Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: a test position for chess programs(fritz could not solve it in a game)

Author: blass uri

Date: 12:37:33 04/03/00

Go up one level in this thread


On April 03, 2000 at 14:32:20, Paulo Soares wrote:

>On April 03, 2000 at 13:14:10, blass uri wrote:
>
>>On April 03, 2000 at 11:47:08, Ernst Walet wrote:
>>
>>>Tascbase 2.1a (the King 2.55) really amazes me by finding Kc2 in 2 seconds at 8
>>>ply.  After entering Kb2 it replies with Qg7+, giving this move a score of +2.12
>>>after 4 seconds at ply 8!
>>>
>>>Can any program beat this?
>>>
>>>Ernst.
>>
>>I said that chessmaster6000 and Rebel have no problem with this position.
>>
>>This fact does not amaze me but the fact that other programs cannot see this
>>simple tactics only because of the fact that the number of plies is big.
>>
>>All the moves are checks replies to check or natural passed pawn moves so strong
>>humans will have no problems to see it in a game if they are not in time
>>trouble.
>>
>>It is amazing that some programs like Fritz are so weak in tactics.
>>
>>Uri
>
>I do not agree with you, in my opinion it's impressive. How The King 2.55 gets
>that? Below the moves that do with the position finishes with 5 pieces.
>
>1... Kb2 2. Qg7+ Kb1 3. Qxa1+ Kxa1 4. h5 b3 5. h6
>b2 6. h7 b1=Q 7. h8=Q+ Qb2 8. Qa8+ Kb1 9. Qxe4+
>
>Paulo

All the moves after Kb2 are checks or replies to check or pusing the unstopped
passed pawn and I expect myself at tournament time control to see all this line.

I guess that the king does push unstoppable passed pawn and check+reply to check
extensions to see this line.

Chessmaster needs to see also the line with 7.h8Q+ Ka2 and here you need
knowledge about pawn endgame to see that white is winning.

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.