Author: Uri Blass
Date: 04:11:31 04/24/02
Go up one level in this thread
On April 24, 2002 at 06:24:21, Brian Katz wrote: >On April 24, 2002 at 03:15:41, Steve Maughan wrote: > >>Stuart, >> >>>That is a question of beauty, I think. But chess is meant (imho) to show which >>>beauty (or playing style) is real and which beauty is false. And that is what >>>the whole competition of chess is about. >>> >>>Therefore, the great playing style is only of true value and beauty, when it >>>is proven over the board. >> >>I admire your search for absolute chess truth! >> >>I think we need to make a distinction here between a single game and all >>possible chess games. There will be a number of games where Hiarcs plays great >>positional chess that is a pleasure to watch - indeed some, or many, of these >>games Hiarcs will 'prove' it's knowledge by winning the games. Then there is >>the question of does Hiarcs always play the best move in every situation - well >>the answer is no but for a huge number of situations we don't know what the >>absolute best move is - and due to the massive number of possible chess >>positions we will never know for certain what the best move is. So in this case >>I'd appreciate many different styles. This is why I buy many different engines. >> >>Regards, >> >>Steve >It would probably make more sense for those of us who are serious chessplayers >on the rated tournament level,club level too, or any level where one wishes to >improve, to purchase Hiarcs as well as Fritz 7. >Fritz is extremely tactical and Hiarcs is very positional. How do you know? I need to get data and definitions in order to say that program X is tactical and program Y is positional. Data can be achieved by doing the following steps: 1)take care to get a result that is 50% between X and Y. If X is better than Y then you should give Y the better hardwrae. 2)analyze the games and count for every program how many games it won thanks to tactics. Use the following defintions for deciding if a game is decided thanks to tactics: A game is considered as a game when the winner won thanks to tactics if the winner was the first side to get a significant fail high and only in the next move or the move after it the opponent got a significant fail low. Significant fail high is defined as a fail high of more than a pawn relative to the evaluation in early iterations. Significant fail low is defined as fail low of more than 0.5 pawns relative to the evaluation in early iterations. Early iteration can be every iteration that the computer needed at least 1% of the total time that it calculated in order to complete it. The position before getting the significant failed high should be evaluated by both engines as not more than one pawn advantage for the winner. 3)The program that is relatively better in tactics is the program that wins more games thanks to tactic by my definition. Uri
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.