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

BGonline.org Forums

Rollouts of Robertie's 501 problems: Pos 129 Snowie Eval

Posted By: Timothy Chow
Date: Thursday, 25 March 2010, at 7:02 p.m.

In Response To: Rollouts of Robertie's 501 problems: Pos 129 Snowie Eval (Chris Bray)

I agree that the variance looks too large here to trust. I don't own XG but I'd prefer a minimum of 5184 games, at 4-ply/4-ply if possible. At the risk of sounding like a broken record, I will point out that the confidence intervals reported by the bots are only estimates based on certain assumptions about Gaussian distributions, independence, etc., that are at best true asymptotically as the number of trials gets very large. In layman's terms, I think claims of "99% confidence" and so forth should be taken with a grain of salt.

But at least XG's absolute equities are in the same ballpark as GNU's (0.800 to 0.900); the same cannot be said about Snowie's evals (0.500 to 0.600).

As for Chuck's question, GNU 2-ply certainly believes (correctly, I think) that cubing is called for after either double hit. I haven't checked what it thinks about what to do after other plays.

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.