Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Stupid chess programs!!! Hiarcs732 can't FINISH M in 74

Author: Tina Long

Date: 04:03:41 12/17/99

Go up one level in this thread


On December 17, 1999 at 06:38:27, blass uri wrote:

>On December 17, 1999 at 04:42:42, Tina Long wrote:
>
>>On December 15, 1999 at 13:08:20, Enrique Irazoqui wrote:
>>
>>>On December 15, 1999 at 11:44:52, Torstein Hall wrote:
>>>
>>>>Take the following position and setup on your PC:
>>>>
>>>>4b3/1r4k1/p3pr2/1p1pPp2/PPpP1Pp1/2P3Pp/K6P/8 w - - 0 1
>>>>
>>>>Every human would play 1.a5 without thinking, but your chessprogram plays
>>>>1.exf6+??? It may be wrong to call a program stupid but.....
>>>>
>>>>Torstein
>>>>
>>>>( So far I have only tried it on Fritz 6 and Hiarcs 7.32 )
>>>
>>>I tried for about half a minute most of the programs I have and they all take
>>>the rook. They are incredibly stupid, but they think the same of us for not
>>>seeing instantly that this position is a mate in 74.
>>>
>>>8/8/8/8/2R1k3/2K5/1P6/3r4 b - - 0 1
>>>
>>>Enrique
>>
>>Hi Enrique,
>>I pasted the above into H732 with tablebases turned on.
>>0 seconds, announces mate in 74, and thought/searched no further.
>>
>>In Analysis mode I keep pushing MoveNow to watch the mate materialise.
>>From about move 56 onwards (I went to move 96 enthralled) it never proceeds but
>>moves back & forth announcing #18, #18, #17, #17, #18, #18, #17 ........
>>All found in no time, but no further searching goes on.
>>
>>Maybe H732 would finally move on to the actual mate in an attempt to avoid 3fold
>>repetition.
>
>I believe that the problem is that hiarcs732 needs all the tablebases to finish
>the mate and if it has not all the tablebases it does not know that promoting
>the pawn is a winning move.
>
>The CD does not have all the tablebases and hiarcs does not understand that
>it does not get closer to mate.
>
>It is interesting to know how many rating points hiarcs lost in games because of
>this problem.
>
>I guess that not much because in many cases the opponent resigned before the
>mistake happened.
>
>This problem was also a problem of crafty but Bob fixed it.
>
>Uri

Thank's Uri, that's sensible & logical, & now I don't have to worry any more.

I'm gonna stick the F5 "talk CD" back in, my games very rarely involve
tablebases, and I like being insulted in that sultry German accent.

Ta,
Tina



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.