Life In 19x19
http://www.lifein19x19.com/

Another connection question
http://www.lifein19x19.com/viewtopic.php?f=11&t=6909		 Page 1 of 1
Author:  PeterPeter [ Sun Oct 07, 2012 2:33 am ]
Post subject:  Another connection question
Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . . . . . . . . |
$$ | . . Y . X . , . . |
$$ | . . 7 O . . O . . |
$$ | . . Y . O . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]

In a recent game I connected the marked black stones by playing at b4. It worked out well enough in the game, but would this have been an effective move against a strong opponent?
Author:  EdLee [ Sun Oct 07, 2012 2:53 am ]
Post subject: 
Peter,
Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . . . . . . . . |
$$ | . a Y . X . , . . |
$$ | . 2 1 O . . O . . |
$$ | . b Y . O . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
The answer depends on the two cuts, (a) and (b),
and what happens when your opponent decides to cut at one of them, and the follow-ups
which depend on the whole board (as usual :)).

For example, if we remove the limited 9x9 size --
if W cuts at (a) with :w3:, B has 2 choices: connect at (b) or atari at (c) --
which leads to two possible ladders -- can you see the ladders?
Click Here To Show Diagram Code
[go]$$Wc
$$ | . . . . . . . . .
$$ | . c . . . . . . .
$$ | . 3 Y . . . . . .
$$ | . 2 1 O . . . . .
$$ | . b Y . . . . . .
$$ | . . . . . . . . .
$$ | . . . . . . . . .[/go]
Similarly, if W cuts at (b) instead with :w3:, B also has 2 choices
and two ladders (at least) to consider.

Does this make any sense? :)
Author:  EdLee [ Sun Oct 07, 2012 3:09 am ]
Post subject: 
Also, locally, blocking was the only move for B.
Otherwise, W pushes through with :w3: and the result is a disaster for B (broken shape):
Click Here To Show Diagram Code
[go]$$Wc :b2: tenuki
$$ | . . . . . . . . .
$$ | . . . . . . . . .
$$ | . . Y . . . . . .
$$ | . 3 1 O . . . . .
$$ | . . Y . . . . . .
$$ | . . . . . . . . .
$$ | . . . . . . . . .[/go]
Author:  PeterPeter [ Sun Oct 07, 2012 3:25 am ]
Post subject:  Re: Another connection question
I see. Black can win the cutting stone, but then White can trap Black in a ladder. Or if Black connects, White can also trap Black in a ladder. But Black has the option (atari or connect), and so the chance to make a good choice based on the rest of the board.

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . . . . . . . . |
$$ | . a Y . X . , . . |
$$ | . 2 1 O . . O . . |
$$ | . b Y . O . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]

On my 9x9 board, if White cuts at B, Black can atari at B2, and White's ladder fails due to the black stone on e5.

If White cuts at A, Black can connect at B2, and White's ladder fails for the same reason. Or, Black can win the b5 stone, but lose the one on c3 (which is what happened in the game).

Either way, Black seems to have come out better than the split you showed in post 3.
Author:  EdLee [ Sun Oct 07, 2012 3:33 am ]
Post subject: 
PeterPeter wrote:
Black can win the cutting stone...
Yes, W also can choose not to make either cut -- it all depends on the whole board (as usual :)).
PeterPeter wrote:
Either way, Black seems to have come out better than the split you showed in post 3.
Correct -- usually (but not always) letting W push through with :w3: is a disaster for B, but there are exceptions. :)
Author:  EdLee [ Sun Oct 07, 2012 3:48 am ]
Post subject: 
PeterPeter wrote:
On my 9x9 board, if White cuts at B, Black can atari at B2, and White's ladder fails due to the black stone on e5.
Yes, but there are other considerations here. :)
If W pushes with :w5:, there are complications -- can you see them?
Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . . d . . . . . |
$$ | . . X 5 X . , . . |
$$ | . 2 1 O . . O . . |
$$ | . 3 X . O . , . . |
$$ | . 4 . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Variation 1:
If you block at (d) with :b6:, then :w7: is a double atari,
and you have to consider the variations:
Click Here To Show Diagram Code
[go]$$W
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . . 6 . . . . . |
$$ | . 7 X 5 X . , . . |
$$ | . 2 1 O . . O . . |
$$ | . 3 X . O . , . . |
$$ | . 4 . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Variation 2:
And if you submit at :b6: instead, W pushes through with :w7:...
Click Here To Show Diagram Code
[go]$$W
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . 6 7 . . . . . |
$$ | . . X 5 X . , . . |
$$ | . 2 1 O . . O . . |
$$ | . 3 X . O . , . . |
$$ | . 4 . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
...and B ends up with another broken shape (see also the broken shape in Post #3):
Click Here To Show Diagram Code
[go]$$W
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . . Y @ . . . . . |
$$ | . . Y @ Y . , . . |
$$ | . X O @ . . O . . |
$$ | . O X . O . , . . |
$$ | . X . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Author:  PeterPeter [ Sun Oct 07, 2012 9:33 am ]
Post subject:  Re: Another connection question
I don't like how either of those turn out for Black.

Would you say :b6: was a mistake, as it gave White 2 cutting options?

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . . . . . . . |
$$ | . . 6 . 4 . , . . |
$$ | . . . 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Author:  Falcord [ Sun Oct 07, 2012 10:02 am ]
Post subject:  Re: Another connection question
PeterPeter wrote:
I don't like how either of those turn out for Black.

Would you say :b6: was a mistake, as it gave White 2 cutting options?

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . . . . . . . |
$$ | . . 6 . 4 . , . . |
$$ | . . . 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]



I'd say it is a bit of a mistake, yes. That shape (which is called the tower peep) is good for white when there is nearby support (such as the 1 stone), it's good enough that white would want to play the 5 stone last to get it, I think. So doing a bit of crude tewari... Maybe black 6 wasn't the best option.

On the limited space of a 9x9 board though it's hard to say. My gut tells me to answer at C4 or B5 instead but if I'm honest I haven't read it out that much.

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . 8 . . . . . |
$$ | . . . 7 4 . , . . |
$$ | . . 6 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]


I'd perhaps go for something like this. You're cut but you're still alive and the corner is quite big
Author:  illluck [ Sun Oct 07, 2012 10:10 am ]
Post subject:  Re: Another connection question
PeterPeter wrote:
I don't like how either of those turn out for Black.

Would you say :b6: was a mistake, as it gave White 2 cutting options?

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . . . . . . . |
$$ | . . 6 . 4 . , . . |
$$ | . . . 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]


Another possible idea for six with fewing cuts is as below:

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . . . . . . . |
$$ | . 6 . . 4 . , . . |
$$ | . . . 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Author:  jts [ Sun Oct 07, 2012 10:57 am ]
Post subject:  Re: Another connection question
I would say :b4: is the mistake, unless you were actively planning to sacrifice it after :w5: .
Author:  Phoenix [ Sun Oct 07, 2012 11:21 am ]
Post subject:  Re: Another connection question
The main idea here is that Black succumbs to a peep from White.

In such a shape, it's not uncommon for Black to peep:

Click Here To Show Diagram Code
[go]$$Bc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . , . , . , . . |
$$ | . . . W . . . . . |
$$ | . . , a 1 . , . . |
$$ | . . . W . . . . . |
$$ | . . , . , . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]


White typically connects solidly at a, and is strengthened. Peeps are usually a tactical move used to get an outcome that outweighs the strengthening of White. As EdLee pointed out, Black's next move if White ignores the peep will be to push through at a himself (following up on the original threat). This leads to a poor shape for White, with cutting points.

Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . , . , . , . . |
$$ | . . . 1 . . . . . |
$$ | . . , a B . , . . |
$$ | . . . W . . . . . |
$$ | . . , . , . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]


Now what's bad about the diagram above is that where the Black stone is simply there to start with, White :w1: has turned it into a peep against his shape on his own turn. Now he does not have time to avoid Black cutting at a. The fact that White has given Black this opportunity by making Black's stone into a peep without making black expend a turn is a big difference from the first diagram.

Now Black can push through and cut at any time, and White still has to spend a turn to defend at a. If :w1: doesn't achieve something big, it might have been better at a to begin with.
Author:  EdLee [ Sun Oct 07, 2012 12:54 pm ]
Post subject: 
Falcord wrote:
I'd perhaps go for something like this.
Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . 8 . . . . . |
$$ | . . . 7 4 . , . . |
$$ | . . 6 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Falcord, that's very bad for Black.
Click Here To Show Diagram Code
[go]$$Wcm9
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . X . , . X . . |
$$ | . 3 5 X . . . . . |
$$ | . 2 1 O X . , . . |
$$ | . 4 X O . . O . . |
$$ | . . X . O . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Black's shape is very bad (broken):
Click Here To Show Diagram Code
[go]$$Wcm9
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . B . , . X . . |
$$ | . @ @ B . . . . . |
$$ | . B @ @ B . , . . |
$$ | . X B @ . . O . . |
$$ | . . X . O . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]

The original :b6: at C4 was bad. The Keima at B5 is better:
Click Here To Show Diagram Code
[go]$$Wc
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . 2 . , . X . . |
$$ | . . . . . . . . . |
$$ | . 6 . . 4 . , . . |
$$ | . . . 5 . . 1 . . |
$$ | . . X . 3 . , . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]
Author:  Buri [ Sun Oct 07, 2012 1:23 pm ]
Post subject:  Re: Another connection question
Greetings,
I don`t know to what extent the rule of thumb applies on a 9x9 but what Ed has just shown is that first, don`t play two stones diagonally with a space in between. That`s bad shape. But, if you do and your opponent cuts, then the correct response is often a keima, as in the diagram above.
Best wishes,
Buri
Author:  EdLee [ Sun Oct 07, 2012 1:49 pm ]
Post subject: 
Hi Buri,
Buri wrote:
...first, don`t play two stones diagonally with a space in between. That`s bad shape.
Unfortunately, it's not so simple.
Sometimes, it is good shape; sometimes, it is not.
It depends on the exact board position.
Author:  Buri [ Sun Oct 07, 2012 4:51 pm ]
Post subject:  Re: Another connection question
Greetings,
Ed, thanks for the clarification. You are right of course. The shape does crop up and is of no significance in a lot of games. However, without any intention to be disrespectful I would like to reclarify what you say.
As a beginner, I(we) are taught this is -basically- a bad shape. (I am citing here Jennie Shen and Guo Juan). Therefore I respectfully feel that the weighting is a great deal more towards treating it as a bad shape, rather than the sense that it is `sometimes good, sometimes bad` which might suggests to a beginner such as myself that this is a more balanced issue. In Jennie Shen`s lectures on the topic (Guo Juan Internet school) it is treated as essentially bad and that the correct response is almost always a keima. However, the direction of the keima must be selected in relation to the whole board situation.
best wishes,
Buri
Author:  Uberdude [ Sun Oct 07, 2012 5:07 pm ]
Post subject:  Re: Another connection question
The one space diagonal aka elephant jump is usually enough bad shape to be one of Fan Hui's basic 4 bad shapes (Ed will be pleased to see the ripped keima is number 1 ;-) )

http://senseis.xmp.net/?LameDuGo#toc4

Note also the ripped tobi of number 2.
Author:  EdLee [ Sun Oct 07, 2012 6:41 pm ]
Post subject: 
Hi Buri, could you reference the name (or number/date, etc.) of Jennie's (or Guo Juan's) lecture(s), and how many minutes into it, etc.?
I have a feeling there's a misunderstanding/mis-interpretation of what was said in the lecture(s),
so let's first find out what was exactly presented in the lecture(s), then proceed from there. :)


----
Uberdude, :tmbup:
Author:  Mef [ Mon Oct 08, 2012 4:48 am ]
Post subject:  Re:
EdLee wrote:
Hi Buri,
Buri wrote:
...first, don`t play two stones diagonally with a space in between. That`s bad shape.
Unfortunately, it's not so simple.
Sometimes, it is good shape; sometimes, it is not.
It depends on the exact board position.



Perhaps a better way to word this would be to say that the double diagonal (elephant's jump) is a shape with a natural weakness, namely that the opponent can play the middle point and threaten a very severe cut. Whether this natural weakness is going to be a problem will of course be context dependent.

Another such shape is the kick-extend shape -

Click Here To Show Diagram Code
[go]$$W Kick-Extend
$$ . . . . . . .
$$ . . a . . . .
$$ . . 1 . 4 . .
$$ . b . 3 2 . .
$$ . . . . . . .
$$ . . . . . . .
$$ --------------[/go]


This shape locally has two natural weak points for white (A and B), however there are still some times where the shape works and is effective.
Author:  PeterPeter [ Mon Oct 08, 2012 9:21 am ]
Post subject:  Re: Another connection question
Thank you for all the advice on this page, which has been at just the right level. As a beginner, it seems like there is a big gap between understanding the rules, and understanding why some natural-looking moves are bad, and this has narrowed it a little.
Page 1 of 1 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/