Author: John Merlino
Date: 11:46:11 02/09/03
Go up one level in this thread
On February 09, 2003 at 09:25:43, David Rasmussen wrote: >On February 08, 2003 at 20:03:21, John Merlino wrote: > >>>> >>>I can't see why it is a blunder, but PostModernist agrees. >> >>The only thing I can see in common between these two positions is that, in both >>of them, White is forced to make a move that loses material. In the first >>position, it's 36.Rf3. In the second, it's 41.Rxf4. I would doubt that Chezzz >>has a specific problem with this, but I would need to see the PV for 32...Qa2 to >>verify this theory. > >After 32... Qa2: > >time nodes depth score pv >00:00:14.22 8768kn 13/26/29 --------------------------------------------------- >00:00:21.21 13Mn 14/29/30 -1.920 33. Ra1 Qb2 34. Rb1 Rxg2 > 35. Rxb2 Rxb2 36. Rf3 Nxf3 > 37. Qxf3 Rxb4 38. g5 Rb1+ > 39. Kg2 Rb2+ 40. Kg3 Rc2 (13235836) >00:00:26.94 16Mn 14/30/31 --------------------------------------------------- Ok, this very LAST PV matches, up to and including 38.g5, what CM9000 sees at depth 7 and scores at -2.33. Later, at depth 10, it changes that move (and the previous move) to: 37...Rc8 38.Qf1 Rcc2 39.g5 Rxb4 40.Qf6+ Kg8 with a score of -2.89. So, assuming that The King is correct, it appears that Chezzz does not see 37...Rc8 and the eventual win of the b-pawn. The White queen will not have a perpetual, and cannot go pawn hunting because of the mate threats from the two rooks. So, perhaps Chezzz does not see the full potential of the two rooks and the mate threat? Or it does not see that the White queen is forced to prevent the mate threat? Hmmmm.... EVENTUALLY, Black will get a passed b-pawn, but I think that's a bit too far away to see from this position.... It looks like Chezzz' eval difference between 32...Rhc8 and 32...Qa2 is very small (about 0.3), so perhaps some tweaking will do the trick (and screw up something else, of course.... :-) jm jm
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.