Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Amateur programs vs Commercial ones

Author: Steffen Jakob

Date: 23:24:13 03/28/00

Go up one level in this thread


Hi Christophe!

That was a very good posting! I had a deja-vu in almost each of your examples!
E.g. in the first round of the CCT1 tournament Hossa lost his game from Nimzo in
a won position because of a permanent brain bug which caused Hossa to move
immediately although he didn't have any information from the PB phase and so
made a move from a very short search which lost the game at one. After the game
I first thought: "Well, my program had a won position and I prob. can find a
workaround for this stupid problem in half an hour. If I only had done this
before my engine would have won against one of the strongest program on earth in
my first serious tournament!". I somehow felt like the true winner of this game
but then realized that exactly the reason why my program lost this game is the
most important difference between commercials and amateurs. Indeed I implemented
this workaround in a very short time, but it didnt solve all problems. It turned
out that I had a elementary design flaw which I fixed later... which took
several days. If I say "days" I mean days where I could use some hours of my
free time.

For me as an amateur programmer it is more interesting to spend the time I have
to work on Hossa to implement new things... to realize my own ideas... it's not
so much fun to cleanup code, and being an amateur the whole thing is basically
motivated by having fun. So I have a lot of code with lots of bugs. Very often
Hossa would play stronger if he didn't have this code at all! Your list also
gives an idea how complex the area of computer chess programming is. You have to
consider so many things to get a competitive engine which is an extremely time
intensive issue. After the first CCT1 round I discussed the difference of
amateurs and commercials with Chrilly Donninger and he had the same opinion as
you have: the main difference is stability.

This is the main reason why I froze my current version of Hossa last week and
started to rewrite my program giving myself as much time as it takes to do
everything as carefully as possible. I am very curious how the playing strength
will change :-)

Best wishes,
Steffen.




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.