Author: stuart taylor
Date: 00:45:52 09/24/02
Go up one level in this thread
On September 22, 2002 at 20:22:50, stuart taylor wrote: >On September 22, 2002 at 20:08:30, Timothy J. Frohlick wrote: > >>Chessfun's games tell us that CT 15 is probably better than CT 14. They are >>really too small a sample for any conclusions as to real strength. >The fact that Tiger never failed to get 50% against all comers, and it was >strictly 10 games against each, no excuses, and they were (presumably) the FIRST >10 games without choosing and cheating, all this indicates that Tiger 15 is >every thing you could hope for-no weaknesses which other programs don't have. >So by buying that program you have the best of everything to date! >And, I would say that intuitively mathematically, it may be even higher in elo >than estimated at present. At any rate, it looks like it may well be ballanced, >and the most authentic chess so far in computers. >>We thank her >>for her efforts. >> >>Tim >> >>If I have Junior 6a and CT 15 won three and lost three with the remainder of >>games being draws then why should I buy CT 15 other than liking Christophe >>Theron? Ans. Because it's there! >It would probably end up with a plus score, and even if it remains even, the >playing quality of Tiger might be seen to be the superior one. Since it's good >against all the other engines too, that may signify that it has much better >taste to it, and may even be best against humans, as it seems to have quite a >wide range of expertise regarding computers. >Or was there something missed out? (Junior 8?) >S.Taylor Again, It's a shame that none can ever comment (intelligently or at all) on this kind of mathematical estimation, which may be much more revealing even than ssdf ratings, and with less games too. 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.