Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Extending Checks

Author: Stuart Cracraft

Date: 07:04:58 09/11/04

Go up one level in this thread


On September 11, 2004 at 05:34:37, Richard Pijl wrote:

>On September 10, 2004 at 21:35:58, Stuart Cracraft wrote:
>
>>I read, somewhere, and I forget who, about
>>if 1 legal move, extend 2 ply,
>>2 or more legal moves, then 1 ply.
>>Anyone have any stats on the effects
>>on play of the above instead of
>>always extend 1 legal move. Does it
>>blow up?
>
>I do this in two steps: extend 1 move for giving check, extend 1 move for a
>single reply, so in fact that is extending 2 moves on the single reply on check.
>Because I've spread it out over two moves, no other extensions can fire during
>these two moves (the move giving check, and the move out of check). If you
>extend 2 ply on one move you might extend the next move for e.g. recaptures or
>threats. I also limit the extensions when the tree is getting too deep.
>>
>>How do people get around the cost
>>of determining that there is only
>>1 legal move?
>>
>>For me, that's an expensive operation
>>involving usually dozens of makemove/unmakemove's
>>with a test to see if the king is attacked,
>>at every single node, before doing the search
>>of the 1 move with the increased depth.
>
>It is not really expensive. In a semi-legal move generator you will have to
>iterate over every move in the regular search loop anyway. When in check, I loop
>over every move in the movelist and remove the illegal ones, and determine the
>number of legal replies. When I start the search loop only the legal moves are
>left, that are usually not many.
>Besides determining the number of replies, some other thinks can be done as well
>while determining the number of replies (e.g. ETC, or perhaps also repetition
>checks (when beta < 0 so a repetition would give a beta cutoff))
>Richard.
>
>>Only rarely is it just 1 legal move to get
>>out of check. But the determination of that
>>is not rare. It has to be done for every
>>sweep of the moves at each node.
>>
>
>>Thanks,
>>
>>Stuart

Thanks -- I found your message very well written and understandable.





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.