Author: Uri Blass
Date: 01:17:54 11/14/02
Go up one level in this thread
On November 13, 2002 at 17:00:20, Robert Hyatt wrote: >On November 13, 2002 at 10:24:44, Uri Blass wrote: > >>On November 13, 2002 at 09:43:36, Bob Durrett wrote: >> >>>On November 13, 2002 at 06:48:47, Joachim Rang wrote: >>> >>>>yep, I think you're right. >>>> >>>>This is what I got from DF 7 on depth 15: >>>> >>>>New game >>>>5rk1/p1p1br2/1p1p2qp/4p3/2PPn1pP/3Q2P1/PPN1P1K1/2RR2B1 b - - 0 1 >>>> >>>>Analysis by Deep Fritz 7: >>>> >>>>1. -+ (-3.31): 1...Bxh4 2.gxh4 Rf3 3.dxe5 Rxd3 4.exd3 dxe5 5.Be3 Nd6 6.h5 Qxh5 >>>>7.Rh1 Qf5 8.Ne1 >>>>2. -+ (-3.09): 1...Rf3 2.exf3 gxf3+ 3.Qxf3 Rxf3 4.Kxf3 Nxg3 5.dxe5 Bxh4 6.Rd4 >>>>Qf5+ 7.Kg2 Ne2 8.Ne3 Qxe5 9.Rg4+ Bg5 10.Re1 >>> >>>Obviously, then, 1...Bxh4 is better than 1...Rf3. Right? : ) >>> >>>Bob D. >> >>No >> >>I suspect that the order may be changed if you use longer search. >> >>The line 1...Rf3 2.exf3 gxf3+ 3.Qxf3 Rxf3 4.Kxf3 Nxg3 is forced and >>after it 5.dxe5 is not the best move. >>I expect programs that use singular extensions like Deep Sjeng to find 1...Rf3 >>after a long search. >> >>Uri > > >I think he meant "obviously, since the score is worse for black after Rf3 >than it is for Bxh4, then the program thinks it is better." I don't understand >unless that is "multi-variation output" why the second move was even displayed >since the score is less negative (better for white). It is clear that it was analysis with 2 options. The point is that there are 2 winning moves and it is even unclear which of them wins faster. The fact that most programs find Bxh4 after enough time does not prove that it is the best move and they may change their mind to Rf3 after longer search. 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.