Author: John Merlino
Date: 14:08:30 03/22/01
Go up one level in this thread
On March 22, 2001 at 12:47:49, Mike S. wrote: >On March 21, 2001 at 21:37:06, John Merlino wrote: > >>This is not a bug, at least not a bug in accessing the data. 2...d5 has a 0 >>priority, as do all subsequent moves in the line, meaning they will never be >>played. I don't profess to know WHY they have a 0 priority, but they do. This, >>therefore, means that the program is behaving properly according to the book >>data. > >Chessmaster had the *white* pieces in this game, and played 1.e4 e5 1.f4. Then, >it is out of book after *black* played 2...d5. > >My understanding of 0 priority is, that a program may not play a certain move >itself, but still can access the best replies to it when it has the opposite >color. If not, it would fall into many traps which are incorrect, but difficult >to refute "over the board". Doesn't Chessmaster work like this too (otherwise, >it wouldn't make sense to include a variation up to the 12th move after a 0 >priority move at move 2, when it never gets accessed by both sides)? > >If I understand you correctly, than the zero prioritiy in this variation for >*white's" moves (or for all the moves *after* 2...d5), is wrong? I think this >should be fixed... the opposition could easily prepare for that, also the users >don't get the opposition of the usual quality, when the book ends after 2 moves. > >Regards, >M.Scheidl You are undoubtedly correct. There is no point in having a line continue for several moves if all of the moves have a priority of 0. However, I have no idea if the data is REALLY wrong or not. Perhaps the person who wrote the book initially added the moves for all lines with a priority of 0, then modified the moves that he thought should be played by increasing their priority. This information is lost to the four winds, because this book was created (by my best guess) about 6 years ago. This is what GM Larry Evans is going to be doing for the next version of the program -- going through all of the moves in the book and modifying their priorities, as well as adding moves in lines that are too short. However, there are no plans to fix these kinds of things in a patch for CM8000. The only lines that will be modified are ones that lead to obviously bad positions. jm
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.