Author: Rolf Tueschen
Date: 12:58:11 06/01/02
Go up one level in this thread
On May 31, 2002 at 22:13:55, Dann Corbit wrote: >Rolf raised an interesting point. To him, the SSDF list is not very interesting >unless you can be sure of the connection to human strength. > >From: >http://home.interact.se/~w100107/welcome.htm > >We see that there are now: >14,378 games in the SSDF database. > >If you could convince GM's to play for $100/game with full concentration, that >would amount to $1,437,800.00 > >I suspect that GM's get 5-10x that amount (depending upon their ability). > >If we could convince them to play for $350.00 per game, it would cost about five >million dollars. $350 is not a lot of money, but with a huge number of games as >a potential, I suspect that a lot of GM's could be talked into it because it >would be a source of steady income. We could have bonus dollars for wins and a >lesser amount for losses to make sure that they were trying really hard. > >Actually, it's not as bad as I thought. I wonder if the chess program >manufacturers might want to cough up some of that cash. With 5 million dollars I can't see a real problem. They pay Kramnik 1 million for 8 PR games against Deep Fritz. Where is your problem? But if you had followed my thought experiment, then you could make it this way and you save some precious money. Give 5 top GM 100 000 dollars. They will create the new anti-computerchess and whoopie the case is solved. Computerchess is solved. Bye-bye, the 2700 Elo numbers, down to earth again at 2350! But I bet 100 dollars that the companies will prefer the other way round, say, the 1 million for eight funny games PR. With many draws and perhaps 1 loss for Deep Fritz! Now this is looking like 2750! Rolf Tueschen
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.