Bki's study journal

Bki · **#41**

EdLee wrote:

:w29: locally your UR corner suffers.

I was aware of that, but at the time it felt like the less bad option. Would atari at M14 a better option? My group is still uncomfortably weak after that, so...

Quote:

:b32: may be a bad habit. (Take.)

You're right.

Quote:

:b44: entirely differnt situation with :b14:

.

I know. But I wasn't totally satisfied with my options here either.

Quote:

:b66: may be a bad habit.

Hum... You may be right. If it got me liberty at least, it may have been worth it. The only benefit was that it made a false eye for white.

EdLee wrote:

Hi Bki,

2 possible reasons:
- the exchange E16-F17 is good for W, bad for you.
- you remove the option of F17 later.

I guess you meant E17-F17?

Wouldn't White answer D18 by E17 himself? (though, again, sente).

Aside from that, I may have overestimated the danger of the B14 descent. Or at least, B15 doesn't seem to be good enough to kill unconditionally. B17 might do the trick though...

Uberdude · **#42**

The push of 10 is bad because it makes white stronger but doesn't make black stronger. You lose options like this one: http://josekipedia.com/#path:pepcndqeqf ... odnencmcnb. Note that as well as taking more territory and eyespace black's outside wall has some cutting point weaknesses in it. Also if white does push up at e17 immediately (a bad move as too slow for a gote) then black shouldn't tenuki quite yet but push in at d16 first to make white's wall have a weakness and strengthen his corner shape; it's a double sente).

Bki · **#43**

Ugh... Exams. I thought they only started tomorrow (But at least I didn't realize that after they took place)...

In addition to the sudden jump in difficulty of 1001 L&D problems when reaching the 5 moves problems (compared to the 3-moves which are not that much harder than the 1-move), I didn't do the 50 exercises I set for myself yesterday. I will try to do at least a couple of page today.

I may or may not play a game later today.

So today I will post a couple of correspondences games I finished recently... Of all three long games I played on OGS, all three ended up in my win after a large kill (Well, in the second I think he resigned prematurely. There might have been possibilities for him to survive)... I would have thought it would be harder to do so in long game. Maybe I'm making more use of the long thinking time than my opponents?

skydyr · **#44**

Bki wrote:

Ugh... Exams. I thought they only started tomorrow (But at least I didn't realize that after they took place)...

In addition to the sudden jump in difficulty of 1001 L&D problems when reaching the 5 moves problems (compared to the 3-moves which are not that much harder than the 1-move), I didn't do the 50 exercises I set for myself yesterday. I will try to do at least a couple of page today.

I may or may not play a game later today.

So today I will post a couple of correspondences games I finished recently... Of all three long games I played on OGS, all three ended up in my win after a large kill (Well, in the second I think he resigned prematurely. There might have been possibilities for him to survive)... I would have thought it would be harder to do so in long game. Maybe I'm making more use of the long thinking time than my opponents?

Different people play correspondence games differently. For me, for example, they're often a sort of slow blitz, where I will rarely spend more than 20 seconds on a move. For other people, they spend minutes or hours analyzing moves throughout the game.

As for these games, there's not that much to say about them that comes to mind. Your opponents seem afraid of fighting to some degree. In the second, for example, black often connects things in a very slow manner rather than fight back in areas of black's influence. The only thing that occurred to me in a quick look at the first game is that it may be better to play a move on the top side one move before black did, instead of playing the block in the center. White can turn the top into something close to territory with one move, and white was ahead of black in the center and didn't really need to worry about a push or a nonexistent hane.

Bki · **#45**

Well, I just did better than winning a lost game. My opponent resigned a game he was winning. I staked the game on the centre, his invasion was not done in the most sensible of way, and so I managed to kill it. But he actually was able to reduce my territory and get quite a bit of profit, to the point that when I counted it was very close despite considering the group dead, and I only made further mistake afterwards!

Bki · **#46**

Not much to say about this game. I tried too hard to kill, missing the opportunity to secure a large territory, but I succeeded at the end, so I won.

Well, it did bring to my mind one thing : I've been playing the low approach to the 3-4 a lot recently, but I'm not at ease with the one space pincer.

Still, just checking the joseki is not exactly a good solution. Taking the time to study them in depth would be better, but I'm of the opinion that things you figure for yourself are better assimilated.

So I've decided to try to consider any possible answer that come to mind (by that, I don't mean every possible moves, but every move I think off that I don't immediately dismiss as "silly", "bad shape", or something), then try to see what the possible continuations might be...

Click Here To Show Diagram Code: [go]$$c One space low pincer. $$----------- $$|.......... $$|.......... $$|..a1.efg.. $$|...,b..... $$|..2.c..... $$|.......... $$|..3d...... $$|.......... $$|.......... $$|..........[/go]

Those are the moves I consider promising. I considered briefly the slide at B17, but that just doesn't feel right with so little space. The taisha at F16 feels like it would end in tears with a pincer so close;

a : attaching on the corner stone to gain a base. It's almost certain to result in both side having some kind of shape. The drawback is that it will strengthen the opponent.

b, c : moving out. b have the additional benefit of pressing black down, but the push and cut must be considered.

d : attaching to the pincer to make shape. hane-crosscut is a possible continuation that come to mind. May be followed by a pincer on the corner stone.

e, f, g : double approach (or pincers?). I would feel more comfortable (read : I would almost certainly play there) :b3:

was at C12 or C11 (or D12 or D11), but it at the very least must be considered.

I will start with a. B hane is the obvious answer. I would even say it feels like the only move.

My instinct is always to cross-cut, so let's see if that's good here?

Click Here To Show Diagram Code: [go]$$Wc One space low pincer, attachment to the 3-4 stone, cross-cut. $$----------- $$|.......... $$|.c23a..... $$|..1X...... $$|..b,...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

a, b and c seems like the only reasonable possibilities.

If a, white will force with c, then play at b or at B16. Black as a ponnuki on the side, but white doesn't really have anything to worry about any-more.

If c, white atari at D16 and descend at B17. I'm not actually quite sure about this one. It feels similar to another variation (if white doesn't cross-cut). Black needs to deal with the aji in :w3:

, but it seems that E16 do suffice for that purpose... On the other hand, white will have the hane at A18 in sente.

Click Here To Show Diagram Code: [go]$$Bc One space low pincer, attachment to the 3-4 stone, crosscut. $$----------- $$|.......... $$|.1XO...... $$|.2OX3..... $$|...4...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

This seems another way to get the same position. Not sure about :w4:

though, but a move here by black would be annoying, so... (if :b3:

at

, then E18 and white capture the corner.)

I feels like this variation favour black a little, but it's not terrible either (just, some other move may be able to get a better results).

Still, there's b left, and I quite like it.

Click Here To Show Diagram Code: [go]$$Bc connect. $$----------- $$|.......... $$|..XOa..... $$|.3OX4..... $$|..12...... $$|.7O.6..... $$|.98....... $$|..X0...... $$|.......... $$|.......... $$|..........[/go]

This isn't a disaster either, as long as the right side is interesting, but a is black's sente, which is very painful. Moreover, another variation leads to a similar result, but without black being able to connect with his pincer stone. So I think this clearly favour black.

:w2:

at

seems another possibility, but after black connect, either white get a minimal corner and a severely damaged approach stone, or a not-totally alive corner that is one move from being sealed in (so he has to take gote).

So the cross-cut isn't good.

Click Here To Show Diagram Code: [go]$$Wc One space low pincer, attachment to the 3-4 stone. $$----------- $$|.......... $$|..X....... $$|.3OX2..... $$|...1...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

So what about :w1:

here? White get shape, black get sente. This isn't bad for either side. Still, what about :b2:

at

?

Click Here To Show Diagram Code: [go]$$Wc One space low pincer, attachment to the 3-4 stone. $$----------- $$|.......... $$|..X....... $$|.2OX...... $$|...1...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

Ko is unreasonable. Connecting unacceptable. So counter atari is the only option

Click Here To Show Diagram Code: [go]$$Wc connect $$----------- $$|.......... $$|..X.b..... $$|.XOX1..... $$|.32O...... $$|.aO.5..... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

Similar result to what we saw earlier, but this time the pincer stone is cut off. It's certainly a win. If black cut at a, he will only damage it further. But he is not alive yet, so he has to either play a (damaging his pincer stone) or b (crawling on the line of defeat).

Still, instead of :b2:

, we might have this instead :

Click Here To Show Diagram Code: [go]$$Wc $$----------- $$|.......... $$|..X23..... $$|.XOX1..... $$|..4O...... $$|.6O.5..... $$|.87....... $$|..X9...... $$|...0...... $$|.......... $$|..........[/go]

This might be good if the right side is interesting. It's similar to what we had with the cross-cut, except that black doesn't have a big, sente endgame play now, so it's clearly better.

if the right side is not promising though, I guess white can connect at :b4:

, and if white cut, can atari his way out before attacking the pincer stone... No instead he extend, then fight.

Click Here To Show Diagram Code: [go]$$Wc Fight $$----------- $$|.......... $$|..X2...... $$|.XOX15.... $$|..3O4..... $$|..O.6..... $$|...7...... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

Black corner might very well die if he isn't careful. So I guess :b4:

isn't reasonable? Then he play at E18, and White is only to happy to let him crawl on the second line.

...

Okay, that's a lot, so it will be enough for now. This is more to write down my own thoughts than anything else, but even then I would appreciate comments (though, no "this move is joseki", because I don't care, rather, "this variation you posted is wrong because black can answer like this and you're in trouble").

skydyr · **#47**

You may want to look at the "Whole Board Thinking in Joseki" books. Volume 1 focuses on the 3-4 point low approach, and the whole book is a series of questions of the type: "given this board position, which play should white choose?" "Which play should black choose in response?" etc. Answers have explanations like "The attachment defends solidly, but has no effect on black." or "The knight's move keeps black from developing this side" etc.

Bki · **#48**

skydyr wrote:

You may want to look at the "Whole Board Thinking in Joseki" books. Volume 1 focuses on the 3-4 point low approach, and the whole book is a series of questions of the type: "given this board position, which play should white choose?" "Which play should black choose in response?" etc. Answers have explanations like "The attachment defends solidly, but has no effect on black." or "The knight's move keeps black from developing this side" etc.

Looking at it, this books seems interesting (and also fit how I think I should approach joseki. Because obtaining a local result is meaningless if it doesn't fit with the board). I will certainly think about it the next time I order go books (which should not be too long to come, as I will soon need more problems books).

Thanks for the recommendation.

Bki · **#49**

To continue, the last possible answer to the hane after the attachment is the double hane.

Click Here To Show Diagram Code: [go]$$Wc Double hane $$----------- $$|.......... $$|.1Xa...... $$|.bOX...... $$|..c,...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

a, b and c are the conceivable answer.

b is unthinkable without the ladder (white connect at c, then after black atari :w1:

, which is the only consistent move after b, white counter atari and capture D17.

Even if black has the ladder, it's not bad for white

Click Here To Show Diagram Code: [go]$$Bc Double hane $$----------- $$|.3........ $$|5OX48..... $$|.1OX6..... $$|..27...... $$|..O9...... $$|...0...... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

That's still pretty good for white all things considered.

I'm not sure, but maybe W could play at :b5:

with

?, so that black has to add another move for his corner to live.

Click Here To Show Diagram Code: [go]$$Bc Double hane $$----------- $$|.......... $$|.OX1...... $$|..OX...... $$|.2.,...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

Pretty comfortable for white.

Click Here To Show Diagram Code: [go]$$Bc Double hane $$----------- $$|.......... $$|.OX....... $$|.2OX...... $$|.413...... $$|..O....... $$|.......... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

That's similar to a 3-3 invasion from a low approach to the 4-4, but without a stone at F16 or F17 (so black group is weaker), but with less space for W group (C15 would be at C14) (so white is a little less comfortable). So I guess it's about equal?

Click Here To Show Diagram Code: [go]$$Bc Double hane $$----------- $$|.......... $$|.OXb...... $$|.2OX...... $$|.3146..... $$|.7O58..... $$|..a....... $$|..X....... $$|.......... $$|.......... $$|..........[/go]

After

we have either :b9:

at a, then white play above 6, capturing the two black stones and getting influence on the top side, or :b9:

at b, then, after forcing at E17 and E18/F19, white capture :b5:

for a ponnuki in the centre, and a wall toward the top side.

Conclusion : Because in each case, black can chose a variation that give white influence on the top side, the attachment is not a good idea if that side isn't interesting for some reason. Otherwise, it tends to lead to two settled groups, but if both players are extremely aggressive there's also the possibility for some fight.

Also, the cross-cut is inferior for white.

Bki · **#50**

So, in this game, I got at quite comfortable position due to an early mistake by my opponent, but then made a mistake and lost a medium-sized group. Thankfully, he had many weak groups which were almost connected at the centre of the board... Almost. So I was able to launch a nice splitting attack, forcing one weak group to live then come back to kill a second, which allowed me to win the game.

Bki · **#51**

To continues on my exploration of 3-4 variations...

Click Here To Show Diagram Code: [go]$$c One space low pincer. $$----------- $$|.......... $$|.......... $$|..a1.efg.. $$|...,b..... $$|..2.c..... $$|.......... $$|..3d...... $$|.......... $$|.......... $$|..........[/go]

Let's continue to look at attachment, only, this time, we will attach to the pincer stone at d.

Click Here To Show Diagram Code: [go]$$c One space low pincer. $$----------- $$|.......... $$|.......... $$|...1...... $$|...,...... $$|..2....... $$|...b...... $$|..34...... $$|..ca...... $$|.......... $$|..........[/go]

I think the only reasonable answer are a, b or c (though c feels submissive to me).

Click Here To Show Diagram Code: [go]$$c Hane $$----------- $$|.......... $$|.......... $$|...X.bd... $$|...,.ce... $$|..O....... $$|...2...... $$|..XOa..... $$|...1...... $$|.......... $$|..........[/go]

This hane is probably the most natural answer (not that it means it's the best). :w2:

is the only reasonable answer, otherwise black will atari there. a might be possible, but after black extends one space right of :w2:

, white can't avoid being cut off. So I don't think it's good unless there's other stones to change the situation.

Now the question, does black have to connect? If he does, then either of b-e seems to be a natural continuation. Lean on one group to gain strength then attack the other. c would be safer, as it is better connected, while the others would likely lead to a fight.

But what if black doesn't connect, and instead strengthen the corner.

Click Here To Show Diagram Code: [go]$$c may also be at a I guess $$----------- $$|.......... $$|.......... $$|...X..1... $$|...,.a.... $$|..O....... $$|...O...... $$|.8XO...... $$|.42X6..... $$|.537...... $$|..........[/go]

black promptly sacrifice that stone and get a wall. White group is fairly safe, and can still get his share of the corner later on. If he has the ladder, then he can also contemplate playing :w6:

at

, but then black can play E10 and sacrifice two more stones, so it's not like that option is that much better.

Other moves than :b3:

seems clearly inferior as black can't save the cut-off stone anyway.

Basically, because of this variation, I'm not so thrilled with this attachment, unless the bottom left corner is such that black's wall can't make an extension and so can come under attack. In such a situation obviously, black would have to be a fool to let white cut, and so we return to a pincer on the corner stone.

Click Here To Show Diagram Code: [go]$$Bc Pincer $$----------- $$|.......... $$|.......... $$|...X.2.... $$|...,ab.... $$|..O....... $$|...O...... $$|..XO...... $$|...X...... $$|..1....... $$|..........[/go]

I will just look at this one for now. It doesn't seems productive to go too far in the variations anyway. This arrive at the moment where you "learn" the variations by reading in game and learning from mistakes you made.

a is the natural answer. b may be possible?

Click Here To Show Diagram Code: [go]$$Bc b $$----------- $$|.......... $$|.......... $$|...X3O4... $$|...521.... $$|..O760.... $$|..8O9..... $$|..XO...... $$|...X...... $$|..X....... $$|..........[/go]

Click Here To Show Diagram Code: [go]$$Bc b (continuation). at a give a ko. $$----------- $$|.......... $$|.37.0..... $$|.24XXOO... $$|.65XOX.... $$|..OXOO.... $$|..OOX..... $$|.8XO1..... $$|.a9X...... $$|..X....... $$|..........[/go]

I'm not sure who would win this. White can at the very least get a ko with double hane after :w10:

This deserves more study... But for now let's look at a :

Click Here To Show Diagram Code: [go]$$Wc a $$----------- $$|.......... $$|.......... $$|...X.O.... $$|...,B1.... $$|..O.270... $$|...O845a.. $$|..XO936... $$|...Xcbd... $$|..X.e..... $$|..........[/go]

After

at a,

at b is met with white c, and the double atari at d means white can escape. If :b12:

at d, then :w13:

at e and again white escape.

So it seems that black can get sealed in, then will follow a fight on whether or not one of white's group can die.

Fun time.

SoDesuNe · **#52**

Just some ideas.

Click Here To Show Diagram Code: [go]$$Bc b $$----------- $$|.......... $$|.......... $$|...X.O.... $$|..1..a.... $$|..O....... $$|...O...... $$|..XO...... $$|...X...... $$|..X....... $$|..........[/go]

Depending on what White's move is, 'a' is more severe.

Click Here To Show Diagram Code: [go]$$Wc a $$----------- $$|.......... $$|.......... $$|...X.O0... $$|...,B18... $$|..O.2679.. $$|...O.54... $$|..XO.3.... $$|...X...... $$|..X....... $$|..........[/go]

Bki · **#53**

SoDesuNe wrote:

Click Here To Show Diagram Code: [go]$$Wc a $$----------- $$|.......... $$|.......... $$|...X.O0... $$|...,B18... $$|..O.2679.. $$|...O.54... $$|..XO.3.... $$|...X...... $$|..X....... $$|..........[/go]

Hum... Interesting.

Click Here To Show Diagram Code: [go]$$Wc a $$----------- $$|.......... $$|.......... $$|...X.O.... $$|...,B1.9.. $$|..O.25.... $$|...O684... $$|..XO73.0.. $$|...X...... $$|..X....... $$|..........[/go]

Running fight? Still, maybe white can get something better...

Click Here To Show Diagram Code: [go]$$Wc at a $$----------- $$|.......... $$|.......... $$|...X.O.... $$|...,B1.a.. $$|..O.20.... $$|...O.94... $$|..XO.3.... $$|...X5..... $$|..X.67.... $$|....8.....[/go]

Bki · **#54**

I played four games today. This one is, I think the game which would benefit the most from a review.

I think :w66:

was the losing move, and should be at N11 (or maybe close to it as another stone that strengthen my group, but N11 or N10 are the ones I'm feeling the best about), as it allowed him to chase my group toward the bottom, inducing him solidifying that side. Without this I would have been able to invade, using the aji of O4 and threatening the bottom left corner. When I decided to invade in the game, that side was too solid for my prospects to be promising. Afterwards... :w82:

struck me by its terribleness, killing off much aji for little gain.

Before black 167, I was already behind, but not by so much (by my quick and approximate count, true) that I felt I had to resign. But I forgot to come back to live at the bottom...

mitsun · **#55**

Not a full review, just some notes on the upper right corner after the opening. Taking the corner with the hane at S16 is large, but misses the vital point. In this position, the hane at O16 is absolutely required. This point is the boundary of W and B influence in the region. Playing there first makes a huge difference to territorial prospects and center strength for both sides. In contrast, the game move is really just yose, making territory but having no other impact on the rest of the game (since B is already settled on that side).

Try laying out two sequences and see if you agree. Sequence A: W hane and connect at S16, B extend to O16, and B continue extending if W keeps blocking. Sequence B: W hane at O16, then keep extending, or maybe even hane again at some point, if B keeps blocking. One side gets to develop on a grand scale, while the other side is compressed to a small overconcentrated position.

Bki · **#56**

mitsun wrote:

Not a full review, just some notes on the upper right corner after the opening. Taking the corner with the hane at S16 is large, but misses the vital point. In this position, the hane at O16 is absolutely required. This point is the boundary of W and B influence in the region. Playing there first makes a huge difference to territorial prospects and center strength for both sides. In contrast, the game move is really just yose, making territory but having no other impact on the rest of the game (since B is already settled on that side).

Try laying out two sequences and see if you agree. Sequence A: W hane and connect at S16, B extend to O16, and B continue extending if W keeps blocking. Sequence B: W hane at O16, then keep extending, or maybe even hane again at some point, if B keeps blocking. One side gets to develop on a grand scale, while the other side is compressed to a small overconcentrated position.

You're right, thanks you for the advice. If I remember correctly, at the time I thought "Well, he didn't block at S17 (which is what I envisioned when I played R17), so I might as well take the corner... I could have waited to connect at S17 (because just making the hane at S17 gote is good enough for now), or maybe simply descended instead? But that's not as urgent as the hane at O16, because when he hane at O17, I had no other choice but to cut, and he settled his group quickly, while I didn't.

Bki · **#57**

2 stones games as white today. I made a game deciding blunder in the last ten move, and my opponent failed to notice the opportunity to kill. Apart from that, a big mistake in the endgame, failing to notice a play was actually a 3.5 point reverse sente (and would have prevented me from being in a situation to make the blunder that nearly cost me the game).

Maybe a mistake of direction with :w9:

in the opening, but it's not like I know how to play a Kobayashi formation anyway, so I simply decided to make the move that felt like had more chances to lead to complications. A few others questionable moves during the middle game, and in the end the reason I won was that my opponent was deadly afraid of ko, not even daring to make the first capture!

As an aside, I'm still working through 1001 L&D problems, though I've slowed down a little due to headaches, less time, and the 5-moves problem taking longer to solve. I'm still trying to do at least one page a day, though.

EdLee · **#58**

Hi Bki,

E16 seems only move.

:b18:

Toothpaste .

:b22:

F13 ?

strange. Locally, at least M9.

Bki · **#59**

Hi Edlee,

EdLee wrote:

Hi Bki,

E16 seems only move.

Playing E16 after :w9:

seems really strange to me... Which must means that :w9:

was a mistake and should have been at D16. Even if it was correct locally, it certainly was a mistake of direction given that I want to develop the top with my Kobayashi formation (but I have no idea how to play the Kobayashi, apart from the "turn that side into a framework" part).

Quote:

:b18: Toothpaste .

I already know about this, but I didn't associate it with my moves upon reviewing that game... This certainly made me realize how much I screwed up that corner.

Bki · **#60**

EdLee wrote:

:b22: F13 ?

was not a ladder block (as much as would have liked for it to be one, there was no way to play a natural move that would also break the ladder)

Quote:

:w33: strange. Locally, at least M9.

I thought briefly about M9 upon reviewing. It does make my bottom group weaker though, so that's must be why I didn't play it (whether it is justified or not). Moreover, that group doesn't need a reinforcement at the moment. And while :w33:

decrease the severity of the black hane, M9 would work better for that purpose.

Bki's study journal

Who is online