Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: To bitboard or not to bitboard?

Author: Uri Blass

Date: 05:03:48 08/31/03

Go up one level in this thread


On August 31, 2003 at 06:34:54, Tord Romstad wrote:

>On August 30, 2003 at 16:56:07, Uri Blass wrote:
>
>>On August 30, 2003 at 14:37:44, Tord Romstad wrote:
>>
>>>Today, my board representation consists of a 256-byte array of squares
>>>(visualised as a 16x16 board with the real board in the left half of
>>>the middle 8 ranks) and a Phalanx-style piece list.  At every node in
>>>the tree, I generate "attack tables" similar to those found in Rebel,
>>>Phalanx and Diep (the tables are not updated incrementally).
>>
>>How do you know the attack tables of diep?
>
>Vincent has described them several times in this forum before.
>
>Tord

I remember now that he posted something about counting number of attackers but I
did not consider it as important.

I have other questions.

You say that the tables are not updated incrementally.
I also remember that your evaluation is the most expensive part of your program.

1)Do you update your attack tables after every move for things like better order
of moves?
2)Do you consider the time of updating your attack tables as part of the time of
your evaluation?

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.