Author: Vine Smith
Date: 00:38:18 04/19/02
Go up one level in this thread
On April 19, 2002 at 00:35:52, Uri Blass wrote: >On April 18, 2002 at 21:40:18, Slater Wold wrote: > >>On April 18, 2002 at 20:31:24, Vine Smith wrote: >> >>>On April 18, 2002 at 19:52:25, Timothy J. Frohlick wrote: >>> >>>>Rex, >>>> >>>>You might be correct Sir. However, Smirin is playing drawish chess. >>>>Nevertheless, Tiger can certainly kick my butt. >>>> >>>>Tim, a satisfied Tiger 14.0 owner >>>> >>>>On April 18, 2002 at 19:42:42, Rex wrote: >>>> >>>>>Tiger is the only program the GM's hope to keep up with. >>> >>>"Smirin is playing drawish chess"?! You've got to be kidding! There was nothing >>>drawish about the drastic defeat he inflicted on Shredder, and he certainly was >>>not to blame for Hiarcs' choice of the Exchange Variation of the King's Indian >>>in the second game (where he got a much better position anyway). And Junior must >>>be running on a lucky processor to have escaped losing a piece down versus >>>Smirin in the last game. If it weren't for the truncated time control, Smirin >>>could be leading 3-0. >>>Nonetheless, Tiger may present some new problems. Not only its "anti-human" >>>setting, but also the book, both work together to avoid closed positions. When a >>>couple of Argentine GMs tried to get away with the Old Benoni last year, Tiger >>>just played 5.f4 (or at move 6, I can't exactly remember) and reopened the >>>position, winning both times. The f4 move wasn't objectively best according to >>>theory, but it was best for a program playing a human, and this is the kind of >>>approach needed against Smirin, who obviously knows a few things about typical >>>program weaknesses. >>> >>>Regards, >>>Vine Smith >> >>If Christophe or Ed, or even Noomen has a hand in this, the book will be 100x >>better than what we've seen thus far. > >I guess that they are going to have hands in the book in the tiger's games. >I also think that the mistakes that the programs did were after book. > >I also think that the quality of the engines is more important than the book. > >> >>CT will be a tough opponent. But it does share it's fair chance of losing. >>Depends on the opening, but I'd guess (if Smirin knows anything about CT) it'll >>be drawn. > >If christophe changed CT enough then smirin may know nothing about it. > >Uri Aren't you the one who observed that Shredder did well at Paderborn because of the programmer's personal involvement, especially in choosing books? Of course, even the best book possible won't help a defective engine, but here we're talking about the top programs available, and it seems very important to give them the right kind of positions versus their formidable opponent. Obviously, game 1 was a complete failure from this point of view, because Shredder was completely adrift as early as move 6, and I'm not sure many other engines could have responded much better (for the right reasons, that is -- the real test would be if an engine could see that it had to retreat Ne8 and get ...f6 in before the kingside starts advancing). Against a top GM like Smirin, the right kind of preparation is crucial, just as it would be for any human opponent facing him. Regards, Vine
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.