Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Are comments about Crafty 16.6 to harsh or just accurate observations?

Author: James Robertson

Date: 21:04:38 01/10/00

Go up one level in this thread


On January 10, 2000 at 23:31:34, Roger wrote:

>Are you concerned with Crafty as it was discussed in de Kook's article, or just
>with version 16.6. If with version 16.6, well...it's a little old to be talking
>that, so...I'll assume it is with de Kook.
>
>Below was my critique of the article. The content of the article is totally
>weird given it's putative purpose. Anyone can see that. Don't you think then,
>that his observations might be made for a political purpose, even if true?
>
>I don't think any of the programmers would object to having their programs
>critiqued CONSTRUCTIVELY...what folks objected to was a political smear
>campaign. You would have to have very poor interpersonal skills...No, you would
>have to be interpersonally oblivious not to see that the title of your post
>seems to align you with people who would stoop to such smear campaigns.
>
>Either that, or you would have to be eager to smear someone yourself. Your
>disingenuousness doesn't pass for sincerity. You're not pulling it off. You're
>transparent.
>
>Roger
>
>
>
>I took a look at this article, and it seems like total BS. First, it opens by
>saying that it's a review of the "best commercially available programs." Then it
>IMMEDIATELY heads into a long ANTI-CRAFTY portion, when we all know that Crafty
>is free for the download!!!!  That says something about the author and his
>qualifications.
>
>The author DOES SAY that Crafty is often included as an engine with commercial
>programs, but even so....Why in the world would you BEGIN AN ARTICLE DEVOTED TO
>A REVIEW OF COMMERCIAL PROGRMS with a review of one of their SUBSIDIARY ENGINES.
>You wouldn't.
>
>He goes on to present a tournament he designed to determine the strength of
>Crafty ("To compare Crafty with the others..."), ostensibly to provide objective
>support for his observation that "Crafty is extremely strong in tactics, but it
>makes positional errors and misplays some endgames. It doesn’t understand piece
>mobility. Despite its high rating on the Internet, Crafty is a weak program
>compared with the best commercial chess programs." The tournament shows Crafty
>getting 2.5 points of 10 against Nimzo, Fritz, and the like.
>
>But since Crafty is free for the download, and since it's a BONUS ENGINE, why
>does he NEED to design a tournament explicitly to evaluate Crafty? Hint for the
>author: "Sir, you should not head directly into a tangent after explicitly
>stating the theme of your article." This is UTTERLY STUPID, because it seems to
>make Crafty the primary focus of an article entitled "Commercial Chess Programs:
>A Comparison."
>
>Worse, later he goes on to IGNORE HIS OWN RESULTS by presenting the outcome of
>another tournament he made for fun in which Crafty participated, at the same
>time controls. Here are the results, taken from the webpage:
>
>ChessBase home competition. Swiss style 7 rounds, 12 players.
>Automatic tournament 30 minutes per player per game;
>1 Junior 5 x ½ 1 1 1 1 1 1 6½
>2 Crafty 16.6 ½ x 0 1 ½ ½ 1 1 4½
>3 Junior 4.6 0 1 x 0 1 ½ 1 1 4½
>4 Fritz 6 0 0 1 x ½ 1 1 1 4½
>5 Nimzo 7.32 0 ½ 0 ½ X 1 1 1 4
>6 Hiarcs 7.32 ½ ½ 0 x ½ ½ 1 1 4
>7 Hiarcs 4 0 0 ½ x 0 1 0 1 2½
>8 Hiarcs 6 0 ½ 1 x 0 0 ½ ½ 2½
>9 Fritz 5 0 0 0 0 1 x 1 ½ 2½
>10 Fritz 5.32 0 0 0 1 0 x ½ 1 2½
>11 Comet B09 0 0 0 1 ½ ½ x 0 2
>12 LG2000 0 0 0 ½ ½ 0 1 x 2
>
>These data DIRECTLY REFUTE his earlier conclusion that "Despite its high rating
>on the Internet, Crafty is a weak program compared with the best commercial
>chess programs." In these data, it ties Fritz 6, and beats Nimzo 7.32, and
>Hiarcs 7.32, and Fritz 5.32. He then acknowledges that "The only reliable
>ranking is that of the so-called Swedish rating list," but neglects to mention
>that Crafty places higher in his little tournament than Hiarcs 7.32 and Fritz
>5.32, former list champions.
>
>Arguments about the relative strength of various programs could go on forever,
>of course, but based on the DATA THE AUTHOR ARGUES FROM, he is cannot exactly be
>called bright.
>
>Roger

I read your review again, and I really had to laugh... it is well written
humorous, and very true. :)

James



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.