|
BGonline.org Forums
PR xg-gnu-snowie-bgblitz
Posted By: Daniel Murphy In Response To: PR xg-gnu-snowie-bgblitz (Michael Petch)
Date: Wednesday, 29 January 2014, at 6:03 p.m.
Yeah, the 1.4x figure is in the manual, based on somebody's study of 300 matches. But there's discussion in the gnubg buglist about that, with Herr Anton mentioning that 1.4x seemed wrong, and 2x seemed more correct.
XG PR should on average be slightly higher (that is, worse) than Snowie ER, because, among other differences, XG counts errors per unforced move, instead of all moves. And XG PR should be very close to GNUBG ER divided by 2, since XG and GNUBG calculate error rates in nearly the same way.
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.