Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A test position for chess programs(counting attackers-defenders)

Author: blass uri

Date: 09:28:04 01/14/99

Go up one level in this thread



On January 14, 1999 at 12:13:42, James T. Walker wrote:

>On January 14, 1999 at 11:35:34, blass uri wrote:
>
>>
>>On January 14, 1999 at 10:44:31, Laurence Chen wrote:
>>
>>>On January 14, 1999 at 10:12:44, blass uri wrote:
>>>
>>>>
>>>>On January 14, 1999 at 08:49:37, James T. Walker wrote:
>>>>
>>>>>On January 14, 1999 at 07:36:07, blass uri wrote:
>>>>>
>>>>>>
>>>>>>5rk1/1B6/4p1pp/2b3p1/3qP3/6P1/1P2QPP1/5RK1 w - - 0 1
>>>>>>
>>>>>>This is a position from the game Junior5-FM Boris Mariasin(The result was draw)
>>>>>>
>>>>>>Junior5 and other programs evaluating it as an advantage fo white because they
>>>>>>do not know to count Attackers-Defenders
>>>>>>
>>>>>>Is there a progam with a more realistic evaluation (small advantage for black or
>>>>>>at least less than 0.2 pawn for white)?
>>>>>>
>>>>>>Uri
>>>>>Hello Uri,
>>>>>I'm a very bad chess player so I don't understand what is happening in this
>>>>>position.  I see that white is up a pawn which is about a point.  So where is
>>>>>the advantage for black?
>>>>
>>>>The advantage of black is that the black pieces attack f2 and all the white
>>>>pieces must defend f2.
>>>>
>>>>White has no plan to do a progress in the position when black can play h5,g4,h4
>>>>with an attack.
>>>>
>>>>I am not saying that black is winning but the evaluation should be at least
>>>>equality.
>>>>
>>>>The problem of computers is that they do not count attacker-defemsors of f2 and
>>>>cannot understand that the fact that white is passive give black at least
>>>>equality
>>>>
>>>>Uri
>>>Uri can you prove this position is equal? Really, you are saying that because of
>>>the opposite color bishops in the position. Yet, let me ask you then, what can
>>>Black do? Black is down a pawn, and White has a passed pawn on the Queenside,
>>>that is why the +1.xx evaluation. So the position is not equal, the assessment
>>>is correct, because White can break the pin of f2 easily, and if Black attempts
>>>to advance the Kingside pawns, it gets in big trouble, it exposes the King to
>>>White pieces.
>>
>>White cannot break the pin easily
>>The game was drawn after
>>26.Ba6 Kg7 27.Bb5 h5 28.Bc4 e5 29.b3 g4 30.Bd5 h4 31.Kh2 Qc3 32.f3 gxf3 33.gxf3
>>hxg3+ 34.Kxg3 Be3 35.Rh1 Bf4+ 36.Kg2 Rc8 37.Qf2 Be3 38.Qa2 Qc2+ 39.Qxc2 Rxc2+
>>40.Kh3 Bf4 41.Bc4 and the sides agree to a draw.
>>
>>What white could do?
>
>*Fritz has this at +.50 for white.  Nimzo calls it +.58 for white.  Since white
>has a pawn up and poor position this seems correct to me.  The black e pawn is
>weak and needs defending after Ba6 and then either Qc4 or Bc4.  Qc4 may force
>swapping queens. Bc4 followed by b3 saves the b pawn.  There seems to be a lot
>of chess to be played yet with options on both sides.  A +.50 advantage is not
>winning,  it's just an advantage.  The fact this game ended in a draw is not a
>surprise to me but in the mean time white still has the extra pawn.
>>
>
>>The only plan for white to get out of the pin is g4 g3,Kg2 but this weakens the
>>white king.
>>
>*Very seldom in chess is there a "only plan".  I don't understand the Kg2 since
>g2 has a pawn on it.

kg2 is only after moving g3 to g4 and g2 to g3.
Do you have another plan for white to win the game?

>
>>After g4 Junior wants to play Rb8 and black gets the pawn back.
>
>*If the rook moves off the f file that kills the pin(Attackers vs Defenders).
>Jim Walker

I agree
Rb8 is what Junior wants to play after g4 and it does not have to be the best
move.
I prefer h5 gxh5 gxh5 and now Junior's evaluation after some minutes is that the
2 best moves of white are g3 0.20
and Qc2 0.12

if Qxh5 Rxf2 and black has a draw.

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.