Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: triangular array vs. walking the hash table

Author: Michael Henderson

Date: 21:37:19 09/20/04

Go up one level in this thread


On September 21, 2004 at 00:15:42, Stuart Cracraft wrote:

>On September 20, 2004 at 00:56:26, Michael Henderson wrote:
>
>>>>1. When you make a two-square pawn move, and the opponent cannot make an ep
>>>>capture, do you hash-xor the "target" ep square in the hash table?
>>>
>>>Currently no special handling is done for the hash table in regards to
>>>enpassant. That information is not recorded in any way.
>>
>>What I meant to say was do you hash in any ep target stuff into the hash key
>>when you make a two-square pawn move, whether opponent can capture ep or not?
>>
>>Michael
>
>I don't do this and hadn't considered it.
>
>What side effects would you expect from the absence of this?

consider two positions: A, and B.  Positions A and B are identical in every
respect but en-passant availability.  Position A has a white pawn at e4 and a
black pawn at d4. In position B white has just played e2e4 and black has a pawn
at d4 -- so en passant capture.

Now Search():
Then black comes to position A --does a search, whatever-- and finally hash
stores.

At some other node in the tree, black comes to position B and uses position A
from the hash table to cutoff that node.  That does not take into account any ep
captures you have at position B, so your search can screw up.

Michael



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.