Author: Jeremiah Penery
Date: 07:05:10 10/15/99
Go up one level in this thread
On October 15, 1999 at 03:25:00, Shep wrote: >On October 14, 1999 at 15:14:28, Jeremiah Penery wrote: > > >>Most programs can find at least a couple of them (#11, maybe #10, some may find >>#3, but probably for 'positional' reasons. #1 and #5 are possible, but a bit >>more difficult, and would probably take longer.) in reasonable time. Maybe even >>in 10 minutes. > >Many programs find #1 in under 10 minutes. Chessmaster 5555 finds #2 in under 6 >hours on a P6-233 (but I don't know how much under) with a correct score. When these programs find #1, do they choose Nxh6 only for a very small positional reason (instead of choosing Qd4 or some other move)? Do they actually find that this move is winning? I think the key move to position #1 is after 1. Nxh6 c3, and the move is 2. Nf5. IMO, this position isn't solved unless A) 2. Nf5 is found and B) a winning score is returned for white. Otherwise, the engine will play 1. Nxh6 and still lose, therefore defeating the purpose of finding the 'brilliant' move. That's a nice result from CM5555 on position #2. I know DT2 solved this one in a couple minutes, with correct score, but we can hardly expect to match their results for this test. :) Jeremiah
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.