Life In 19x19 http://www.lifein19x19.com/ |
|
Does this kill? (techniques for large scale capture/escape) http://www.lifein19x19.com/viewtopic.php?f=15&t=673 |
Page 1 of 1 |
Author: | prokofiev [ Fri May 28, 2010 9:25 am ] |
Post subject: | Does this kill? (techniques for large scale capture/escape) |
Does this kill, or can black escape/live? I played this in a game and it did in fact kill, but at my level that has little to do with the facts of the matter. I'm always queasy about going for the kill as there's a lot that can go wrong. Does black have any tactics to escape here? (Apologies if this looks like I'm attempting to brag about a big kill. I don't mean to be --- I'm very interested in tactics that go into this sort of situation and whether black has anything here.) The game sequence from here (to critique if you wish): |
Author: | Jordus [ Fri May 28, 2010 9:36 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
I can see black living there. ![]() Even in the case of the actual game sequence. if black plays G18, white plays h18 to prevent escaping, black plays g16, now it looks possible to make two eyes. Another thing to look at, whites groups at m15 area is in trouble. Blacks 4 stones at m12 are still alive. Black plays m11, white id forced to play n10, black plays m10 now the group is escaping, and black can pressure white. The white group(s) in m15 is now in trouble since it has less liberties than the originally questioned black group. |
Author: | HermanHiddema [ Fri May 28, 2010 9:52 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
Personally, the first question I'd ask myself is: Should black even try to live? Black has a lot of points, taking sente now with F5 and leaving that group for aji seems a sufficient strategy in this position, IMO. If I definitely did need to live, I'd start by asking white some tricky questions, like this: Note that it is very acceptable for black to lose his tail (the ![]() Regardless of whether Black sacrifices part or all of the group, it is still valuable for getting large sente endgame, like this: |
Author: | gaius [ Fri May 28, 2010 9:55 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
prokofiev wrote: Your move looks quite thin: Black escapes here by sacrificing two stones, which means that white cannot seal him in. If you want to kill you must play solid moves that do not leave behind weaknesses. It also requires that you are super-thick in the area. In this game I think you are thick enough to kill, so you should play the most solid moves, denying black any aji to exploit: If black now cuts at 'a', the marked stone can be easily sacrificed. This is the way to kill. |
Author: | Solomon [ Fri May 28, 2010 9:56 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
I would rather die as Black and leave some aji in that position than try to live there. |
Author: | freegame [ Fri May 28, 2010 10:07 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
I would be looking at moves like A for black to live. But if I was black I don't think I would want to live there. white gets way to strong and will take the whole left side, while black is left with a group with 2 points if he is lucky. I would use the aji of the stones to try and get a solid base on the left to make sure white does not get too much points. If black lives on the left side, the top right will probably be enough points to win. A move like B looks nice. If white spends another stone to kill the four black stones that are in danger, black can make a two-space jump to get a solid group and good endgame at C. If white plays something like C10 trying to kill everything, the extra black stone on the left will provide a lot more chances for black to live. *edit. wow lots of post all of a sudden. *edit @ gaius white will play 2 at 3 in your first diagram. this makes things a lot harder. @ hermanHiddema doen't white just nobi (4@5) ? instead of allowing a cut |
Author: | prokofiev [ Fri May 28, 2010 10:38 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
Thanks for all the (quick!) replies. Very helpful. This idea that even if black can live, he can only do so by making white quite strong makes sense. I guess that means white should in any case make sure to stay thick when responding even if the kill slips away. Given this, my attempts at followups to your moves: (Herman's): Now if black a, white b is very thick on the outside and black still isn't out. If black b, white a looks like it should end up capturing, but this seems preferable to black as the aji is very bad for white? (gaius's): (This seems to be what freegame was referring to) To gaius: The small knight's move instead for white does seem good. I guess even if white isn't trying to kill, leaving less aji like that is better. (freegame's): As you suggest, allowing the extension does seem bad. I still can't read properly what black should be doing next, though. Maybe 'a' to force white to capture and move out? |
Author: | prokofiev [ Fri May 28, 2010 10:47 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
HermanHiddema wrote: Regardless of whether Black sacrifices part or all of the group, it is still valuable for getting large sente endgame, like this: Thanks, this is a nice point. I don't think about things like this enough. This ties in to that whole "being good at the endgame is really being good at the middle game" idea that I think John Fairbairn posted in some thread about the endgame. |
Author: | prokofiev [ Fri May 28, 2010 10:52 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
Jordus wrote: I can see black living there. ![]() Even in the case of the actual game sequence. if black plays G18, white plays h18 to prevent escaping, black plays g16, now it looks possible to make two eyes. Another thing to look at, whites groups at m15 area is in trouble. Blacks 4 stones at m12 are still alive. Black plays m11, white id forced to play n10, black plays m10 now the group is escaping, and black can pressure white. The white group(s) in m15 is now in trouble since it has less liberties than the originally questioned black group. Yes, black's actual play seemed fairly submissive. Regarding making two eyes: Black tried that at the end and only made one eye. I can't myself see a way to make two for black, at least after the game sequence. Regarding m15 in trouble: black's m12 stones are netted (unless you mean at some earlier point?). |
Author: | freegame [ Fri May 28, 2010 11:42 am ] |
Post subject: | Re: Does this kill? (techniques for large scale capture/esca |
I was indeed thinking about something like that. if 4 @ A, black can cut. After having a closer look it might indeed not really provide the forcing moves you need. |
Page 1 of 1 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |