Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty 17-10 v Fritz 6a. Nunn 1.....1 min, 2 min, 3 min, 5 min

Author: blass uri

Date: 23:45:14 04/23/00

Go up one level in this thread


On April 24, 2000 at 00:42:47, Christophe Theron wrote:
>
>
>I can defend opposite ideas (without saying I 100% believe in them, it's just to
>explain why the situation is not so simple):
>
>
>Searching deeper let's you understand some positional concepts, even if you do
>not have them in your evaluation function.
>
>It is also possible to think that a program that has more knowledge built in its
>evaluation function will understand things at blitz, things that its opponent
>will never understand (because it would understand only if it could search
>deeper).

I thought about knowledge that searching one ply or two plies cannot detect.

I am not sure if Crafty knows more this kind of knowledge

I test Crafty against Fritz5(16 bit) in the nunn match and I do not see
improvement at longer time control and the results now are small advantage for
Fritz at all time control that I tested

11:9(5 minutes per game)
11:9(5+5 blitz)
6:3(15+15 blitz)

Maybe I need more games to see the improvement.

Inspite of these results there are clearly cases when I see that programs have
stupid evaluation when they cannot practically see it by search because they
cannot see 60 plies forward.

I think that knowing this kind of knowledge is important in long time control.



>
>So I could argue that more accurate knowledge will be much more important in
>blitz games.
>
>But these arguments lead to nothing.
>
>
>I think we can say that a given amount of knowledge (without taking search into
>account) gives you X elo points. Then the search (actually the depth of the
>search) gives you Y additional elo points.
>
>For a given program, X is fixed and does not depend of the time control. Y
>depends on the speed of the computer and the search time.
>
>So what's left to discuss is "how much do you gain from the search when you
>increase the search time".
>
>Actually this is very closely related to what we call the "effective branching
>factor".

I do not think that the "effective branching factor" is important because you
can increase the "effective branching factor" by prunning and you are not going
to have a better program.

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.