Author: Uri Blass
Date: 16:00:14 07/24/05
Go up one level in this thread
On July 24, 2005 at 07:19:51, Gian-Carlo Pascutto wrote: >On July 24, 2005 at 06:15:54, George Tsavdaris wrote: > >> >>http://www.chesscafe.com/mig/mig.htm > >The strongest program (Shredder) is no longer sold by ChessBase --> time to put >the propaganda machine in action. > >So, now, beating other programs is not important anymore and if Fritz gets >clobbered by an 180k free program this is because it's 'smarter'! It's because >it has so much 'knowledge'! > >The new Fritz is not or almost not stronger than the old one where you payed >money for? No really, you can 'feel' it's playing more 'humanlike' instead! > >Also handy that the claim of being stronger against humans is almost impossible >to objectively verify. > >The really sickening thing is that people will eat this ChessBase propaganda by >the boatloads and report is as facts. > >-- >GCP I think that you missed the following paragraph when they claim that Fritz does well against other programs and Methias Feist was sure that more knowledge would also make it strong against other programs: "The danger was that if you stop tuning to play computers, you are going to lose against them; at least that’s what we were concerned about. Everyone was worried about that, except for one person, Fritz co-programmer Mathias Feist. He was sure that more knowledge would also make it strong against other programs. Mathias has been proven right. Since we’ve switched to the new data structure it’s still playing pretty well against other programs." I think that every intelligent person understand that more knowledge can help programs to play better against other programs. One of the ways to make programs stronger against other programs is simply by improving the evaluation and most of the improvement in fruit from Fruit1.0 to Fruit2.1 based on Fabien is thanks to improvement in the evaluation. If somebody tell you that having more knowledge in the evaluation is not productive against computers he does not know what he is talking about. The problem is of course what knowledge to add because it is easy to add counter productive knowledge but I have no doubt that adding correct knowledge in the evaluation is one of the ways for improvement. 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.