Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Human vs. Chess Engine EMail Chess Match

Author: Stephen Ham

Date: 15:23:11 11/30/99

Go up one level in this thread


Dear Bruce,

Those are good questions, one of which is answered at the URL's listed. In
short, there will be NO HUMAN INTERFERENCE with the computer's play. We intend
for this to be a "clean" experiment to test pure machine versus human play at
the higher levels of correspondence play. In that respect, ONLY the moves
selected by the computer will be chosen.

The question regarding time allotment is less clear, even to me, and thus
potentially a point for controversy. We have no set reflection period since we
are both (Don Maddox and I) busy businessmen and my only e-mail access is on my
computer here at the office. Thus we can't transmit moves to each other on
weekends since I can't receive them during that period. Still, it's my
understanding that the computer/chess engine will calculate its move starting
from the point Don Maddox is done with the computer (i.e. 5PM on week days)
until he needs it again (i.e. 8AM on weekdays). On Friday night, it could run
continously until Monday since we are not counting weekends in the reflection
time. Hence it seems the chess engines should have approximately 15 hours of
available computing time per 24 hour period on weekdays alone. As to the
specific logistics, this is up to Mr. Don Maddox, President of ChessBase USA.
Since I believe that he has a vested interest in having his chess engines
perform well, I trust he will not short them of their available computing time.
Don will report the time used per chess engine/move so let's see what happens. I
think we can safely conclude though that the chess engines will have plenty of
available reflection time. Meanwhile, you will see that I move fairly rapidly.
Again, although no set reflection limit exists, you will see that I play well
with the standard ICCF time parameters of 10 moves/30 days. Thus I think we
should have a fair comparison of human versus computer/chess engine in
correspondence chess, at least the best that can be obtained from the present
competitors given the logistics involved.

I hope this arrangement proves satisfactory to you viewers.

Sincerely,

Steve Ham



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.