Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The reason that gandalf is a good program for analysis

Author: Amir Ban

Date: 12:26:44 11/23/00

Go up one level in this thread


On November 23, 2000 at 12:33:12, Peter Fendrich wrote:

>On November 22, 2000 at 12:10:18, Uri Blass wrote:
>
>>I found from my experience that Gandalf does not do preprocessing.
>>
>>I asked the programmer about it to know if this is always the case(because I
>>suspected that maybe there are some rare cases when it is a preprocessor) and he
>>told me that Gandalf does not care about the root position.
>>
>>I think that it is very good for analysis because I can generate a tree and use
>>the scores of the program to decide about the move when I cannot do it with
>>other programs because the score is not made from the same point of view.
>>
>>I remember also that Amir claimed that Junior does a little bit of preprocessing
>>but he does not like it and thinks to change it
>>
>>I am interested to know about the behaviour of other top programs.
>>I think that it is possible to divide top program by the reply to the question
>>if they do preprocessing.
>>
>>I already know that Rebel and Tiger do preprocessing and I also know that
>>shredder5 is a preprocessor by my definition(I do not have it but I asked the
>>programmer about it and he told me that shredder5 is not a preprocessor by the
>>average definition of it but but there is some preprocessing done at the root).
>>I also know that crafty is a preprocessor(not by the average definition that I
>>do not know).
>>
>>I am also interested to know what is the average definition.
>>My definition is that if there is some preprocessing done at the root then the
>>program is a preprocessor.
>>
>>Uri
>
>There are always problems to find clear and simple definitions.
>I can see a problem even with this definition.
>
>Suppose that the program stores a table saying that knights are generally doing
>worse in the corners than in the center of the board and this table is not
>changed during the game. This is not preprocessing by your definition.
>
>Suppose that the program changes that table once in the root when we reach the
>endgame. This is preprocessing by your definition. Let's say that you ignore
>this case because the table is changed only once. Then you have the case that
>all pieces have a similar table each that changes only once when endgame is
>reached. Is that preprocessing? If not, suppose that you change these tables
>whenever the game reaches different stages in the game (opening, middle game,
>endgame etc).
>Is this preprocessing?
>

It is preprocessing, and it is not minor at all. It will mix up your program's
behavior in stage transitions. It will swap pieces to reach a stage, or avoid
doing that, all for the wrong reasons. And when it sleepwalks into the next
stage, the program will wake up with a surprising new eval.

Amir


>And so on...
>
>In my program Terra, there are some bitboard tables working like that but I
>don't consider Terra being a preprocessor.
>
>//Peter



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.