search

master2be1 / pychess

Automatically exported from code.google.com/p/pychess
GNU General Public License v3.0
0 stars 0 forks source link

Pychess+Darwin #57

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
The pychess eval function is important in many ways.

It contains a lot of hardcoded point values for different things.

* We took the engine out of context, and made it add a little random to all
pointvalues.
* We made it play a lot of times against it self
* Everytime the engine won, the values was frozen, and in the next game,
the two engines evalfunctions would be the ones of the winning engine + a
little random.

Original issue reported on code.google.com by lobais on 25 Oct 2006 at 5:19

GoogleCodeExporter commented 9 years ago

Original comment by lobais on 31 Oct 2006 at 7:51

GoogleCodeExporter commented 9 years ago 
So, what needs to be done? What is the issue?

Original comment by G.nius...@gmail.com on 1 Jul 2013 at 5:31

GoogleCodeExporter commented 9 years ago 
The evaluation function of a chess engine has several component. They has some 
constant values which need to be work together to produce a good final value. 
Tuning these constant manually is a hopeless task, so developers need some 
method to find the optimal values. One method can be some "darwin-like" system, 
Thomas was talking about. Last time I read another promising one here: 
http://remi.coulom.free.fr/CLOP/

Original comment by gbtami on 1 Jul 2013 at 7:47