Author: David Blackman
Date: 05:54:47 06/25/99
Go up one level in this thread
On June 25, 1999 at 08:08:20, Leon Stancliff wrote: >The top ten humans in the U.S.A. average 2595 on the Fide scale. > > > >The top ten micro programs average 2550 on the Hallsworth Selective Search scale >when operating at 200 mhz. We would expect at least a 50 point increase when >operating at 400 mhz. Therefore, the micros would average 2600. Is there any reason to believe the Hallsworth Selective Search scale and the Fide scale are the same? Or even close? Games between computers and GMs under proper competitive conditions are not all that common. I would guess the Hallsworth Selective search was set up using a small sample of games played a long time ago. (I know this is the case for SSDF.) So both rating systems have had time to inflate or deflate and we probably have no idea at all how they correlate today. I wouldn't be surprised if the two systems are out of sync by 200 points or more. And i wouldn't like to guess which way.
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.