Yesterday's rule dispute in Korea

Pio2001 · **#121**

Pio2001 wrote:

On the other hand, the French rule forgets to tell that the dead stones must be removed if pass stones are used to count the game !
[...]
they forgot to reintroduce the part about removing dead stones there !

In fact, I'm wrong. I misunderstood the rule : there is no need to tell that dead stones must be removed, even counting territory + prisoners + pass stones, since, thanks to the pass stones, their capture costs nothing.
In the french rule, the removal of dead stones appears in the commentary, but is not part of the rule's statements.

Bill Spight · **#122**

RobertJasiek wrote:

I realise that I do not know: what is the Capture Game?

The first player to capture one or more stones wins.

Quote:

Click Here To Show Diagram Code: [go]$$Wcm16 No Pass Baduk $$ ---------- $$ | . O X X . | $$ | 4 O O X 2 | $$ | 5 O X X 1 | $$ | 6 O O X 3 | $$ | . O X X . | $$ -----------[/go]

Why is the lower right eye worth 2 points?

Click Here To Show Diagram Code: [go]$$Wc No Pass Baduk $$ ---------- $$ | . O X X . | $$ | O O O X X | $$ | . O X X O | $$ | O O O X O | $$ | . O X X . | $$ -----------[/go]

The left side is worth -3 + 2 = -1, right? The right side, I claim, is worth 1 + 2 - 2 = 1. In which case the total score is 0, so whoever moves first loses.

Click Here To Show Diagram Code: [go]$$Wc White first $$ ---------- $$ | . O X X . | $$ | O O O X X | $$ | 1 O X X O | $$ | O O O X O | $$ | . O X X 2 | $$ -----------[/go]

Click Here To Show Diagram Code: [go]$$Wc White first, cont. $$ ---------- $$ | . O X X . | $$ | O O O X X | $$ | O O X X 4 | $$ | O O O X 3 | $$ | . O X X X | $$ -----------[/go]

White to play loses, as advertised.

Click Here To Show Diagram Code: [go]$$Bc Black first $$ ---------- $$ | . O X X . | $$ | O O O X X | $$ | . O X X O | $$ | O O O X O | $$ | . O X X 1 | $$ -----------[/go]

Click Here To Show Diagram Code: [go]$$Bc Black first, cont. $$ ---------- $$ | . O X X . | $$ | O O O X X | $$ | 4 O X X 2 | $$ | O O O X 3 | $$ | . O X X X | $$ -----------[/go]

Black to play loses, as advertised.

RobertJasiek · **#123**

Oh. I expected an eye-local evaluation...!

Bill Spight · **#124**

RobertJasiek wrote:

What is Bill Frazor's fractional territory position and your evaluation of it?

Fraser's position is a ko position, OC, since the ko takes place after the dame are filled. IIRC, each play loses ⅓ pt. on average. IOW, each play gains ⅔ pt. on average by area scoring. I'll see if I can dig it up.

RobertJasiek · **#125**

"The result of this exchange is 16 - 10 = 6 pts. This is the result regardless of who plays first, and is therefore the value of the ko under these conditions. When Black is komaster and White's reply elsewhere gains 10 pts. the value of the ko is 6 pts. If White's reply gains 4 pts. the value of the ko is 12 pts. If White's reply gains 0 pts. (White fills a dame) the value of the ko is 16 pts. If White's reply loses 1 pt. (White fills in territory) the value of the ko is 17 pts. How the value of the ko changes with the value of plays elsewhere when one player is komaster"

Why do you call it the value of the ko and do not call it the value of the ko ensemble?

Bill Spight · **#126**

RobertJasiek wrote:

Oh. I expected an eye-local evaluation...!

Well, it's there in the variations. In CGT terms, {1|2||} = 2.

Bill Spight · **#127**

RobertJasiek wrote:

"The result of this exchange is 16 - 10 = 6 pts. This is the result regardless of who plays first, and is therefore the value of the ko under these conditions. When Black is komaster and White's reply elsewhere gains 10 pts. the value of the ko is 6 pts. If White's reply gains 4 pts. the value of the ko is 12 pts. If White's reply gains 0 pts. (White fills a dame) the value of the ko is 16 pts. If White's reply loses 1 pt. (White fills in territory) the value of the ko is 17 pts. How the value of the ko changes with the value of plays elsewhere when one player is komaster"

Why do you call it the value of the ko and do not call it the value of the ko ensemble?

I was explaining komaster values. Ko ensemble is a term from my theory before I learned Berlekamp's komaster theory.

RobertJasiek · **#128**

Ok, but what then is a komaster value? A count? In which sense? Of what? How to interpret it in the global context?

Bill Spight · **#129**

RobertJasiek wrote:

Ok, but what then is a komaster value? A count? In which sense? Of what? How to interpret it in the global context?

A komaster value is a count. In some cases it depends on the global temperature. At those temperatures the ko mast is not vertical, but inclined.

Example: Suppose we have a so-called half point ko where nothing else is left except many dame and Black is komaster. Black to play can fill the ko for 1 pt. The value of the ko is 1 pt. in that case.

Now, the mast value of the ko is the value above which the mast is vertical, and it is ⅔ in that case. Sometimes people informally refer to the mast value simply as the value, which can be confusing.

RobertJasiek · **#130**

I am afraid I understand nothing. Thermographs do not help my understanding at all.

Bill Spight · **#131**

RobertJasiek wrote:

I am afraid I understand nothing. Thermographs do not help my understanding at all.

But you do understand how, in the example I gave, the result will be 1, regardless of who plays first, right?

RobertJasiek · **#132**

Yes if territory scoring is used and the ko is a basic endgame ko (not a real half point ko).

Bill Spight · **#133**

RobertJasiek wrote:

Yes if territory scoring is used and the ko is a basic endgame ko (not a real half point ko).

OK. Example 2: Suppose we have a ko with Black, komaster, where Black to play could win the ko in 1 move for 6 pts., and if White won the ko with two moves the local score would be 0. Suppose also that there are many plays that gain 1 pt. and nothing else that gains more, except the ko. How do we value the ko by komaster theory?

RobertJasiek · **#134**

I would want it to be 6 but you will say 5.

Bill Spight · **#135**

RobertJasiek wrote:

I would want it to be 6 but you will say 5.

Not just me, but Professor Berlekamp.

You get it.

And if there are only dame left, the value is 6.

And after the dame stage, if Black has an unremovable ko threat but White has no threat, the value is 7. (Unless the rules require Black to win the ko at the dame stage, OC.

)

jaeup · **#136**

Pio2001 wrote:

How can it be ? I was under the impression that under japanese rules, if the only thing left on the board is an unfinished ko, the intersection that is inside is not a point of territory (because of the unwritten rule that says that if we lift ko bans and give the sente to the opponent, the ko may be recaptured, therefore one of the stones around the empty intersection is dead).
Why is it different in this game ? According to the same reasoning, the white stone A16 should be dead, therefore the intersection A15 is dame, therefore White has no territory in A13 until she fills A15.

In the hypothetical play, after Black's ko capture, White can afford to make a pass to recapture the ko and eventually capture the whole Black group. Korean rule does not mandate a pass for the ko capture, but there is no problem doing so. Thus, in both rules, the whole Black group is captured at the end of the hypothetical play. (Technically, Black fails to create a new stone uncapturable for White.)

There are a few possible Black strategies, but none is successful. If you still have trouble following the hypothetical play, I will give you a few diagrams.

Bill Spight · **#137**

Going back to the original position, I have made an SGF file for no pass go with prisoner return, no group tax, using the method of exiting the ko to evaluate it. The file clearly shows that Black's threat is not to win the ko, but to convert it to a regular ko. But when there are no ko threats, White is komaster of that ko. And that makes White komaster of the approach ko, as well. As komaster White wins the approach ko, but at the cost of 1 pt. so the original position is worth 1 pt. more for White.

Matti · **#138**

Bill Spight wrote:

Matti wrote:

In some cases the area and territory scoring are fundamentally different. See the first diagram at http://www.harryfearnley.com/go/bestiary/msiivola/siivola.html. With territory scoring the position is seki, but with area scoring once the dame have been filled, white can start filling a big eye with no extra cost, and force balck to select which white group to capture.this leaves the other white group alive with two eyes and the adjanced black group dead.

With apologies to the late John Rickard, it is a matter of the rules. Yes, this is a standoff, and the J89 rules permit Black to pass without cost while White fills the big eye. However, as I believe Honinbo Shuwa understood, the standoff "should" be resolved when plays cost 1 point, as with capturing dead stones, when you don't let the opponent pass for free. In that case White plays first and Black replies on the board. The result is 9 pts. for White after Black makes the last play. No Pass Go with Prisoner Return indicates the same value, and it uses territory scoring.

If the Japanese could agree on Honinbo Shuwas interpretation, the proböem would be solved. If not the difference could be escalated by making the groups bigger.

Quote:

John Fairbairn · **#139**

Nothing to do with modern rules or disputes, but there is interesting food for thought in something Go Seigen said.

I have been working on the book of the Go vs Iwamoto ten-game match. One game ended in a jigo. In the post-game comments Go made the suggestion that the 芇 Japanese character for jigo should be changed to the Chinese character 和. The Japanese didn't go that far, and they now tend to use katakana anyway, but if characters are insisted upon, 持碁 is now normal (though that risks some confusion with 持 = seki).

But the interesting aspect is why 芇 was chosen in the first place. I assume it came in via Chinese. The ancient Chinese Shuo Wen dictionary gives it as meaning 'to hit' but it was obscure even then. It seems to have acquired the meaning 'to lose' but in a gambling context (i.e. to be hit??) before acquiring a go sense which the old book 通玄集 gives as meaning 'neither side wins or loses.'

I don't know the date of the 通玄集 (can anyone help?) but the apparent transition from 'lose' (at least in some games) to 'jigo' may hint at a change in old Chinese rules.

Pio2001 · **#140**

jaeup wrote:

(Technically, Black fails to create a new stone uncapturable for White.)

...and White, after being captured, can play a stone that can't be captured by Black. I see !

Thanks for the explanation. The sequences were ok for me. It was the part "capturing them would enable a new stone to be played that the opponent could not capture" that I was missing.
I didn't think about that because I had associated this part of the rule with snapback. It didn't come to my mind while thinking about a ko.

Yesterday's rule dispute in Korea

Who is online