Author: Harald Faber
Date: 05:18:29 07/10/01
Go up one level in this thread
On July 10, 2001 at 07:05:28, Uri Blass wrote: >On July 10, 2001 at 06:10:32, Harald Faber wrote: > >>On July 10, 2001 at 05:51:49, Fabio Barrettone wrote: >> >>>On July 10, 2001 at 04:41:36, Harald Faber wrote: >>> >>>However, I don't know why the people at Chessbase has given it a book that >>>doesn't suite at all its style of play. They have given J7 a book that reach >>>almost all the times simple positions and this is as to say "destroy me" with >>>engines as Chess Tiger considering the style of J7. >> >>AFAIk the book is by Boris Alterman and he is GM and member of the Junior team. >>Correct me if I'm wrong. So the book should certainly be better than the Tiger >>or Shredder-book which only have been assembled out of a database. > >1)I am going to be surprised if the Tiger and the Shredder books are really only >based on database. Shall I quote the answer of a member of the ChessBase team? I do not want to do that, it was a private e-mail. But you can be ensured that it is a trusted and experienced member and tester who wrote to me that the Shredder book was built from a small database with games of ELO >2450 or >2400. And in a German board someone who also has the ChessBase Tigers compared the two books, Shredder 5.32 and CB-Tiger, and found that they have the same book. >I know that the programmers are interested in being number 1 in the ssdf list so >I can expect them to give their program a better book than it and you do not >have to be a good player in order to do it in the first moves. I know, but maybe time was fading away? Remember that both programmers only have given their engines to the distributor. So my explanation for these books is a) they are not willing to build an appropriate opening book b) they didn't have the time c) they think that the engines are strong enough to compensate some lacks and the work to build an own book is not worth it. >Example:If you watch your program and find that it does relatively bad results >with black after 1.e4 e6 you can teach it not to play e6 against 1.e4 > >It can learn to do it from results in the ssdf games but it is going to lose >games and rating points in order to do it. You know that, I know that and many others know that too. But look e.g. at my tourney how well Shredder 5.32 did with this book. See me surprised too. >2)I do not believe that Alterman's book is better than other books. Who knows? I don't know how much and what kind of work he spent. If he only looked at some chess informant or so, then you are right, it may not be better. >You need to waste a lot of time in it and find opening that are good for Junior7 >and I doubt if Boris alterman did it. OK, we doubt it, what else can we guess? ;-) >I know that Junior lost one game against Fritz in WCCC thanks to Boris alterman. >He decided to go for a line that Junior did not understand and after the game >explained an idea that no computer that I know could find at least at that >time(I did not test it again). Now you understand what championships are good for, because such mistakes only appear in important championships. :-) I think the best method to find good and bad lines is to take a look at the games. Start where the book line ends and look what happens in the next 5-10 moves. If the -1.2 arises to a justified +0.2 then the line cannot be bad just because the engine came out of book with -1.2. And so on and so on. >Uri
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.