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

BGonline.org Forums

Don't understand this cube problem

Posted By: Timothy Chow
Date: Sunday, 6 November 2011, at 10:29 p.m.

In Response To: Don't understand this cube problem (mtuhtan)

I think this should be a take. Not sure about the double. Blue still has work to do to escape his checkers from behind White's five-prime. Some of Blue's good rolls are duplicated (e.g., 6's to make a point or escape, 1's to step up to the edge of the prime or make a point). White does lose quite a lot of gammons but I think she wins enough to compensate.

How do you analyze positions like this one?

This may be a good opportunity for me to elaborate on why I'm not a huge fan of playing out positions over and over again. The first point is that it's time-consuming, and the second point is that if you are playing it out without a specific purpose in mind, you may still be left in the dark.

Here are the steps I'd take in analyzing a position like this one.

1. I'd use GNU's temperature map (does XG2 have something similar?) to make sure I understand how to play all of Blue's immediate rolls. If something seems surprising, I'd edit the position manually to "make" the recommended move and then examine White's temperature map.

2. Next, I would explicitly articulate my own theory about the position that led me to my evaluation. Say for instance that I think it's barely a double but the computer says it's a massive double. I'd then try to list potential market losers that I may have overlooked. Of course I use the bot to see if they're really market losers. Or say that I think that White wins a lot because of her prime. I would then try varying the prime (e.g., repositioning the spares, moving it forwards or backwards a pip, changing the locations of Blue's checkers behind it) to see how this affects the valuation. If changing the prime makes very little difference then I try to come up with a new theory. Maybe the prime isn't important because White keeps dancing too long for it to matter. So then I try weakening Blue's front position to see how that affects things.

The basic principle is to use a crude version of the "scientific method." Formulate a hypothesis. Make a prediction. Test it with the bot. Change your hypothesis if necessary and repeat. This way you're actively engaging with the position and gathering information with a purpose. This is much better than rolling out (interactively or not) without any clear goal in mind.

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.