Author: Uri Blass
Date: 05:32:43 08/20/03
Go up one level in this thread
On August 20, 2003 at 05:18:18, Jeroen Noomen wrote: >Hi all, > >I have just completed reading Christian Kongsted's book >'How to use computers to improve your chess'. I must say that >I like this book. It gives a lot of valuable information, not >only for those who want to improve his/her playing strength, >his/her chances of beating the silicon beast, but also for >programmers. > >Kongsted gives a lot of interesting positions in which computer >programs are not able to come up with the right evaluation. So I >think it is a very interesting manual for programmers as well. > >Of course I was very interested in what the book would say about >the program's opening books. Unfortunately this topic gets less >than two pages, so I was a bit disappointed by this fact. Also I >don't agree with Kongsted's evaluation of the opening books of >modern top programs. Let me quote some of them: > >1. Opening books are often made as the last thing before a release. > >This is not true. People like Alex Kure, Sandro Necchi, Dan Wulff >and me spent the whole year to improve it and to adjust it. Hundreds >of hours are spent to test the book and adjust it. The question is if there is some proof that your book is superior. The question is if the top programs do better against strong amateurs when both sides use book and not when the sides play the nunn match. Looking at the results of Deep Sjeng or Gandalf at WBEC I am not convinced about it. Based on results they seem to be not better than programs like Aristarch or Ruffian. I do not know about a proof that Ruffian and Aristarch are better than Deep Sjeng and Gandalf in the nunn match(the opposite and I know that Leo posted some results when Deep Sjeng beated Ruffian in the nunn2 match. I can also add that the book of movei was really small (few kbytes and there was a beta tester who helped me by copying few lines that movei did well in them but I believe that he did not play games against Deep Sjeng) Movei is also a significantly weaker engine than Deep Sjeng1.5 so I expected the worse from the match against Deep Sjeng and I was surprised when Movei could get 1.5/4 including a win in the line 1.e4 c6 2.d4 d6 I am sure that the books that are used in world championship are different but it seems that at least the book of the commercial product is not superior(I can add the fact that Kurt who tested Ruffian against the commercial programs when both sides use the same book found relatively better result for Ruffian when every program used its book) Uri
This page took 0.01 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.