Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The secrets of Rebel

Author: Uri Blass

Date: 13:37:52 12/30/02

Go up one level in this thread


On December 30, 2002 at 16:26:31, Russell Reagan wrote:

>On December 30, 2002 at 03:17:38, Martin Giepmans wrote:
>
>>Imagine a programmer who has worked for years to improve his
>>engine. He has discovered many ideas that Ed also discovered,
>>or similar ones. These ideas gave his engine an edge and
>>of course he has never revealed them to anyone.
>>
>>Then, one day he logges on to CCC and ...
>>How would this programmer feel?
>>Not too happy, I guess!
>
>The problem here is not with the Ed Schröders of the community. The problem is
>the large portion who don't share their ideas. If you don't share your idea, you
>get to think you're the only one doing it and believe that you have a "secret"
>advantage. The down side is that everyone independently solves the same problem
>and almost everyone wastes their time. If everyone shared their ideas, it would
>probably be ridiculously suprising how much farther advanced we would be. But it
>doesn't work if only one or two share their ideas. It's not a coincidence that
>there are more and more Crafty level engines on the scene than there used to be.
>But how many surpass Crafty convincingly? Not many. That's no coincidence
>either. Before long we will see more engines closer to Rebel's level, but few
>will pass it, because that requires thinking on your own.
>
>A different, but similar, situation happened this past semester in a unix class
>I was taking. We had labs to do involving solving various problems, writing
>scripts to accomplish certain tasks, and so on. My professor was an old school
>unix guy, and so we did a lot of the projects either in pairs or as a class, and
>I was amazed at how differently people solved the same problem. Many of the
>approaches people took are methods I never would have considered on my own. The
>result is that now I can look at a problem from many different angles and
>generally find a better solution. That wouldn't have happened if we never shared
>ideas.

There are common ideas but different people think about different ideas and
there are cases that one person can know that other programs do not use some
idea.

One example:
chessmaster solve mates faster than other programs and I am not talking about a
small factor that can be explained by optimization so the programmer can know
that his program use some different algorithm that other programmers do not use.

Uri



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.