Now that lots of people have weighed in, I gave the position to Crazy Stone for fun (2 million playouts). Its top choices are listed in alphabetical order (that is, it thinks a is best), but all four are very close in evaluation. It assesses the current position as approximately W+5.5 (assuming 7.5 komi).
Where to play at the end of the Fuseki (practical example)?
-
DrStraw
- Oza
- Posts: 2180
- Joined: Tue Apr 27, 2010 4:09 am
- Rank: AGA 5d
- GD Posts: 4312
- Online playing schedule: Every tenth February 29th from 20:00-20:01 (if time permits)
- Location: ʍoquıɐɹ ǝɥʇ ɹǝʌo 'ǝɹǝɥʍǝɯos
- Has thanked: 237 times
- Been thanked: 662 times
- Contact:
Re: Where to play at the end of the Fuseki (practical exampl
dfan wrote:Now that lots of people have weighed in, I gave the position to Crazy Stone for fun (2 million playouts). Its top choices are listed in alphabetical order (that is, it thinks a is best), but all four are very close in evaluation. It assesses the current position as approximately W+5.5 (assuming 7.5 komi).
Thank you.
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).
-
Actorios
- Lives with ko
- Posts: 130
- Joined: Mon Apr 26, 2010 9:03 pm
- Rank: KGS 3k
- GD Posts: 0
- KGS: Actorios
- Universal go server handle: Actorios
- Has thanked: 28 times
- Been thanked: 8 times
Re: Where to play at the end of the Fuseki (practical exampl
Thank you so much for all those detailed and valuable answers ! Very enlightening!
I'm amazed by how great is this online go community...
P.S. I ended up playing the slack move of C10 (iron pillar) and my opponent invaded at D14 as a result.
I'm amazed by how great is this online go community...
P.S. I ended up playing the slack move of C10 (iron pillar) and my opponent invaded at D14 as a result.
-
Uberdude
- Judan
- Posts: 6727
- Joined: Thu Nov 24, 2011 11:35 am
- Rank: UK 4 dan
- GD Posts: 0
- KGS: Uberdude 4d
- OGS: Uberdude 7d
- Location: Cambridge, UK
- Has thanked: 436 times
- Been thanked: 3718 times
Re: Where to play at the end of the Fuseki (practical exampl
Actorios wrote:DrStraw wrote:Two others to consider are O4 (better than P5) and C15.
That's interesting. I tend not to know when to shoulder hit (or not) and didn't even consider it.
I would have thought it would give a lot to Black and I would be afraid not to get a proper payback from the territory he would be able to make.
I believe you say it is better than P5 because it does a better job at flattening white, am I correct in stating that?
As I didn't see anyone make clear the o4 vs p5 difference: p5 is typically used against a shimari that has 2 distant extensions, the idea being you have miai of attaching on whichever side of the shimari black doesn't defend. However here black only has 1 extension which is close, so for sure he would defend the lower side. But then white doesn't have anything particularly good against the r7 stone, for example black can connect up and white didn't achieve much below. As you say the o4 shoulder hit actually flattens black more, pushing him down to 3rd rather than 4th line.
FWIW o4 and c15 were my 2 main thoughts in that position. And regarding John's thoughts that the right side white group is not thick but potentially weak, I would call it thick as it's a big solid chain with no peeps/cuts/other weaknesses (but with little interest in extended discussions of Japanese terms) but also be mindful that in the future it could be attacked as a whole so o4 has a tinge of prophylaxis, extending a helping hand to the white right side group before the stakes are raised (at this point black is unlikely to want to take a big loss on the lower side for some speculative attack of doubtful profit on the white group, but maybe later he would).
-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Where to play at the end of the Fuseki (practical exampl
I'm intrigued by the view that White's right-side detached iceberg is thick, and especially by the supporting view that it is a solid chain with no reference to its usefulness.
Of course there is always scope for interpretation, but I think "thick" here is being mixed up with "safe" - is this a Jasiek definition?
And I don't think that even "safe" here is an acceptable view. The group as it stands has no guaranteed eye either in the centre or on the side, and is exposed to profitable and strengthening Black ijime (bullying) moves on both side skirts and probably in the centre. That looks like at least 5 or so free points for Black.
To repeat: what's the point of a "thick" group if it can't do the work of any thickness? What's its point if it gives the opponent good, free moves?
It's already been accepted that the White shoulder hit provides some succour for the iceberg above, but how can a group be called thick if it needs help?
A prime requirement for thickness is that it should be a tower of strength, not just a tower. It has to be efficient. Thickness is as thickness does. If it ends up being attacked itself, tant pis, and just avoiding attack doesn't make it thick.
Below is a position from a Japanese book under the heading "Looks thick but is really thin."
This large White group at least really does have a guaranteed eye, and it even has an extension. Yet the text describes it as "unexpectedly thin". It adds that "to an amateur's eyes it appears extremely thick, but Black has two forcing moves at A and B and White does not really have two eyes yet. Despite appearances, the group is actually thin."
One reason pros don't have to count so much to see who is ahead is that they can see inefficiencies like this at a glance and so know who must be behind. Of course, I may be wrong in that this group, like any iceberg may hide 8/9ths of its value below the surface, but it looks to me like enough go-board warming has already taken place.
Of course there is always scope for interpretation, but I think "thick" here is being mixed up with "safe" - is this a Jasiek definition?
And I don't think that even "safe" here is an acceptable view. The group as it stands has no guaranteed eye either in the centre or on the side, and is exposed to profitable and strengthening Black ijime (bullying) moves on both side skirts and probably in the centre. That looks like at least 5 or so free points for Black.
To repeat: what's the point of a "thick" group if it can't do the work of any thickness? What's its point if it gives the opponent good, free moves?
It's already been accepted that the White shoulder hit provides some succour for the iceberg above, but how can a group be called thick if it needs help?
A prime requirement for thickness is that it should be a tower of strength, not just a tower. It has to be efficient. Thickness is as thickness does. If it ends up being attacked itself, tant pis, and just avoiding attack doesn't make it thick.
Below is a position from a Japanese book under the heading "Looks thick but is really thin."
This large White group at least really does have a guaranteed eye, and it even has an extension. Yet the text describes it as "unexpectedly thin". It adds that "to an amateur's eyes it appears extremely thick, but Black has two forcing moves at A and B and White does not really have two eyes yet. Despite appearances, the group is actually thin."
One reason pros don't have to count so much to see who is ahead is that they can see inefficiencies like this at a glance and so know who must be behind. Of course, I may be wrong in that this group, like any iceberg may hide 8/9ths of its value below the surface, but it looks to me like enough go-board warming has already taken place.
-
Knotwilg
- Oza
- Posts: 2432
- Joined: Fri Jan 14, 2011 6:53 am
- Rank: KGS 2d OGS 1d Fox 4d
- GD Posts: 0
- KGS: Artevelde
- OGS: Knotwilg
- Online playing schedule: UTC 18:00 - 22:00
- Location: Ghent, Belgium
- Has thanked: 360 times
- Been thanked: 1021 times
- Contact:
Re: Where to play at the end of the Fuseki (practical exampl
Hi John
For my own sake, I usually translate "thickness" as "strong influence", the former being about the safety of the group, the latter about its usefulness towards the rest of the board. We wouldn't call a two eyed group in the corner "thick".
Still the absence of defects is what determines the "thickness": it is strongly connected and can find eyes easily (therefore influences the rest of the board, because stones in its vicinity will find themselves under attack rather than attacking the group).
On closer inspection, Black's R7 already reduces some of the thickness (especially the ability to find eyes) but it is offset by at least two features of the group: the "thick" turn at the right side, which will instantly offer an eye in gote, and the ponnuki in the middle which will only become a false eye through great effort by Black.
There are 2 reasons why I concur with O4 now:
* it is likely to keep sente
* there is no single good move at the left side
If O4 would strengthen the group in gote, or there would be a single good move to develop the left, I'd definitely not play O4. The additional strength it gives to the "iceberg" is only the 3rd reason to play it.
I could overestimate its thickness and thickness of groups in general.
For my own sake, I usually translate "thickness" as "strong influence", the former being about the safety of the group, the latter about its usefulness towards the rest of the board. We wouldn't call a two eyed group in the corner "thick".
Still the absence of defects is what determines the "thickness": it is strongly connected and can find eyes easily (therefore influences the rest of the board, because stones in its vicinity will find themselves under attack rather than attacking the group).
On closer inspection, Black's R7 already reduces some of the thickness (especially the ability to find eyes) but it is offset by at least two features of the group: the "thick" turn at the right side, which will instantly offer an eye in gote, and the ponnuki in the middle which will only become a false eye through great effort by Black.
There are 2 reasons why I concur with O4 now:
* it is likely to keep sente
* there is no single good move at the left side
If O4 would strengthen the group in gote, or there would be a single good move to develop the left, I'd definitely not play O4. The additional strength it gives to the "iceberg" is only the 3rd reason to play it.
I could overestimate its thickness and thickness of groups in general.
-
Uberdude
- Judan
- Posts: 6727
- Joined: Thu Nov 24, 2011 11:35 am
- Rank: UK 4 dan
- GD Posts: 0
- KGS: Uberdude 4d
- OGS: Uberdude 7d
- Location: Cambridge, UK
- Has thanked: 436 times
- Been thanked: 3718 times
Re: Where to play at the end of the Fuseki (practical exampl
John, when I use thick it doesn't imply safe, indeed I would say that white group is less safe than it is thick! That example you gave with the peeps is exactly the sort of peep of a net in a somewhat-thick-but-not-so-much-really group I was thinking of which is absent from white's group. This talk actually reminds of Lee Sedol vs Gu Li jubango game 4, in which Lee made a thick group in a black area that couldn't really do much and a lot of people's initial judgement was it was a poor result for him. Gogameguru's Relentless has a whole page dedicated to judgement of this position, which is available in the free sample (page 208 of https://gogameguru.com/i/go-books/sampl ... sample.pdf). I seem to recall we have discussed that position on this forum before, with JF calling it a Chernobyl group that just sits there and lays waste to the lower side, it doesn't have to do anything with its thickness. In that case thickness is as thickness is, not does, so why the difference? Both Lee's group and this one have a ponnuki to the centre, but one that is not strictly a real eye yet, plus a black stone on the 3rd line 2 spaces away. In Lee's game Gu later attacked at a, and Lee tenukid and Gu later attacked again at b and Lee ended up making a small life in sente. Despite getting 2 tenukis from it Lee later said he should not have ignored the initial attack but jumped to c to preserve the power of the thickness. Now there are of course some differences in the positions, one major one being Gu's lower side groups is somewhat thin, and thus white's thick group exerts some power in dissuading black from invading the lower left white corner as that would damage his side, so that is indirectly made larger/stronger by white's group.
Anyway, I don't think of thick as a yes/no question but a sliding scale, and one of many useful analogue concepts that are not orthogonal. So to me it's more thick than not thick, but less thick than say the wall after a 3-3 invasion under a 4-4 (but even that is not invincible and doesn't have secure eyes yet and might get attacked later). But it also has some heaviness so would bear both these qualities in mind when planning what to play next.
Anyway, I don't think of thick as a yes/no question but a sliding scale, and one of many useful analogue concepts that are not orthogonal. So to me it's more thick than not thick, but less thick than say the wall after a 3-3 invasion under a 4-4 (but even that is not invincible and doesn't have secure eyes yet and might get attacked later). But it also has some heaviness so would bear both these qualities in mind when planning what to play next.
-
Kirby
- Honinbo
- Posts: 9553
- Joined: Wed Feb 24, 2010 6:04 pm
- GD Posts: 0
- KGS: Kirby
- Tygem: 커비라고해
- Has thanked: 1583 times
- Been thanked: 1707 times
Re: Where to play at the end of the Fuseki (practical exampl
"Thick" is something you see depending on strength. What a pro may see as thick may not seem that way at 5k, and vice-versa.
I see White's group as thick and strong, and find it inefficient to protect (or attack) it right now.
Pro may or may not see it differently, but to find out, a pro needs to comment. Even then, you get the pro's answer, but it's not clear that this will help the OP.
I see White's group as thick and strong, and find it inefficient to protect (or attack) it right now.
Pro may or may not see it differently, but to find out, a pro needs to comment. Even then, you get the pro's answer, but it's not clear that this will help the OP.
be immersed
-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Where to play at the end of the Fuseki (practical exampl
I see White's group as thick and strong, and find it inefficient to protect (or attack) it right now.
So how do you see the White position in the Japanese example I gave?
-
Knotwilg
- Oza
- Posts: 2432
- Joined: Fri Jan 14, 2011 6:53 am
- Rank: KGS 2d OGS 1d Fox 4d
- GD Posts: 0
- KGS: Artevelde
- OGS: Knotwilg
- Online playing schedule: UTC 18:00 - 22:00
- Location: Ghent, Belgium
- Has thanked: 360 times
- Been thanked: 1021 times
- Contact:
Re: Where to play at the end of the Fuseki (practical exampl
John Fairbairn wrote:I see White's group as thick and strong, and find it inefficient to protect (or attack) it right now.
So how do you see the White position in the Japanese example I gave?
With three immediate forcing moves against it and the right part mostly void of influence I see it as rather safe still but less influential than the group in our example or the Gu-Lee game.
I would not call it "thin" though because the aji in it is fairly straightforward.
-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Where to play at the end of the Fuseki (practical exampl
I've been too tied up to come back on this, but if I may try to move things along again...
First, what I am detecting here is a conflation of the two kinds of thickness, atsumi and atsusa. But there are important differences. Simplifying a bit, atsumi (lines of stones facing outwards) belongs to the opening and atsusa (groups already with two eyes) belongs to the middle game. Strategic and tactical objectives in each of those phases tend to differ markedly.
Second, I am also detecting a view that "it looks like thickness so it must be thick." This is like the old problem with good shape (katachi). It's static. The alleged superiority of haengma is that it incorporates a dynamic element (though Japanese covers that with the formula katachi + suji = haengma). Almost every amateur kyu game you look at, but dans could be included in this, contains dozens of moves we play just because they look pretty. No surprise in that: it's inherent in go being a game of pattern recognition, compounded by too many blitz games where there is no time to think. But this is not "go." It's stasis. And thickness has the same requirement to be dynamically efficient. What it looks like is neither here nor there. If it isn't acting as thickness then it's not thickness. A lump of clay is not a cup. It's hardly even a cup if it's moulded into a cup shape. For proper use it has to be refined and made fit for purpose - fit to drink from. So long as it does that it can be thimble-size, jumbo size and decorated anyway you like.
Now no-one has demonstrated to me how the White iceberg in the first game here will ever be able to function adequately as thickness at this stage of the game. Yes it looks thick. But how will it ever act as thickness? Telling me it may have two eyes and can ignore one or even two forcing moves is telling me absolutely nothing about its dynamic function as thickness.
I gave an example from a Japanese book of a group that looks like archetypal thickness to the conflaters. But not only does the pro say the White group there is not thick, he says it's actually surprisingly thin. I must confess, the reason I remember this example so well is that I can just about accept it's not thick (has no function as thickness) but I baulk at the idea of it being thin - but I'm one of those who assume the pros know more than the amateurs, so I'm not saying I'm right, just still ignorant.
Now let me go the opposite extreme, from the iceberg to the icicle.
Explaining why White A here is bad, the pro (Kobayashi Satoru) says "Because the triangled stones are thick, White A is too greedy."
The triangled stones are weak: they have cutting points, few liberties, are subject to forcing moves. But they are thick because they do the work as thickness (i.e. Black can push the White invader towards them).
If I saw the OP position and did not know the provenance, I would not hesitate to say it was an amateur game. If I saw the Gu Li game without knowing the provenance I would hesitate. As uberdude points out, there is business going on to the left that has to do with whether White can make his right hand wall function as thickness or not. The fact that it is unclear how this business will turn out perhaps puts a question-mark over White's strategy, which is precisely why it elicited comments by other pros.
So to summarise what I am recommending:
- learn to differentiate the types of thickness and their proper contexts
- accept that thickness is truly only thickness if it works as thickness
- ergo accept that the White group in the OP position is just an ugly blob.
(For those who think uberdude's point that the ugly blob might be acting as a Go Seigen group and that counts as working as thickness, let me remind them that my original article on GSG groups was (a) a piece of journalistic entertainment, and (b) insofar as it was legitimate theorising, it pertained specifically to the areas on the sides of the board adjacent to the corners, which are notoriously difficult to handle. It was based on insights from Go that had confounded other pros. At no point has it ever been associated directly with thickness, but to the extent it can be it would be with atsusa (and thus with the later middle game), NEVER EVER with atsumi. The games being discussed here are still in the fuseki or late fuseki stage.)
First, what I am detecting here is a conflation of the two kinds of thickness, atsumi and atsusa. But there are important differences. Simplifying a bit, atsumi (lines of stones facing outwards) belongs to the opening and atsusa (groups already with two eyes) belongs to the middle game. Strategic and tactical objectives in each of those phases tend to differ markedly.
Second, I am also detecting a view that "it looks like thickness so it must be thick." This is like the old problem with good shape (katachi). It's static. The alleged superiority of haengma is that it incorporates a dynamic element (though Japanese covers that with the formula katachi + suji = haengma). Almost every amateur kyu game you look at, but dans could be included in this, contains dozens of moves we play just because they look pretty. No surprise in that: it's inherent in go being a game of pattern recognition, compounded by too many blitz games where there is no time to think. But this is not "go." It's stasis. And thickness has the same requirement to be dynamically efficient. What it looks like is neither here nor there. If it isn't acting as thickness then it's not thickness. A lump of clay is not a cup. It's hardly even a cup if it's moulded into a cup shape. For proper use it has to be refined and made fit for purpose - fit to drink from. So long as it does that it can be thimble-size, jumbo size and decorated anyway you like.
Now no-one has demonstrated to me how the White iceberg in the first game here will ever be able to function adequately as thickness at this stage of the game. Yes it looks thick. But how will it ever act as thickness? Telling me it may have two eyes and can ignore one or even two forcing moves is telling me absolutely nothing about its dynamic function as thickness.
I gave an example from a Japanese book of a group that looks like archetypal thickness to the conflaters. But not only does the pro say the White group there is not thick, he says it's actually surprisingly thin. I must confess, the reason I remember this example so well is that I can just about accept it's not thick (has no function as thickness) but I baulk at the idea of it being thin - but I'm one of those who assume the pros know more than the amateurs, so I'm not saying I'm right, just still ignorant.
Now let me go the opposite extreme, from the iceberg to the icicle.
Explaining why White A here is bad, the pro (Kobayashi Satoru) says "Because the triangled stones are thick, White A is too greedy."
The triangled stones are weak: they have cutting points, few liberties, are subject to forcing moves. But they are thick because they do the work as thickness (i.e. Black can push the White invader towards them).
If I saw the OP position and did not know the provenance, I would not hesitate to say it was an amateur game. If I saw the Gu Li game without knowing the provenance I would hesitate. As uberdude points out, there is business going on to the left that has to do with whether White can make his right hand wall function as thickness or not. The fact that it is unclear how this business will turn out perhaps puts a question-mark over White's strategy, which is precisely why it elicited comments by other pros.
So to summarise what I am recommending:
- learn to differentiate the types of thickness and their proper contexts
- accept that thickness is truly only thickness if it works as thickness
- ergo accept that the White group in the OP position is just an ugly blob.
(For those who think uberdude's point that the ugly blob might be acting as a Go Seigen group and that counts as working as thickness, let me remind them that my original article on GSG groups was (a) a piece of journalistic entertainment, and (b) insofar as it was legitimate theorising, it pertained specifically to the areas on the sides of the board adjacent to the corners, which are notoriously difficult to handle. It was based on insights from Go that had confounded other pros. At no point has it ever been associated directly with thickness, but to the extent it can be it would be with atsusa (and thus with the later middle game), NEVER EVER with atsumi. The games being discussed here are still in the fuseki or late fuseki stage.)
-
Kirby
- Honinbo
- Posts: 9553
- Joined: Wed Feb 24, 2010 6:04 pm
- GD Posts: 0
- KGS: Kirby
- Tygem: 커비라고해
- Has thanked: 1583 times
- Been thanked: 1707 times
Re: Where to play at the end of the Fuseki (practical exampl
John Fairbairn wrote:- accept that thickness is truly only thickness if it works as thickness
Why don't you think the OP example works as thickness? Can you explain this more precisely? Is it because there aren't nearby groups to attack? Or are there other characteristics you're looking for?
It can be argued that those who think the group is thick must explain why they think the group works as thickness. Conversely, it can be argued that those who think the group is not thick must explain why they think the group does not work as thickness.
Precise definitions are difficult, and sometimes a little bit of intuition is involved. I suspect this is the case both for those that think the group is thick and for those that think it is not thick.
be immersed
-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Where to play at the end of the Fuseki (practical exampl
Conversely, it can be argued that those who think the group is not thick must explain why they think the group does not work as thickness.
I already did, so I regard this request as trolling (again) - or at least argument for argument's sake, which you seem very fond of.
-
Kirby
- Honinbo
- Posts: 9553
- Joined: Wed Feb 24, 2010 6:04 pm
- GD Posts: 0
- KGS: Kirby
- Tygem: 커비라고해
- Has thanked: 1583 times
- Been thanked: 1707 times
Re: Where to play at the end of the Fuseki (practical exampl
John Fairbairn wrote:Conversely, it can be argued that those who think the group is not thick must explain why they think the group does not work as thickness.
I already did, so I regard this request as trolling (again) - or at least argument for argument's sake, which you seem very fond of.
Accusing me of trolling isn't very helpful - that's not what I'm trying to do.
Anyway, I'm just asking for clarification. We can't really discuss the topic if we don't know what the points are that we're discussing. We may even agree and not realize it if we're arguing different points. I'm trying to prevent disagreeing about something that you're not trying to say, hence the request for clarification.
The only argument that seems to have been given as to why this group is *not* thick is that there aren't nearby groups to attack. Unless I'm missing something, which is why I asked for clarification.
And if that's the only reason that it's *not* thick, I don't think it's enough of a reason. I think a group can be thick, even if there's not a nearby group to attack in the immediate future. In fact, I think a group can be thick even if it's not *immediately* useful right now, but has potential to be useful later.
I suppose there's also this:
As objective reasons for calling it potentially weak, it does not have any guaranteed eyes at all yet and is bounded by strong Black groups on both sides. If Black plays a move against it, White probably has to defend (or else accept a truly weak group), but his defensive move would be purely that - there is no move to make territory while defending.
But even if white doesn't make much territory when white eventually defends (if white even needs to defend against another black move in the area), black isn't getting much from the attack, either. While white doesn't get many points from the group, black doesn't seem to have a good way to make many points attacking, either.
In fact, I think that's often common in thick shapes: They may not seem to make a lot of points now, but are potentially useful later.
be immersed
-
Shaddy
- Lives in sente
- Posts: 1206
- Joined: Sat Apr 24, 2010 2:44 pm
- Rank: KGS 5d
- GD Posts: 0
- KGS: Str1fe, Midorisuke
- Has thanked: 51 times
- Been thanked: 192 times
Re: Where to play at the end of the Fuseki (practical exampl
I'd consider that white group thickness (by which I mean I'd call it thickness if I were talking about it in Chinese). It's strong enough that after I invade the bottom side, Black will not be able to find compensation by attacking it.