shape move

[jaca]

Dont assume all are [new]

thank you for reminding me that i should have stated at the outset that the study problem this thread is about, was not and is not written for old hacks, but primarily for people fairly new to Go, and oldies (like me) who never really got the hang of the game, but still live in hope of being able to see a bit more clearly -

so even if you have only just learned the rules, it is my intention that you should be able to get something out of my explanation of the author's explanation - which, when it eventually comes, will fill in a few gaps the problem author's own explanation takes for granted, in the hope of making it comprehensible to everyone.

Go is a bit like Rubick's Cube, in that it is at one and the same time, very simple and impossibly complex. But it is also unlike Rubick's Cube, in that you don't need to know everything to get anywhere, so, like tiddley-winks or croquet, you can enjoy it even if you know hardly anything at all about it.

I am aiming to explain in everyday language what i mean by that seemingly self-contradictory statement when it comes to putting together and taking apart all the different answers that people of a fairly wide range of abilities have put forward, in the light of the problem author's own answer and explanation of it.

black to play

this problem comes from a book; if you know the author's answer, please don't spoil it for others by telling

18.png (207.96 KiB) Viewed 16305 times

[Uberdude]

g14 is my instinct, and then driving tesuji

[jaca]

please can you spell out the driving tesuji for anyone who may not know what that is? here's an sgf you can use; it doesn't have the moves in the game order as i just placed them by eye from the book diagram.

18.sgf

(4.26 KiB) Downloaded 695 times

[jlt]

Uberdude probably meant this:

capture.png (171.83 KiB) Viewed 16277 times

On the other hand, this problem reminds me of similar problems in which the answer was

P7.

This move separates two white groups and attacks R7.

[Uberdude]

Indeed jlt.

I did also think about something on the right side, though rather than cap I would go for the q8 q10 combo if I wanted to attack the top white group, and q9 q7 if I wanted to attack the lower group, but the top one is weaker so prefer the former.

[jaca]

g14 is my instinct

instincts are, by definition, instinctive, so hard to explain exactly how they come about - but can you do some post-hoc rationalisation, so we can get an idea of why g14 was the first thing that came to mind?

[Uberdude]

Well, my thought process on seeing this board is first to parse the board into groups/territories. Then it's looking for the hot area, and the white g15 stone stands out because it's the only stone/group on the board with few (2) liberties. Is it a junk stone we don't care about? No, so it's probably quite an urgent area of contact fighting. In fact my first instinct was black f14, connecting black groups, and other moves in the area that popped up on my policy network were j14 and h15, but g14 looks good to get the black group out and separate k16. After that I also noticed the white lower left group is close to death with a2 hane, but then I saw with g1 sente it lives. So actually g14 was not my instant reaction, but needed maybe 5-10 seconds of thought.

[jaca]

thanks for that, Uberdude;

a note to others: this isn't a "trick" problem to catch you out, but a genuine tutorial problem illustrating a point the author is making, which i will share with you in due course.

so far we have, i think, 3 or 4 answers by two people, which is already quite a lot to be getting on with, but there is even more we can get out of the problem - it's not a graded problem with one right answer and many wrong answers, but an opportunity to get inside our own minds to discover what we think about when deciding where to play next in a real game - and, importantly, to discover what we don't think about!

So i hope we will get answers from other people too, especially kyus of all grades, for whom the book is aimed at, even if the example comes from a high-level game.

Once all the answers are in, i will try to put them together in an overall picture.

When answering, please indicate your playing grade, and give your reasons for your choice, so we can later see the different ways different grades think. Please use a board picture to illustrate your answer, and ensure it has coordinates marked so readers won't have to scroll back to the first post.

[John Fairbairn]

the white g15 stone stands out because it's the only stone/group on the board with few (2) liberties.

Andrew: Is this (liberty counting) a standard way you look at a position? If so, is it something you ere taught or is it because you are a mathematician?

I ask partly because I remember being taken aback when discussing some position with Charles Matthews where I was convinced about the choice of move. I can't remember the details of my own thoughts but they certainly had nothing to do with numbers. Charles agreed with my choice and - the bit that surprised me - added by way of explanation something like, "Well, that would be so - Black and White have an even number of stones." When I asked whey he thought along those lines he said it was just because he was a mathematician (superior don actually!) and numbers came naturally to him. Maybe others here have that mindset?

The only time I think I ever focus on liberties, outside of a capturing race, is to apply the proverb "five alive."

As to the problem at hand, my first thought was a probe (B17) with a view to seeing whether F14 might work? Did that underlie you first thought, too?

[Uberdude]

I don't actually count the liberties, the area stands out because there are single stones hane-ing against each other, so it's a locally hot position. m17 only has 3 liberties and is another single stone in contact with a single stone, but because of the nearby support it's not so hot (though n17/n16 wedge is still warm). I rarely consciously count liberties outside a semeai (and even then I tend to read the moves rather than count liberties, probably I should learn those shortcuts; I can barely remember how many liberties the big eyes have!). And with liberties getting short in fighting where "5 alive" comes in handy I'm not explicitly counting liberties 1..2..3..4, it's more like just a sense of liberties of chains getting low.

Something I do notice reviewing games with Matthew Macfadyen is very often we end up with lines like "What if this cut? Blah blah variation, here's a semeai, let's count liberties, who wins?", whereas my reviews tend not to go down such fighting lines ending in a semeai and be more positional judgement or what's the right shape kind of questions. A reflection of our different styles of play .

[Bill Spight]

One problem. It appears that it is White's move. I.e., there is a missing White stone.

Edit: Also, does the book call Black's move a shape play?

Edit2: My guess is that the missing White stone is on A-02. Else Black A-02 looks pretty good.

[mhlepore]

One problem. It appears that it is White's move. I.e., there is a missing White stone.

Edit: Also, does the book call Black's move a shape play?

Edit2: My guess is that the missing White stone is on A-02. Else Black A-02 looks pretty good.

White has F-01 descent to cross under or make two eyes.

So we need clarification as to whether there is a stone missing somewhere.

[Bill Spight]

One problem. It appears that it is White's move. I.e., there is a missing White stone.

Edit: Also, does the book call Black's move a shape play?

Edit2: My guess is that the missing White stone is on A-02. Else Black A-02 looks pretty good.

White has F-01 descent to cross under or make two eyes.

So we need clarification as to whether there is a stone missing somewhere.

Count the stones, including the one White stone captured in the top left.

[Bill Spight]

I don't actually count the liberties, the area stands out because there are single stones hane-ing against each other, so it's a locally hot position. m17 only has 3 liberties and is another single stone in contact with a single stone, but because of the nearby support it's not so hot (though n17/n16 wedge is still warm). I rarely consciously count liberties outside a semeai (and even then I tend to read the moves rather than count liberties, probably I should learn those shortcuts; I can barely remember how many liberties the big eyes have!). And with liberties getting short in fighting where "5 alive" comes in handy I'm not explicitly counting liberties 1..2..3..4, it's more like just a sense of liberties of chains getting low.

Well, parrots have been shown to be able to distinguish between 5 and 6 things at a glance. Even humans who are not Rainman can probably make such distinctions without consciously counting.

[mhlepore]

Count the stones, including the one White stone captured in the top left.

Hmmm... Maybe F16 needs a white stone?