[computer-go] cgos komi (was: Odd experimental results)
Jacques Basaldúa
jacques at dybot.com
Wed Sep 27 03:09:43 PDT 2006
David Doshay, wrote:
> komi on CGOS is 7.5
>
> RANGE GAMES BLACK PERC WHITE PERC
> 600 - 799 17546 41.47 % 58.53 %
> 800 - 999 10883 39.36 % 60.64 %
> 1000 - 1199 1515 43.37 % 56.63 %
> 1200 - 1399 5542 43.79 % 56.21 %
> 1400 - 1599 9774 48.23 % 51.77 %
> 1600 - 1799 14706 49.37 % 50.63 %
Let me make a guess: Komi is compensation for white because
she does not make the first move. If players are skilled
moves have a high value, if players are not, moves have much
lower value. So what the table above shows is that komi
7.5 *is correct*. The asymptotic behavior towards 50% as ELO
increases is what proper komi should produce.
Jacques.
More information about the computer-go
mailing list