Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: King, rook pawn and wrong bishop endgames

Author: Dieter Buerssner

Date: 06:18:45 10/19/00

Go up one level in this thread


On October 18, 2000 at 14:14:20, Uri Blass wrote:

>I can give another rule:
>Blacks draw if it is black to move and the distance between the black king and
>the corner in king moves is smaller than the ditance between the corner and the
>white pieces except the bishop(white king in king moves and white pawn in pawn
>moves)
>
>If it is white to move then use distance-1 instead of distance for white.

Uri, thank you very much, for giving your rule. Also thanks to all
the other answers to my questions.

I have made a small experiment. I set up 10000 random positions
with a white pawn on h, the wrong bishop and the kings on two random
legal squares.  Black to move.
Of the 10000 positions, 3716 where draws due to TBs.
I compared the result of your rule, and the result that is returned by TBs. Your
rule had an overall success rate of 90%.
At first sight, this might not look too good, but it will probably be
very good in practical play. I checked some of the positions, where the
"rule" failed. Most positions, where you rule failed, where when it predicted
a loss and the position is draw are trivial, because the Black king can capture
the pawn in the next move. So in a real game, the quiescence search would find
this immediately.

In a few cases, the pawn can be captured the move after the next move,
and these are of course more dangerous. One example

[D] 2B5/8/7P/5k2/8/8/3K4/8 b - - 0 0

But again here, with the help of search, this will be found soon.

Rarely, your rule predicted a draw, when the position is lost. But here the
success rate of your rule was almost 99%. (In 95 positions, your rule predicted
a wrong draw). One example:

[D] 5k2/8/7P/8/8/8/B7/6K1 b - - 0 0

I fear, it is very difficult, to see all those cases.

Of course, as you pointed out, things get much more complicated,
when there are black pawns on the board.

One final question. When the white rook pawn is doubled, it should be
enough, to look at the more advanced pawn. Or are there any exceptions?

Regards,
Dieter




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.