Author: Robert Hyatt
Date: 09:51:13 08/26/05
Go up one level in this thread
On August 26, 2005 at 12:32:03, Gian-Carlo Pascutto wrote: >On August 26, 2005 at 12:24:05, Robert Hyatt wrote: > >>Maybe I don't understand the problem or "bug" I am supposed to have here??? > >Do you have a multipv mode? It would be good to see the evaluations of Rb1 and >Qe2 from the root position, for increasing depths. > >It's easy to find Rc3 after Rb1 because you will generally be nullmoving for the >other player. If you can't see it at the root you might decide at greater depth >that Qe2 is not all that great and switch back to Rb1 instead. This is what >happened in the game. > >Crafty thinks at low depth that Rxd3 wins for black (and hence won't play Rb1), >but this is wrong (it's a draw). The variation looked like a bug, but I forgot >that you don't handle checks in the qsearch at all... > >-- >GCP Check your analysis. That position is a mate in 3 for black. :) Note white king can't evade the Qxe4#. If I did checks in q-search I probably would have seen it was a mate for black, rather than just -3.5... :) I can get multiple PVs. Can you post the PGN for the game and I can get it to display the five best moves and scores via "annotate"...
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.