Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A Little Endgame-Test...

Author: blass uri

Date: 11:35:46 08/07/00

Go up one level in this thread


On August 07, 2000 at 14:17:32, Jari Huikari wrote:

>On August 06, 2000 at 07:12:21, blass uri wrote:
>
>>On August 06, 2000 at 06:54:57, Carlo Meier wrote:
>>
>>>Ra2+  ist not the right move... :)
>>>
>>>greeting:  Carlo
>>
>>Do you mean that Ra2 does not win or that it does not win in the shortest way?
>>
>>I think that rook and a bishop win against bishop and a pawn.
>>White cannot take the pawn but may drive black king into the corner
>>
>>Uri
>
>How about moving Rh1 and then Nf4 ?
>
>					Jari

I also thought about this line but did not see a clear win when I analyzed with
a computer after some minutes.

The problem is that the black bishop can be in the diagnol h5-d1 and I do not
see a way for white to get to the right square with the knigh.

I admit that
1.Ra2+ Kh3 2.Ng5+ Kg3 3.Be5+ f4 is good for black and it is surprising that some
programs(including crafty with no tablebases) believe that a piece advantage
after sacrificing the rook for 2 pawns is more than 4 pawns advantage

I believe that only material counting programs with no tablebases can see a main
line with an endgame with a piece advantage but some programs have evaluation
function that is worse than only material evaluation in this case and they
prefer to go for position with rook for a pawn adavntage for white and not a
knight advantage.

Junior5.9 has better evaluation and it evaluate the position as close to +1 for
white and does not see win for white.

Uri



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.