Author: Dann Corbit
Date: 13:30:23 01/12/01
Go up one level in this thread
On January 11, 2001 at 09:15:04, Steffen Jakob wrote: >On January 11, 2001 at 09:07:15, Jouni Uski wrote: > >>B f F - R a n g l i s t e A m a t e u r e: >> >>Blitz/Aktivschach >>insgesamt > 31.000 Partien >> >>(Cel.525; Book: Noomen-Select, SEE, Nunn1/2-Test) >> >>Program Elo + - Games >> >>Crafty 17.11 : 2488 17 14 1514 >>Crafty 16.6 : 2466 30 38 332 >>Crafty 17.13 : 2456 35 41 257 >>Crafty 17.14 : 2445 41 33 286 >>Crafty 16.15 : 2437 36 35 289 >>Crafty 15.20 : 2423 58 50 130 >> >>Question: Has there been real/significant progress after this statistics?! > >About one year ago I introduced a sophisticated queen mobility eval term in >Hossa. This term is very expensive. Now I see games from time to time where >Hossa captures the opponents queen even in the middle of the board or gives away >a pawn to keep the opponents queen unmobile for a long time. On the other hand >Hossa got slower and overlooks tactical shots more often than before that >change. I didn't notice any significant rating change when I added this feature. >Would you say that I made any progress here? ;-) Two different goals appear: 1. Play good chess. 2. Win lots of games. The two goals are not necessarily identical. With greater CPU speed, it may become more valuable (or more of a hindrance!). Sometimes, I think both kinds of approaches have great value. Having people try different things makes the chess played by the programs more interesting.
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.