error in Davies L&D ?

cyclops · Post by **cyclops** » Thu May 27, 2010 6:07 am

Triggered by the thread Improving Life & Death? I started to reread Davies' L&D to check whether it is that boring. It isn't ;-)

Here, though, a statement I do not understand.

Click Here To Show Diagram Code: [go]$$c Davies: B1 kills $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . a b | $$ | . . , . , X X X 2 | $$ | . . . . . X O O d | $$ | . . , X , X O . c | $$ | . . . . X O O g 1 | $$ | . . . . . e . f . | $$ +-------------------+[/go]

Davies claims that B1 kills unconditionally. ( the board is 19x19 btw )
I don't see an answer to W2.

First let's try Wa. See also Davies' problem #3 ( section 21 )

Click Here To Show Diagram Code: [go]$$c White lives! How? $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . X . | $$ | . . , . , X X X O | $$ | . . . . . X O O . | $$ | . . , . , X O . . | $$ | . . . . X O O . X | $$ | . . . . . . . . . | $$ +-------------------+[/go]

Check the hidden diagrams for the solution.

Bd in the first diagram is desperate and runs at best into a ko.

Click Here To Show Diagram Code: [go]$$c B1@a $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . 4 3 | $$ | . . , . , X X X O | $$ | . . . . . X O O a | $$ | . . , . , X O . 2 | $$ | . . . . X O O . X | $$ | . . . . . . . . . | $$ +-------------------+[/go]

The last moves, Bb & Bc, that interfere with W's escape are answered by W connecting, forcing B to block on the outside, thus allowing We again. Thereafter f & g are miai as in the hidden diagrams.

So Davies is wrong?

amnal · Post by **amnal** » Thu May 27, 2010 6:39 am

cyclops wrote:Triggered by the thread Improving Life & Death? I started to reread Davies' L&D to check whether it is that boring. It isn't
Here, though, a statement I do not understand.
$$c Davies: B1 kills
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . , . , . , . . |
$$ | . . . . . . . a b |
$$ | . . , . , X X X 2 |
$$ | . . . . . X O O d |
$$ | . . , X , X O . c |
$$ | . . . . X O O g 1 |
$$ | . . . . . e . f . |
$$ +-------------------+

Click Here To Show Diagram Code
[go]$$c Davies: B1 kills $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . a b | $$ | . . , . , X X X 2 | $$ | . . . . . X O O d | $$ | . . , X , X O . c | $$ | . . . . X O O g 1 | $$ | . . . . . e . f . | $$ +-------------------+[/go]

Davies claims that B1 kills unconditionally. ( the board is 19x19 btw )
I don't see an answer to W2.

This is an interesting problem, and seems to highlight a subtlety of the L group that I didn't know.

Click Here To Show Diagram Code: [go]$$c The proper way to kill. $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . . . | $$ | . . , . , X X X . | $$ | . . . . . X O O . | $$ | . . , X , X O . . | $$ | . . . . X O O . . | $$ | . . . . . 1 . . . | $$ +-------------------+[/go]

I believe that this is the proper way to kill the L+1 group. Now white's hanes are of no value.

Click Here To Show Diagram Code: [go]$$c Davies: B1 kills $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . a b | $$ | . . , . , X X X 2 | $$ | . . . . . X O O d | $$ | . . , X , X O . c | $$ | . . . . X O O g 1 | $$ | . . . . . e . f . | $$ +-------------------+[/go]

As you say, 2 tries to convert this shape into a separate problem in which white lives. This seems to work, if black answers at a or b. Thus: black answers at e? (e is clearly the real vital point of this shape).

Click Here To Show Diagram Code: [go]$$c White 4 is the only move $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . 5 . . | $$ | . . . . . . . 4 . | $$ | . . , . , X X X 2 | $$ | . . . . . X O O . | $$ | . . , X , X O . . | $$ | . . . . X O O . 1 | $$ | . . . . . 3 . . . | $$ +-------------------+[/go]

Does white live now? Not if black has any stones nearby. But he does if white has stones on this side (and possibly he escapes anyway if there are no stones along all 19 points). I don't know if I'm missing something or not...

Overall, kill this shape by playing the vital hane, not black 1

EDIT: More thinking:

Click Here To Show Diagram Code: [go]$$c Black can block? $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . 9 . | $$ | . . , . , . . 7 8 | $$ | . . . . . . . 6 3 | $$ | . . , . , X X X 2 | $$ | . . . . . X O O . | $$ | . . , X , X O . 4 | $$ | . . . . X O O 5 1 | $$ | . . . . . . . . . | $$ +-------------------+[/go]

Your ko variation is actually better for black than it looks. It seems to me that this is sufficient for black - he can now net white 6 if it pokes out. It's complex and might not work if white has nearby stones, though, so I don't know if Davies actually considered it.

Note that the stone at d3 is important - this wouldn't work nearly so well without it, and it's lacking in the other problem.

My conclusion: In the absence of any other stones, white really is dead in the corner.

judicata · Post by **judicata** » Thu May 27, 2010 6:56 am

If this is right, then maybe add it to the errata page at sensei's library. There is actually not much for this book.

HermanHiddema · Post by **HermanHiddema** » Thu May 27, 2010 7:21 am

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

After

, White is dead unles she can escape with :w6:

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

White cannot live, :w6:

and

are miai

Click Here To Show Diagram Code: [go]$$Wcm6 $$ . . . . . . . . . | $$ . . . . . , . . . | $$ . . . . . . . . . | $$ . . . . . . . 2 . | $$ . . . . . . . . . | $$ . . . . . . . 1 B | $$ . . . . . X X X O | $$ . . . . . X O O O | $$ . . . X , X O . . | $$ . . . . X O O . X | $$ . . . . . X . . . | $$ ------------------+[/go]

Black will respond to :w6:

with

, the

stone is important here, because it can still escape.

Click Here To Show Diagram Code: [go]$$Wcm6 $$ . . . . . . . . . | $$ . . . . . , . . . | $$ . . . . . . . . . | $$ . . . . . . . 2 . | $$ . . . . . . 4 . 3 | $$ . . . . . . . 1 C | $$ . . . . . X X X O | $$ . . . . . X O O O | $$ . . . X , X O . . | $$ . . . . X O O . X | $$ . . . . . X . . . | $$ ------------------+[/go]

Capturing with :w8:

obviously fails

Click Here To Show Diagram Code: [go]$$Wcm6 $$ . . . . . . . . . | $$ . . . . . , . . . | $$ . . . . . . . . . | $$ . . . . . . 4 2 . | $$ . . . . . 5 3 . 6 | $$ . . . . . . . 1 B | $$ . . . . . X X X O | $$ . . . . . X O O O | $$ . . . X , X O . . | $$ . . . . X O O . X | $$ . . . . . X . . . | $$ ------------------+[/go]

Escaping outward with :w8:

and

fails because black pulls out the :bc:

stone.

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

Creative moves like :w4:

also fail, white still cannot live in the corner, and the situation has not been helped.

amnal · Post by **amnal** » Thu May 27, 2010 7:27 am

HermanHiddema wrote:
$$Wcm6
$$ . . . . . . . . . |
$$ . . . . . , . . . |
$$ . . . . . . . . . |
$$ . . . . . . . 2 . |
$$ . . . . . . . . . |
$$ . . . . . . . 1 B |
$$ . . . . . X X X O |
$$ . . . . . X O O O |
$$ . . . X , X O . . |
$$ . . . . X O O . X |
$$ . . . . . X . . . |
$$ ------------------+

Click Here To Show Diagram Code
[go]$$Wcm6 $$ . . . . . . . . . | $$ . . . . . , . . . | $$ . . . . . . . . . | $$ . . . . . . . 2 . | $$ . . . . . . . . . | $$ . . . . . . . 1 B | $$ . . . . . X X X O | $$ . . . . . X O O O | $$ . . . X , X O . . | $$ . . . . X O O . X | $$ . . . . . X . . . | $$ ------------------+[/go]

A shape I didn't know. Thanks!

gonut · Post by **gonut** » Thu May 27, 2010 9:18 am

Cyclops's original diagram adds a Black stone at Q5 that the book does not have. As a result, HermanHiddema's tesuji would not work for the book problem:

[sgf-full](;AB[or][pq][nq][pp][ro][qo]AW[qr][pr][qq][rp][qp]SZ[19]C["Life and Death", Davies ch 21, introductory problem];B[sr];W[so];B[sn]C[Fails]
;W[sp];B[ps];W[rn];B[rl]C[Without a stone on Q5, this tesuji fails.];W[qn];B[po];W[ql])[/sgf-full]

gonut · Post by **gonut** » Thu May 27, 2010 9:25 am

However, Cyclops still seems to have found another error in Davies' book. Instead of a Black kill, Black T2 only seems able to salvage a 1000 year ko.

[sgf-full](;AB[or][pq][nq][pp][ro][qo]AW[qr][pr][qq][rp][qp]SZ[19]C["Life and Death", Davies ch 21, introductory problem];B[sr];W[so];B[rn];W[ps]
;B[sn]
(;W[sp];B[rr];W[rs];B[ss]C[Black has a thousand year ko.])
(;W[sq]C[White only has a direct ko.]))[/sgf-full]

cyclops · Post by **cyclops** » Fri May 28, 2010 5:40 pm

You found your B5 in the last diagram, so you deserves the credit. I think it works.
Also Herman deserves the merit for his nice stopping move that works for my diagram but not for Davies'.

cyclops · Post by **cyclops** » Thu Jun 03, 2010 5:11 am

Click Here To Show Diagram Code: [go]$$c $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . . . | $$ | . . , . , a X X . | $$ | . . . . . X O O . | $$ | . . , . X X O . . | $$ | . . . . X O O 1 . | $$ | . . . . . . . . . | $$ +-------------------+[/go]

Davies second "killing" move ( B1 in the above diagram ) for the first L+1 group also fails if the upperclosure is as weak as in that diagram with the cutting point at a.

Click Here To Show Diagram Code: [go]$$cW is alive because 1&5 cant survive $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . 3 . | $$ | . . , . , . X X 2 | $$ | . . . . . X O O . | $$ | . . , . X X O . . | $$ | . . . . X O O 1 5 | $$ | . . . . . 6 . 4 . | $$ +-------------------+[/go]

Click Here To Show Diagram Code: [go]$$c 1000years ko $$ +-------------------+ $$ | . . . . . . . . . | $$ | . . . . . . . . . | $$ | . . , . , . , . . | $$ | . . . . . . . 5 3 | $$ | . . , . , . X X 2 | $$ | . . . . . X O O 4 | $$ | . . , . X X O . . | $$ | . . . . X O O 1 7 | $$ | . . . . . 8 . 6 . | $$ +-------------------+[/go]

gaius · Post by **gaius** » Thu Jun 03, 2010 5:54 am

Heh, it seems he can escape indeed. Anyway, locally he cannot live and that's the point of the problem

. The Davies book is only about the status of local eyeshapes, this kind of stuff you have to figure out for yourself during a game - no book can help you with that because it's far to dependent on the surrounding position.

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

Black can escape, but he cannot live locally.

Click Here To Show Diagram Code: [go]$$wc $$ . . . . . . . . . | $$ . . . . . , . . . | $$ . . . . . . . . . | $$ . . . . . . . 3 . | $$ . . . . . . 2 . . | $$ . . . . . . . O X | $$ . . . . . 1 X X O | $$ . . . . . X O O O | $$ . . . X , X O . . | $$ . . . . X O O X X | $$ . . . . . O . X . | $$ ------------------+[/go]

Another way to play.

cyclops · Post by **cyclops** » Thu Jun 03, 2010 6:32 am

gaius wrote: ... The Davies book is only about the status of local eyeshapes, ...

Davies states ( see his introduction ) that we should artificially assume the rest of the board empty and big enough to be far from the hidden borders. He almost always takes care to make the outside strong enough. So I don't believe you are correct about his intentions. In any case it seems helpful to identify the few cases in the book were the threat to escape might alter the status. Instead of an error I should have called it a footnote.

Life In 19x19

error in Davies L&D ?

error in Davies L&D ?

Re: error in Davies L&D ?

Re: error in Davies L&D ?

Re: error in Davies L&D ?

Re: error in Davies L&D ?

Re: error in Davies L&D ?

Re: error in Davies L&D ?

Re: error in Davies L&D ?

same problem with second killing move

Re: error in Davies L&D ?

Re: error in Davies L&D ?