Author: Bruce Moreland
Date: 23:29:26 01/22/98
Some of us are trying to redact (fix) ECM. I didn't really want to do this, but I've been trying to help when a machine frees up and I find some time. I have some questions about what constitutes a correct problem. Some of these have been asked before, but I'll ask a whole bunch in the hopes that we can get everything straight in this one thread. 1) What should be done when a problem that was supposed to lead to a decisive advantage breaks down and instead leads to an advantage that is arguable? 2) What should be done when there is more than one move that seems to lead to decisive advantage? In some cases this is easy -- a mate in 5 is better than a score of +3, but what about +2 versus +4, or a case where the score chosen is a matter of taste (one programmer might have a larger trade-down bonus than another)? Should we concern ourselves with trying to determine what a "better" solution is? 3) What should be done about transpositions or near-transpositions, for instance cases where an exchange is initiated, followed by the key move, leading to an approximately similar score? For example, position 146 is the famous "Ba3" Botvinnik position, and I could swear that someone argued at one point that 1. h4 should be added as a cook, because 2. Ba3 was still unavoidable. 4) When a cook appears to be better than the solution, do we throw out the problem, or adopt the cook? 5) What if nobody can make sense out of the problem? Do we throw it out, try to get more information about it, or what? bruce
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.