Re: [gnugo-devel] Hoshi fixes

[Evan Berggren Daniel]

On Sun, 27 Jul 2003 address@hidden wrote:

>
> > My hoshi patch (the second of the two, not yet on the devel page) does
> > exactly that.
>
> The patch makes this an antisuji:
...
> and replaces it with the shoulder hit:

The jump should probably not be marked antisuji, assuming we remove or
change the pattern that marks it J, as discussed elsewhere.

> The problem with the one space jump is that it provokes X
> to make territory. In exchange O gains some power in the
> center, and has to recoup the immediate territorial loss
> later, for example by attacking the hasami stone.
>
> There is no question that the shoulder hit is a better
> move in many game positions.
>
> The file after Evan's patch contains two variations.
> But black (X) has MANY ways of varying the order of
> moves. Black can for example push up immediately
> instead of the two moves in the revised hoshi_keima.sgf.
> In the file joseki03.sgf this would be at M4 and gives this position:
>
>  8 . . . . . . . . . . . . . . . . . . . 8
>  7 . . . . . . . . . . . . . . . . . . . 7
>  6 . . . . . . . . . . . . . . . X . . . 6
>  5 . . . . . . . . . . . . . O . . . . . 5
>  4 . . . X . . . . . + O X . . . X . . . 4
>  3 . . . . . . . . . . . X . O . . . . . 3
>  2 . . . . . . . . . . . . . . . . . . . 2
>  1 . . . . . . . . . . . . . . . . . . . 1
>    A B C D E F G H J K L M N O P Q R S T
>
> Of course W must play M5 and be strong enough to
> handle the cut at L5 if B plays it immediately.
>
> But after Evan's patch, W plays F3, C14 or elsewhere.
> This is of course disastrous.

I can add some variations of move reoderings and the like.

Evan