On Mon, Oct 11, 2010 at 8:21 PM, Adrian Petrescu <[email protected]> wrote:
> I don't think it's unreasonable. Look at the ever-growing gap between the > top chess players and the top chess computers, and consider: > > a) There's no reason to believe the strongest chess computers are as good > as "God" yet. The low draw rate in computer chess tournaments supports this. > I think computers are still quite far away from perfect play in chess. By analogy, a far far simpler game is checkers and computers are just now approaching perfect play but still not quite there in actual over the board play. In other words they are still losing games to other checkers programs. However checkers is ridiculously simple compared to chess and the same could be said for chess as compared to go. I believe to a perfect chess player the top humans or computers would look like frequently blundering fools. > b) Top chess players only come as close as they do because they themselves > train using computers, a luxury that Go professionals do not yet have. > > So, imagine the gap between today's top chess computer and the best * > pre-computer* player, and add to that the gap between the computer and > "God". The result is a first approximation of how far away Go professionals > are from God, if you accept the assumption that Go professionals are as good > at Go as Chess professionals were at Chess before computers ;) > > Of course, this is all being silly... > - Adrian > > On Mon, Oct 11, 2010 at 5:12 PM, Darren Cook <[email protected]> wrote: > >> > Rin Kaiho also said a long time ago that he would need 4 stones from >> > god. So kudos to Rin. >> > >> >> There was a study about 10 or 15 years ago that used the measured >> >> variance >> >> in score to extrapolate perfect play (with zero variance), and it got 4 >> >> stones better than the top pros. That's where this estimate comes >> from. >> >> So, this is actual handicap stones, not pro ranks? So "God" is not 12- >> or 13-dan pro, but something like 21-dan pro? (assuming 3 pro ranks per >> handicap stone; 28-dan pro if 4 ranks per handicap stone) >> >> That's a lot of mistakes Rin thinks the top pros are making. >> >> Darren >> _______________________________________________ >> Computer-go mailing list >> [email protected] >> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go >> > > > _______________________________________________ > Computer-go mailing list > [email protected] > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go >
_______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
