Author: Fabien Letouzey
Date: 04:24:30 03/11/04
Go up one level in this thread
On March 11, 2004 at 06:12:14, Djordje Vidanovic wrote: >Hello Fabien, >Fruit is _really_ promising. However, I'd think twice about adding evals before >cleaning up other stuff and adding tablebase access. May be your move generator >needs some polishing first. And do give Fruit a bit of book. Illiterate as it >is, it is doing quite well! Its style is very refreshing, fun to watch. A bit of basic endgame knowledge is sure not going to replace TBs, but I think it will make their (TBs) addition much less urgent. I often see the same basic endgames come again and again and Fruit does not know they are dead draw. Those can be encoded without TBs, and I will see if more knowledge is really necessary. What do you mean with "move generator" problems? Do you mean it plays really bad moves at time? Moves that no other engine would make? The plan is to deliver Fruit 1.0 "as is" ASAP (after basic testing). After that I have so much to add that I will have to choose carefully. Book is not a priority in the code. Maybe users can create books for Fruit, to be used in GUIs. I would gain time to make other additions instead. Fabien.
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.