Author: Robert Hyatt
Date: 05:52:55 09/16/04
Go up one level in this thread
On September 16, 2004 at 07:56:30, Günther Simon wrote: >On September 15, 2004 at 19:51:11, Robert Hyatt wrote: > >>Please think about what is happening. If I only include 1. d4 for Crafty to >>play, I just eliminated a _lot_ of potential opening theory. No one can force >>me to play anything but 1. d4, unless they replace my book. I play d4 a lot to >>avoid lots of tricky book analysis done by the pros, because 1. d4 certainly >>limits black's choices, where I can attempt to control things better. I don't >>have to tune for _every_ position, which is my point. I can construct my book >>so that I only play into positions where things work well... > >Does this mean you don't suggest book random 1 and book width 2 >anymore against computers for the crafty rc file? Depends on where. In an ICCA tournament? Nope. On ICC? Yes. There randomness is more important that playing the very best opening, to avoid repeating the same game too often and getting "cooked". In real tournaments I very often use book random 0, so that the program will choose from the book lines based on what it "likes" itself... >(You suggested this in the past!) >With that setting your book plays either 1.d4 or 1.e4... No it won't. If you have (as I do) 1. d4! in the books.bin or bookc.bin, that is all you will get. The randomness only counts when there are no ! moves pre-selected by the book preparer... That is why I added the "flags" and "play percent" to the book in fact... > >Günther
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.