|
The computer played the well known Bxh2+ type sacrifice. This sacrifice must be calculated exactly. The game can go either way. It is almost impossible for a human player to see every variation to the end, but humans often play it anyway because they know that it often works.
The critical point came on move 16. Kasparov could have played 16. g3. The computer would have two possible moves. It could check with Nh2+. However, then the White king could just walk away with moves like 17. Kf2 Ng4+ 18. Ke1.
Kasparov said after the game that he was afraid of Qh2 instead. However, then with care Kasparov could push up his pawns with moves like f5 and e4, blocking the development of his own pieces and creating space for his king to run.
It seems that with best play Kasparov had a win. However, out of respect for the awesome calculating power of his opponent, and perhaps forgetting that the computer is limited in how far ahead it can see, Kasparov headed into a variation which guaranteed a quick draw.
The final game will be on Friday. That game will decide the match. ESPN will be covering the game on TV, a first. See http://sports.espn.go.com/sports/tvlistings/index?date=20030207 Let us hope against hope that Kasparov will pull out and win. However, realistically, the odds favor either the computer or a draw.
Kasparov will be paid $500,000 by FIDE for playing Deep Junior. He can earn an additional $300,000 if he wins the match. The event is sponsored by X3D Technologies Corporation, which makes special 3D glasses useful for watching events such as this chess match. http://www.x3dworld.com . Kirsan Ilyumzhinov, the President of FIDE and Kalmykia, is present. The event is moderated by Grandmasters Maurice Ashley and Yasser Seirawan. Frederic Friedel and ChessBase are providing Internet coverage.
Sam Sloan was there too. Take a look at the pictures on http://www.tcf.net/game5.html
Sam Sloan
|
[Event "FIDE Man-Machine WC"] [Site "New York USA"] [Date "2003.02.05"] [Round "05"] [White "Kasparov, Garry"] [Black "DEEP JUNIOR"] [Result "1/2-1/2"] [ECO "E48"] [WhiteElo "2847"] 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 O-O 5.Bd3 d5 6.cxd5 exd5 7.Nge2 Re8 8.O-O Bd6 9.a3 c6 10.Qc2 Bxh2+ 11.Kxh2 Ng4+ 12.Kg3 Qg5 13.f4 Qh5 14.Bd2 Qh2+ 15.Kf3 Qh4 16.Bxh7+ Kh8 17.Ng3 Nh2+ 18.Kf2 Ng4+ 19.Kf3 Nh2+ 1/2-1/2