Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Questions about Lazy Eval.

Author: Ryan B.

Date: 14:34:33 04/11/05

Go up one level in this thread


On April 10, 2005 at 23:44:59, Scott Gasch wrote:

>On April 10, 2005 at 17:38:11, Ryan B. wrote:
>
>>Most programs I’m sure have some form of a bad trade value.  On what side of the
>>Lazy Eval should the Bad trade value be factored in?  Is Lazy eval really safe
>>to do before king safety and board control are factored in?  In my program board
>>control can be worth up to 3.2 pawns and king safety up to 6 pawns.  Should I
>>make the margin large for Lazy eval or just use it to save time on all the small
>>tactical adjustments?
>>
>>-Ryan
>
>If you use too large of a margin in lazy eval then it will never fire.  It's a
>balancing act.  One trick is to make sure you do the large positional terms
>early on in your eval.  Another trick is to have more than one exit point from
>eval; at each exit point you know more about the position than you did at the
>last one so that at each exit point you can use a more and more narrow margin.
>One final (somewhat obvious) point  related to this stuff: don't do _anything_
>you don't have to until after your first lazy decision.  If you can put off the
>initialization of variables, the construction of an attack table, etc... do so.
>
>Good luck,
>Scott

Thank you very much.  I really like the idea of having more than one exit point.
   Also the idea of calculating the largest bonuses first may really help me
because until now I have been calculating all tactical knowledge first then
positional knowledge.

-Ryan



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.