Author: Peter Berger
Date: 08:41:11 03/01/04
Go up one level in this thread
On March 01, 2004 at 01:55:40, Uri Blass wrote: >On February 29, 2004 at 18:03:20, Peter Berger wrote: > >>First one to win 6 games would win the match. Time control 45min/game + 10sec >>increment/move. >> >>Junior 8 played on PIV1.8GHz, 288MB Hash, 3-4-5 men tablebases >>Postmodernist 1.010 played on PIV3.2HT, ~400MB Hash, 3-4-5 men tablebases >> >>The match was won by Junior 8 by 6-3 with 2 draws. >> >>There is one major conclusion - can't one of the hobby bookcooks help poor PM a >>little? It sure needs a better, or at least reasonable opening book. > >This is a surprising result and I expected Junior8 to win more convincingly >after remembering that Crafty had big problem to beat Junior with clearly bigger >hardware advantage and only got 6-5 with many draws). Well, before someone jumps on us, of course this match (actually both) was *much* too short to reach any statistically reliable result, not even close - so this is all just wild guesses. But I do agree that Postmodernist did much better than Crafty when it is about the actual games and don't think Junior would have much of a chance against it with a 1-10 hardware disadvantage, if PM were not too unlucky with its book, although this would have to be tested. One reason might be that Postmodernist is much more aware of king safety issues and compensation for a pawn than Crafty, so it didn't get surprised by Junior's sacs so often. >I wonder if postmodernist made a big improvement relative to 1008 to be better >than Crafty(at least if we do not consider the book factor and play the nunn >match that I guess that both Crafty and postmodernist are not specially tuned >for it). > >Another possibility is that postmodernist simply has good results against >Junior8 but not against other programs. > I think Postmodernist is an engine that is just not so well-known and popular, but that is really pretty strong. Some time ago I played another little match against Hiarcs 9 at 60/10 with similar hardware difference, that I stopped at 3.0-3.0. Maybe it just likes my computers :). Peter
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.