Author: Robert Hyatt
Date: 14:48:39 01/07/00
Go up one level in this thread
On January 07, 2000 at 12:54:47, Bertil Eklund wrote: >On January 07, 2000 at 12:20:53, Robert Hyatt wrote: > >>On January 07, 2000 at 10:15:10, Graham Laight wrote: >> >>>On January 07, 2000 at 05:35:35, Enrique Irazoqui wrote: >>> >>>>Tiger beat Century 12-8, +6 -2 =12. Games will be posted this evening on >>>>http://www.computerschach.de/tourn/ >>>> >>>>Enrique >>> >>>Congratulations to Tiger for an excellent score! >>> >>>This result is bad news for those who wish to put forward the argument that the >>>Rebel Century vs GM series gives us a good picture of the strength of today's >>>top programs, I'm afraid. >>> >>>-g >> >> >>And why is that? 12-8 is not a huge victory Elo-wise. it is the same as a 3-2 >>victory which is barely 50 Elo points different. And the next question is how >>does this affect computer vs human discussions? I point back to several >>'computer killers' over the years, that were very strong against computers, >>but which were terribly weak against humans. Lexentrique from the late 70's >>comes to mind as one. >> >>We will see Tiger vs two humans soon in Ed's 6 game match vs humans. That >>will tell us more. > >Hi! > >It´s 72 points and not barely 50. >Yes of course TWO MATCH-games tell us more than 20 games against Tiger. > My math was bad and you are correct. However, as to the last part of your post, if you _really_ believe that 20 games vs a computer gives _any_ indication of how a program will do against a _human_, I have some prime ocean-front property in Wyoming I'd like to sell you. Computer vs Computer has little to do with predicting computer vs human. _very_ little. You only have to log on to ICC, use the "search" command, to see this very clearly. >Bertil
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.