Author: martin fierz
Date: 11:24:05 10/14/02
Go up one level in this thread
On October 14, 2002 at 09:09:00, Chessfun wrote: >On October 14, 2002 at 07:29:44, George Sobala wrote: > >>On October 14, 2002 at 00:49:55, martin fierz wrote: >> >>>On October 13, 2002 at 22:08:01, robert flesher wrote: >>> >>>>I believe that kramnik was afraid of the calculation power of the computer in >>>>the 4 of 3 majority endgame. These endgames even though theory suggests a draw >>>>with best play are very complex and can often lead to a win for the majority. >>>>Many possibilities could have happened. Perhaps a queen and pawn versus >>>>queen....OUCH then there is trouble when its the tablebases that scare the hell >>>>out of anyone. Kramnik cracked he is human. >>> >>>4-3 with rooks or queens should be a draw. i don't think kramnik was afraid of >>>that! >>> >>>aloha >>> martin >> >>But this ending had knights too - and computers are pretty scarey opposition >>with queens AND knights. Calculating endings for Knights, Bishops or Rooks is >>all the same to a computer, but even SuperGMs can struggle to visualise the >>tactical threats posed by a Q+N pair in 5 or 6 moves time. Q+N positions are a >>lot less "logical" than those involving just Rooks and Bishops. > > >[D]6k1/N4p2/1p4pp/q3n3/4Q3/4P1PP/P4PK1/8 b - - 0 33 > >Kramnik could simply have played 33...Qxa7 34. Qxe5 Qxa2 35. Qb8+ Kh7 36. Qxb6 >That assumes he wanted to simply get the knights off. In fact it seemed to me, >simply another way for the hoover to get into a drawing endgame. > >It still seems to me this would have been far better for Kramnik than 33...Qxa2 >34. Nc8 Qe6 35. Nxb6 where the knights are still on. > >Sarah. good point! i didnt take any chess books with me to honolulu, so... does anybody have a book on endings which discusses 4-3 with only queens? i would have thought it was pretty drawish, but if kramnik didnt go for it he must have had a different opinion! aloha martin
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.