Re: [gnugo-devel] go-go joseki

[Arend Bayer]

Paul wrote:

> Gunnar Farneb�k wrote:
> > Paul wrote (on June 18):
> > > Here is a bunch of josekis for go-go point (5-5).  Prepared based on
> > > Kogo's joseki dictionary.  Revisions are welcome as my joseki skills
> > > are not too good.
> >
> > My 5-5 joseki skills are nonexistent but I think we should use it as
> > is; it can't very well be any worse than the current total lack of
> > knowledge.
> 
> So, uh, maybe Dan could commit it?  The patch involves changes to
> `Makefile.am'.

Oops, sorry this got lost for so long.

The patch below fixes the Makefile bugs, and also adds the approach
moves from your earlier patch (without them gogo.sgf doesn't do
anything).
I have added one more and devalued the hoshi approach move, as I believe
it is bad.

Arend

Index: patterns/Makefile.am
===================================================================
RCS file: /cvsroot/gnugo/gnugo/patterns/Makefile.am,v
retrieving revision 1.31
diff -u -p -r1.31 Makefile.am
--- patterns/Makefile.am        9 Sep 2005 22:40:39 -0000       1.31
+++ patterns/Makefile.am        9 Sep 2005 22:55:16 -0000
@@ -72,10 +72,10 @@ GGBUILTSOURCES = conn.c patterns.c apatt
                  fuseki9.c fuseki13.c fuseki19.c josekidb.c\
                 read_attack.c read_defend.c handipat.c oraclepat.c
 
-DBBUILT = gogo.sgf hoshi_keima.db hoshi_other.db komoku.db sansan.db \
+DBBUILT = gogo.db hoshi_keima.db hoshi_other.db komoku.db sansan.db \
          mokuhazushi.db takamoku.db
 
-DBBUILT_INPUT = -i gogo.sgf -i hoshi_keima.db -i hoshi_other.db -i komoku.db \
+DBBUILT_INPUT = -i gogo.db -i hoshi_keima.db -i hoshi_other.db -i komoku.db \
                -i sansan.db -i mokuhazushi.db -i takamoku.db
 
 DB_TO_TAG = aa_attackpats.db attack.db barriers.db conn.db defense.db\
Index: patterns/Makefile.in
===================================================================
RCS file: /cvsroot/gnugo/gnugo/patterns/Makefile.in,v
retrieving revision 1.50
diff -u -p -r1.50 Makefile.in
--- patterns/Makefile.in        9 Sep 2005 22:40:39 -0000       1.50
+++ patterns/Makefile.in        9 Sep 2005 22:55:16 -0000
@@ -149,11 +149,11 @@ GGBUILTSOURCES = conn.c patterns.c apatt
                 read_attack.c read_defend.c handipat.c oraclepat.c
 
 
-DBBUILT = gogo.sgf hoshi_keima.db hoshi_other.db komoku.db sansan.db \
+DBBUILT = gogo.db hoshi_keima.db hoshi_other.db komoku.db sansan.db \
          mokuhazushi.db takamoku.db
 
 
-DBBUILT_INPUT = -i gogo.sgf -i hoshi_keima.db -i hoshi_other.db -i komoku.db \
+DBBUILT_INPUT = -i gogo.db -i hoshi_keima.db -i hoshi_other.db -i komoku.db \
                -i sansan.db -i mokuhazushi.db -i takamoku.db
 
 
Index: patterns/fuseki.db
===================================================================
RCS file: /cvsroot/gnugo/gnugo/patterns/fuseki.db,v
retrieving revision 1.44
diff -u -p -r1.44 fuseki.db
--- patterns/fuseki.db  12 Jun 2005 09:34:14 -0000      1.44
+++ patterns/fuseki.db  9 Sep 2005 22:55:17 -0000
@@ -481,6 +481,74 @@ Pattern F18
 :8,FeEaj,shape(2.98)
 
 
+
+Pattern F19
+|..xxX     go-go invasion at san-san
+|....x
+|..*.x
+|.....
+|.....
++-----
+
+:/,FeEaj, shape(1)
+
+|..abX
+|....c
+|..*.d
+|.....
+|.....
++-----
+
+; x_alive_somewhere(a) + x_alive_somewhere(b)
+; + x_alive_somewhere(c) + x_alive_somewhere(d) <= 1
+
+
+Pattern F20
+
+|...????    go-go invasion at hoshi
+|.....??
+|....X.?
+|...*..?
+|.......
+|.......
+|.......
++-------
+# I think this is suboptimal, thus the worse shape. /ab
+
+:/,FeEaj, shape(0.94)
+
+
+Pattern F21
+
+|..?xx??    go-go invasion at komoku
+|.....??
+|....X.?
+|..*...?
+|......?
+|.......
+|.......
++-------
+
+:8,FeEaj, shape(1)
+
+
+Pattern F22
+
+|..o???
+|..oo??
+|......    5-3 approach move to go-go
+|......
+|..*.X.
+|......
+|......
+|......
+|......
++------
+
+:8,FEaj,shape(1)
+
+
+
 ###################
 # Corner enclosures
 ###################