TsumegoFromGames64/Attempts

Sub-page of TsumegoFromGames64
[Diagram]

Attempt. Moves 1-10

Herman Hiddema: Hmmm, looks bad for white. Here's an attempt which results in ko...

[Diagram]

Attempt, moves 11-20

Herman Hiddema: [90] Black can also take the ko with B3. White can then play W4 as a ko threat, and after B5 white retakes the ko and we get basically the same result.

[Diagram]

Zinger's idea

zinger: [100] Egads, black is risking alot by doing this! If I were black, I would just play B3 at W4 in the first diagram, then work on the center with a - it looks like enough to win. Sorry for giving a counting reply to a tsumego problem, but after all, the whole board was posted :)

[Diagram]

Zinger's idea

[110] tderz the left side is now worth

  • lower white group: 14 white stones + 13 territory = 41
  • separating black stones: 11B + 2T = 24
  • upper white group: 9W + 10T = 28
    • the sum is 41 + 24 + 28 = 93 points, which is substantially bigger than the meagre centre of 12-16 points.

This variation is not playable with B5! (B5 instead of a)

[Diagram]

Zinger's idea

[120] tderz: better immediately the ko, as White could not connect with W6, she would be lost. However, this ko is so big, that only internal ko threats could be valid?

[Diagram]

Zinger's idea

[130] tderz: perhaps something like this (if now m or c, then something like n)
and living with the lower group, or simply c as threat.

[Diagram]

Zinger's idea?

Chris Hayashida: I got the idea that this was the sequence that Zinger suggested. While he did mark a on the second diagram, I think he was referring B3 and W4 on the first diagram.

I'm not sure that this would be enough for Black. a, b, and c are all big points, and White has sente. When I was playing, I counted that White was ahead if his group survived. I probably need to re-evaluate this position to see if this is still the case.

zinger: Yes, this is what I meant. I count black ahead by about 10 points on the board, but then there's komi, and it's white's turn. So maybe it's about even. Still, I would do this rather than risk everything, and rely on my endgame. Besides, it looks like white lives anyway. Chris: did you win in the end? By how much?


My solution

Chris Hayashida: This is what I played during the tournament. I don't want to list it as a solution until it stands up against scrutiny from the other deshi. If it does, I'll move it to the solution subpage.

[Diagram]

Game

Chris Hayashida: [10] This took me five minutes to read out (during the tournament) and I finally played this sequence. It is a similar sequence to what was posted above, but I played atari with W6 and saved the other moves for local ko threats.

[Diagram]

Variation 1

Chris Hayashida: [20] If B1, then White throws in at W2 and catches the marked black stones in oiotoshi for a second eye. This is what happened in the game.

[Diagram]

Variation 2 - Ko

Chris Hayashida: [30] This was the variation that I wasn't as sure about. B1 leads to ko, but I think White has a huge number of local threats, starting with W2.

[Diagram]

Variation 2 - Ko

tderz: [32] a and b are miai - hence White cannot live without ko and must exploit (internal) ko threats c, d, else or in combination.

[Diagram]

Black resolves the ko

Chris Hayashida: [40] If Black plays B1 to resolve the ko, then W2 again catches the marked black stones in oiotoshi for the second eye.

[Diagram]

Variation 2 - Ko resolving

tderz: [42] Above B1 (over)kills a dead white group. White lives, hence Black did not achieve much else than putting stones in his own territory.

[Diagram]

White resolves the ko

Chris Hayashida: [50] If Black responds to White's threat with B1, then White will retake the ko with W2. I think White can ignore almost any threat and play W4. If Black ignores W4, then both groups live with W6. I believe this is bigger than any other threat on the board.

[Diagram]

White starts another ko

Chris Hayashida: [60] I think B1 is forced. Answering the ko threat at a is not big enough, so White must play the ko with W2. I don't think Black has a big enough ko threat.

Tderz Just comparing the two diagrams [10] and the one above (I did not check the game at all, which would be necessary to decide which strategy is eventually necessary), Black could (better?) consider to play this B1 immediately in above diagram [10] (see below [70]), thereby taking much aji out of the position, capturing White and accepting the upper result , but not playing a ko about the lower..

[Diagram]

cont. from above dia. 10

[70] tderz Black allows White with dia. [20] to live with the bigger group, as can be seen in dia. [40]. Black doesn't want to give White such a choice and should therefore (IMO) connect directly with B1.

Chris Hayashida: By my count, this variation still has White ahead (including 6.5 komi.) What do you think?

tderz I didn't count - as I stated before - and I do not dare to. Perhaps I should then not give any commentary at all, and as soon as ko is involved, commenting would not make much sense.

Above I just compared two diagrams, this [70] with [20] and [30].

Analysis: white+circle seems bigger than white+square. white+circle is perhaps still bigger than the white upper left corner, even if it survives by capturing two stones ([50], [60] where Black loses everything). >Hence, I wanted to conclude that capturing the bigger group makes more sense than any ko [30] where one overkills the already dead & smaller group or capturing the smaller white+square in exchange.

If White still is leading, so be it! Often analyzing and going through games can show many options & alternatives, but not a clear way to win. Most probably there were many options to make easier better moves earlier on.

[Diagram]

cont. from above dia. 10

[80] tderz: this group is quite big. It seems worth to try living. White could perhaps consider another move instead of W6 in [10] above.

What is with the direct cut at W1? If Black were connecting solidly at a ( a point of interest indeed), White circle wraps up nicely and lives with two groups.


This is a copy of the living page "TsumegoFromGames64/Attempts" at Sensei's Library.
(OC) 2009 the Authors, published under the OpenContent License V1.0.
[Welcome to Sensei's Library!]
StartingPoints
ReferenceSection
About