It is currently Mon Jul 24, 2017 5:40 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 7 posts ] 
Author Message
Offline
 Post subject: Simplified Japanese Rules Applied
Post #1 Posted: Thu Nov 05, 2015 3:10 pm 
Tengen

Posts: 4398
Liked others: 0
Was liked: 584
Quotation reference: viewtopic.php?p=195159#p195159

Pio2001 wrote:
RobertJasiek wrote:
http://home.snafu.de/jasiek/sj.html

If I understand them properly, there is no point in filling the teire if the players passed their turn, the empty intersection in the last ko, if not connected, is worth one point of territory, and if a four bent remains together with a small seki, the four bent, being proven dead, is removed while the seki remains on the goban ?


Before the players end the alternation by passing, usually they fill enough dame to motivate the opponent to fill teire so that the opponent does not have territory on teire intersections because of surrounding them with his independently alive strings. (This presumes that, if necessary, the analysis would be performed competently. The rules permit incompetent performance. There is also the possibility of value-symmetries where unfilled teire would "accidentally" create the same score.)

If there is exactly one unfilled basic endgame ko at the end of the alternation, it depends on how the players perform the analysis. Competent play in the analysis without ko fight means that the ko is either filled by the defender moving first in the analysis or captured-then-filled by the attacker moving first in the analysis. The defender's ko stone would thus be proven as independently alive while the attacker can establish his two-eye-formation but thereby not prove any ko stone of his as independently alive in the position at the end of alternationn because the attacker does not have any such string in the ko in that position. Incompetent play can lead to a different assessment of the defender's ko stone. Since there is such variety of considerations involved including a ko fight during the analysis, one or both players can be expected to prefer avoiding this by winning and filling the ko already during the alternation. Positional superko means that rare positions can occur in which not filling the last basic endgame ko is correct strategy because the opponent does not have any legal play; then the defender can postpone filling the ko until the analysis to prove the ko stone's independent life and the empty ko intersection's territory in the position at the end of the alternation in only these rare cases; such a game has been reported once for 9x9 (and might occur there at times) but is inconceivable for 19x19. EDIT: There is also the possibility of winning the ko fight during the alternation, the opponent passing because of having no dame and not wishing to fill territory (losing 1 point by doing so), during the analysis the opponent may still not recapture due to positional superko, the player creates independent life and so proves one extra territory intersection in the ko. This case occurs in ca. every 20th game and means that the rules are not for traditionalists; they might prefer to avoid superko (for various other reasons, too).

Bent-4 and seki can lead to different results depending on the exact shape and the global ko threat situation. I am too tired to work out now whether some such shapes should not be dissolved during the alternation.

EDITED.

Top
 Profile  
 
Offline
 Post subject: Re: Simplified Japanese Rules Applied
Post #2 Posted: Fri Nov 06, 2015 5:15 pm 
Lives with ko

Posts: 186
Liked others: 0
Was liked: 31
Rank: kgs 6 kyu
KGS: Pio2001
Hi,
Here is an example of conditional life involving a bent four and a seki (this is an imaginary game) :

Image

The bent four is in the top left corner. The only threat left is the seki in D5, but it is too small to overcome the bent four.

Japanese rules : top left : Black is dead; center : Black and White live in seki. White wins.

Under area scoring, the stones must be actually captured, which leads to the following position.

Image

In order to win the ko, White has to sacrifice the seki, and Black wins !

The interesting points in this position are
-The status of the four bent can't be discussed : it is dead under both real and hypothetical play.
-Black can't prove that the White chain in the seki is dead. There is no way for her to capture it if White doesn't cooperate.

Under area scoring, Black must demand the strict application of the rule in order to claim her victory : there is no agreement about so-called "dead groups". Each player has 1 point per stone + 1 point per intersection surrounded by the stones, end of story.

The french rule allows this by carefully avoiding to state that a disagreement should be about the status of a group of stones. Any disagreement allows to resume the game.
On the other hand, the rule doesn't state that the game has to go on until the dead stones are actually captured. It is just implicit in the definition of the score.

I don't know the details of the AGA rule about this.

Now, if I follow your simplified japanese rule, what happens if White passes during the analysis, and everything stays the same ?
The analysis failed to create a two-eye formation in the top left corner in the position of the A13 white chain or in the position of the E13 black chain. The whole region is surrounded by only one independently alive chain, that is white's in H13.
the rule says "They consider the connected regions that are adjacent only to one player's independently alive strings and that consist of intersections being empty or having opposing not independently alive strings on, which they remove".
This region doesn't match this definition : it also has a not independently alive string of the same colour. Therefore nothing is removed from it (*). Black wins.

On the other hand, if white captures the bent four during the analysis, then a two-eye-formation has been formed in the top and another one in the center. The bent four is removed as well as the seki, and Black wins.

The way I see it, you avoid all the complications of the japanese rule by giving up the tradition of the teire points (that lead to the new definition of seki in the 1989 rule), the tradition of having the players paying the price of resolving kos and the tradition of having the four bent dead (that lead to the different rules in alternation and analysis in the 1989 rule). Thus returning to a strict territory scoring.
I like this.


(*) I interpret the rule this way so that this position remains seki :

Attachment:
seki.png
seki.png [ 11.27 KiB | Viewed 971 times ]

Top
 Profile  
 
Offline
 Post subject: Re: Simplified Japanese Rules Applied
Post #3 Posted: Fri Nov 06, 2015 5:26 pm 
Lives with ko

Posts: 186
Liked others: 0
Was liked: 31
Rank: kgs 6 kyu
KGS: Pio2001
It's late and I must go, but there is something still unclear to me : in the simple seki attached in the above post, isn't the region from E3 to J3, plus E2 to J2, plus E1, plus J1 (thus excluding the three white stones), a "connected region that [is] adjacent only to one player's independently alive strings and that consist of intersections being empty or having opposing not independently alive strings on" ?
Thus, shouldn't the black stones be removed before counting ??

Top
 Profile  
 
Offline
 Post subject: Re: Simplified Japanese Rules Applied
Post #4 Posted: Fri Nov 06, 2015 8:00 pm 
Lives in gote

Posts: 635
Location: Littleton, CO
Liked others: 227
Was liked: 187
Rank: KGS 4k
Universal go server handle: jeromie
I can't answer the more complicated questions, but stones in seki are not considered "independently alive." Thus the white string at G1 is not independently alive, and the black string is not adjacent to only independently alive strings of the opposing color.

Top
 Profile  
 
Offline
 Post subject: Re: Simplified Japanese Rules Applied
Post #5 Posted: Fri Nov 06, 2015 10:19 pm 
Tengen

Posts: 4398
Liked others: 0
Was liked: 584
Pio2001 wrote:
if I follow your simplified japanese rule, what happens if White passes during the analysis, and everything stays the same ?
The analysis failed to create a two-eye formation in the top left corner in the position


If, during the analysis, the players only pass, then (usually) it does not create any two-eye-formation anywhere on the board, so none of the strings on the board is independently alive, so there is exactly no territory on the whole board.

Pio2001 wrote:
in the simple seki [...] shouldn't the black stones be removed before counting ??


Opposing not independently alive strings are removed only from "connected regions that are adjacent only to one player's independently alive strings and that consist of intersections being empty or having opposing not independently alive strings on". In a simple seki, any region consisting of one player's string(s) and its adjacent empty strings is not adjacent only to one player's (i.e., the opponent's) independently alive strings because the opponent's seki strings are not independently alive strings because the analysis does not create any two-eye-formation on at least one intersections of any of the opponent's simple seki strings.

Top
 Profile  
 
Offline
 Post subject: Re: Simplified Japanese Rules Applied
Post #6 Posted: Fri Nov 06, 2015 11:10 pm 
Judan

Posts: 6208
Liked others: 1465
Was liked: 2367
BTW, normally, in a game with this bent four and this seki and no other ko threats, under rules in which this combination would be played out at the end of play, within a certain temperature range White should start the ko and answer the Black threat and then gain the value of a play elsewhere in the ko exchange. In this case under area scoring that range is 4 points to 7 1/3 points by area scoring, 3 points to 6 1/3 points by territory scoring. :)

_________________
"Drooling Banjos"

Top
 Profile  
 
Offline
 Post subject: Re: Simplified Japanese Rules Applied
Post #7 Posted: Sat Nov 07, 2015 4:47 am 
Lives with ko

Posts: 186
Liked others: 0
Was liked: 31
Rank: kgs 6 kyu
KGS: Pio2001
jeromie wrote:
and the black string is not adjacent to only independently alive strings of the opposing color.


RobertJasiek wrote:
In a simple seki, any region consisting of one player's string(s) and its adjacent empty strings is not adjacent only to one player's (i.e., the opponent's) independently alive strings [...]


Aaaah ! The region must have nothing else adjacent to it ! I understand now.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 7 posts ] 

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 2 guests


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