Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CCT5 planning started

Author: Russell Reagan

Date: 21:13:39 11/11/02

Go up one level in this thread


On November 11, 2002 at 16:07:15, Dieter Buerssner wrote:

>But somebody, who
>really wanted to cheat could do it nevertheless. I believe (but I have neither
>the will, nor the intention to prove it) that I could write an engine version,
>where the operator totally takes over the time management while playing under
>WB. More sophisticated cheating may be possible, by running 2 computers, one
>with n-best mode, and one normally. One would even have a PV ready for kibitzing
>all the time.

You'd be suprised how far some people would go. In online 3D video games
cheating is out of control. People rewrite video drivers so that they can
control the transparency of walls and see enemies coming. The enemy never has a
chance. Aimbots are also popular. It does the 3D vector math and moves your
mouse for you so you always have perfect aim.

I suspect that when kibitzing a PV becomes a requirement, there will be someone
who will take advantage of that and pull out the move to ponder and essentially
play with double the time of his opponent since he will almost always ponder the
right move. You could take the score in the kibitzed line, and if the score is
significantly different than what your engine sees, spend more time searching.
If your opponent is stronger than you (like my engine vs. Fritz, for example) I
could just play whatever move Fritz thinks is best, and let it do the work for
me. Right now, since we are only starting to make this requirement, it probably
won't be a significant problem. But cheaters are clever animals. It won't be
long.



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.