Author: Uri Blass
Date: 18:00:07 02/03/03
Go up one level in this thread
On February 03, 2003 at 19:39:53, Peter Berger wrote: >On February 03, 2003 at 17:14:41, Peter Berger wrote: > >>On February 03, 2003 at 16:29:46, Robert Hyatt wrote: >> >>>On February 03, 2003 at 14:15:12, Matthew Hull wrote: >>> >>>>On February 03, 2003 at 12:32:19, Amir Ban wrote: >>>> >>>>>On February 03, 2003 at 02:12:30, J Mike wrote: >>>>> >>>>>>On February 03, 2003 at 00:13:00, Robert Hyatt wrote: >>>>>> >>>>>> >>>>>>>This is an opening that worked 20+ years ago. David Levy used it to beat >>>>>>>Chess 4.x, and then Cray Blitz, in his much-publicized "wager". But it is >>>>>>>no longer something that a program can't handle, if it understands anything >>>>>>>about blocked pawns at all. >>>>>>> >>>>>>>He was fortunate that this didn't go nuclear in his face... >>>>>>> >>>>>>>I'd be surprised if he tries it again in game 6. >>>>>> >>>>>>The comp actually played the position very badly. h3 and latter a3??? What the >>>>>>hell was that? Even an 1800 won't play moves like h3 and a3 ... Kasprov should >>>>>>have tried b5 after a3 or maybe b5 one move before a3 .. The comp was lucky that >>>>>>this didn't go nuclear in it's face ... >>>>>> >>>>> >>>>>I was wondering when anyone would notice that Deep Junior was offering a piece >>>>>for two moves in a row. >>>> >>>> >>>>Sound or unsound (i.e. bug)? >>>> >>>>Matt >>>> >>> >>>Sound. Most programs agreed. >> >>Crafty did, others didn't - I haven't analyzed the position in depth but it >>looked _very_ dubious IMHO. > >OK - at least some variation should be given here - 17...b5? (signed by Mueller) >18. cxb5 Qxc3 19. Qxc3 Rxc3 20. Bd4 is discussed in the comments of GM Mueller >in the German comments on chessbase.de and said to be winning for white - that >is debatable, to say the very least .. > >But after Kasparov played 17. .. Rfe8 and Junior opted for 18. a3?! , ..b5!? is >definitely even more interesting . If now 19. cxb5 Qxc3 20. Qxc3 Rcx3 white >looks simply miserable IMHO, whatever prog evals give. > >Of course I realize that better players discussed these positions at length - >but at least they didn't manage to convince humble me yet. > >Peter 18.cxb5 or 19.cxb5 is not forced in these lines. I believe that in both cases the move cxb5 is a losing blunder. I think that the fact that GM Mueller said that after 17...b5 18. cxb5 Qxc3 19. Qxc3 Rxc3 20. Bd4 is winning for white only show that we cannot trust GM's evaluation and GM Mueller like Fritz8 seem to overevaluate passed pawns. I have no proof that 17...b5 is losing but the fact that Fritz8's eval goes dow every iteration clearly suggest it. I was surprised by Amir Ban's post that Junior was offering a piece for 2 moves in a row because the piece offer is only in a line that was not played. I believe that black could get a small positional advantage by 17...b5 or 18...b5 when I do not assume capturing b5 but offering black some positional advantage is not offering a piece. 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.