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

BGonline.org Forums

Prime Time revisited at score with cube

Posted By: neilkaz
Date: Thursday, 24 December 2009, at 12:28 a.m.

In Response To: Prime Time revisited at score with cube (Nack Ballard)

Nack, with all due respect, I find Snowie to be a nearly useless tool for studying such a match score oriented decision. My discussion will only concern the top two plays.

Snowie doesn't do checker play or cube action according to match score in its rollouts. This is position is very match score oriented since the trailer will rewhip to 4 as soon as anything might possibly go favorably for him. However, that most certainly isn't what is happening in the RO that Snowie is playing according to money. Additionally the trailer won't care if he is then G'd on 4.

Note that the live cube values of the two plays are basically tied, yet Snowie's adjustment has one play much better than the other. My experience indicates that "$2 + that adjustment can get you a cup of coffee at Starbucks"..ie Snowie's adjustment of cubeless results is absolutely worthless for a lopsided match score situation like this.

Also, we'd find it nice if you can export from Snowie to show us the winning and G chances here for both sides. (Someone can tell you how to do so)

Perhaps Stick can extend his GNU RO to at least 1296, however, Ken can shed more light here if he does a RO on a higher setting like XG 4 ply. This isn't that long a game and many plays will be simple, so I suspect Ken could bang out several thousand or more overnight on 4 ply. (Forgive me as my comps are both tied up)

... neilkaz ...again stressing that unfortunately Snowie RO results can't be trusted here.

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.