Author: Robert Hyatt
Date: 15:11:35 11/05/97
Go up one level in this thread
On November 05, 1997 at 10:42:21, Amir Ban wrote: >On October 30, 1997 at 13:05:13, Robert Hyatt wrote: > > > >>4. stupid rule giving slower machines 10 extra minutes. Can't someone >>figure out that this doesn't work *if* everyone is thinking on the >>opponent's >>time? If machine equality is desired, make it a rule. Don't fudge >>things in >>a way that doesn't work. >> > >This was not about machine equality. It was to correct the wrong that >was inflicted on those who came to Paris with a promise of a K6/233 to >discover a K6/200 on-site. Nobody thought the Alphas should be >penalized, but even those who had an Alpha 767 were entitled to a >performance advantage relative to a K6/233, not to a K6/200. > >Amir I don't disagree with you, I simply believe, however, that the basic premise is flawed in a basic way. With thinking on the opponent's time, this extra 10 minutes is not 10 extra minutes. It is much less, in reality, when I can use it to think about a response to your move. If I'm right 50% of the time, then 50% of your "longer searches" are totally negated, because I used your extra time as well... It would have made just as much sense to shorten the alpha times by 10 minutes, but would have produced the same net result... there's not a good way to handicap like this, unless thinking on the opponent's time is disallowed... Perhaps a better solution would have been to install a process on each alpha that computed for 1 second and then slept for 9 seconds, slowing the machine down by a factor of 10% roughly...
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.