gnugo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gnugo-devel] trevor_1_14.5


From: Daniel Bump
Subject: Re: [gnugo-devel] trevor_1_14.5
Date: Sat, 17 Nov 2001 12:29:57 -0800

I put up a patch owl_1_14.1 which addresses comments by Gunnar and
Trevor about patterns to be removed. Since this is a step in the
direction Gunnar and Trevor are advocating I assume it's not
controversial and added it to the CVS.

> A1: Gunnar's defended already - thanks, so perhaps it can be salvaged.

Added as a constraint that the O stone on the 4th line has 3 liberties
or less. If this is true escape is a real possibility and the solid
connection seems probably best.

> A303: Could be useful, though one space to the left of a is prob. better.
> A304: B should never try *, so no need to defend it.

OK, the patch removes A304.

> A1004: I couldn't see why this made sense.

Sorry, it was A1004 not A1001 that was removed earlier.

> D211:  I saw the following match:

I modified the pattern so that match can't occur:

|..O
|..O
|.OX
|.*x
|.x?

The comment (try before hanging connection) and pattern value 
suggests why the pattern is there---I believe this came from 
a case where the owl code found a non optimal defense, a hanging
connection on the first line where popping out into the open is 
obviously bigger.

> constructed.  We could debate the relative merits of various patterns 
> for quite a while.  It seems to me more productive to let the test 
> cases speak for themselves.

For the most part I agree. But referring, for example, to A1, there
are probably a few owl patterns for which there apparently isn't a
test case in the db but which are obviously needed.

Dan






reply via email to

[Prev in Thread] Current Thread [Next in Thread]