Author: Dann Corbit
Date: 10:42:58 09/22/03
Go up one level in this thread
On September 22, 2003 at 13:35:56, Pete Waring wrote: >I was comparing the results from a few engines for analysis purposes (trying to >decide what engines I like to help with blunder checking), and got some strange >answers for the following position: > >r1bqr1k1/p4p1p/2p1n1p1/3p4/1B3Q2/P2B4/5PPP/2R3K1 w - - 0 1 > >Some engines, such as Fritz 6, Ruffian 1.0.1, and Little Goliath 3.6, show the >position as being equalized after Qf3, while others, like Shredder 6.02, Crafty >19.03 and Yace Paderborn show black as clearly winning after Qf3. Two examples: > >Analysis by Fritz 6: > >1. = (0.00): 29.Qf3 >2. -+ (-6.31): 29.Qe5 c5 30.Bc3 Nd4 31.Qg3 Nf5 32.Qf4 c4 33.Bxf5 Bxf5 >3. -+ (-6.34): 29.Qd2 a5 30.Bc3 d4 31.Bb2 c5 32.Bc4 Rb8 33.Re1 Qf6 >4. -+ (-6.50): 29.Qe3 a5 30.Bd2 c5 31.Qg3 Ba6 32.Bxa6 Rxa6 33.Qg4 Qd6 > >Analysis by Shredder 6.02: > >1. -+ (-8.14): 29.Qd2 Bb7 30.Qd1 Rc8 31.Bc3 Nf4 32.Bf1 c5 33.Bd2 Qg5 >2. -+ (-8.28): 29.Qh6 Bd7 30.Bd2 c5 31.Rd1 Rb8 32.Kh1 Rb3 33.Bc1 Nd4 34.Be3 >3. -+ (-8.31): 29.Qf3 c5 30.Bd2 Nd4 31.Qd1 c4 32.Be3 Nf5 33.Bxf5 Bxf5 34.Qd4 Re4 >35.Qc3 >4. -+ (-8.21): 29.Qe3 a5 30.Bd2 c5 31.Re1 Ba6 32.Bxa6 Rxa6 33.Qc3 Qh4 34.Qd3 Qc4 >35.Qxc4 dxc4 > >No engine, after actually executing Qf3, still thinks the position is equal. >Any clues as to why any engine would think the position is equal following this >move? Makes me wonder about using some of these for something as simple as >blunder checking... > >Thanks for any thoughts. Let them run a long time. At least 6 minutes. Always run several engines. That way, you can test many different ideas. Another good thing to try is to make the suggested moves and analyze those.
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.