Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: De Gorters review of Crafty 16.6 (integral text, English translation)

Author: James Robertson

Date: 11:26:54 04/26/00

Go up one level in this thread


On April 26, 2000 at 06:22:21, Jeroen van Dorp wrote:

>Read the text, don't interprete it.
>
>He doesn't base his judgement on the results in the tournament. He based it on
>Crafty's type of play.

Ok. Let's stop again a minute and think. You are now saying that his opinion on
Crafty is not based on the results of the tournaments, but just on the type of
play he saw in Crafty. In other words, he looked at 20 games and ignored the
results because he was just analyzing the type of play. That is fine. He can say
if he does not like the way Crafty handles mobility. That is his opinion.

But he didn't stop there. He made a strength judement on Crafty.

"Crafty is a weak program compared with the best commercial chess programs."

You are saying he based this conclusion soley on Crafty's type of play, which he
 judged inferior (remember, you are saying he didn't take game results into
account).

Do you realise how little sense this makes? Obviously, if out of those 20 games
against the "best commercial chess programs" Crafty scores a nearly even score,
its "type of play" in those 20 games MUST be almost as good as that of the "best
commercial chess programs".

What could his conclusion be based on?
It can't be the tournament results (Crafty scores on the top 50% of the time).
It can't be the internet ratings (Crafty has higher ratings).
It can't be the type of play (Crafty scored nearly 50%, so its type of play must
be nearly as good).

What other options are there? It must be based on preconceived ideas.

James

>
>The high rating on the net are no proof for him.
>He thinks high rating on the net comes from superiour hardware; that's why he
>wanted to test on equal hardware.
>
>Jeroen ;-}



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.