Author: Marcus Kaestner
Date: 00:27:08 12/09/98
Go up one level in this thread
> >Of course it was a serious tourney but some things obviously are left behind the >curtain. Did all participants play with their own settings or with original >ones? > Some people played original, some with its own settings. But it is also on every tournament, isn´t it? Also the programmers are changing their settings from round to round depending on the next opponent or on the results of the last rounds. If you want, no result in no tournament is correct, even in sweden because they often makes mistakes which is normal and happens to everyone who operates lots of programs. But let me give one more comment to the hardware: Surely, if you compete one program against itself, this one which have a double-speed hardware has a favor because it gets an extra ply out of the hardware-speed. But if you play different programs against each other the things are changing dramaticly. Sure, the program which has the faster hardware is also in favor, but very often cannot get a result out of this. For example: If I play my Nimzo 3.5 with my special settings and implemented knowledge against itself, the double-speed Nimzo will score about 6,5:3,5. But if I play against Rebel 10, the score is 5:5 doesn´t matter on which machine. If I play Fritz 5, the score is 6:4 for me, doesn´t matter which machine, If I play Rebel 9 the score is 6:4 for me doesn´t matter which machine and so on. The key is the using of the opening variations, settings and knowledge. In the right opening, when my program has good knowledge he will win the game (or draw) surely, doesn´t matter on which hardware. When a program can play against the style of the other, the hardware is not important except in cases of blundering. Take a look at Genius 5 in this tournament. On P60 he made a great score because he always played good openings. The points or half-points it loss, it loss of course of blunders (slow hardware!!). On the other side Genius 5 on the very fast PentiumII-450 played bad openings and made a very bad score also he looked about 3 plys deeper than Genius on P60!! It´s always the same: The hardware is the less important component of all in a round tournament! But let genius 5 play against itself on P60 and P450 about 100 games, the score will be about 70:30 for the fast machine. But the hole discussion is fruitless because some people do not want to understand. But I don´t let it be that somebody says to me that we are fools and have no idea of computerchess. This persons can bet with me: I compete every person with CM6000 against my Rebel 10 on half-speed hardware about a 10 games match about $1000.-- I´m really not a Rebel fan, but I´m sure I will win every bet against CM6000 also on much slower hardware. Marcus
This page took 0.01 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.