Author: Sune Larsson
Date: 14:18:09 02/20/01
Go up one level in this thread
On February 20, 2001 at 15:52:28, Robert Hyatt wrote: >On February 19, 2001 at 09:50:42, Sune Larsson wrote: > >>I think a good try for a human vs a computer is the "Chepukaitis setup". >>Chepukaitis is a 65 years old, very original and talented master. >> >>As black you open with d6 and g6 - after that put pawns on a6,b6,e6,h6. >>The bishops are placed on b7 and g7 - the knights on d7 and e7. >> >>As a little example this is a rated game vs CM8K personality GM Fischer. >>40 moves in 40 minutes PIII 800. > >David Levy used to play this against computers. It worked great until he >ran into a later version of Deep Thought where he got smashed by playing >so passively. > >I think he lost 4-0 if I recall. Correct, the match was played in 1989 with the result mentioned above. Though Levy didn't play Nge7 in any of the 2 games as black, but rather an early and "normal" Ngf6. I also found out that Radjabov used this "Chepukaitis setup" in the last round of a Dutch tournament this year, playing for a win vs a lower rated opponent. Hm, must have been in Corus B vs Vink. Sune > > > >> >> >>(3) Fischer,R (2861) - Larsson,S (2694) [B06] >>Chessmaster 8000 Rated Game 40/40 Kil, 2001 >> >>(This is the latest game, played just a couple of days ago. Facing Fischer this >>time ;) playing GM:s the alpahabet down...) >> >>1.e4 g6 2.d4 d6 >> >>(I have no opening preparation and don't want to play special "anti-computer" >>openings, so the following idea popped up OTB. To the memory (oh well, he's >>still alive and kicking!) of Master Chepukaitis.) >> >>3.Nf3 Bg7 4.c3 a6 5.Bd3 b6N (5.-Nbd7) 6.0-0 e6 7.Bf4 h6 8.e5 Bb7 9.Nbd2 d5 >>10.Nb3 Nd7 11.Qc2 Ne7 12.Be3 0-0 13.Qd2 Kh7 14.Rae1 Nf5 15.Bf4 c5 16.dxc5 bxc5 >>17.c4 >> >>(Of course. I was very pleased with the opening, but now calculation starts.) >> >>17...dxc4 18.Bxc4 Bxf3 19.gxf3 Nb6 >> >>(This was the only varaiation I looked at. Playing 3 times faster than normal >>tournament level tends to shrink the "tree". Simply missed the strength of >>19.-Qh4, but the plan performed is also good.) >>[19...Qh4 20.Qxd7 Qxf4 21.Re4 Qg5+ 22.Rg4 Qh5 23.Re1 Rad8 24.Qc6 Nd4 25.Nxd4 >>cxd4 26.Qe4 d3 27.Bxd3 Rd5 -1.14 Fritz] >> >>20.Qxd8 Rfxd8 21.Re4 Nd4 22.Nxd4 cxd4 23.Bb3 Rac8 >> >>(The clock is ticking... There were other promising moves) >> >>24.Rd1 Nc4 25.Rexd4 Rxd4 26.Rxd4 Nxe5 27.Kg2 Rc5 >> >>(And here a total standstill. Just didn't know how to continue. The move played >>carried some idea of protecting the a6 pawn by a push to a5...(27.-g5) -0.72 >>Fritz >> >>28.Be3 Rc7 29.h3 Nd7 30.Rc4 Rxc4 31.Bxc4 Bxb2 32.Bxa6 Ba3 = Fritz >> >>(Still 8 moves to the time control. Knew I had lost the grip and was worried >>about a white king walk to b3...) >> >>33.f4 Kg7 34.Bb5 Nc5 35.Kf3 f5?! zeitnot >> >>(Doing nothing would lose, I thought - but this move gives away a pawn. Slipping >>downhills...) >> >>36.Be8 Ne4 37.Bd4+ Kh7 38.Bf7 Nc5 >> >>(Fingers crossed for him to take on c5 and e6...;) >> >>39.Ke2 Ne4 zeitnot (39.-Bb4) 40.Bxe6 Nc5 >> >>(First time control and still alive) >> >>41.Bc4 g5 42.fxg5 hxg5 43.f4 Kg6 44.fxg5 Kxg5 45.Kf3 Ne4 46.Bd5 Nd6 47.Be3+ Kh4 >>48.Kg2 Ne4 49.Be6 Nd6 50.Bd7 >> >>(Somewhere around here I got the illusion that the next time control was after >>60 moves... Spent a lot of time...) >> >>50...Nc4 51.Bf2+ Kg5 52.h4+ >> >>(Trying to calculate the effects of 52.-Kg4. I saw some mates after Bb5 or that >>the king would be driven away from the h-pawn. So finally...) >> >>52...Kg6 (52.-Kg4!?) 53.Bd4 Nd6 54.Bc6 Ne4 55.Be8+ Kh6 56.Be3+ Kg7 57.h5 Nd6 >>58.h6+ Kh7 >> >>(Here a flash of insight made me realize that the next time control was after 80 >>moves...;) >> >>59.Bd7 Kg6 (Time) >> >>Another hard loss. Maybe there still are chances but these endings are very hard >>to play. Need more time so maybe a change to the new FIDE time controls would be >>good. 1-0 >> >>Sune
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.