How consistent is the 2.5-3 pts per stone rule? I had heard that, but I assumed it applied to walls on the 3rd or 4th line. A wall on the 6th line must be worth less, right?
Not a whole lot less. The effect of a wall falls off as some function of the inverse of the distance. ( I suspect that the function is one term and something between x^1 and X^2 ) In this case it is like having a wall on the fourth rank and having someone chop off rows 's' and 't'. Or maybe a better way to consider it is a fourth-rank wall with a few stones on the opposite side of the board, which would locally neutralize the effect of the wall 11 or 12 stones away.
How much is the difference in value ( to the east ) of this wall:
Presumably, whatever effect the wall might have on the two most eastern rows is zero, as it is neutralized by the white stones there. But, in practice, would you consider the wall worth sustantially less?
What then, is the difference between the wall above and the one below?
I'm also curious as to how you would work things like extensions into that. Presumably a wall that already has a good extension, or that completes a moyo, is worth more, but do you have any sense of how much more?
Well, since it takes 2 moves to play those stones, and the current temperature is maybe 10 *shrug*, then it might follow that the stones - if perfectly placed - would raise the value of the wall about 20 points. But they are not perfectly placed. The one at J17 is too close to the wall and too far away from its neighbor. It should be at L16 or L17, I think. That imperfection is hard to evaluate.
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 4:33 am
by Kirby
I'm not sure I totally agree with Joaz's last choice.
If he wants a diagram like this (taken from his explanation, with marks added):
I do not see why he needs the marked stone exchange. The exchange seems to add to the number of stones sacrificed, while allowing black to fix that local area. I'm not sure what good it does for white. If he wants to go that route, I'd think it'd be better to skip the exchange, and just play what he wanted to:
Yes, the marked area, above, will probably not be totally useful for white right now, but why make the exchange to allow black to close off that area?
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 4:58 am
by topazg
I'm not entirely convinced by his reasoning either, although I think the result is close to the same. However, I think this is a really poor overall final result on the left for White:
Sure, it requires , but is so huge that I can't see it being omitted, and it's fairly obviously urgent after . It's a really really huge corner and side.
Posted: Mon Apr 04, 2011 5:05 am
by EdLee
Kirby,
I think the order matters: if first, then later if W turns at (a), B will not make the empty triangle at (b); B will be able to block at (c):
$$W $$ --------------- $$ | . . . . . . . $$ | . . . . . . . $$ | . . . . . O . $$ | . . . X . . . $$ | . . . . X . . $$ | . . X X O O . $$ | . . O O X 1 . $$ | . . a X 2 . . $$ | . . c b . . . $$ | . . X , . . . $$ | . . . . . . .
This doesn't seem too bad. I'm capturing some stomnes and making fourth-line territory. It seems likely that Joaz is going to use forcing moves against my stones here to take a bit of thickness here, but I think that's doable. E14 looks unlikely to link solidly in sente, so his thickness will either be flawed or gote, in which case I can start reducing. If I reduce, I'll have to consider either capping one of the low middle under-the-star-point stones on the top and bottom, or just leaning on his framework from the right side.
I am also a little concerned that he could steal my corner here. While he could do so in gote with a fairly safe joseki, he could also do something aggressive like this. My concern here is the number of liberties I have on the D14 group. This is enough of a concern that I may play to protect the corner soon, if he doesn't get there first. Or at least threaten F17, so he cannot take the corner without sacrificing that area.
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 9:07 am
by Kirby
@EdLee:
I can see that playing inside first gives b an empty triangle, but it also absolutely ensures that black encloses the territory, which is what confuses me. In other words, though black has an empty triangle, I feel the benefit of solidly walling off territory seems greater than white's benefit.
How does white benefit from an exchange leading to an empty triangle if the empty triangle is part of a now solidly connected wall?
If the exchange had not been made, isn't there the small possibility of a ko threat in the area, which has now been extinguished?
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 10:34 am
by hyperpape
@Kirby
I don't know if Joaz's sequences work, but they won't work if back doesn't have that empty triangle--the white is stuck at two liberties.
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 10:42 am
by Joaz Banbeck
@Chew Terr: If you play 28, I play 29. If you don't like 28, that's ok by me too.
Annoyingly, it feels like I need to spend a move in the corner to button it up, removing aji. This gets me some points, but at this stage of the game it feels slow. However, I really don't like the idea of my opponent getting the corner invasion in sente because of threats against D16. So I'll play the slow defensive move that looks like it works (shown above), and hope to be able to reduce my opponent sufficiently after that. At least I should have a big chunk of points on the left. I feel like my handicap is reduced, but not entirely gone.
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 12:55 pm
by Joaz Banbeck
That's a mistake by MHO, I fear.
The attachment is sente, threatning the sequence below, which in addition to undermining white, solves much of black's shape problems in the corner:
But with the line chosen, I have sente! Bwaahahahaha...
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 1:15 pm
by Chew Terr
One of the pieces of advice for improving and playing well is to count often. Specifically, it's recommended to count fully at least three times during the game, starting at turn 30ish. I'm trying to work on my counting (I'm terrible at it), so here goes:
$$Wcm21 Move 27 $$ --------------------------------------- $$ | b b b . . . w w w w w w w . . . . . . | $$ | b b b 0 . . w w w w w w w . . . . . . | $$ | b b b . . O w w w O w . . O . . . . b | $$ | b b b X . . . w w , . . . . . X b b b | $$ | b b b . X O . w . . . . . . . . . b b | $$ | b b X X O O . . . . . . . . . . X b b | $$ | b b C C X O . . . . . . . . . . . b b | $$ | b b C X X . . . . . . . . . . . . b b | $$ | b b b X . . . . . . . . . . . . . . . | $$ | b b X , . . . . . , . . . . . , . . . | $$ | b b b . . . . . . . . . . . . . . . . | $$ | b b b . . . . . . . . . . . . . X . b | $$ | b b b . . X . . . . . . . . . . . b b | $$ | b b b X . . . . . . . . . . . . . b b | $$ | b b b . . O . . . . . . . . . . X . . | $$ | b . . X . . w . . , . . . . O , . . . | $$ | . . . . . O w w w O . . w w w . O . . | $$ | . . . . . . w w w w w w w w w w w w w | $$ | . . . . . . w w w w w w w w w w w w w | $$ -------------------------------------
[go]$$Wcm21 Move 27 $$ --------------------------------------- $$ | b b b . . . w w w w w w w . . . . . . | $$ | b b b 0 . . w w w w w w w . . . . . . | $$ | b b b . . O w w w O w . . O . . . . b | $$ | b b b X . . . w w , . . . . . X b b b | $$ | b b b . X O . w . . . . . . . . . b b | $$ | b b X X O O . . . . . . . . . . X b b | $$ | b b C C X O . . . . . . . . . . . b b | $$ | b b C X X . . . . . . . . . . . . b b | $$ | b b b X . . . . . . . . . . . . . . . | $$ | b b X , . . . . . , . . . . . , . . . | $$ | b b b . . . . . . . . . . . . . . . . | $$ | b b b . . . . . . . . . . . . . X . b | $$ | b b b . . X . . . . . . . . . . . b b | $$ | b b b X . . . . . . . . . . . . . b b | $$ | b b b . . O . . . . . . . . . . X . . | $$ | b . . X . . w . . , . . . . O , . . . | $$ | . . . . . O w w w O . . w w w . O . . | $$ | . . . . . . w w w w w w w w w w w w w | $$ | . . . . . . w w w w w w w w w w w w w | $$ -------------------------------------[/go]
Black: Left:48 Right: 12+5=17 Total: 65 White: Bottom: 33 Top: 21 Sente: 10? Total: 64 Of course, this doesn't give a lot of points that could easily happen. I'm being conservative on what I call territory. For example, if I got amazing shoulder hits and leaning, maybe I could keep most of the top to third line territory, but he could jump into and wreck my corner easily enough. Considering the fact that it's white's sente, let's give him an additional 10ish points to account for that extra move. Overall, the game is playable for both sideds. I might be a LITTLE ahead, as my groups feel quite stable and, if he jumps into my corner, I'll get something for it, in most cases. I have less framework potential, but my points seem more solid (I have options like L16, L4, N3, G4, and so on, that all actually carry a bit of threat). So I don't feel like I can count accurately, but the game seems OK so far. Not sure how much of my handicap I've lost, but I don't think I'm actually losing, yet.
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 1:56 pm
by jts
@joaz -
Thanks for the explanation, that was helpful. I realize now that part of what confused me was bad accounting; even if a wall on the 6th line is almost 3pts per stone, you're still giving up an extra 2pts per stone to your opponent, relative to a 4th line wall. That only matters if he had no claim to side territory before you built your wall, of course.
Re: Malkovich 104 - Joaz vs Chew Terr
Posted: Mon Apr 04, 2011 2:12 pm
by Joaz Banbeck
jts wrote:@joaz -
Thanks for the explanation, that was helpful. I realize now that part of what confused me was bad accounting; even if a wall on the 6th line is almost 3pts per stone, you're still giving up an extra 2pts per stone to your opponent, relative to a 4th line wall. That only matters if he had no claim to side territory before you built your wall, of course.
Yes, the eastern value of my wall is worth the about same, 4th or 6th, no matter what happens on the back side of it. The absolute value of my wall is the eastern value minus whatever improvement he gets in the west. His improvement is the value of his current position in the west minus the value of his pre-wall position in the west.
Ultimately, was the attachment and building the wall a good idea? I'm beginning to think not. His position on the left looks rather strong if he plays the upper left a bit more agressively. It felt like he still would have had the better part of his two stone lead after 30 moves. I think that I made a mistake someplace back there...maybe F5 should have been at F6?
Anyway, I have sente, and he has weaknesses in the upper left. I think that makes it about an even game right now.
Posted: Mon Apr 04, 2011 3:48 pm
by EdLee
Kirby,
As hyperpape also pointed out, W gains one lib from the exchange; otherwise B can kill it with one move:
$$B $$ | . . . . X . . . . $$ | . . X X O O . . . $$ | . 1 O O X . . . . $$ | . . . X . . . . . $$ | . . . . . . . . . $$ | . . X , . . . . .
[go]$$B $$ ------------------- $$ | . . . . . . . . . $$ | . . . . . . . . . $$ | . . 8 . . O . . . $$ | . 7 6 X . . . . . $$ | . X 4 5 X . . . . $$ | . O X X O O . . . $$ | . O O O X 2 . . . $$ | . . O X 3 . . . . $$ | . 1 . X . . . . . $$ | . . X , . . . . . $$ | . . . . . . . . .[/go]
So W can turn at in sente (B can reply with ); otherwise, W can live as above, the corner aji is not good for B, and W gains many outside sente moves (e.g. (a)...(e)) which can be very useful in the future:
$$W $$ ------------------- $$ | . . . . . . . . . $$ | . . . . . . . . . $$ | . . . 4 . O . . . $$ | . . . X . . . . . $$ | . . . . X 3 . . . $$ | . . X X O O . . . $$ | . . O O X a c . . $$ | . . 1 X b d . . . $$ | . . . 2 . e . . . $$ | . . X , . . . . . $$ | . . . . . . . . .
[go]$$B $$ ------------------- $$ | . . . . . . . . . $$ | . . . . . . . . . $$ | . . 8 . . O . . . $$ | . 7 6 X . . . . . $$ | . X 4 5 X . . . . $$ | . O X X O O . . . $$ | . O O O X 2 . . . $$ | . . O X 3 . . . . $$ | . 1 . X . . . . . $$ | . . X , . . . . . $$ | . . . . . . . . .[/go]
So W can turn at in sente (B can reply with ); otherwise, W can live as above, the corner aji is not good for B, and W gains many outside sente moves (e.g. (a)...(e)) which can be very useful in the future:
$$W $$ ------------------- $$ | . . . . . . . . . $$ | . . . . . . . . . $$ | . . . 4 . O . . . $$ | . . . X . . . . . $$ | . . . . X 3 . . . $$ | . . X X O O . . . $$ | . . O O X a c . . $$ | . . 1 X b d . . . $$ | . . . 2 . e . . . $$ | . . X , . . . . . $$ | . . . . . . . . .
[go]$$W $$ ------------------- $$ | . . . . . . . . . $$ | . . . . . . . . . $$ | . 9 5 . . O . . . $$ | . 4 6 X . . . . . $$ | 0 3 2 . X . . . . $$ | . 1 X X O O . . . $$ | . 7 O O X W . . . $$ | . . O X B . . . . $$ | . 8 . X . . . . . $$ | . . X , . . . . . $$ | . . . . . . . . . $$ | . . . . . . . . .[/go]
Actually, I didn't consider white living inside - so if he can, I might see the benefit of the exchange. If he can't, though, it seems better not to do it (to me).
, but 
. It's a really really huge corner and side.
first, then later if W turns at (a),
Bwaahahahaha...
exchange earlier):
in sente (B can reply with 
);