gnugo-devel
[Top][All Lists]
Advanced

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

Re: [gnugo-devel] Nonjoseki


From: Gunnar Farnebäck
Subject: Re: [gnugo-devel] Nonjoseki
Date: Thu, 29 Jul 2004 05:56:07 +0200
User-agent: EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/21.3 (sparc-sun-solaris2.9) MULE/5.0 (SAKAKI)

Arend wrote:
> > Lower left corner:
> > D4 is not joseki but I suppose E5 is an acceptable response. After
> > black's tenuki G4 seems too mild. I would have played E2 to try to
> > force black into bad shape.
> 
> In fact, the correct answer is E2 immediately.

Do you mean at move 8 with immediately, or are you still at move 10?

> >
> > Upper left corner:
> > K17 is just too wide to be considered a pincer. I don't think D17 is
> > joseki but maybe it's playable. The result after C15 and tenuki does
> > not look good. Is there any better alternative to C15?
> 
> C18 is mandatory after after D18. Things get a little complicated after
> Black B18 then.

The joseki database would be happy to have some of those moves. :-)

> C15 is a bit of a desaster, and after having played it, White should
> at least settle the shape with D14 in sente before tenuki.

That's what I thought too.

Below follow some test cases from later in the same game. These can
also use some reviewing.

# K4 is overconcentrated. Which moves more than D14 are good here?
loadsgf games/olympiad2004/19x19/int-gnu.sgf 42
??? reg_genmove black
#? [!K4]

# With the black stone at R9, R6 is locally better than Q6.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 46
??? restricted_genmove black Q6 R6
#? [R6]

# P3 does not look urgent here.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 50
??? reg_genmove black
#? [!P3]

# First live on the left edge.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 60
??? reg_genmove black
#? [C10]

# B10 not sufficient to live.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 62
??? reg_genmove black
#? [!B10]

# B11 not sufficient to live.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 64
??? reg_genmove black
#? [!B11]

# T13 is an overplay. T14 works better and also solves the problem
# with bad aji in the upper right corner.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 96
??? reg_genmove black
#? [T14]

# Q10 does not make much good. More interesting to play some move in
# the center.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 104
??? reg_genmove black
#? [!Q10]

# K19 is an overplay. After black M18 white has to abandon K19 or risk
# a ko in the corner.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 110
??? restricted_genmove black K19 M18
#? [M18]

# T18 dies right out if white finds the killing move at R19. P19 and
# R19 both live, but K19 and M19 must be sacrificed. N19 lives with
# everything except T17 and possibly T19, which are less important.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 114
??? reg_genmove black
#? [N19]

# Owl test to be placed in owl1.tst. Features an eye space with a
# cutting point. 
loadsgf games/olympiad2004/19x19/int-gnu.sgf 115
??? owl_attack T19
#? [1 R19]

# M3 is not necessary. Blocking in the center is big.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 132
??? reg_genmove black
#? [H8]

# G10 is locally clearly better than G9 as it threatens to cut.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 162
??? restricted_genmove black G9 G10
#? [G10]

# Can't just give up the ko. T16 and L17 should be big enough ko
# threats. G5 might be barely big enough as well.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 170
??? reg_genmove black
#? [T16|L17|M17|G5]

# O15 is bigger than H4.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 178
??? restricted_genmove black H4 O15
#? [O15]

# N13 is worthless.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 194
??? reg_genmove black
#? [!N13]

# H6 is completely worthless. The other atari is clearly better.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 196
??? restricted_genmove black H6 G5
#? [G5]

# The ko can at least be captured once before connecting.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 202
??? restricted_genmove black C6 E7
#? [C6]

# We are ahead. Even if owl misreads, the thrashing dragon
# suppression should generate a move at L3 or K3.
loadsgf games/olympiad2004/19x19/int-gnu.sgf 242
??? reg_genmove black
#? [K3|L3|L1]

/Gunnar




reply via email to

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