Author: Chris Carson
Date: 06:49:35 02/09/03
Go up one level in this thread
On February 09, 2003 at 09:04:30, Sune Larsson wrote: >On February 09, 2003 at 08:36:33, stuart taylor wrote: > >>On February 09, 2003 at 08:14:36, Sune Larsson wrote: >> >>>On February 09, 2003 at 06:50:10, stuart taylor wrote: >>> >>>>On February 08, 2003 at 21:39:20, Sune Larsson wrote: >>>> >>>>>On February 08, 2003 at 20:50:54, stuart taylor wrote: >>>>> >>>>>>E.g. Smirin and another who has best chance against machine should play, perhaps >>>>>>3 vs DF and 3 vs DJ each, then, if no one gets much improvement on Kasparov and >>>>>>Kramniks results, then we will know that the competition is over, and that man >>>>>>does not dominate machine in chess. >>>>>>S.Taylor >>>>> >>>>> >>>>> How many games needed for the result to be statistically valid ? >>>>> >>>>> /s >>> >>>>It's not that system. We just want to know if there is still plenty of life left >>>>in human chess, which computers cannot yet overcome, and that this life still >>>>CAN result in the humiliation of computers in a show exhibition. >>> >>>>S.Taylor >>> >>> >>> Aha, I was just about to write some words like "of course there is...", when >>> I got to the final two words in your answer - *show exhibition*... Everything >>> is then revolving around marketing, media interests and steering up people >>> in order to make them buy some products. For those chess players involved, >>> it's very easy money of course. And the results will be an effect of the most >>> vital thing here: To make the wheels keep on spinning - the show must go on... >>> Offer $ 10.000, for a 6-game match, to IM/GM:s > 2450 ELO and your mailboxes >>> will be overflooded and tele-lines blocked... ;-) >>> >>> /s >> > >>I'm interested in what "of course there is" evidence that computers are still >>greatly lacking in what it takes to equal the best of humans. > > > This was just my very personal opinion based on games/positions between > computers-humans from the last two years. I think it would be possible > to get evidence by arranging matches like those I mentioned above. > 6-10 games, classical time level, $10-15.000 for the human if he/she wins > the match - otherwise just expenses paid - maybe a starting fee etc. This > will produce a lineup of highly motivated IM/GM:s > 2450 ELO. > I like this idea. But then again, I like Human vs Human and Human vs Comp and Comp vs Comp games. How about offering $X's for any Human to win a Y game match vs the prog/hw? This would have to be better defined, but the idea is good. Also, it could be at any Human vs Human event, just have the prog/hw there and let the games begin. My guess would be that the computer would win/draw the vast majority, but a few players could win a short match. Another idea might be to just let the computer be in any Human event. The computer win/draws/loss count. Seems to solve all motivation issues. Ofcourse Chess Tiger type wins by the comps would be common even at events like Dortmund these days. As a side note, I wonder how many chess players would put up a fee (refunded if you win) in case of a draw or loss? One's own money is a big motivator.
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.