Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rebel trick one margin questions

Author: Uri Blass

Date: 03:33:11 01/04/03

Go up one level in this thread


On January 04, 2003 at 05:34:23, Ed Schröder wrote:

>On January 02, 2003 at 08:21:51, Uri Blass wrote:
>
>>This trick is from Rebel programming stuff:
>>
>>MARGIN = 3.00                              // 3 pawns safe-guard value
>>  MARGIN + highest hanging piece value       // Queen=900, Rook=500 etc.
>>  MARGIN + 9.00 when own_king_was_in_check_before_make_move
>>  MARGIN + 6.00 when the opponent can promote the next ply
>>  if (SCORE-MARGIN > BETA)                   -> return TRUE
>>  else return FALSE
>
>>1)Why score-margin>beta?
>
>Because it works.

The point is that I already prune when score>beta so pruning when
score-margin>beta save no nodes for me when margin is positive and change
nothing.

score+margin<alpha fit the idea to prune when there is probably no capture to
change the score.

when the score of the side to move is bigger than beta qsearch usually return
beta.

I will give an example

suppose that I search the following position and I go to the qsearch

[D]rnbqkb1r/pppppppp/8/7n/4PP2/8/PPPP2PP/RNB1KBNR b KQkq f3 0 3

suppose beta =0

I say:score of the side to move is bigger than beta(black has a big advantage
and it is black to move) so I do not need to search capturing f4.



I will try the idea with different values but for me the idea means
score+margin<alpha and not score-margin>beta.

I already change my move generator(otherwise finding the biggest capture takes
me significant time inspite of the fact that movei generates all the legal moves
after every make move).

changing my move generator also made movei 5% faster.

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.