Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Alpha/Beta adjustments via the hashtable or why Fritz 7 has messy PV's

Author: Uri Blass

Date: 02:28:39 12/03/01

Go up one level in this thread


On December 03, 2001 at 04:34:03, Gian-Carlo Pascutto wrote:

>On December 03, 2001 at 03:55:56, Uri Blass wrote:
>
>>I do not understand it and I will be happy to see an example of a tree
>>that leads to pruning the main line.
>
>It should happen in any branch where there was some form
>of forward pruning leading to a lower bound on the score,
>and where the branch was researched later on.
>
>There are probably other cases, but this is the obvious
>example.

I still prefer to see an example to understand and it means a chess position
and the nodes that the program searches before showing the wrong line.

I do not understand much about the search of chess programs.
The following is a summery of what I know about it:

I understand the reason for using two bounds alpha beta by the following example
that I invented

line 1.e4 score 0.1 pawns for white

line 1.d4 Nf6 score 0.12 pawns for white

line 1.d4 d5
I can have 2 bounds 0.1 and 0.12.
If the score for white is less than 0.1 then I know that 1.d4 is illogical(e4 is
better and the difference is not important)

If the score for white is more than 0.12 then I know that d5 is illogical(Nf6 is
better)

If the score for white is between 0.1 and 0.12 then the exact score is important
for me and I need to calculate it.

I also understand the idea of null move pruning that means that you prune moves
with no threat when finding if there is a threat is done by searching to reduced
depth.

>
>>I do not agree
>>If I analyze a position for many hours for depth 19 and the program found an
>>interesting sacrifice then I do not want to analyze also the position after the
>>sacrifice for many hours to get depth 18 in order to understand the reason for
>>the sacrifice.
>
>Play out the move. Enough data should still be in the hashtables
>that 18 ply can be reached fairly fast.

In this case I think that it may be a good idea to do a search for reduced depth
for the opponent before pondering because in this way Fritz is not going to
ponder for the wrong move.

Uri



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.