It is currently Fri Dec 03, 2021 12:09 pm

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 391 posts ]  Go to page Previous  1 ... 12, 13, 14, 15, 16, 17, 18 ... 20  Next
Author Message
Offline
 Post subject: Re: Japonese counting
Post #281 Posted: Mon Sep 13, 2021 7:43 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
Let's continue on this thread, Kvasir.

kvasir wrote:
Gérard TAILLE wrote:
It seems not clear how you handle this famous double-ko "abuse"
Let's take a slightly different position:
Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O . . . . .
$$ | a O . . . . .
$$ | X O . . . . .
$$ | . B O O O . .
$$ | B B B B O . .
$$ | O B . B O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

In this position, due to the liberty at "a" and the double-ko "abuse" then black marked stones become uncapturable (=> alive). How do you correct this result?



First, in step 1 of the first iteration we get that the marked stones are alive.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O . . . . .
$$ | . O . . . . .
$$ | X O . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . W X W X . .
$$ | W W W W X . .
$$ | . W X X X . .
$$ | W W X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



The intention was that the marked stones can not be removed in the second iterations because they are marked alive in the first iteration, so now we can show that the black stones are dead in step 1 of the second iteration.

Click Here To Show Diagram Code
[go]$$W :b4: :b6: :b8: pass to not help white
$$ | . . . . . . .
$$ | O O . . . . .
$$ | 1 O . . . . .
$$ | 5 O . . . . .
$$ | 3 X O O O . .
$$ | X X X X O . .
$$ | 9 X 7 X O . .
$$ | 2 W X W X . .
$$ | W W W W X . .
$$ | . W X X X . .
$$ | W W X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Maybe we should take it to another thread, as was so kindly suggested, but I hope I am communicating my meaning.

Edit ==========
Now I realize that we get a double-ko seki because some of the black stones are marked alive in iteration 1. I am not sure this is a problem or unexpected, how about white needing to play the approach move to capture? Anyway it was a good point.


Oops a approach move seems a very big contraint.
Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X . . . . O .
$$ | X O O O O O .
$$ | . B O O O . .
$$ | B B B B O . .
$$ | O B . B O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]
You cannot really force white to play all tha approach moves?

BTW did you look at the other suggestion: alternating systematically a "ko ban requiring an explicit pass-for-ko" and a "normal ko ban"?

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #282 Posted: Mon Sep 13, 2021 9:08 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
Gérard TAILLE wrote:
You cannot really force white to play all tha approach moves?


Why not? It is a capturing race, it makes sense that liberties have to be filled for the privilege of not having to fight the ko. The same thing happens if you replace the double-ko with a bent-4, you need to fill certain amount of liberties to win the capturing race.

The only problem I see is that it is quite tricky to deal with this kind of capturing races when ending a game, and it is not clear if you require N approach moves in some position that the Nihon Kiin would not require M moves instead.

If the aim is to not require the approach moves, then I have some ideas, but I don't think that is the idea with Japanese rules. Once you don't require approach moves you are actually changing something important. The Japanese rules are in my opinion meant to give a "typical" or "default" result of actual play with area scoring. For example, it is usually not possible to win large ko fights at the end of the game in area scoring, therefore Japanese rules do not allow you to come back to ko shapes that still exists after the game. If you don't require the approach moves it is akin to giving one player a free ride in the capturing race.

Gérard TAILLE wrote:
BTW did you look at the other suggestion: alternating systematically a "ko ban requiring an explicit pass-for-ko" and a "normal ko ban"?


I am not sure what systems you mean, but I briefly considered some other ideas. Mostly discarded, because it would be complex or allow ko fights using ko threats during confirmation.

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #283 Posted: Mon Sep 13, 2021 9:14 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
How works "alternating ko ban" with double ko?

Click Here To Show Diagram Code
[go]$$B
$$ +-------------------
$$ . . . O X O . O . |
$$ . . . O X X O O O |
$$ . . . O X . X O X |
$$ . . . O X X X X . |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]


Click Here To Show Diagram Code
[go]$$B
$$ +-------------------
$$ . . . O X T 1 O . |
$$ . . . O X X O O O |
$$ . . . O X . X O X |
$$ . . . O X X X X . |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]
after :b1: white needs a pass-for-ko to play at :wt:

Click Here To Show Diagram Code
[go]$$B
$$ +-------------------
$$ . . . O X T 1 O . |
$$ . . . O X X O O O |
$$ . . . O X . X O T |
$$ . . . O X X X X 2 |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]
after :w2: white and black needs a pass-for-ko to retake the ko by :wt: or :bt:

Click Here To Show Diagram Code
[go]$$B black :b3: pass for ko at :w2:
$$ +-------------------
$$ . . . O X T X O . |
$$ . . . O X X O O O |
$$ . . . O X . X O . |
$$ . . . O X X X X O |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]
Now black is able to retake the ko but white not

Click Here To Show Diagram Code
[go]$$W white :w4: pass for ko at :b1:
$$ +-------------------
$$ . . . O X . X O . |
$$ . . . O X X O O O |
$$ . . . O X . X O . |
$$ . . . O X X X X O |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]
Now black and white are able to retake their ko

Click Here To Show Diagram Code
[go]$$B
$$ +-------------------
$$ . . . O X . X O . |
$$ . . . O X X O O O |
$$ . . . O X . X O 5 |
$$ . . . O X X X X T |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]
after :b5: white needs a pass-for-ko to play at :wt:

Click Here To Show Diagram Code
[go]$$B
$$ +-------------------
$$ . . . O X 6 T O . |
$$ . . . O X X O O O |
$$ . . . O X . X O X |
$$ . . . O X X X X T |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]
after :w6: white and black needs a pass-for-ko to retake the ko by :wt: or :bt:

Click Here To Show Diagram Code
[go]$$B :b7: pass-for-ko :w8: pass-for-ko
$$ +-------------------
$$ . . . O X O . O . |
$$ . . . O X X O O O |
$$ . . . O X . X O X |
$$ . . . O X X X X . |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]

this loop on 8 moves is very known.
But now is the difference. Because all ko have already used the pass-for-ko then now, the normal ko ban has to be used.

Click Here To Show Diagram Code
[go]$$B :b11: pass
$$ +-------------------
$$ . . . O X O 9 O . |
$$ . . . O X X O O O |
$$ . . . O X . X O X |
$$ . . . O X X X X 0 |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]

In this last diagram :b9: and :w10: are normal ko => :b11: is a normal pass allowing white to play :w12: elsewhere (=> the loop has really been broken).

Click Here To Show Diagram Code
[go]$$Bm11 :w12: elsewhere
$$ +-------------------
$$ . . . O X 4 X O . |
$$ . . . O X X O O O |
$$ . . . O X . X O 3 |
$$ . . . O X X X X O |
$$ . . . O O O O X X |
$$ . . . . . . O O O |
$$ . . . . . . . . . |
$$ . . . . . . . . . |[/go]

After :b13: :w14: then black :b15: will be again a normal pass allowing white to play :w16: elsewhere.

After all that we return to the very beginning with the use of pass-for-ko etc.

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #284 Posted: Mon Sep 13, 2021 9:20 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
kvasir wrote:
Gérard TAILLE wrote:
You cannot really force white to play all tha approach moves?


Why not? It is a capturing race, it makes sense that liberties have to be filled for the privilege of not having to fight the ko. The same thing happens if you replace the double-ko with a bent-4, you need to fill certain amount of liberties to win the capturing race.

The only problem I see is that it is quite tricky to deal with this kind of capturing races when ending a game, and it is not clear if you require N approach moves in some position that the Nihon Kiin would not require M moves instead.

If the aim is to not require the approach moves, then I have some ideas, but I don't think that is the idea with Japanese rules. Once you don't require approach moves you are actually changing something important. The Japanese rules are in my opinion meant to give a "typical" or "default" result of actual play with area scoring. For example, it is usually not possible to win large ko fights at the end of the game in area scoring, therefore Japanese rules do not allow you to come back to ko shapes that still exists after the game. If you don't require the approach moves it is akin to giving one player a free ride in the capturing race.

Gérard TAILLE wrote:
BTW did you look at the other suggestion: alternating systematically a "ko ban requiring an explicit pass-for-ko" and a "normal ko ban"?


I am not sure what systems you mean, but I briefly considered some other ideas. Mostly discarded, because it would be complex or allow ko fights using ko threats during confirmation.


It seems there is here some misunderstanding. White has an eye and black has not => it is not a capturing race and black is expected to be dead without the need for white to add a move.
I tried in another post to clarify what means "alterning ko ban".

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #285 Posted: Mon Sep 13, 2021 10:43 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
jann wrote:
Gérard TAILLE wrote:
jann wrote:
This is a longstanding question in J89 which has two possible answers:
  1. Maybe passing for each ko is only required once for each player (see here)
  2. Or the problem was overlooked by J89 authors, and will be fixed in the future (with some hack like requiring a normal pass of either player before passing for a ko a second time)

1)
If you use "pass once per ko" then black can pass immediatly and wins the game because the two white marked stones are dead in the confirmation phase

2)
Click Here To Show Diagram Code
[go]$$B
$$ +---------------------–-
$$ | 2 X X O X O . . . O X O . O . |
$$ | X O 1 O X O . . . O X X O O O |
$$ | O O X X X O . . . O X . X O X |
$$ | O X X O O O . . . O X X X X . |
$$ | . O X O . . . . . O O O O X X |
$$ | . O X O . . . . . . . . O O O |
$$ | O X X O . . . . . . . . . . . |
$$ | O O O O . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . |[/go]
without using "pass once per ko", in order to win, black must continue at least one move.
After :b1: :w2: then black passes and wins the game.

Nice position, thank you. Even if harmless here since the result is the same (B can force the reinforcement with the same play either way), this does show a case where the attacker might need to play the same ko twice in confirmation.

Quote:
Alternating systematically a "ko ban requiring an explicit pass-for-ko" and a "normal ko ban" seems to resolve the problem.

That idea sounds more complicated and conceptually less sound than my #2 alternative above. Anyway, the Nihon Ki-in is probably aware of the problem by now and will choose an official solution in their next version.


I agree that the idea may appear complicated but it is only a theoritical complexity. In practice it is really very simple. The idea is only to break the loop in double ko and the associated two pass-for-ko.
In practice you have only to know that pass-for-ko applies to all ko except for double ko where normal ko ban is used. In addition I doubt you will often meet a position like the one I found. Maybe another wording may be found (to avoid a loop made only of ko and pass-for-ko) to show it is not that complex, but I am not a expert in wording ;-)

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #286 Posted: Mon Sep 13, 2021 10:49 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
Gérard TAILLE wrote:
It seems there is here some misunderstanding. White has an eye and black has not => it is not a capturing race and black is expected to be dead without the need for white to add a move.


Well, bent-4 shapes also have an eye and even big-eye can lose to an eyeless group. BUT I do agree that no one is going to fill the liberties in the last example you gave. It does beg the question "how do we know?" :D

The rules text doesn't actually require arguments in a specific format to determine the life and death, it is worth thinking about.

I will suggest one thing, the iterative confirmation process I gave could be updated to allow filling dame from groups marked as alive without alternative play (i.e. if a white group is determined to be alive, then white stones can fill all the dame connected to this group before the next iteration) but I fear this might not work: might destroy some sekis, allow a shortcut in capturing races, require detailed rules about which dame is actually allowed to be filled, and possibly contradict some of the examples. There are examples where the status or territory is conditional on some dame being played, so treating dame liberally would risk contradicting these examples.

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #287 Posted: Mon Sep 13, 2021 11:15 am 
Lives in gote

Posts: 400
Liked others: 0
Was liked: 35
Gérard TAILLE wrote:
The idea is only to break the loop in double ko and the associated two pass-for-ko.

That is not the idea, that is the DESIRED EFFECT. I keep saying that rules need to have convincing theoretical explanations, to be robust vs anomalies. Remember the 4 cases of local/remote bent4/moonshine positions, and logical explanations for all but your rules? For ko passes:

J89 traditional: ko pass needed each time
REASON: kos are frozen continuously to force playing them independently

J89 alternative: ko pass for first recapture only
REASON: kos are frozen in relation to the original stop position, but can be thawn later and revert to normal play

my #2 hack above: normal pass of either player before repeated ko pass for same ko
REASON: kos are frozen even deeper if taken and retaken after ko pass

your idea: ko pass for odd (1st, 3rd etc) recapture only
REASON: ... ?

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #288 Posted: Mon Sep 13, 2021 12:51 pm 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
jann wrote:
Gérard TAILLE wrote:
The idea is only to break the loop in double ko and the associated two pass-for-ko.

That is not the idea, that is the DESIRED EFFECT. I keep saying that rules need to have convincing theoretical explanations, to be robust vs anomalies. Remember the 4 cases of local/remote bent4/moonshine positions, and logical explanations for all but your rules? For ko passes:

J89 traditional: ko pass needed each time
REASON: kos are frozen continuously to force playing them independently

J89 alternative: ko pass for first recapture only
REASON: kos are frozen in relation to the original stop position, but can be thawn later and revert to normal play

my #2 hack above: normal pass of either player before repeated ko pass for same ko
REASON: kos are frozen even deeper if taken and retaken after ko pass

your idea: ko pass for odd (1st, 3rd etc) recapture only

REASON: ... ?


Oops I am not sure it is my idea Jann.

when I saw your post below
jann wrote:
This is a longstanding question in J89 which has two possible answers:
  1. Maybe passing for each ko is only required once for each player (see here)
  2. Or the problem was overlooked by J89 authors, and will be fixed in the future (with some hack like requiring a normal pass of either player before passing for a ko a second time)


It seems to me that "alternating ko ban" maybe a interpretation of your point 2. I just pointed that this interpretation looks to work well. In fact it does not look a stupid idea. "Normal ko ban" do not allow to handle ko independantly but "ko ban requiring a pass-for-ko is too strong because it creates an unexpected loop in double ko situation where "normal ko ban" would be better. So why not a mixed ko ban?
Yes Jann it is not a reason but the result looks good with such idea.

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #289 Posted: Tue Sep 14, 2021 12:02 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
Gérard TAILLE wrote:
Oops a approach move seems a very big [constraint].


Dame is empty intersections that is adjacent to stones that are alive. How about using this definition to permit dame to be filled out of order?

The dame-filling rule is that one or more empty intersections connected to only alive stones of one color and either alive or dead stones of the other color, can be filled during confirmation, and this can be done out of normal playing order.


Using this position.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O . . . . .
$$ | . O . . . . .
$$ | X O . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


We start by noting that the marked stones are alive, therefore white is allowed to fill the dame without interruption from black.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | W W . . . . .
$$ | V W . . . . .
$$ | X W . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Now we can show that the marked stones are dead like this:

Click Here To Show Diagram Code
[go]$$W :b4: pass
$$ | . . . . . . .
$$ | O O . . . . .
$$ | V O . . . . .
$$ | X O . . . . .
$$ | 1 X O O O . .
$$ | X X X X O . .
$$ | 5 X 3 X O . .
$$ | 2 O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



Now for your other position we need to apply the dame-filling rule twice, because the tail must be captured twice.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X . . . . O .
$$ | X O O O O O .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



White can fill dame because the marked white stones are alive, and play :w1:.

Click Here To Show Diagram Code
[go]$$W
$$ | . . . . . . .
$$ | . . . . . . .
$$ | W W W W W W .
$$ | V V V V V W .
$$ | X X X X V W .
$$ | X V V V V W .
$$ | X W W W W W .
$$ | 1 X W W W . .
$$ | X X X X W . .
$$ | O X . X W . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



Black plays :b2: .

Click Here To Show Diagram Code
[go]$$W
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | V V V V V O .
$$ | . . . . V O .
$$ | . V V V V O .
$$ | 2 O O O O O .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Now white can use the dame-filling rule again to capture with :w3:

Click Here To Show Diagram Code
[go]$$W
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | V V V V V O .
$$ | V V V V V O .
$$ | V V V V V O .
$$ | X O O O O O .
$$ | 3 X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Now the life and death can be determined in same way as before.

Click Here To Show Diagram Code
[go]$$W :b6: pass
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | V V V V V O .
$$ | V V V V V O .
$$ | V V V V V O .
$$ | . O O O O O .
$$ | 3 X O O O . .
$$ | X X X X O . .
$$ | 7 X 5 X O . .
$$ | 4 O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



Maybe this is workable to solve the dame filling problem in general?

If so then I am thinking about how to make an iterative process actually work, that is to allow one to make the claim that the outside stones are alive first and then use that to invoke dame filling.


Edit =====
Actually, one may say that :w3: is capturing a ko stone but I think that could be fixed by defining it away by saying something about ko-shapes.


Last edited by kvasir on Tue Sep 14, 2021 12:23 am, edited 1 time in total.
Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #290 Posted: Tue Sep 14, 2021 12:18 am 
Lives in sente
User avatar

Posts: 1216
Liked others: 11
Was liked: 140
Rank: German 1 Kyu
"Reductio ad absurdum" is a common practice in mathematics.

:w1: In the example, you first claim that Black's stones are alive.

:w2: Second, you apply procedures, which would be valid if the claim was true.

:b3: Third, you show that it results a contradiction with your claim. ==> The claim cannot be true.

_________________
The really most difficult Go problem ever: http://igohatsuyoron120.de/index.htm
Igo Hatsuyoron #120 (still unresolved by professionals, maybe solved by four amateurs, really solved by KataGo)

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #291 Posted: Tue Sep 14, 2021 2:48 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
kvasir wrote:
Gérard TAILLE wrote:
Oops a approach move seems a very big [constraint].


Dame is empty intersections that is adjacent to stones that are alive. How about using this definition to permit dame to be filled out of order?

The dame-filling rule is that one or more empty intersections connected to only alive stones of one color and either alive or dead stones of the other color, can be filled during confirmation, and this can be done out of normal playing order.


Using this position.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O . . . . .
$$ | . O . . . . .
$$ | X O . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


We start by noting that the marked stones are alive, therefore white is allowed to fill the dame without interruption from black.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | W W . . . . .
$$ | V W . . . . .
$$ | X W . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



The idea is interesting Kvasir but I am not sure it could work here.

In the position above taking the context of your proposal, the black stones are also alive because they are simply not capturable (if in the confirmation phase white approachs the ko then black starts the loop in the double ko)

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | W W . . . . .
$$ | a W . . . . .
$$ | B W . . . . .
$$ | . B O O O . .
$$ | B B B B O . .
$$ | O B . B O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

As a consequence the dame at "a" cannot be filled because it is adjacent to an alive white groupe and to an alive black group.

BTW in Go it is quite difficult to recognise if a dame can be filled or not (I mean without side effects). For that reason the rule requests that these dame must be filled during normal play. If they are not filled then one of the player may well have a very bad surprise with the result of the confirmation phase (and the other player a good surprise).

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #292 Posted: Tue Sep 14, 2021 3:54 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
Gérard TAILLE wrote:
In the position above taking the context of your proposal, the black stones are also alive because they are simply not capturable (if in the confirmation phase white approachs the ko then black starts the loop in the double ko)

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | W W . . . . .
$$ | a W . . . . .
$$ | B W . . . . .
$$ | . B O O O . .
$$ | B B B B O . .
$$ | O B . B O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

As a consequence the dame at "a" cannot be filled because it is adjacent to an alive white groupe and to an alive black group.


What I wrote is probably not as clear or precise as it could be, but the intention is that the color that has alive stones can fill the dame irrespective of the other color. That is white can fill dame from an alive group against an alive black group, against a dead black group, or against a black group that has unclear status. However, white can not fill dame if there is a dead white stone, which is the case of ko stone in the double-ko.

kvasir wrote:
The dame-filling rule is that one or more empty intersections connected to only alive stones of one color and either alive or dead stones of the other color, can be filled during confirmation, and this can be done out of normal playing order.


The requirement for the dame filler to be alive irrespective of the other color is to allow "a" to be filled, but forbid
"b" from being filled in a double-ko seki.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O . . . . .
$$ | a O . . . . .
$$ | B O . . . . .
$$ | . B O O O . .
$$ | B B B B O . .
$$ | O B . B O . .
$$ | b O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



As per example 25 the marked stone is dead and therfore "b" can not be filled.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O O . . . .
$$ | X X O . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | P X . X O . .
$$ | b O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]



Gérard TAILLE wrote:
BTW in Go it is quite difficult to recognise if a dame can be filled or not (I mean without side effects). For that reason the rule requests that these dame must be filled during normal play. If they are not filled then one of the player may well have a very bad surprise with the result of the confirmation phase (and the other player a good surprise).


That is true. I haven't tried a full treatment of this, I think there are some issues related to different kinds of life stones. In general issues with keeping the rest of the J89 definitions unchanged.

Not sure if I should go into details, I have taken note of some consequences of using this dame-filling rule as-is with J89. For example, unless different treatment is introduced for different types of life stones then seki must consist of dead stones that can't be removed because they are not surrounded by alive stones; this is different from the examples in J89 even though the territory would appear to be the same.

Like you said, it might not work, at least it would be different from the the J89 framework.

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #293 Posted: Tue Sep 14, 2021 4:32 am 
Lives in sente
User avatar

Posts: 1216
Liked others: 11
Was liked: 140
Rank: German 1 Kyu
Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O O O . . .
$$ | C C C O . . .
$$ | X X X O . . .
$$ | X O O O . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

During actual play, Black needed THREE external liberties ...

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O O O . . .
$$ | . 4 2 O . . .
$$ | X X X O . . .
$$ | X O O O . . .
$$ | 1 X O O O . .
$$ | X X X X O . .
$$ | O X 3 X O . .
$$ | 5 O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

... to win the semeai (ko-fight excluded).

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O O O . . .
$$ | M M M O . . .
$$ | X X X O . . .
$$ | X O O O . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

The current J89 makes sure that Black's group with THREE external liberties at most will be judged "dead" in status confirmation, while White's group is "alive" (FOUR external liberties would result in "seki").
This complies with the Japanese understanding of such positions, as Black's fate depends on a ko, while White's group has a solid eye.

_________________
The really most difficult Go problem ever: http://igohatsuyoron120.de/index.htm
Igo Hatsuyoron #120 (still unresolved by professionals, maybe solved by four amateurs, really solved by KataGo)


Last edited by Cassandra on Tue Sep 14, 2021 6:14 am, edited 2 times in total.
Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #294 Posted: Tue Sep 14, 2021 5:57 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
Cassandra wrote:
The current J89 makes sure that Black's group with TWO external liberties at most will be judged "dead" in status confirmation, while White's group is "alive".
This complies with the Japanese understanding of such positions, as Black's fate depends on a ko, while White's group has a solid eye.


This is what I had in mind before, it seems to follow from the j89 rules and I have seen it stated before when people discus rules. There is however no direct example of this in the examples that I can find. I'd be curious to see actual examples of this from games, because either someone adds these moves at the very end of the game or they don't.

It is actually a prudent question, does anyone actually play by the j89 rules?

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #295 Posted: Tue Sep 14, 2021 6:15 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
For what is worth. GoPanda2 score estimator follow this pattern as long as it is white to move. When it is black to move it is different.


Attachments:
rules.png
rules.png [ 270.31 KiB | Viewed 893 times ]
Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #296 Posted: Tue Sep 14, 2021 6:45 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
kvasir wrote:
For what is worth. GoPanda2 score estimator follow this pattern as long as it is white to move. When it is black to move it is different.

Now I see where is the misunderstanding. You have just analysed the position in normal play while I was in confirmation phase and still on you explanation of the four steps and the iterative process you illustrated in https://lifein19x19.com/viewtopic.php?p=267334#p267334.

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | O O . . . . .
$$ | . O . . . . .
$$ | X O . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X . . . . O .
$$ | X O O O O O .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Do you agree that in the current J89 CONFIRMATION PHASE the two positions above lead to the same result : seki.
If it is true the number of liberties for the black group of stones at the top is irrelevant isn't it?

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #297 Posted: Tue Sep 14, 2021 7:35 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
You said this...

Gérard TAILLE wrote:
It seems there is here some misunderstanding. White has an eye and black has not => it is not a capturing race and black is expected to be dead without the need for white to add a move.


and now you say this...
Gérard TAILLE wrote:
Do you agree that in the current J89 CONFIRMATION PHASE the two positions above lead to the same result : seki.
If it is true the number of liberties for the black group of stones at the top is irrelevant isn't it?



I think it depends on if you wish to take the j89 pass-ko literally and if you take the view that eternal pass-ko cycles in confirmations don't exist. I can see someone refusing both or accepting any combination of the two. It also depends on if you are willing to consider what will be the actual ruling on this position.

I don't think I can take the view in PRACTICE that either of these positions is seki in confirmation, but this is complicated because dame is supposed to be filled before finishing confirmation and it is not clear to me how to best handle irregularities like this in practice. Basically, there is a difference between our best understanding of the words in j89 and the actual procedure of ending a game played using j89 rules. So I can agree that if we were to hold the view that most of j89 is to be taken literally and that eternal pass-ko cycles are not excluded, then these positions (and I think many more) are seki; but I can not unconditionally adopt the premise of that view.

We do seem to have crosstalked by not stating our premise and assuming it would be understood.

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #298 Posted: Tue Sep 14, 2021 8:21 am 
Lives in sente
User avatar

Posts: 1216
Liked others: 11
Was liked: 140
Rank: German 1 Kyu
Gérard TAILLE wrote:
Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X . . . . O .
$$ | X O O O O O .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Do you agree that in the current J89 CONFIRMATION PHASE the two positions above lead to the same result : seki.
If it is true the number of liberties for the black group of stones at the top is irrelevant isn't it?

Click Here To Show Diagram Code
[go]$$B
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X . . . . O .
$$ | X O O O O O .
$$ | 1 X O O O . .
$$ | X X X X O . .
$$ | O X 2 X O . .
$$ | 3 O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

Click Here To Show Diagram Code
[go]$$B :b7: pass-for-ko
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X 4 6 . . O .
$$ | X O O O O O .
$$ | X X O O O . .
$$ | X X X X O . .
$$ | 5 X O X O . .
$$ | X O . O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

Click Here To Show Diagram Code
[go]$$Bm7
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X 4 O .
$$ | X O O 2 . O .
$$ | X O O O O O .
$$ | X X O O O . .
$$ | X X X X O . .
$$ | X X O X O . .
$$ | X O 3 O X . .
$$ | O O O O X . .
$$ | 5 O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

White is NOT uncapturable.

------------------

Click Here To Show Diagram Code
[go]$$W
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X 3 . . . O .
$$ | X O O O O O .
$$ | 2 X O O O . .
$$ | X X X X O . .
$$ | O X 1 X O . .
$$ | 4 O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

Click Here To Show Diagram Code
[go]$$W :b8: pass-for-ko
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X O 5 7 . O .
$$ | X O O O O O .
$$ | X X O O O . .
$$ | X X X X O . .
$$ | 6 X O X O . .
$$ | X O . O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

Click Here To Show Diagram Code
[go]$$Wm9
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . 3 . O .
$$ | X X X X 1 O .
$$ | X O O O . O .
$$ | X O O O O O .
$$ | X X O O O . .
$$ | X X X X O . .
$$ | X X O X O . .
$$ | X O 2 O X . .
$$ | O O O O X . .
$$ | 4 O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]

Black is uncapturable.

------------------

Concluding result:
Black is alive; White is dead.

However, there is NO Black territory, due to the DAME.
The position is NOT a seki, so...

==> Black will request resumption of the game, and take White's stones off the board.

_________________
The really most difficult Go problem ever: http://igohatsuyoron120.de/index.htm
Igo Hatsuyoron #120 (still unresolved by professionals, maybe solved by four amateurs, really solved by KataGo)

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #299 Posted: Tue Sep 14, 2021 8:44 am 
Lives with ko

Posts: 255
Liked others: 17
Was liked: 30
Rank: panda 4 dan
IGS: kvasir
Cassandra wrote:
Click Here To Show Diagram Code
[go]$$B :b7: pass-for-ko
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X 4 6 . . O .
$$ | X O O O O O .
$$ | X X O O O . .
$$ | X X X X O . .
$$ | 5 X O X O . .
$$ | X O . O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Obviously, black has time to connect. Me not thinking :scratch:

Top
 Profile  
 
Offline
 Post subject: Re: Japonese counting
Post #300 Posted: Tue Sep 14, 2021 10:21 am 
Lives in sente

Posts: 826
Liked others: 16
Was liked: 37
Rank: 1er dan
kvasir wrote:
We do seem to have crosstalked by not stating our premise and assuming it would be understood.


OK Kvasir let's try to be clearer.

Click Here To Show Diagram Code
[go]$$B Position 1
$$ | . . . . . . .
$$ | O O . . . . .
$$ | . O . . . . .
$$ | X O . . . . .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Click Here To Show Diagram Code
[go]$$B Position 2
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X . . . . O .
$$ | X O O O O O .
$$ | . X O O O . .
$$ | X X X X O . .
$$ | O X . X O . .
$$ | . O X O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Assume we are in CONFIRMATION PHASE

First of all in the curent J89 (with its flaws ;-)) the two above positions are seki.

What about your proposal which may correct some flaws (I do not know).
I understood that in position 1 black is dead but I was not sure about position 2? You mentionned that external dame may be filled and that sounded an interesting idea. When I assume dame are filled then I see that black group is always dead. It looks strange OC but that does not harm : in any case it is an unfinished position and black will not stop the game.
Why not concluding black is dead in any case (without reagrding the number of liberties (dame)? It is fine with only one liberty (and it corrects J89 flaw) and it does not harm with more liberties because the position become unfinished.
I guess now it was not your proposal but it was not that stupid.

kvasir wrote:
Cassandra wrote:
Click Here To Show Diagram Code
[go]$$B :b7: pass-for-ko
$$ | . . . . . . .
$$ | . . . . . . .
$$ | O O O O O O .
$$ | . . . . . O .
$$ | X X X X . O .
$$ | X 4 6 . . O .
$$ | X O O O O O .
$$ | X X O O O . .
$$ | X X X X O . .
$$ | 5 X O X O . .
$$ | X O . O X . .
$$ | O O O O X . .
$$ | . O X X X . .
$$ | O O X . . . .
$$ | X X X . . . .
$$ | . . . . . . .[/go]


Obviously, black has time to connect. Me not thinking :scratch:


In the above position the white group is alive because :w4: and :w6: are in the "enable" region (in J2003 they are in local2)

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 391 posts ]  Go to page Previous  1 ... 12, 13, 14, 15, 16, 17, 18 ... 20  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group