Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: WMCCC clock handling

Author: Christophe Theron

Date: 14:18:59 10/14/97

Go up one level in this thread



I don't see why your opponent could demand your clock to be
accurate. I plan to use a simple clock sheme: my program knows the time
controls, and it knows the time I take to manipulate (I shall use 10
seconds for this tournament). This 10 seconds cushion is enough, I know
it
by experience, to take care of clock and/or operator problems, because
my
real operating time is about 4-5 seconds per move.

I have no way to adjust the program's elapsed time, so the only thing I
can
do is correct the time control. So if I see I'm really going to fall on
time control (due to abnormally long operating time), I decrease the
time
control. For example on a 1h game, I could correct it and say 57mn time
control, because of a 3mn time error. I never adjust in the other way,
that
is, giving more time to the program because I did the moves so quickly.

I do not display the clocks (except an estimation of the remaining time
till next time control, displayed before the start of the search)
because
this version is prototype and has no graphical interface. Anyway, the
remaining time displayed is enough information. I took part of many
computer or human tournaments and never had any problem.

And it's TOO LATE for me to modify my program to
1) display time controls for both side
2) implement a way to correct the elapsed/remaining time

If the tournament rules are changed this way, it's like throwing me out
of
this event.

One other thing: do you realize what it is to have your opponent
demanding
a clock adjustment when you are in deep time problems in a tight ending?

My opinion: if you want a simple rule, here it is: NO TIME ADJUSTMENT.
Everybody is responsible for giving its program a good cushion, and not
falling asleep in the middle of the game. This could make the
tournament director's life easier, too. And no cheating.

In special cases, there could be exceptions, but there must be a very
good reason for this (power supply interrupted for example, or
earthcake). Simply being in time crisis because of poor operator time is
not a good reason. In that case, your opponent wins. Period.


Regards

	Christophe Theron



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.