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

BGonline.org Forums

Pair of moves???

Posted By: Maik Stiebler
Date: Friday, 29 June 2012, at 11:09 p.m.

In Response To: Pair of moves??? (Timothy Chow)

The main difference there, I think, is that GNU excludes "trivial" decisions. If it were as you say, then I think the GNU rating would be more than twice the Snowie ER, which it typically isn't.

I don't think so. Note that gnubg ER does count (non-trivial) cube decisions in the denominator, while Snowie only counts checker plays (and dances).

Ok, I'm not sure and it does seem that Daniel agrees with you. So here's a test game (sorry about my nactation inability, the 42 is played split+cover 24/22 9/5)

41$-32Z-42?-55K-F-D-P

According to gnubg there is one "real" cube decision for each side. There are two checker play decisions and one forced move for player 1, and two checker play decisions, no forced move for player 2. On 2-ply analysis, total EMG equity loss is 0.136 for player 1 and 0.029 for player 2.

What is the gnubg ER and the Snowie ER for both players?

Including forced moves, there is a total of five checker plays, so Snowie ER is 27.2 for player 1 and 5.8 for player 2.

Number of decisions is 1(cube)+2(checker play)=3 for player 1 resulting in a gnubg ER of 45.3. For player 2 it's 1+2=3 => ER=9.7.

Apart from rounding errors, my results are consistent with gnubg's output, which does not prove much, but yes, I'm confident that I got it right.

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.