Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: VOCCT 1

Author: Vasik Rajlich

Date: 02:07:01 02/24/05

Go up one level in this thread


On February 23, 2005 at 16:33:46, Anthony Cozzie wrote:

>I'm not sure how to explain this, so I am just going to fumble around and hope I
>can get my point across without being an ass.  I simply don't like operators at
>big tournaments. Having thought for a while, my problem is that an operator
>invests nothing compared to an author.
>
>Zappa contains roughly 13,000 lines of C++ code (not counting Eugene's stuff).
>If I plug that into COCOMO (months = 2.4*KLOC^1.05), I get a development time of
>3 years.  At standard US Goverment contracting rates (60K/yr, multiplier 2.5)
>that equates to $450,000.  Any good chess engine will require a similar amount
>of time, and I'm sure the amount of time I have spent on Zappa is dwarfed by the
>amount Stefan works on Shredder.  SOS, Diep, Fritz, Crafty, Shredder, Junior,
>Hydra: all of them are the product of YEARS of work.  Even "amateur" engines
>like Gothmog or Amateur have a lot of man hours behind them.  The same goes for
>opening book creators.  Necchi, Arturo, Kure, Alterman, and Noomen have all
>worked on their books for years.  So if Arturo & I show up at the WCCC this year
>or the next, the total value of our entry is probably pushing $1M, and this
>would be dwarfed by some of the other contestants.
>
>I say this not to brag about my accomplishments, but to show just how much work
>it is to compete at high levels in computer chess.  It is just like any other
>sport; everyone knows that world class swimmers or tennis stars work on their
>game every day. The only difference is that instead of getting to the gym for
>squats and sprints we hang out at the computer lab improving our move ordering
>:)
>
>An operator shortcuts all that by the simple expedient of purchasing a program
>and opening book for 50$, which requires about 0.001% the effort of developing
>it from scratch.  This has several effects.  First, they simply aren't "in the
>club", and cannot reasonably expect to be.  Secondly, they have none of the
>knowledge gained from the years of work, which makes conversing with them much
>less interesting.  Third, people attach a lot of importance to things they spent
>a lot of time on.  If I win a tournament with Zappa I feel good about my effort,
>if I win using a program I bought chessbase.com for 50$ I feel . . . not much.
>If I participate with Zappa and lose to someone who bought Shredder for 50$, it
>is only natural that I will resent that.
>
>This is why all the big computer chess tournaments have insisted on "authors
>only".  In fact, only the events on the playchess server work differently.  If
>you want to compete as an author, write a program.  It isn't easy, but neither
>is it impossibly hard.  It doesn't take more than a few months to get something
>that can play legal moves.  Having written this essay, I realized that it really
>says almost nothing that isn't in the subject line :)
>
>anthony

Nice post. Ok - it's not really fair to give $$ estimates to what you do for
fun, otherwise my neighbor's garden is _really_ expensive - not to mention the
neighborhood basketball game - but the point is quite right.

Just to add a little bit - we also want CC to be a sport. The issue unique to CC
is that anybody can buy all the top programs, and run their own tournaments. How
are these tournaments different from a WCCC? Ok - the book is different - but is
that all?

To keep the sporting aspect, what we need is a way to point to a small number of
tournaments and say "this is it - this is special and you should follow it".
Author presence is a pretty good way to draw that line.

Vas



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.