Author: blass uri
Date: 02:06:45 10/22/99
Go up one level in this thread
On October 22, 1999 at 04:24:14, Paulo Soares wrote: >On October 22, 1999 at 02:55:23, blass uri wrote: > >>On October 21, 1999 at 23:33:20, Paulo Soares wrote: > >(snip) >>> >>>I don't know any chess program that is good in this type >>>of analysis. All they make the same mistakes that Fritz6 did in >>>the example that you presented. >> >>No >>Junior and Hiarcs does not suffer from the same problem >>The evaluation is not changed drastically after cxb6. >> >>Uri > >Uri, as James Walker says, as of habit, you are right. >What do you find of missing a little to leave the routine? >The doctors say that to leave the routine it is good for the >health (a joke). >I didn't evaluate the problem correctly, same because, >to be sincere, I didn't think of the drastic change of the >evaluation after cxb6. But I continue thinking, independent >of any thing, I don't know any program that makes good analyses >without a human's help . I agree that humans can help programs to do a better analysis but I think the team human+program will do better result with a program that you do not need to waste time for finding mistakes like the mistakes of Fritz6. The team is going to do better when the program can learn from previous search. Fritz does not know to learn from previous search but Hiarcs7.32 knows to do it and Junior6 will also know to do it so you can analyze with these programs and the program is going to improve its knowledge when you go back to the root position. 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.