Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: History heuristic

Author: Tord Romstad

Date: 02:43:07 03/19/04

Go up one level in this thread


On March 19, 2004 at 02:17:28, Sergei S. Markoff wrote:

>Hello!
>
>>Interesting.  I've always thought that checks should be given some form of
>>priority in the move ordering list.  Of course they will be expensive to
>>calculate but could be well worthwhile.  Please do let us know the results.
>
>Ok, seems to be good. I'm trying now to implement a lot of knowledge in move
>orderering -- it's my favorite way to integrate knowledge anywhere :) The checks
>with SEE>=0 seems to be good in middlegame. Also pawn pushes with SEE>=0 after
>castling and excluding pawns of king shelter.

I am not sure I understand the logic behind the last rule above.  Is there
any reason to believe that safe pawn pushes are better (in general) than
most other moves?

>Also I'm trying to implement some
>attacks info -- "forks" e.t.c. Hint: expensive knowledge can be implemented when
>remaining depth >2*INCPLY or >3*INCPLY e.t.c.

Yes.  It's strange that so few people seem to realize this.  Apparently,
almost everyone uses exactly the same move ordering techniques at all nodes,
regardless of the remaining depth.  It makes sense to use much more expensive
move ordering knoledge when the remaining depth is big.  If the expected
size of the subtree is millions of nodes, it is clearly a good idea to
spend a lot of effort to make sure the best moves are searched first.

>For example SEE for non-capturing moves.

I do this everywhere in the main search, even close to the leaves.

>In endgame we must keep in view pawn pushes when SEE>=0 and target piece
>is defended or opponent's king is out of square. Checks in endgame usually good
>if 1) it also attacks other piece with threat to capture it with SEE>0 2) there
>are no evasion moves that going king to the center. There are a lot of knowledge
>that can be used here. I will report community about results.
>
>>PS I ask this every time - but when will the UCI version of SmarThink be
>>completed? I wait with anticipation
>
>:) It will be. But I'm waiting for ST that will be significantly stronger than
>Ruffian 1.05.

Please don't wait so long! :-)

I once asked you whether there was any chance that you would port your
engine to Linux and/or Mac OS X, and you replied that you would make an
attempt as soon as the next version was finished.

Version 0.17a has the perfect strength as a sparring partner for my own
engine.  The last time I tried, my engine lost a 100-game match by 47.5-52.5.
An engine significantly stronger than Ruffian would just be too depressing
to play against, I'm afraid.  :-)

Tord



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.