It seems to me that White’s move 8 at A11 creates a false eye (assuming it goes to ko), where as A10 would make two valid eyes.
1 Like
@mark5000
I’m having trouble verifying that black wins this semeai:
For example this continuation seems losing for black:
What am I missing?
1 Like
Sorry, the diagram isn’t that clear. There are two black stones at A9 and A10 before white captures at A11 with move 8, so white can’t play at A10, but I don’t think white is in any trouble. With the throw-in at D6 white can take away almost all of black’s liberties.
1 Like
This is actually why A5 as a local ko threat extending that group’s liberties is so important, it makes the trapped black stones has fewer liberties than it, which make the white group loop around and creates one giant eye regardless if there is a false eye shape on the edge.
1 Like
This can lead to at least a seki for black if white tries to remove liberties from the outside and black can also remove liberties of the white group from B3 below and keep the shared liberties in C5 D5 unfilled (if white is not careful, that white group can die as well even the group above). And a seki here is more effective in reducing the white territory.
3 Likes
Would White to C6 for Move 6 in the above variation ruin Black’s plans? It seems like no matter which way Black responds to that (C5, C7, B6) White can connect and cause Black problems with a shortage of liberties. Haven’t worked it out in a demo, so just working it out in my head…
1 Like
Because black connect the wrong way wC6 bC7 wC5 bB6, it will be a semeai black is faster with one more liberty.
I suppose white can still start a ko out of the white group, but I feel the ko is good for black, haven’t worked out all the ko threats yet, and honestly this whole exercise is about whether it is better to continue with a reduction, or a ko fight. There are so many different variations that would lead to ko, and they are all trade off need to be calculated.
1 Like
Can you please show the variation to a seki (with coordinates in the diagram)?
1 Like
Let’s assume white is already alive above and black’s outside is already connected (which technically is still not, the double cut H8 H6 can still be exploited)
2 Likes
I’m not sure I understand. I tried playing out this variation and - as long as White plays A5 after Black B6, it wins the semeai. What am I doing wrong?
2 Likes
That’s the ko variation I added in the reply. When black played B3 next.
Come to read it again, that’s a very bad ko for white, and I think there is a way for black to completely ignore it and turn it into a semeai.
2 Likes
I take a look at the ko variation, and the “power triangle” at move 10 always bothers me, it feels wrong, after quite a bit of trying to work out the ko threats, they always involved local ko threat at D6, and gain white more liberties on the bottom.
But today as I’ve tried to work through the rabbit hole, where one of them involved seki, it led me remembered with the earlier reply for possible seki, where D6 is crucial for the seki (keep D5 shared liberty), and the assumption of assuming the white group on the top is alive is not the case if white turns at B10, that would remove any chance of white living locally, and has to win the ko, or compromise by sacrificing the stones below. If white tries to save the stone below, the upper part of white group will be ko to live.
I haven’t work through all the possible options yet, can anyone help explore and verify them?
I think move 10 should be either directly throw down to D11, but there are other options for creating the ko.
2 Likes
Why does any of this stuff in the lower left take precedence over C18, which seems like sente. Is it really more urgent?
3 Likes
Making that C18 exchange (assuming it is sente for black) gains secure points for black, but it also reduces aji/potential black might have on the upper side (aji-keshi).
So I think it’s essentially a good move, but I find it difficult to tell if the timing is right.
3 Likes
C18 isn’t sente, in my opinion.
4 Likes
The reason (I suppose) for C18 not being definite sente ATM, is that black doesn’t really move in that deeply, even if white ignores it.
4 Likes
Isn’t it better for white to just make the ko directly with 10 and keep A-E as internal ko threats?
If white really has that many internal threats, then maybe the ko is difficult to play for black. Black needs to have many big threats elsewhere, so maybe blacks need to prepare those before starting the ko?
1 Like
What about on the other side at E18 first?
How realistic is for black to lean on L18 to live inside and break the whole upper side? And just more threats for creating a ko on the left side. I don’t have an answer to that though.
I was thinking of black playing 5 inside, maybe at G17, and force white to cut it off from the corner. With this additional forcing move inside, black might have enough potential to live.
Black doesn’t have to answer with A10 for B9, but A11, and creating step ko, it will be much harder for white to win the ko
1 Like