Author: Vincent Diepeveen
Date: 03:09:30 08/17/00
Go up one level in this thread
On August 16, 2000 at 18:51:57, Marcus Kaestner wrote: Let's quit the yes no arguments. Both are strong engines and both have a high ranking on my 'good chances to get this wmcc title'. Whether shredder has a % more chance or junior a small edge, that's only interesting for SSDF, as a single % there will get you #1. In WMCC you play against an opponent at most 1 game. 3 results - win - lose - draw At 9 games, a single point more can matter 80 points initially, but considering they end high, a single point matters up to 150 rating points of difference. So suppose the next theoretical scenario after round 8: - junior 7.0 - shredder 7.0 (though 7 out of 8 is unlikely score nowadays in computerchess, this is a theoretical example) Now, obviously they already played each other. If shredder gets last round Xinix or something, and junior has the bad luck to play for example crafty last round, then there is a chance junior loses and a less likely chance that shredder even draws. However for tournament rating achievement the easy point of shredder gives shredder a much better TPR if junior loses the game, which simply is a possibility, as you play 1 game, not 10. So whether a program is 10 or 20 points better or even 50 points better as it's opponent, in a tournament that doesn't count! In a tournament it's about being there at the right time, and whichever gets a winner, it will be a heavy fight to get it! >>>remember last year. everyone was sceptic when i predicted shredder to win the >>>title. now everyone is sceptic when i predict chessbase NOT to win the title. >>>we will see... >> >>Yes, it was a surprise. >>How did you guess? > >because i was sure that shredder was far the best at that time in a 7-round >tournament. >why? >because >1. on 7 rounds the booklearner is not important >2. on that machine (under nt, with 500mb ram) shredder hardly looses a game >3. shredder hardly throw away an advantage against weak opponents >4. stefan and i prepared the engine and the book for the wcc. our tests showed >similiar results as they occured in he tournament. so we both were very sure >that shredder will win the title. > >> >>It is clear that shredder was not the best because other programs like Ferret > >shredder was the best program that time for a 7-round tournament. > >>and the chessbase programs had hardware advantage that could give them more than >>50 elo and the difference between shredder4 and the top programs of that time on >>equal hardware is not bigger than 50 elo(for example the difference between >>shredder4 and Fritz5.32 is probably less than 50 elo). > >hardware is much less important than good preparation. > >> >>Do you believe that stefan forced luck by good opening prepration? > >we tried. but unfortunately there were too many holes in the book. but the lines >which was played generelly did not fit very well to the style of the opponent. >so in the end they were not able to win, even when they had an advantage after >the opening. > >> >>Did stefan planned the openings to push the opponents to positions that they do >>not understand? > >for example. > >marcus
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.