Author: Ernst A. Heinz
Date: 07:11:35 01/22/98
Hi all,
"DarkThought" finished its run with the original ECM at 90 sec
per position and I continue to analyse the results. In addition
to the inconsistencies already pointed out by Amir,
"DarkThought" differs on some more positions among the #100-#199
(even after 20 min analyses each).
ECM #135
original solution = [ Nxb7 ]
"DarkThought" says = [ a5 ] hxg3 fxg3 f4 Nxb7 ... is equally good
(score = +1.3 after iteration #13,
looks like a transposition)
ECM #182
original solution = [ Qxe3 ]
"DarkThought" says = [ Nd5 ] Nxf1 Ne7+ Rxe7 Qxe7 Ne6 Rxf1 ...
(score = +1.7 after iteration #12,
score of Qxe3 is +1.9)
ECM #196
original solution = [ Rxd5 ]
"DarkThought" says = [ c3, Nbd4 ] are equally good
(score +1.1 after iteration #14, it thinks
Black loses the weak pawn on d5 anyway)
What do others think about these alternatives?
=Ernst=
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.