Author: stuart taylor
Date: 18:28:58 04/23/01
Go up one level in this thread
On April 23, 2001 at 13:09:14, Jorge Pichard wrote: >On April 23, 2001 at 11:20:24, stuart taylor wrote: > >>On April 23, 2001 at 08:10:59, Vincent Lejeune wrote: >> >>>On April 23, 2001 at 06:37:11, Daniel Chancey wrote: >>> >>>>Is this the biggest margain of victory ever in the SSDF matches? I'll bet that >>>>a Chessmaster personality that did well in earlier games against Deep Fritz will >>>>score 16-24 or mabye 20-20. >>>> >>>>Castle2000 >>> >>>IMHO, The main problem is not the personality, but is that CM8000 doesn't learn >>>, that means that Deep Fritz try to repeat opening where it play better than >>>CM8000 >> >>If that's the case,I'm not too bothered by the results, as the actual >>calculation and tactics and knowledge will still yield about as good as any >>computer program, for personal use, which is what I want it for. I mean, I don't >>need a program for the purposses of it entering a tournament, through my >>auspicies. >>As far as I'm concerned, therefore, it still might be the strongest program on >>earth at the present time, only that it repeats its few mistakes and loses to a >>machine which repeats its few victorious lines. Then if you correct that little >>problem, and run the test again, it might beat Fritz by a large margin. >>S.Taylor > >Sorry to tell you this, but if you look at all the games no more than 10 were >repeated games, in which CM8000 at the most would have won probably 4 more. How >does 11.5 to 28.5 look still Pretty Bad. That would be relatively decent. It only has to be the strongest CM so far, as far as I'm concerned, which would automatically make it quite competitive against anyone or thing. S.Taylor
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.