Author: Uri Blass
Date: 12:51:07 12/31/03
Go up one level in this thread
On December 31, 2003 at 15:41:19, Uri Blass wrote: >On December 31, 2003 at 14:17:48, Steven Edwards wrote: > >>On December 31, 2003 at 14:06:59, Uri Blass wrote: >> >>>This can be solved easily by extending every stupid check that is not first >>>check by almost 2 plies. >> >>http://dictionary.reference.com/search?q=ad+hoc&db=* > >with normal extensions latest beta version of movei needs almost 1/2 minute on >A1000 to see mate in 15 and needs 9 plies(see the game beloiw for the line). > >Note that I expect strong humans not to play Qh5+ because it is a practical >mistake to start to calculate a long line when there is a simple winning >move(exf6 in this case) because the target is not to win as fast as possible and >it is better to be 99% sure of winning and not 98% even if you win faster and >humans should give at least 2% chance that they miss something if they calculate >a long line in OTB game. > >Uri Or maybe it is not so simple. I was influenced by the score of chess programs when they see exf6 but the line after exf6 is also not simple and have sacrifices. exf6 Nxf6 fxe6+ Bxe6 and now white needs to play Rxf6 or Qh5+ Kg8 Rxf6. 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.