Author: Omid David
Date: 08:58:18 09/17/02
Go up one level in this thread
On September 17, 2002 at 11:44:56, David Rasmussen wrote: >On September 17, 2002 at 11:05:58, Omid David wrote: > >>On September 17, 2002 at 08:44:31, David Rasmussen wrote: >> >>>Is there a list somewhere of the errata of test suites? Specifically, I would >>>like the errors in 1001 Brilliant Ways To Checkmate. >>> >>>/David >> >>Do you mean "1001 Winning Sacrifices in Chess" (WAC)? > >Nope. I am talking about Fred Reinfeld's book. The suite is called bwtc.epd. > >>I'm currently working on >>them. They have much less errors than the ECM. >> > >OK. > >>I couldn't believe the number of errors I found in ECM! Are there any >>corrections available? > >I don't know, but _I_ couldn't believe the number of errors I have found in >1001BWTC. I wonder if this "classic" is at alle suited for tactical training of >lower rated players (like myself). I mean, everytime I find a solution, or not, >I can't be sure at all if the book is right or not. > >/David The only solution might be a deep analysis of all test suites :-( I tried to use ECM to compare the tactical strength of different algorithms, but the result was pure nonsense. On many occasions the engine found the "recommended" move at say ply 8, and then found an even better move at ply 10. Or even worse: sometimes the "recommended" move is found at ply 8, and then refuted at ply 10!!!
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.