Author: Christophe Theron
Date: 18:12:13 06/04/98
Go up one level in this thread
On June 03, 1998 at 17:05:32, odell hall wrote:
>I am curious do programmers know how to defeat there own programs. It
>would seem reasonable since they would have specific knowledge about the
>strengths and weakness of their own programs. For instance i would love
>schroeder to tell me the best way to beat rebel. Or is this question
>rediculous?? Can I please get some comments on this from various
>programmers
I go to a "human" club every Saturday to test Tiger.
Last Saturday, for the first time, I played myself against my program. A
friend of mine, who has tested Tiger against other programs, operated
the "monster".
Well... The "monster" ran on a 386sx 20MHz laptop. 20MHz is already very
slow, but you also have to consider that Tiger is a 32 bits program, and
the 386sx uses a 16 bits path to communicate with the memory.
On this computer, Tiger reaches the incredible speed of 700 nodes per
second!!!
I was completely crunched in less than 10 moves. I did not try a second
game, as everybody around was laughing loud at me. Usually these guys
have a hard time to beat Tiger on this computer. Tiger generally plays
at 15 min per game at the club.
My loss was no exception. I have never beaten Tiger (on this computer)
since a long time.
As a chess programmer, I have the great pleasure to feel proud of me
even if I loose a game... against my own program.
The human players took some time (maybe 2 years) to understand how to
beat my program.
At the begining they used to fight just to go (and win) the endgame, but
it's now difficult due to strong improvements in this field.
Now they have found something else: close the position, bring all your
pieces in front of Tiger's castle, then make some sacrifice to open the
fortress.
On such a slow computer, Tiger has problems with this. But sometimes the
sacrifice is wrong, and my program can get a draw with some tactical
tricks. I'm now considering to write a special "anti human mode":
1) penalize closed positions
2) detect slow attacks against castle
This should not be very difficult.
Christophe
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.