Author: stuart taylor
Date: 17:48:54 09/27/01
Go up one level in this thread
On September 25, 2001 at 19:48:21, Dann Corbit wrote: >On September 25, 2001 at 19:44:26, stuart taylor wrote: > >>On September 25, 2001 at 19:30:16, Dann Corbit wrote: >> >>>On September 25, 2001 at 18:15:40, john c cook wrote: >>> >>>> junior 7 beat chess tiger 14 10 5 5 on two 1.4 ghz Amd >>>>but on my old 800mhz piii chess tiger 14 beat junior 7 8 5 3 >>>>all game at 40/2hr 144 meg hash for junior 192 meg hash for chess tiger >>>>on the piii i belive chess tiger 14 was the best it beat all my other prorgam >>>>at 40/2hr it look to me that junior 7 on a fast computer see more but i like >>>>the way chess tiger play chess has any one else come up the same way >>> >>>Sorry, but your conclusion does not follow from the data presented. >>>It might be true, but you certainly don't present convincing evidence. >>> >>>With a few hundred games, you might have a better argument. >> >>It's OK for him to say it anyway, because he ALSO gives the details from which >>we can understand whatever we want. >> I'm inclined to observe indications of things, even when there is no conclusive >>evidence yet. > >Keep in mind that I did not say he was *wrong* -- only that he does not have >enough data to reach a firm and sound conclusion. > >I am aware that many people are ready to judge the strength of a program on >limitied data. I have seen claims to judge strength based not only on a single >game, but on a single move! >;-) > >Does not mean that those who make such predictions are wrong either. Only that >it is not mathematically convincing yet. Whether or not a person feels evidence >is compelling is up to the individual. I suppose I can appreaciate that the natural feelings of a mathematician would be more strictly mathematical. 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.