Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Analysing while retracting moves

Author: Dann Corbit

Date: 18:28:17 11/26/01

Go up one level in this thread


On November 26, 2001 at 19:34:48, Tom Kerrigan wrote:

>On November 26, 2001 at 15:12:59, Robert Hyatt wrote:
>
>>If it were one move, you would be right.  But it typically isn't.  It is
>>_many_ moves.  Say 10 moves or 20 plies.  And the score doesn't get backed up
>>the the _real_ problem move at all...
>
>Ah, here's where we fundamentally disagree. Do you have any evidence that
>"typically" mistakes can only be found with 20+ plies of search? 20 sounds to me
>like a number that was pulled from somewhere unpleasant.

It's pretty obvious that is it a ball-park figure, because of the modifier 'Say'

I expect that it will vary wildly with game phase.  You are not going to get 20
plies deep on the second move of the game, and if you only get 20 plies deep
when you have 7 chessmen left on the board, then something is very wrong.

Of course, how you have null-move pruning operating will make a big difference
too.  If you have null-move turned on with R=3 and with 7 chessmen left, you
might search very deeply and still miss important tactical gaffes.

In short:
1.  The sort of errors missed
2.  The probability of an error of a given type being missed at a given depth
3.  The probability that a move is ideal (IOW, gives the same game-theoretical
outcome as the best possible move)

will definitely be a function of the exact search method used.



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.