Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fruit, trying to answer various missed posts

Author: Fabien Letouzey

Date: 07:25:34 03/15/04

Go up one level in this thread



Uri,

>I think that fruit can be improved by better weights(probably by increasing the
>values for knight and bishop to be better than rook and pawn)

Thanks for the detailed explanation.  This is one of the numerous problems I am
well aware of, it appears very often :)
As you have guessed, I use only basic material values (as yet).  Fruit does not
even try to avoid exchanges when in a bad position ...

>I also think that checks in the first ply of the qsearch together with
>R=3 may be superior to crafty.

I started adding checks in first ply of QS "after null move", which I consider a
must have (or some mate detection).  Then I switched it to "always" because of
the lack of tactical extensions (recapture, etc ...).  I did not test it, but it
seemed like the "safe" thing to do.

>It may be also interesting for me if you use hash in the qsearch and if yes how
>do you do it(today I do not hash the qsearch).

Not yet, but I will give it a try.  I need to think carefully about it because
the code has to be adapted to the quiescence search "selectivity".  I expect my
quiescence to get more and more complex in time, so at some point hashing will
probably become beneficial.  Also this might help me putting checks after the
1st ply as well.

Fabien.




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.