Author: Will Singleton
Date: 08:15:11 11/03/02
Go up one level in this thread
On November 03, 2002 at 05:03:33, Uri Blass wrote: >On November 03, 2002 at 03:10:34, Walter Eigenmann wrote: > >>>I don't know about this test suite.Where did you get the positions? >> >>History of chess... :) >> >>>Several look wrong, like 2,4,11, etc. >>>Or are these just particularly difficult? >>>Several appear too easy, like 7,24,31 etc. >>>At 30 sec, crafty gets 20/40, Knightdreamer the same, >>>and Amateur gets 18/40. >> >>There are no "wrong" positions: > >I think that will meant wrong solutions > >>"avoid move" = NOT to move ! >> >>Regards: Walter > >I think that there was no misunderstanding about it. > >I looked at the positions that were described as very easy and movei avoided the >move in 0 seconds in most of the cases but the difference in score after making >the wrong move and analyzing was only a positional difference. > >I also gave it to analyze position number 2 and movei does not like White's >position but it does not prevent it to play Qxh1(score 0.33 for black at depth >14) > >main line begins with Qxh1 e3+ Kg1 Rxh1+ Kxh1 e2 Kh2 Qe4... > >Again the difference seems to be positional difference because after making Qe3 >Movei says +0.38 for black at depth 12 and +0.41 for black at depth 13. > >Uri You are right, Uri. It's always interesting to try "am" suites, but about half of these have no clear solution, with only small positional differences. I identified about 15 positions where the "am" move was bad, but there was no real alternative, and 5 where the move was avoided in 0 secs, or never considered. However, about half seemed to be good am candidates, so the author has the basis for a good suite. WIll
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.