Author: Robert Hyatt
Date: 21:20:26 03/09/01
Go up one level in this thread
On March 09, 2001 at 21:25:26, Sune Larsson wrote: > > [D]8/2p1p3/B1p3p1/4PkP1/5P2/5K2/8/1b6 w - - 0 1 > > Melnichenko,E 1979 > > Allright Evgenij(?), let's have a look at what you got here. > Pretty normal position for once...Bishop ending with same colors, > 3 vs 4 pawns, but black has a double farmer on the c-file. On the > other hand they are passers. If white's king moves to the queenside > the f and g pawns are easy to chop off. Doesn't look too bad for > black, eh? Did you say *normal* position?! Just forget it! > > > Test: White to move and win. The Russian solution further down. > > > Sune > Interesting position. Crafty plays the correct move instantly. But for at least a minute on my PIII/750 notebook, it sees a score of 0.00. But if you play the game at blitz speeds, it follows your analysis perfectly but it takes several moves before it begins to see this is winning. I think these kinds of positions are interesting. The right moves are essentially forced (maybe not to a human, but to a computer, since it finds every move very quickly). Yet the ultimate outcome is not seen correctly for quite a while. Of courese, you really don't have to understand how good the move is if you can 'devine' that the move is the right move, even if for the wrong reason... This happens in your other test (below) as well, at least for Crafty. > > > > > > > >Melnichenko,E >1979 > >1.Bc8+ e6 2.Ba6 Be4+ 3.Ke3 Bb1 4.Bc4! Bc2 5.Ba2! c5 6.Bc4 Bb1 7.Bb3 c6 8.Bc4 Bc2 >9.Ba2 c4 10.Bxc4 Bb1 11.Bb3 c5 12.Bc4 Bc2 13.Ba2 c4 14.Bxc4 Bb1 15.Bb3 Be4 >16.Bxe6+ 1-0
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.