Author: Mark Rawlings
Date: 15:34:55 03/08/99
Go up one level in this thread
On March 08, 1999 at 18:07:24, Thoralf Karlsson wrote: >On March 08, 1999 at 14:12:28, José de Jesús García Ruvalcaba wrote: > >>On March 06, 1999 at 16:48:28, Lin Harper wrote: >> >>>Since the vast majority of the rated games for the SSDF rating list are >>>comp-comp games, we can assume there will be no rating changes with faster >>>processors because they all benefit equally (well, almost equally). >> >> ¿Vast majority? I think *all* the SSDF games are comp-comp games (if I am wrong >>I would like to be corrected by an authoritative source). > > >The *level* of the list is based on 337 games played against humans in serious >tournaments between 1987 and 1991. Just a thought... I wonder if it's time to "recalibrate" the list? I assume by now we have a lot more human-comp data. > >The relative ratings are based only on comp-comp games, played on the tournament >levels. > > >> If you get a faster processor, the same program will (most likely) get a higher >>rating. I doubt SSDF will rate the same program on two different processors, but >>technically it is posible (they would be separate entries with different >>ratings). > >When we change hardware we use to play 5-6 programs on the old as well as on the >new hardware, in order to get information about how many rating points the >hardware in average gives. Hopefully we can do so next time also. > >Thoralf Karlsson
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.