Author: Roy Brunjes
Date: 12:44:54 01/28/01
Go up one level in this thread
On January 28, 2001 at 14:41:40, Uri Blass wrote: >On January 28, 2001 at 11:40:04, Thorsten Czub wrote: > >>On January 28, 2001 at 10:01:17, Roy Brunjes wrote: >>>Century 3 missed a golden opportunity. It's Be7 followed by h4 let white close >>>the king-side up. Much better would have been 20. ... h4! That would have >>>forced the issue. White is practically lost after that move. >>> >>>Roy >> >>thats true. > >My Junior5.9 has no problem to see h4 in 0 seconds and I guess that the same is >truth for Junior6. > >It only proves again that saying that the fast searchers are stupid is not >correct and they can often find good moves that slower searchers cannot >find(Junior found h4 because of evaluation and not because of search and only >later it fails high on this move again and again) > >Uri Quite right Uri. I tested Junior 6a and learned a bit about differences between Fritz and Junior in the process. Backing up a few moves, Fritz 6e (PIII-500) sees h7-h5 in about 2 minutes (cannot remember the exact time now). Junior 6a will not play h7-h5 in any reasonable time. It's King Safety code seems to prevent that. If I turn off the King Safety (engine parameter), Junior 6a eventually (about 7 mins) sees h5 as a good move. Fast forward to the next move and it is Junior 6a that sees g5! almost instantly, while Fritz 6e will not play g5 even after many minutes of search. Curious. Fritz 6e and Junior 6a both see h4 almost immediately (with Junior being the quickest). What does this prove? I'm not sure! It's interesting to play with though. My somewhat tweaked Century 3 personality with King Safety set to 125 and Attractiveness and Attacking both set to 125 (and a few other minor tweaks) sees h4 after about 2 minutes. While this tweaked personality is almost certainly weaker vs other programs, my guess (and only a guess -- no data to use here) is that it would be somewhat stronger vs humans as many humans seem to have a tougher time defending vs an aggressive computer opponent. The tactical situations brought about are, in my opinion, more likely to favor the computer, hence the reason why I use it for games. It's also a bit better at spotting spectacular moves by GMs in their games (as I analyze them). Roy Roy
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.