Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: For Dann Corbit ---or

Author: Dann Corbit

Date: 08:55:30 04/02/99

Go up one level in this thread


On April 01, 1999 at 21:38:36, Phil Dixon wrote:
[snip]
>I have been thinking along the lines of depth of search, position evaluations,
>and adding/changing cell values.  Asking what if? questions. Could a person save
>hours of computer time analysis with the "what if" question?  Statistical
>analysis of x piece on y square number of times.  My intention would be to
>scrutinize the openings with the idea of deriving advantages from the openings.
>Isn't that where the game begins?  I would appreciate your ideas, positive or
>negative.  I am just in the thinking stage and will admit that I have a LOT to
>learn.
OK, I see what you are after.  I plan something very similar using SQL and a
move generator.  Right now, we have about 300K positions analyzed at 12 minutes
or more of PII 300 MHz CPU time equivalent.  We also have about 30 million
positions analyzed at 2 seconds.  This includes every move ever played during
any part of the game five or more times and every position taken (no matter how
many times) out to 30 plies.  I plan to use SQL to refine the pathways and
develop pv's with 30 or more full moves in them.  You might read the project
Ghost games section of the C.A.P. FAQ.



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.