Author: Uri Blass
Date: 04:22:32 08/25/01
Go up one level in this thread
On August 25, 2001 at 01:38:23, Mig Greengard wrote: >On August 25, 2001 at 00:04:53, K. Burcham wrote: >> >>i cannot imagine how stefan felt watching this game. shredders score >> quickly went up 5 points, and maintained a five point advantage for quite >> some time. stefan had to feel very confident. somewhere about move 60 >> stefan probably witnessed his score drop tremendously. there must have >> been a lot of surprised chess enthusiasts in the game room. > >I talked about this during this game and a few others I commented live at >KasparovChess.com. I enjoy comparing how computers play and solve problems >compared to how a strong human would go about it. Shredder had an undeniable >advantage for most of the game, and a clearly winning advantage after it won the >exchange. It then started doing things that a strong human would never consider >for purely pragmatic reasons. I mentioned at the time, while Shredder was still >well in the plus, that it was playing with fire. (Gromit did the same thing, >turning a simple draw into a wild ride that it lost to Junior.) > >Your first obligation when you have such a position is to eliminate counterplay. >This isn't 100% possible against a strong opponent, but you certainly work to >simplify the position in order to keep control. The last thing you want is to >see the position sharpen so much that, despite your objective advantage, one >error in calculation can cost you the half or full point. > >Which is exactly what happened to Shredder. It's eval did not swing so much >because it evaluated the position incorrectly when it was at +5 (although it >does tend to be overoptimistic, like many programs, which is why I have so much >admiration for Junior's more conservative, and generally more accurate, evals), >but because it allowed the position, despite being advantageous, to get so sharp >that each half-ply meant life or death. This is often fatal for humans and >computers alike. > >Mikhail Tal would stir up unfathomable complications on the board because he >trusted his instincts and because he knew he could calculate better than anyone >in the world. But even Tal knew when it was time to simplify into a winning >endgame! No matter how well you calculate, you don't give your opponent a batch >of passed pawns (if you can help it) if only because nobody is perfect and you >just might have missed something. > >Programs are vulnerable here because they always play the "best move," and not >the most practical move. They can't see the inherent complexity of a coming >position, just the eval. If they have the right eval they can see the complexity of the position in the evaluation and reduce the score. If they see that thei eval is dropping inspite of being more than +5 they should understand that something may be wrong with their evaluation and change their evaluation function Here is shredder analysis at move 60(it is on slow p200 but I gave it a lot of time) You can see that Shredder eval is 5.87/7 5.79/8 5.69/9 5.65/10 5.46/11 5.52/12 5.34/13 5.33/14 It dropped almost every iteration Deep Junior 7 - Shredder [D]4r2k/7p/2q5/p2p1PP1/5N2/2b1p1PK/4B3/3Q4 b - - 0 1 Analysis by Shredder 5.32: 60...d4 -+ (-6.18) Depth: 1/2 00:00:01 60...d4 61.Ne6 -+ (-6.17) Depth: 2/4 00:00:01 60...d4 61.Ne6 Qe4 -+ (-6.00) Depth: 3/6 00:00:01 60...d4 61.Ne6 Qe4 62.g4 -+ (-6.11) Depth: 4/8 00:00:01 60...d4 61.Ne6 a4 62.Kg4 -+ (-5.86) Depth: 5/10 00:00:01 60...d4 61.Ne6 Qd5 62.Bc4 Qxc4 63.Qxd4+ Bxd4 -+ (-5.86) Depth: 5/10 00:00:01 60...d4 61.Ne6 Qe4 62.Bd3 Qd5 63.Kg4 -+ (-5.73) Depth: 6/12 00:00:02 12kN 60...Re5 61.g4 d4 62.Ba6 -+ (-5.74) Depth: 6/12 00:00:02 20kN 60...Re5 61.Kg4 d4 62.Qc2 Qe8 63.Bc4 -+ (-5.80) Depth: 6/12 00:00:03 31kN 60...Re5 61.Qc2 d4 62.Kg4 Kg8 63.Qb3+ Kg7 64.Bf3 -+ (-5.68) Depth: 7/14 00:00:05 69kN 60...d4 61.Ne6 Qe4 62.Nxd4 Bxd4 -+ (-5.69) Depth: 7/14 00:00:05 76kN 60...d4 61.Ne6 Qe4 62.Bd3 Qe5 63.Qh5 Rc8 -+ (-5.87) Depth: 7/14 00:00:06 105kN 60...d4 61.Ne6 Qd6 62.Qd3 Rb8 63.Qc4 Qe5 64.Kg4 -+ (-5.79) Depth: 8/16 00:00:10 196kN 60...d4 61.Qb3 Qa8 62.Kg4 a4 63.Qf7 Rf8 64.Qd7 Qe4 -+ (-5.69) Depth: 9/18 00:00:35 944kN 60...d4 61.Qb3 Qd7 62.Kg4 a4 63.Qb5 Qxb5 64.Bxb5 Ra8 65.Bxa4 Rxa4 -+ (-5.65) Depth: 10/20 00:01:02 1771kN 60...d4 61.Qb3 Qc7 62.Kg4 Rc8 63.g6 Rf8 64.gxh7 Qb8 65.Qe6 Kxh7 66.Kf3 -+ (-5.46) Depth: 11/22 00:02:55 5261kN 60...d4 61.Qb3 Qd7 62.Kg4 a4 63.Qb6 Rf8 64.Qc5 Ra8 65.Qe5+ Qg7 66.Qd5 Ra5 -+ (-5.52) Depth: 12/24 00:09:22 17679kN 60...d4 61.Qb3 Qd7 62.Kg4 a4 63.Qb6 Rf8 64.Qe6 Qg7 65.Qd5 Rc8 66.Bf3 Qe7 67.Ne6 Kg8 -+ (-5.34) Depth: 13/26 00:25:19 49607kN 60...d4 61.Qb3 Qd7 62.Kg4 Bb4 63.Qc4 Re5 64.Ne6 Bc5 65.Bd3 Bb6 66.Kf4 -+ (-5.33) Depth: 14/28 02:18:32 295630kN Don't blame Shredder for the Rd8 blunder; that's the >symptom, not the disease that all comps face. Rd8 blunder is not the only problem in the game because even without this blunder shredder did not have the advantage when it played Rd8. Uri
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.