Question for Chess Players...is this rule true?

Discussion of chess software programming and technical issues.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
mjlef
Posts: 1427
Joined: Thu Mar 30, 2006 12:08 pm
Contact:

Question for Chess Players...is this rule true?

Post by mjlef » Sun Apr 20, 2008 9:12 am

A piece becomes more powerful the less pieces of lower value the opponent has.

Is this true? For example, the fewer minor pieces an opponent has (knights and bishops), the more powerful its rook becomes.

Testing seems to show this, but I am way too lazy to wait for the thousands of games I might need to prove this.

Mark

User avatar
hgm
Posts: 23772
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Full name: H G Muller
Contact:

Re: Question for Chess Players...is this rule true?

Post by hgm » Sun Apr 20, 2008 9:11 pm

Perhaps we should organize a huge joint effort to address fundamental questions like this? If we all donate a week of CPU time to play 3500 or so 40/1' games, and 30 persons participate, we can do 100k games.

If we use top-engines, the Pawn-odds score is over 70% (i.e. a Pawn corresponds to a 20% excess score), and the standard error in 400 games is only ~2%. So we could measure 250 judiciously chosen piece combinations (all close to equality) to a precision of ~10 cP.

User avatar
michiguel
Posts: 6388
Joined: Thu Mar 09, 2006 7:30 pm
Location: Chicago, Illinois, USA
Contact:

Re: Question for Chess Players...is this rule true?

Post by michiguel » Mon Apr 21, 2008 1:34 am

mjlef wrote:A piece becomes more powerful the less pieces of lower value the opponent has.

Is this true? For example, the fewer minor pieces an opponent has (knights and bishops), the more powerful its rook becomes.

Testing seems to show this, but I am way too lazy to wait for the thousands of games I might need to prove this.

Mark
No, there is no "rule" such as that one for a carbon chess player. However, it makes sense indirectly. A rook is stronger when there are useful open files. The more minor pieces present, the less useful may become because the open files can be controlled. For instance, a Re1 is not to great if you have a Bd6 and Bd7 that controls the access.

Miguel

Harald Johnsen

Re: Question for Chess Players...is this rule true?

Post by Harald Johnsen » Mon Apr 21, 2008 7:36 am

mjlef wrote:Testing seems to show this, but I am way too lazy to wait for the thousands of games I might need to prove this.

Mark
What did you test ? I don't understand what you are talking about, how did you translate 'powerfull' in your engine eval() ?
You are talking about equal material positions right ?

HJ.

User avatar
Kempelen
Posts: 620
Joined: Fri Feb 08, 2008 9:44 am
Location: Madrid - Spain
Contact:

Re: Question for Chess Players...is this rule true?

Post by Kempelen » Mon Apr 21, 2008 8:08 am

mjlef wrote: Is this true?
Maybe you can conclude a general rule which assure that, but remember what a grandmaster would answer you: "it all depends on the position!"

mjlef
Posts: 1427
Joined: Thu Mar 30, 2006 12:08 pm
Contact:

Re: Question for Chess Players...is this rule true?

Post by mjlef » Mon Apr 21, 2008 12:52 pm

There are a lot of way of implementing this. My first attempt is just to count the number of opponent minor pieces. The fewer the opponent minor pieces, the larger the rook value becomes. For queens I count number of opponent rooks and minor pi9ecesm, but the minor pieces count twice since exchange with them is worse for the queen, so whatever squares they control are more significant and restrictuive for the queen.

So I guess "powerful" in my case means "worth more". So this means a R and pawns vs say Knight and pawns would be worth more than +2, where say in an opening, a R vs N difference would be worth less than this.

CThinker
Posts: 388
Joined: Wed Mar 08, 2006 9:08 pm
Contact:

Re: Question for Chess Players...is this rule true?

Post by CThinker » Fri Apr 25, 2008 11:15 pm

mjlef wrote:A piece becomes more powerful the less pieces of lower value the opponent has.

Is this true? For example, the fewer minor pieces an opponent has (knights and bishops), the more powerful its rook becomes.

Testing seems to show this, but I am way too lazy to wait for the thousands of games I might need to prove this.

Mark
If you consider that most chess engines score mobility by counting the number squares that a piece can go to, then your statement has some merit to it (as far this type of scoring goes). A mobile queen is better than a mobile knight.

Now, if you consider scoring mate patterns (see Ed Schröder's chess programming article, from which I copied a lot of ideas from), then it goes the other way. You have less likelihood of getting into positions of mate patterns if you have less pieces to make the necessary combinations. A KQ-KQ is a draw, while a KQRBPPP-KQRBPPP has a greater possibility of being a non-draw.

Post Reply