Author: Matt Frank
Date: 15:20:03 01/29/99
Go up one level in this thread
> I don't think we're that far apart on agreement that at 500 pts defferential > the lower rated player winning is very rare at best but what about computers? > This is the crux of the question and the curiosity that prompted the original >thread to begin with . Perhaps Enrique or some other testers could shed light on > this? Does the rating differential lessen with computers as suggested in an >earlier thread to say around 375 or does it indeed correlate to the 500pt.range. It is well known that the bigger the difference in ratings the wider the margin of error expected, that doesn't mean though that if an infinite number of games could be played that the results would be any diifferent however (whether computer vs. computer or human vx. human). In other words if there is a true, consistent, and stable rating difference of 500 points the results will be very close to 19-1. And again, SSDF or Selective Search may have such data at their disposal although they do try to avoid contests between hardware/software configs with greater than a couple of 100 pts difference. Enrique may very well have data on this. Matt Frank
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.