Author: Jeremiah Penery
Date: 22:27:38 03/16/00
Go up one level in this thread
On March 15, 2000 at 21:29:22, Joshua Lee wrote: >On March 15, 2000 at 00:20:39, Jeremiah Penery wrote: > >>On March 14, 2000 at 13:45:08, Joshua Lee wrote: >> >>>Nolot pos 1 Nxh6 31:51 10/30 >>> 10 Rxf7 6:50 9/29 >> >>Can you post the evaluations for these moves, and the line given? >> >>I'm not at all convinced that it found the solution to #1. It probably gives >>something like Nxh6 c3 Bxc3 etc., rather than the winning Nxh6 c3 Nf5!! cxb2 >>Qg4! etc. >> >>The second one is likely correct, but I'd still like to see the evaluation. For >>it to have truly been solved, the score needs to be a significant + score. >You know the most important thing is to find the right move in the first place. That is only the first step. > it said -/+ -0.73 Nxh6 Then it most likely didn't have the correct answer. The reason I think the line/score matter is that even if it plays the correct first move, it can still easily lose the game if it doesn't follow up properly. Specifically for the Nolot test, the point is to find _winning_ combinations. If the program doesn't see the combination, or see that it is winning, then it has NOT solved the problem.
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.