Author: Thom Perry
Date: 04:41:00 12/08/98
Go up one level in this thread
On December 08, 1998 at 04:44:18, Harald Faber wrote: >On December 07, 1998 at 17:31:48, Heiko Mikala wrote: > >>>>>>MCP8 have an astounding opening book. >>>>>>It have 6.291.456 positions in 61,44 MB hash. >>>>>>It follows the book generally in 15-30 moves. >>>>>>In the game below, it's in the book for 40 moves! >>>>> >>>>>Now the all decisive question: with which evaluation did MCP8 come out of >>>>>book? >>>>>If more than +2.00 then we can almost be sure to have a killer line. >>>> >>>>MCP8 gives 41.Kf1 Qxf3+ mate in 8. >>> >>>OK, THIS IS A KILLERLINE. Probably taken from an autoplayer game Sandro >>>and/or Marty added to the book... >> >>Rebel 8 has such lines against MChess too. In one of my tournaments it played a >>french opening against MChess Pro 5 completely out of book up until the point >>where MChess saw the mate against itself... > >Wow! Could you please give me that game? >BTW I remember when Ed mentioned s.th. like he did implement also one (or some >more?) killerline because he wasn't satisfied with MCP's opening book full of >killerlines. He asked Marty if he found them. :-) >So probably YOU have found at least one of them. :-) >As far as I get MCP8 I'll check some other loosing lines too to see if Sandro >took them out. > >>By the way, I love both programs, these are my favourites, so please don't >>misunderstand me. No war please ;-) > >Don't get me wrong, I love MCP since version 5 and already ordered v.8. But I >like Rebel10 more than Rebel8. :-) >And I don't want to see killerlines, no matter of whom. Personally, I don't see anything wrong with killerlines. If you have a good program like MCP with the learning enabled, you should only lose one game to the killerline <g>.
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.