Does it follow that (a) is quad-sujiba or sujiba4 : Skipping (b), tri-sujiba or sujiba3 :He calls it double sujiba, and it is worse than normal.
Shape for DDKs -- and thickness, too!
-
EdLee
- Honinbo
- Posts: 8859
- Joined: Sat Apr 24, 2010 6:49 pm
- GD Posts: 312
- Location: Santa Barbara, CA
- Has thanked: 349 times
- Been thanked: 2070 times
-
Charles Matthews
- Lives in gote
- Posts: 450
- Joined: Sun May 13, 2012 9:12 am
- Rank: BGA 3 dan
- GD Posts: 0
- Has thanked: 5 times
- Been thanked: 189 times
Re: Shape for DDKs -- and thickness, too!
OK, I might be getting there.John Fairbairn wrote:No, not quite, because the empty inside point in an empty triangle is not a sujiba (it was until the offending extra stone was added).So sujiba = non-nose, in old money?
is not worthwhile in terms of liberties. The black chain goes up from eight to nine, which is a "so what?" result.Quite different here, where
is typically fine. There is no gain in liberties, but clearly the play is much more effective.-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Shape for DDKs -- and thickness, too!
Effectiveness and efficiency are two different things. Yoda mentioned only efficiency. He also, as I recall, made no mention (or just a passing one) of liberties.Quite different here, where
You could try looking at his blog. There are four decent-sized articles replete with diagrams, which I suspect you could follow without text.
I did think of naughtily translating the blog, relying on the grounds it would help sales of his book, but there is not enough life in this forum any more to justify that. Also, I'd prefer to wait until my copy of the book arrives.
-
Charles Matthews
- Lives in gote
- Posts: 450
- Joined: Sun May 13, 2012 9:12 am
- Rank: BGA 3 dan
- GD Posts: 0
- Has thanked: 5 times
- Been thanked: 189 times
Re: Shape for DDKs -- and thickness, too!
In quest of said articles, I found the Japanese Wikipedia page on Yoda:John Fairbairn wrote:You could try looking at his blog.
https://ja.wikipedia.org/wiki/%E4%BE%9D ... 0%E5%9F%BA
With a simple tachi example for sujiba , not a million miles from the first one above. That text is CC-by-SA, anyway. The blog is in the external links, but I didn't immediately understand the navigation.
A combinatorial explanation would cut through the linguistic difficulties,
-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Shape for DDKs -- and thickness, too!
I think you would get more out of http://kisei.yomiuri.co.jp/column/shinan_yoda/01.htm and the three subsequent pages.
For me the linguistic difficulty there is that I haven't the remotest idea what you are talking about.A combinatorial explanation would cut through the linguistic difficulties,
-
Charles Matthews
- Lives in gote
- Posts: 450
- Joined: Sun May 13, 2012 9:12 am
- Rank: BGA 3 dan
- GD Posts: 0
- Has thanked: 5 times
- Been thanked: 189 times
Re: Shape for DDKs -- and thickness, too!
We each have our own ways of researching, and thinking out loud.
I was happy just to get the kanji from the section title. Led me to a Youtube video of Yoda commenting on shape in a pro game; and his Twitter feed. Google Translate is a bit hazardous, but I now understand "useless clogging" FWIW.
I was happy just to get the kanji from the section title. Led me to a Youtube video of Yoda commenting on shape in a pro game; and his Twitter feed. Google Translate is a bit hazardous, but I now understand "useless clogging" FWIW.
-
Charles Matthews
- Lives in gote
- Posts: 450
- Joined: Sun May 13, 2012 9:12 am
- Rank: BGA 3 dan
- GD Posts: 0
- Has thanked: 5 times
- Been thanked: 189 times
Re: Shape for DDKs -- and thickness, too!
To cover the connection between making empty triangles and liberties, here is a fairly generic chain. Let's hope I get this right.
Some of the liberties of the chain, those marked "a", are such that Black on playing them makes a new empty triangle. The "b" points make two empty triangles in three of the cases. There is a quibble about one of the "b" points, because it fills in an empty triangle, or moves it one to the right, however you look at it.
The "a" points add a liberty. The "b" points do not add a liberty.
The remaining liberties are of type "c" and "d". The type "c" is what I would call a nose point. Type "d" is the sort of abject play that adds no liberties to the chain: it in fact loses one, the point on which it is played. That is in contrast to the plays at "c" that add two.
So far so good. Without too much ado, plays that "create an empty triangle" are those that add at most one liberty?
Some of the liberties of the chain, those marked "a", are such that Black on playing them makes a new empty triangle. The "b" points make two empty triangles in three of the cases. There is a quibble about one of the "b" points, because it fills in an empty triangle, or moves it one to the right, however you look at it.
The "a" points add a liberty. The "b" points do not add a liberty.
The remaining liberties are of type "c" and "d". The type "c" is what I would call a nose point. Type "d" is the sort of abject play that adds no liberties to the chain: it in fact loses one, the point on which it is played. That is in contrast to the plays at "c" that add two.
So far so good. Without too much ado, plays that "create an empty triangle" are those that add at most one liberty?
-
Charles Matthews
- Lives in gote
- Posts: 450
- Joined: Sun May 13, 2012 9:12 am
- Rank: BGA 3 dan
- GD Posts: 0
- Has thanked: 5 times
- Been thanked: 189 times
Re: Shape for DDKs -- and thickness, too!
To continue the thought, I imagine a tutorial game could be set up, involving chains of both colours and their shared liberties. This is to emphasise the aspect of "struggle for liberties" hinted at above.
I don't have time to research background right now, but I'm pretty sure the game aspect is not that new.
Anyway, imagine that points A, B, C, ... are on a go board and are empty but adjacent to a black chain and a white chain. The players play them out in turn, and then count their total liberties.
The strategy is clearly to start with the "best" plays. Each point P falls into one of four or five classes as above, considered just for one colour. Actually there are probably more. For tutorial purposes one wouldn't try to be exhaustive, but to set up instructive examples.
So ... this is on-topic for DDKs learning about shape. And I'm pretty sure it is related to core elements of what Yoda is talking about, but coming from a different perspective, naturally.
I don't have time to research background right now, but I'm pretty sure the game aspect is not that new.
Anyway, imagine that points A, B, C, ... are on a go board and are empty but adjacent to a black chain and a white chain. The players play them out in turn, and then count their total liberties.
The strategy is clearly to start with the "best" plays. Each point P falls into one of four or five classes as above, considered just for one colour. Actually there are probably more. For tutorial purposes one wouldn't try to be exhaustive, but to set up instructive examples.
So ... this is on-topic for DDKs learning about shape. And I'm pretty sure it is related to core elements of what Yoda is talking about, but coming from a different perspective, naturally.
-
EdLee
- Honinbo
- Posts: 8859
- Joined: Sat Apr 24, 2010 6:49 pm
- GD Posts: 312
- Location: Santa Barbara, CA
- Has thanked: 349 times
- Been thanked: 2070 times
Curious abiut アキ三角 -- seems to be the empty triangle.http://kisei.yomiuri.co.jp/column/shinan_yoda/01.htm
A quick search: アキ三角 = 空き三角...
Code: Select all
http://ja.wikipedia.org/wiki/形_(囲碁)
-
John Fairbairn
- Oza
- Posts: 3724
- Joined: Wed Apr 21, 2010 3:09 am
- Has thanked: 20 times
- Been thanked: 4672 times
Re: Shape for DDKs -- and thickness, too!
I have been reading Yoda's Theory of Sujiba book as opposed to his blog, and although I have some way to go, many interesting points have emerged.
One is that the focus is not really on empty triangles. although they are used early on for examples. For one thing, an empty triangle has one liberty in the crook of the L, but has two sujiba there, and the number of sujiba is important. Also, it's misleading to think about strings. Yoda's focus is on straight lines of stones (so an empty triangle seems best viewed as having two overlapping lines of stones, hence it has two sets of sujiba).
But more important is that (as you would expect) the book goes much further than the blog, and the main focus is on suji rather than katachi, hence the choice of name sujiba (suji points). Suji is used in various ways in go, and is often translated as 'style', but it is the dynamic counterpart to static katachi ([good] shape), and shoud be thought of as describing the local flow or movement of stones. Suji + katachi = Korean haengma, but there is a case that the Japanese division into two components is a more powerful way of understanding it.
What flows from Yoda's theory is the following situation.
White 1 is bad suji (it's on a sujiba). It should be at A. Black 2 creates an equal shape. White has thus missed an opportunity to create an advantage.
White 1 is fine here. There was no scope to get an advantage and he maintains the equality of symmetry.
Black 1 is poor suji because White 2 makes the position in suji terms slightly favourable for him. Black 1 is poor because Black A creates an inferior empty triangle next, whereas White A, which we can assume is best answered at Black B, would merely create an equal situation.
As this example will show it's not an easy theory to grasp (and of course there's quite a lot more to it, such as a 2-on-1 rule not mentioned in the blog), mainly, in my case at least but I suspect generally, because the usual but inferior way of looking at shapes is too ingrained and this theory really does require looking at positions through fresh eyes. But the examples he gives are convincing, even when also startling. There are situations where the theory fails BTW, but Yoda lists them himself.
In passing he also gives a way of viewing tesuji (the connection is the suji portion of course) which appears to offer an easy way to find tesujis. I haven't checked it out properly yet, partly because there is a later chapter devoted to "The first definition of tesuji in go history" I need to study, but early signs are that it works. The difference in his first throw-away definition is that he talks about forcing moves, which I already understand, but the second definition is based on sujiba, which I haven't fully absorbed yet.
Yoda's writing style has some of the bluster of Kajiwara, and like all Japanese books lacks the clarity that a Sujiba for Dummies would have, but at this stage I am rather inclined to accept that this book may justify its blurb "A way of playing that will change 400 years of go history."
One is that the focus is not really on empty triangles. although they are used early on for examples. For one thing, an empty triangle has one liberty in the crook of the L, but has two sujiba there, and the number of sujiba is important. Also, it's misleading to think about strings. Yoda's focus is on straight lines of stones (so an empty triangle seems best viewed as having two overlapping lines of stones, hence it has two sets of sujiba).
But more important is that (as you would expect) the book goes much further than the blog, and the main focus is on suji rather than katachi, hence the choice of name sujiba (suji points). Suji is used in various ways in go, and is often translated as 'style', but it is the dynamic counterpart to static katachi ([good] shape), and shoud be thought of as describing the local flow or movement of stones. Suji + katachi = Korean haengma, but there is a case that the Japanese division into two components is a more powerful way of understanding it.
What flows from Yoda's theory is the following situation.
White 1 is bad suji (it's on a sujiba). It should be at A. Black 2 creates an equal shape. White has thus missed an opportunity to create an advantage.
White 1 is fine here. There was no scope to get an advantage and he maintains the equality of symmetry.
Black 1 is poor suji because White 2 makes the position in suji terms slightly favourable for him. Black 1 is poor because Black A creates an inferior empty triangle next, whereas White A, which we can assume is best answered at Black B, would merely create an equal situation.
As this example will show it's not an easy theory to grasp (and of course there's quite a lot more to it, such as a 2-on-1 rule not mentioned in the blog), mainly, in my case at least but I suspect generally, because the usual but inferior way of looking at shapes is too ingrained and this theory really does require looking at positions through fresh eyes. But the examples he gives are convincing, even when also startling. There are situations where the theory fails BTW, but Yoda lists them himself.
In passing he also gives a way of viewing tesuji (the connection is the suji portion of course) which appears to offer an easy way to find tesujis. I haven't checked it out properly yet, partly because there is a later chapter devoted to "The first definition of tesuji in go history" I need to study, but early signs are that it works. The difference in his first throw-away definition is that he talks about forcing moves, which I already understand, but the second definition is based on sujiba, which I haven't fully absorbed yet.
Yoda's writing style has some of the bluster of Kajiwara, and like all Japanese books lacks the clarity that a Sujiba for Dummies would have, but at this stage I am rather inclined to accept that this book may justify its blurb "A way of playing that will change 400 years of go history."
Last edited by John Fairbairn on Fri Jun 26, 2015 5:25 pm, edited 1 time in total.
-
EdLee
- Honinbo
- Posts: 8859
- Joined: Sat Apr 24, 2010 6:49 pm
- GD Posts: 312
- Location: Santa Barbara, CA
- Has thanked: 349 times
- Been thanked: 2070 times
Hi John, could you elaborate the above sentence; I don't follow.Black 2 is poor because Black A creates an inferior empty triangle next
I can see
at (a) makes an empty triangle for B:
But what did you mean by "Black 2" ? Did you mean at 
?In which case I don't see any empty triangles for B soon: Or maybe my brain is not parsing correctly. If others understand what I'm missing, please chime in.
Wow.at this stage I am rather inclined to accept that this book may justify its blurb "A way of playing that will change 400 years of go history."
-
ez4u
- Oza
- Posts: 2414
- Joined: Wed Feb 23, 2011 10:15 pm
- Rank: Jp 6 dan
- GD Posts: 0
- KGS: ez4u
- Location: Tokyo, Japan
- Has thanked: 2351 times
- Been thanked: 1332 times
Re:
I haven't seen anything on Yoda's approach except this thread but from more general theories of efficiency note that none of the marked stones are necessary to capture an original stone on 'a' (and one is unnecessary to make an eye). So 'quad-sujiba' might fit pretty well!EdLee wrote:Does it follow that (a) is quad-sujiba or sujiba4 : Skipping (b), tri-sujiba or sujiba3 :He calls it double sujiba, and it is worse than normal.
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21