Author: Don Dailey
Date: 19:18:48 01/19/98
Go up one level in this thread
>Suits me. How about we first fix the current ECM suite then. Either >decide that a position needs an additional key move, or that a position >should be discarded. > >I want to see the following: > >1. no positions that are solved positionally at very shallow plies, >because >a program gets credit for playing the right move for the wrong reason. > >2. no positions that have one key move given, but another key move >produces >an eval of +12 or something. I'd like to consider +12 as correct, >unless we >can demonstrate that +12 is wrong and a deeper search proves it.. > >3. later I'd like to see easy ones discarded because running this >entire >thing for 1 or 2 minutes will take over half a day at one minute, over a >full day at 2 minutes. That's too long... I do not think we have agreement on your first condition. But I would like to emphasize that I think it's an important one. The set won't be nearly as useful if every program plays the right moves for the wrong reasons. - Don
This page took 0.01 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.