Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Further testings with Gambit, joking a bit with the settings

Author: Fabio Barrettone

Date: 08:32:42 05/03/01

Go up one level in this thread


On May 03, 2001 at 10:07:18, Uri Blass wrote:

>I understand reducing SS because if increasing SS is counterproductive then
>reducing SS may be a good idea but I do not understand why to set More Gambit to
>1.
>
>Uri

Following what Christophe said less Selectivity = more speed and more pruning,
and I've tried this. Giving GT 1 minute exactly for move in a starting position
without book with default selectivity, it will finish ply 13 and play, but with
selectivity = 3 it reach the second move on ply 14 before moving (this in my
computer naturally).

So, watching the style "aggressive" play I noticed that in almost all the cases
it needed only a bit more to think (and in many cases only very very little, as
for example the game I posted last time here against Junior, where only a move
more in the search and instead of losing it should have won) to play much better
and find winning ideas. In fact with style aggressive Gambit Tiger II is much
more concerned about King's attack and open files and resemble more the version
1.0, but with still more knowledge. When you attack the opposite King almost all
the times is more useful to look at three, four key moves to their fullness than
to look at more move variations that one can play.

This is not the case (or less) with style normal, that seems to be a bit 'lazy'
and sometimes enter in a strange "phantom cicle", so to say, in the sense that
it feels itself quite confortable when in fact it isn't, and it is less willing
to attack (it resembles the figurative mexican :-)); so more speed doesn't
improve too much the engine than less pruning to find alternative moves.

Sincerely,
Fabio.



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.