Tooveli wrote:Re: 67
$$Bcm61 Prisoners: white 2 black 0
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . O . . . . |
$$ | . . . . . . . . . . X . . O . O . . . |
$$ | . X O O . . . . . . . . O O . O O X . |
$$ | . . X , O . . O . X . X . . O X X . . |
$$ | . X . . . . . O X . . . . X X . . X . |
$$ | . . . X . . . . X . . O . . O . . . . |
$$ | . . . . X . O . . . . . . . . X . . . |
$$ | . . Q . . . . . X . . X . O . X . . . |
$$ | . . . . . . . . . X X O O . . . . . . |
$$ | . . . , . . . . . , O X . . . , . . . |
$$ | . . . Q . . . . . . X X O O . . . . . |
$$ | . . . . . . . . . . . a . . . . . . . |
$$ | . . . . . . . . . . . . O . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . 5 . . . . . . . . . . . . O . . . . |
$$ | . X 4 O . . . . . , . . O . O X X . . |
$$ | a 6 8 . . O . b . . . . . O X . . . . |
$$ | . 7 . . . . . . . . . . . . X . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------
- Click Here To Show Diagram Code
[go]$$Bcm61 Prisoners: white 2 black 0
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . O . . . . |
$$ | . . . . . . . . . . X . . O . O . . . |
$$ | . X O O . . . . . . . . O O . O O X . |
$$ | . . X , O . . O . X . X . . O X X . . |
$$ | . X . . . . . O X . . . . X X . . X . |
$$ | . . . X . . . . X . . O . . O . . . . |
$$ | . . . . X . O . . . . . . . . X . . . |
$$ | . . Q . . . . . X . . X . O . X . . . |
$$ | . . . . . . . . . X X O O . . . . . . |
$$ | . . . , . . . . . , O X . . . , . . . |
$$ | . . . Q . . . . . . X X O O . . . . . |
$$ | . . . . . . . . . . . a . . . . . . . |
$$ | . . . . . . . . . . . . O . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . 5 . . . . . . . . . . . . O . . . . |
$$ | . X 4 O . . . . . , . . O . O X X . . |
$$ | a 6 8 . . O . b . . . . . O X . . . . |
$$ | . 7 . . . . . . . . . . . . X . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]
I was taught 67 as joseki. It's meant to be a good probe and good timing in this sequence.
It is hard for White to descend at a because of the weaknesses it creates (eg. Black plays at b). Forcing White to play 68 early leaves connecting underneath in the corner as big endgame for later.
Yes, the 67 clamp probe is standard (it is played immediately in about 300 of 500 games in GoGoD with this corner shape), but it is not
always a good move according to Guo Juan. The point is it asks white whether he wants to connect or descend.
Actually white has a 3rd choice, tenuki, say to play a pincer on the left side, forcing black to capture the stone and make eyes and corner profit so black would usually be content with this; if he is not (e.g. because the white pincer makes some huge left side) then he should extend down the side first as the clamp is his privilege: white would usually make a kick/block/checking extension to stop black extending further down the side, and then black could clamp.
$$ Descent: good points, bad aji
$$ | . . . . . . . . . . .
$$ | . . . f . . . . . . .
$$ | . . . . . . . . . . .
$$ | . 3 . b . . . . . . .
$$ | . 1 2 O . . . . . , .
$$ | 6 4 c . . O . a . . .
$$ | . 5 . d . e . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------
- Click Here To Show Diagram Code
[go]$$ Descent: good points, bad aji
$$ | . . . . . . . . . . .
$$ | . . . f . . . . . . .
$$ | . . . . . . . . . . .
$$ | . 3 . b . . . . . . .
$$ | . 1 2 O . . . . . , .
$$ | 6 4 c . . O . a . . .
$$ | . 5 . d . e . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------[/go]
If white descends then black loses yose (in yose black could exchange 6 for 5), in exchange for white having bad aji from the cut left at
c. As others mentioned, one use of this aji is the point
a, if white makes a pincer on it then black has the placement at
d which makes miai of the cut at
c and connecting at
e. Another use of this aji is the attachment at
b, if white pushes though then black immediately cuts at
c and white is short of liberties so either loses the corner or black gets to block white's push, turning it into a bad move. Thus white usually defends one point below
c and black makes a thick connection to left left of
b. Sometimes white doesn't defend straight away as then black may treat
b lightly (e.g. jumping to
f), so would keep attacking on the outside, and then black connects, white defends, and black runs.
$$ Connect: good aji, bad points
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . 3 . . . . . . . . .
$$ | . 1 2 O . . . . . , .
$$ | a 4 6 . . O . . . . .
$$ | . 5 b . . . . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------
- Click Here To Show Diagram Code
[go]$$ Connect: good aji, bad points
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . 3 . . . . . . . . .
$$ | . 1 2 O . . . . . , .
$$ | a 4 6 . . O . . . . .
$$ | . 5 b . . . . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------[/go]
If white connects, then black can no longer aim at
a or
b from the previous diagram, but instead has the connection at
a. This is surprisingly big and is a difficult timing question. If white answers at
b it does not make any eyes for black, but black may be pleased to get some sente yose.
$$ Timing question
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . 3 . . . . . . . . .
$$ | . 1 2 O . . . . . , .
$$ | 7 4 6 . . O . . . . .
$$ | b 5 8 . . . . . . . .
$$ | . . a . . . . . . . .
$$ ------------------------
- Click Here To Show Diagram Code
[go]$$ Timing question
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . 3 . . . . . . . . .
$$ | . 1 2 O . . . . . , .
$$ | 7 4 6 . . O . . . . .
$$ | b 5 8 . . . . . . . .
$$ | . . a . . . . . . . .
$$ ------------------------[/go]
The important of white blocking at
above can be seen in this game I am playing on OGS:
http://www.online-go.com/games/board.php?boardID=291183 (lower left). After exchanging h4 for j5, I block in gote at c2 as I judged it was needed for the safety of my group, and also to weaken black. This was proved correct when black later made the surprisingly severe attack at h2, using the power of his ponnuki to the right. I only just managed to live, and this was because the descent to
a threatens to make 1 eye (not 2 remember!) with the throw in at
b.
If white ignores black's first line connection black can make one eye in gote in the corner and has some followups against white's shape for later: if white blocks
at
a then black
b makes miai of
c and
d, so white usually has to answer
at
b, giving black some gain in yose and reduction of white's eyespace.
$$ Ignore the connection
$$ | . . . . . . . . . . .
$$ | . . . . . . . 2 6 . .
$$ | . . . . . . . . . . .
$$ | . X . . . . . . . . .
$$ | . X O O d . . . . , .
$$ | 1 O O c b O . . . . .
$$ | . X 3 4 7 . . . . . .
$$ | . . 5 a . . . . . . .
$$ ------------------------
- Click Here To Show Diagram Code
[go]$$ Ignore the connection
$$ | . . . . . . . . . . .
$$ | . . . . . . . 2 6 . .
$$ | . . . . . . . . . . .
$$ | . X . . . . . . . . .
$$ | . X O O d . . . . , .
$$ | 1 O O c b O . . . . .
$$ | . X 3 4 7 . . . . . .
$$ | . . 5 a . . . . . . .
$$ ------------------------[/go]
Even if white answers the clamp with the timid, safe connection instead of the greedy descent, it is not always a good exchange for black. This is a rather subtle point, but it makes black's 2 marked stones on the outside slightly heavier (harder to sacrifice). The reason for this is white has been strengthened so black has lost some other options. For example, say it is unplayed and later in the game black invades with the attachment as below. If white is strong around and wants to kill everything,
is a strong move but as black has not clamped he can get
in sente which helps a lot to make eyeshape.
$$ Sabaki without clamp
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . # . . . . . . . . .
$$ | . # O O 4 6 . . . , .
$$ | . O . . 3 O 1 . 2 . .
$$ | . . 8 7 . 5 . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------
- Click Here To Show Diagram Code
[go]$$ Sabaki without clamp
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . # . . . . . . . . .
$$ | . # O O 4 6 . . . , .
$$ | . O . . 3 O 1 . 2 . .
$$ | . . 8 7 . 5 . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------[/go]
If black has clamped and then later sacrificed the marked stones then
is less sente as it only threatens to connect to the circled stone (which is not 100% alive) as opposed to cutting which is.
$$ Sabaki with clamp slightly harder
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . # . . . . . . . . .
$$ | . # O O 4 6 . . . , .
$$ | . O O . 3 O 1 . 2 . .
$$ | . B . 7 . 5 . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------
- Click Here To Show Diagram Code
[go]$$ Sabaki with clamp slightly harder
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . . . . . . . . . . .
$$ | . # . . . . . . . . .
$$ | . # O O 4 6 . . . , .
$$ | . O O . 3 O 1 . 2 . .
$$ | . B . 7 . 5 . . . . .
$$ | . . . . . . . . . . .
$$ ------------------------[/go]
In conclusion, I would play the clamp in this position, but there is plenty to consider when doing so and it is far from automatically a good move.
is aji keshi. I see it as a probe to force white to commit to either letting black have a nice endgame attach or letting black have powerful aji to draw on (if white chooses to block). If black waits, it may be too easy of a choice for white.
is bad in my mind. What had me think negatively of this move instinctively in the first place? Simply that I haven't seen this move played before, either in my own or in professional games? That it just didn't "feel" right? I'm not sure.
