J2003's global ko pass (unblocking all kos for both players at once) were mentioned recently. I'm not sure if even J2003's explicit localization can make that work. The theoretical defect of such rule is it may allow passing for two remote ko fights in one move, for non-Japanese results:
J2003 problem
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: J2003 problem
I understand the idea Jann but this example does not seem to work.jann wrote:J2003's global ko pass (unblocking all kos for both players at once) were mentioned recently. I'm not sure if even J2003's explicit localization can make that work. The theoretical defect of such rule is it may allow passing for two remote ko fights in one move, for non-Japanese results:
In confirmation phase all white stones are dead by simply beginning by and black wins the capturing race.
This position above is better but black seems still to be one move ahead:
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: J2003 problem
Yes OC Jann it works now.jann wrote:Ok, then what if the race is longer?
Good point!
-
RobertJasiek
- Judan
- Posts: 6273
- Joined: Tue Apr 27, 2010 8:54 pm
- GD Posts: 0
- Been thanked: 797 times
- Contact:
Re: J2003 problem
In Japanese rules bent4 is dead (OC) due to the pass-for-ko rule, then the seki collapses. In confirmation the only allowed ko threat is a pass, so B can start capturing from left. This is a case of line #2 of my earlier rules classification table (bent4 with nearby unremovable threat).
How do you evaluate the statuses (for middle and right W stones for example) under J2003?
How do you evaluate the statuses (for middle and right W stones for example) under J2003?
-
RobertJasiek
- Judan
- Posts: 6273
- Joined: Tue Apr 27, 2010 8:54 pm
- GD Posts: 0
- Been thanked: 797 times
- Contact:
Re: J2003 problem
1) Referring only to bent-4 does not do this position justice. Bent-4 is not a nuclear bomb overriding the rest of the board.
2) For at least months, I lack time to apply J2003 to non-trivial positions.
2) For at least months, I lack time to apply J2003 to non-trivial positions.
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: J2003 problem
Why not taking simply the well known L&D exemple 7.2. (to avoid an unknown non-trivial position).RobertJasiek wrote:1) Referring only to bent-4 does not do this position justice. Bent-4 is not a nuclear bomb overriding the rest of the board.
2) For at least months, I lack time to apply J2003 to non-trivial positions.
If you replace the move
pass-for-ko by a move ko-pass the result is completly changed.Re: J2003 problem
I explained how Japanese rules with pass-for-ko work here, including the rest of the board.RobertJasiek wrote:Bent-4 is not a nuclear bomb overriding the rest of the board.
I'm not sure how your rules work, but it seems when B starts the bent4 ko, W can respond by starting another ko on the right, then pass for both simultaneously (which is a nonsense from a Japanese viewpoint), thereby keeping some W stones uncapturable with a left-or-right tactic.
Gérard: the official example is different, these rules use local enable, that's why I wrote they may or may not get away with the theoretical defect of this kind of ko pass (apparently not).
-
RobertJasiek
- Judan
- Posts: 6273
- Joined: Tue Apr 27, 2010 8:54 pm
- GD Posts: 0
- Been thanked: 797 times
- Contact:
Re: J2003 problem
It is not how J rules work but how J1949 and WAGC worked for only the precedent positions.
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
-
Cassandra
- Lives in sente
- Posts: 1326
- Joined: Wed Apr 28, 2010 11:33 am
- Rank: German 1 Kyu
- GD Posts: 0
- Has thanked: 14 times
- Been thanked: 153 times
Re: J2003 problem
J2003 does NOT have GLOBAL "enable".Gérard TAILLE wrote:Why not taking simply the well known L&D exemple 7.2. (to avoid an unknown non-trivial position).RobertJasiek wrote:1) Referring only to bent-4 does not do this position justice. Bent-4 is not a nuclear bomb overriding the rest of the board.
2) For at least months, I lack time to apply J2003 to non-trivial positions.
If you replace the move
Thus, both corners are L&D judged seperately, and independent of each other. The result will be the same.
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)
Igo Hatsuyōron #120 (really solved by KataGo)
Re: J2003 problem
Yes J2003 tries to use local enable, which is probably why this flaw was not discovered until now. But local enable <> local capturability. Btw, bent4 + mannenko is slightly different beast anyway: the bent4 player can also flip the thousand year ko before starting the bent4 ko, as threat removal to delay/make harder the start of the losing ko there.
Looks nicer and still seems to work. Shrinked and color-balanced: This right side ko shape really deserves its own name, it seems valuable component. The original (used in my first post above) was from Matthew Macfadyen apparently, and that corner shape gives W the additional ability to make temporary seki and recreate the ko later. This simpler one is more direct but nicely similar to bent4, even in behavior almost like an "inverse bent4" - one side can wait and start a losing ko anytime.Gérard TAILLE wrote:Why not this simplier position
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: J2003 problem
Yes Jann I agree. Sorry for having made a bad mixure with J89 and J2003 (too quick answer).jann wrote:Yes J2003 tries to use local enable, which is probably why this flaw was not discovered until now. But local enable <> local capturability. Btw, bent4 + mannenko is slightly different beast anyway: the bent4 player can also flip the thousand year ko before starting the bent4 ko, as threat removal to delay/make harder the start of the losing ko there.
-
Gérard TAILLE
- Gosei
- Posts: 1346
- Joined: Sun Aug 23, 2020 2:47 am
- Rank: 1d
- GD Posts: 0
- Has thanked: 21 times
- Been thanked: 57 times
Re: J2003 problem
Remember Jann, this right side ko shape was already presented by le_4TC in the post viewtopic.php?p=266251#p266251jann wrote:Yes J2003 tries to use local enable, which is probably why this flaw was not discovered until now. But local enable <> local capturability. Btw, bent4 + mannenko is slightly different beast anyway: the bent4 player can also flip the thousand year ko before starting the bent4 ko, as threat removal to delay/make harder the start of the losing ko there.
Looks nicer and still seems to work. Shrinked and color-balanced: This right side ko shape really deserves its own name, it seems valuable component. The original (used in my first post above) was from Matthew Macfadyen apparently, and that corner shape gives W the additional ability to make temporary seki and recreate the ko later. This simpler one is more direct but nicely similar to bent4, even in behavior almost like an "inverse bent4" - one side can wait and start a losing ko anytime.Gérard TAILLE wrote:Why not this simplier position