The purpose of a handicap games is to allow a 50% chance of either contestant winning.
If one program loses 100% of games against another, it would take a vast improvement to advance to 90% or 80%. If the same two programs have an appropriate handicap, and a base of 50% odds, it should be easier to see the beneficial impact of various changes. Contrast "my winrate against a motley assortment of opponents, which changes month-to-month, is 3% higher than yours" with "my program can give yours two stones, and still win 50% of the games." Programs do not care, but their human designers might have an incentive to find a way to leapfrog over their opponents. Terry McIntyre <terrymcint...@yahoo.com> Any system of entrusting the government to judge and correct its own abuses is the same as appointing the accused criminal as his own judge and jury: don't expect many convictions. -- Allen Thornton, Laws of the Jungle ________________________________ From: Christoph Birk <b...@ociw.edu> To: computer-go <computer-go@computer-go.org> Sent: Friday, June 5, 2009 2:59:07 PM Subject: Re: [computer-go] New CGOS On Fri, 5 Jun 2009, Don Dailey wrote: > Handicap games opens a can of worms. The last time we discussed it, > it was difficult to get any kind of reasonable agreement on how to do it. Handicap games are for humans ... they get frustrated losing over and over. Computers have no problems with that. I agree with Don, adding a handicap server is an unnecessary complication. Christoph _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
_______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/