Author: Sune Fischer
Date: 13:17:22 09/13/04
Go up one level in this thread
On September 13, 2004 at 15:37:04, Sandro Necchi wrote: >The only way to evaluate the games is to analyse them. One needs to analyse all >games; won, drawn and lost. > >To create a book from a database from the score percentage if you see a move >with 99% out of 100 games it means that it is almost winning for sure or that >finally the opponent found the way to kill that move? > >The only way to find out is to check the moves and find out. > >When you analyse a game you should be able to answer the following questions: > >1. Did I won because I played better or because the opponent blundered? >2. Could I play better? >3. Where? >4. Why did I play not at the best? >5. Did I select a variation not suiting my style and or my chess knowledge? >6. Was this line more suiting my opponent? >and so on. > >Of course to be able to answer them needs "a lot" chess understanding and this >can be achived only reading very many chess books (I think I have/had about 400) >and playing chess yourself at the club and International tournaments too. > >The chess programs are good too to practise, but the best way to get the most >from them is to play against them like you would play normally and not >anti-computer chess. >I mean if you want to improve you need to face them on their best and not to >weak them in order to get the tough opposition... > >To understand everything from scores/percentages...is simply a joke! I wouldn't call the only sure proof of progress a joke :) Every change however logical it may seem must be tested, but in looking for explicit features to add to the evaluation I think you are right that games should be analyzed for inspiration. -S. >Sandro
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.