[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Bot evals (XGR bug: no, but surely need to be improved)

Posted By: Matt Cohn-Geier
Date: Wednesday, 14 December 2011, at 9:31 p.m.

In Response To: Bot evals (XGR bug: no, but surely need to be improved) (eXtreme Gammon)

Being off in terms of absolute equity is understandable; however, in position #2 it's WAY off. It's so far off that it actually values position #2 as significantly better for White than position #1. However, even 1-ply and 2-ply correctly evaluate position #2 as a RD/P.

1-ply is indeed breaking the 6pt to hit loose in #2, which explains why it values that position as better than #1. But 2-ply doesn't break the 6pt with aces and appears perfectly competent of playing it out. 2-ply still makes some errors, but it also makes errors in position #1, and the kind of errors that it makes in either position are nothing that would indicate a .800 swing in equity. So, I can understand why XGR and XGR+ would be off here. But still puzzled as to why XGR++ is so far off the mark.

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.