Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Chessbase

Author: Mike S.

Date: 13:00:15 03/12/03

Go up one level in this thread


On March 12, 2003 at 03:55:34, enrico carrisco wrote:

>(...)
>
>One effective way is simply viewing times per move using the "smoves" command.
>Book moves that should be nearly automatic for a player of a certain "level"
>should not take 3-5 sec consistently.  Also, simple piece trades and perpetual
>check situations also should not take 3-5 sec.

This would effectively lead to wrong accusations.

Don't you ever take time during a seemingly obvious exchange sequence, looking
for traps, sacrifice ideas, choosing between two options to take back, etc.?

For a player of a certain level, there are NO automatic moves. Maybe he'll see
quickly if there is only one option which make sense, but certainly no automatic
moves. In perpetual check situations for example, on might take some time to
see, if it 's really forced or how can I escape the perpetual, as long as there
is more than one legal move...

Also, during openings I may take some time considering which theory variant I
like to play today (or what it is at all, if I don't know that opening very
well). So that neither indicates comp use.

Furthermore, using a nearly constant thinking time *always* doesn't prove at all
that somebody is cheating (more the opposite IMO, as programs are very flexibel
with their time usage).

That detection idea is faulty... The first priority should always be, to *avoid
wrong accusations* caused by silly ideas of how human play "should look like
normally". The bandwith of human behaviour in chess is very wide after all.

Regards,
M.Scheidl



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.