Life In 19x19

After a lot of work I managed to build a position for which the win need a special strategy I have never seen before.
Can you solve the following problem:

AGA rules, komi 14.5

Click Here To Show Diagram Code

[go]$$B Black to play and win
$$ ------------------------
$$| . . . . O X O O O . X |
$$| O O O O O X O O O X X |
$$| X X X X X X X X O . X |
$$| . . . . . X X O . O O |
$$| . . . . . . X O O O O |
$$| . X . . . . X X X X X |
$$| X X X . . . X O O O O |
$$| O O X . . . X O . O . |
$$| . O X X X X X O O X O |
$$| X O O O O X O O X X X |
$$| . X X . O X O X . X . |
$$ -----------------------[/go]

Edit :look at https://lifein19x19.com/viewtopic.php?p=279399#p279399 for a far better problem.

Gérard TAILLE wrote:AGA rules, komi 14.5

May I ask the point of setting Komi to 14.5? Is it just to to ensure that a win vs loss depends on the solution?

It's funny. Another superko beast.

hzamir wrote:

Gérard TAILLE wrote:AGA rules, komi 14.5

May I ask the point of setting Komi to 14.5? Is it just to to ensure that a win vs loss depends on the solution?

Basically the komi is here to help you. If black plays the correct sequence black will win, otherwise black will lose.
I could have only said "what is the best result for black" but in that case you have no clue to know if you are close to the solution.

dany wrote:It's funny. Another superko beast.

Yes Dany it is a really a beast.
Assuming white is allowed to play "unuseful" moves (e.g. by using the double ko when it is not necessary) but is not allowed to play "bad" moves (a move which loses points), how many moves are necessary before black can make the decisive capture (I mean a capture which is not an exchange in the double ko in the bottom right corner or in the sending 2 return 1 sequence)?

I have just discovered a dual solution for the problem I showed in this topic. By just adding a black stone and a white stone in the upper left corner this dual issue is fixed:

AGA rules, komi 14.5

Click Here To Show Diagram Code

[go]$$B Black to play and win
$$ ------------------------
$$| X . O . O X O O O . X |
$$| O O O O O X O O O X X |
$$| X X X X X X X X O . X |
$$| . . . . . X X O . O O |
$$| . . . . . . X O O O O |
$$| . X . . . . X X X X X |
$$| X X X . . . X O O O O |
$$| O O X . . . X O . O . |
$$| . O X X X X X O O X O |
$$| X O O O O X O O X X X |
$$| . X X . O X O X . X . |
$$ -----------------------[/go]

After more work I have discovered the black attack is far stronger than what I believed initialy. As a consequence the win for black in my previous positions is not very interesting because almost any black moves (including a pass!) is winning.
Though the results I claimed are good it is not satisfactory and I have to change the position in order to have a quite interesting solution. Here it is with a more usual komi:

AGA rules, komi 6.5

Click Here To Show Diagram Code

[go]$$B Black to play and win
$$ ------------------------
$$| . O . O . O . O . O . |
$$| O O O O O O O O O O X |
$$| X X X X X X O . O . . |
$$| . . . . . X X O . O . |
$$| . . . . . . X O O O O |
$$| . X . . . . X X X X X |
$$| X X X . . . X O O O O |
$$| O O X . . . X O . O . |
$$| . O X X X X X O O X O |
$$| X O O O O X O O X X X |
$$| . X X . O X O X . X . |
$$ -----------------------[/go]

This time I am quite happy. The position seems quite simple, without weaknesses (with japanese rules it is even an endgame position) but with AGA rules, this apparently losing position is a winning one due to a very surprising sequence!
Take pleasure.

Gérard TAILLE wrote:endgame position

I think you mean "settled position".

Gérard TAILLE wrote: Black to play and win

I must be missing something, what is the number of prisoners on each side?
At first sight, assumming no prisoners, black has 22 points, white has 14, so with komi 6.5 black won already.

RobertJasiek wrote:

Gérard TAILLE wrote:endgame position

I think you mean "settled position".

Yes Robert. I mean it is a position for which both players can easily decide to pass and go in the confirmation phase instead of playing the loop made of a single pass followed by an exchange in the double ko.

sorin wrote:

Gérard TAILLE wrote: Black to play and win

I must be missing something, what is the number of prisoners on each side?
At first sight, assumming no prisoners, black has 22 points, white has 14, so with komi 6.5 black won already.

I assume AGA rules that means area counting. The number of prisoners is irrelevant.

As an example if

Click Here To Show Diagram Code

[go]$$B :b3: pass, :b5: pass, :w6: pass
$$ ------------------------
$$| . O . O . O . O . O 4 |
$$| O O O O O O O O O O X |
$$| X X X X X X O . O . 2 |
$$| . . . . . X X O . O . |
$$| . . . . . . X O O O O |
$$| . X . . . . X X X X X |
$$| X X X . . . X O O O O |
$$| O O X . . . X O . O . |
$$| . O X X X X X O O X O |
$$| X O O O O X O O X X X |
$$| 1 X X . O X O X . X . |
$$ -----------------------[/go]

black count : 38 stones + 24 territory = 62 points
white count : 45 stones + 12 territory = 57 points
with a 6.5 komi black loses the game by 1.5 points

If I'm right with my solution idea, it's a pretty nasty superko beast. A kind of "pass fight", and at the same time an example of the difference between situational superko and positional superko, isn't it?
And I think maaany moves are necessary before black can make the decisive capture...

jmeinh wrote:If I'm right with my solution idea, it's a pretty nasty superko beast. A kind of "pass fight", and at the same time an example of the difference between situational superko and positional superko, isn't it?
And I think maaany moves are necessary before black can make the decisive capture...

You are 100% right jmeinh.
Do you think you have discovered the right sequence allowing to apply your idea?

Well, I think so.

jmeinh wrote:Well, I think so.

I'm not good at creating images, so I use coordinates.
B A3, W A1, B L4, W J1, B A2.
Now White has a choice:
a. W pass, B A3 , and W A1 is forbidden.
b. W L9 (or elsewhere), B A3 (repeating the loop until W cannot make a move without damaging himself).
c. W L3, B H1 (SSK allows this, in contrast to PSK), W L9 or elsewhere (SSK forbids passing - is that right? I'm not an SSK expert...), B A3 and the same idea as b.

The basic idea is correct and I confirm the SSK allows such sequence. Congratulation for that ajmeinh.
Unfortunetly the execution is not correct.