Author: leonid
Date: 07:24:07 04/20/02
Go up one level in this thread
On April 20, 2002 at 10:13:46, Gian-Carlo Pascutto wrote: >On April 20, 2002 at 10:05:37, leonid wrote: > >>>PV: d7c6 a8c6 f3e4 f4e4 c5e4 c6b5 b2b5 c4c5 b5c5 e6c5 c3c5 >> >>It is interesting to see that your program solved this position by selective in >>11 moves. My selective also was not able to find any shorter mate that 11. >>Probably our selective have similar search pattern. >> >>Shortest mate is just one move below. > >You have to be carefull how you interpret what the PV means. > >What it displays is what Sjeng considers to be the 'line of most >resistance'. In a depth-based program (like a normal chessprogram), >this would be the shortest mate. > >For Sjeng, this is not true, especially if 'quiet' moves are possible. > >For example, in the positions Ed posted, Sjeng will display a PV that's >three moves long in a position that's a mate in >10. This is because >Sjeng considers the quiet move to be 'harder' than the forced checks. Aha! I just counted number of your moves and they are 11. This is where I was sure that we found mate at same depth. But at what exactly depth your program found mate? If it found it at 10 moves, it worked better that mine of this position. Leonid. >-- >GCP
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.