Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Further Announcement: Game Play Modified

Author: Graham Laight

Date: 12:19:14 02/01/03

Go up one level in this thread


On February 01, 2003 at 09:48:25, Russell Reagan wrote:

>On February 01, 2003 at 06:59:08, Uri Blass wrote:
>
>>1)Does it learn from previous games or can a player win and play another game
>>and win by the same moves(Note that it is possible also to prevent players to
>>repeat previous game by generating some random noise)?
>
>It does not learn from previous games. I believe that the first 6 decisions are

True. If you want more learning effect, go for a longer game!

>from a random number generator, and that affects the rest of the game. Also
>whether you choose correctly or not affects the rest of the game, so you
>shouldn't be able to win by the same moves unless you are very lucky and get the
>same 6 decisions on the first 6 moves from the random number generator.

Even then, I have built in a little bit of random noise.

>>2)Does the computer use the time that the human thinks on his moves on it's
>>decisions?

No - it only "thinks" when you click on a button. We're not playing chess here.
A lot of processing is done, and JavaScript is about 100x slower than well
optimised compiled C (it is interpreted by your web browser rather than
compiled) - but you still get several million lines of code per second executed,
and that seems to be sufficient for the pattern-matching work being done here.

>It is pretty quick. You can look at the javascript source code by doing a 'view
>source' in your web browser.

Agsin - true  :)

-g



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.