It’s actually higher than 7.5, since black 184.5 with seki would be a win for black in the Chinese rules with 3.75 komi (184.5 > 184.25).
And technically with the positional superko rules, but in practice, they are pretty hard to implement in amateur games, and if they happened, often just as draw, and the Swiss System used can often find the winner even if there are draw.
It was more like customs than rules, and the official “regulations” still use black 184 win. But instead of writing the details, it simply says “for example, there can be black 184 wins, and other details can be defined/regulated by each tournament”
正式比賽採用黑棋貼點制度。如貼六點，則黑184子為勝。其他貼點方式，悉由競賽規程另行規定之。
Sure, I meant 8.5 (which 185 B win literally seems to mean) is still “effectively” the same as 7.5 unless there is an odd seki, which is rare enough to ignore in a simplified view.
So even though a “real” 8.5 komi would not be parity-safe value (only odd+0.5 is), it can act as practical substitute for 7.5 (which is safe) - unlike 7.0, where parity and the extra stone could actually flip lose/draw in common games.
With 8.5 we actually mean 7.5, so if the extra stone rounds a common B+8 territory board to B+9, we are happy with that since we originally wanted to say 7.5 to begin with (so B+8 territory should win for B, unlike with real 8.5).
Ya, and we’ve known this for a long time, and when we first tried to move away from the 184 threshold to match the result with the Chinese rules, we were using 184.5 as a threshold. But in practice, since the odd shared liberty happened so rarely, the extra 0.5 added to the threshold became somewhat unnecessary and confused quite a bit of the DDK player. And in practice, we don’t even need to record the actual winning margin (we often just record the black area count and who wins, like B wins at 185 “黑185勝”), the added half a point making it a bit redundant in the end.
Hi people, just coming back to the novice stuff for a second, to check if it’s all really understood. Person A claims that his 2 stones inside the territory of Person B can live and wants to continue the game. Person B says: “No, they are dead. Accept it or suffer the consequences, fool”.
Situation A: They continue the game and “B” was right. The score is calculated based on the board position before all the moves that happened when the threat occurred. (I imagine two beginners trying to play according to the japanese rule taking a picture of the board before the invasion begins so they can restore the previous position). Person A suffers the consequences and will regret it forever.
Situation B: They continue the game and “A” was right. The score is calculated normally after all the moves are played (there’s no going back), since “A” created a territory within “B”'s territory, right? “A” earns the right to say the sentence “Who’s the fool now, uh?” for a couple of months at random occasions.
In practice, I think that’s a way it could reasonably go without breaking too much? But Japanese Rules do weird things with adjucation, so depending on the situation (would it be in all situations, since there have been 2 passes, or only some? I don’t know), what determines the Life and Death might not be the players’ playing it out, even hypothetically, but rather hypothetical perfect play observing the weird ko rules and stuff
tldr: I don’t understand Japanese Rules, and I’m probably just straight wrong on half of the above paragraph, but I feel like a good bet when faced with an interpretation of Japanese Rules that seems relatively simple, is that it’s a misinterpretation. @yebellz ?
PS: I assume you’re talking irl. On OGS one should not continue play because there’s no mechanism to restore the position
Yes, irl. So OGS doesn’t implement the japanese rules per se, as I understood, because if there’s a disagreement, the person who passes and knows the opponent’s stones are dead just let her continue play until the end and the score is calculated on the end. By the way, he just keeps passing and capture all the opponent’s stones.
Well IRL if you both aren’t sure and can’t agree, take a pic of the game in case of someone else may help later and start play a new one!
Soon enough you ll be able to determine when a game is finished.
If you stay confused, I ll be happy to comment it here, if you have the courage to reproduce it on a OGS demo board (so that we can look on variations)
When these novices play in a real life club, some more experienced player can help them out when there is a group status disagreement. IME these sort of situations rarely happen after new players have played a couple of dozens of games.
When the novices play in real life and nobody is around to help them out in case of a group status disagreement, I’d recommend them to use area scoring (or perhaps stone scoring), because it allows them to play it out without affecting the score. Although IME real novices (children at least) are not aware or don’t care about losing some points by capturing dead stones (they love capturing stones).
Mind you that using area scoring won’t solve all the scoring problems that novices encounter when there is nobody around to help. The most common one is that they left some gaps in territory boundaries (such cases are posted frequently on Reddit). Choosing a more beginner-friendly scoring method won’t help in such cases. The only thing that helps to avoid such situations is experience.
Without going into specifics of current Japanese rules, the basic idea with territory scoring is that you ALWAYS revert the position first, regardless of what was found in playout (because the playout was only done to answer the question “what COULD happen”). This is what hypothetical play means, as opposed to real play.
However, you can always resume play as well, ie. ask to continue playing for real, from any stopped position. This is not Japanese or territory scoring specific, but is true for nearly all human rulesets. So if you find something interesting during hypothetical play (or even under area scoring, you see something on an autoscore screen), that could actually become true soon. (With the interesting question of who plays first, which again is ruleset-specific - and assuming OGS supports resumption, which it probably does but I’m not completely sure.)
Btw, while territory scoring normally needs this “revert the position” option which is rarely provided online, there is a possible workaround or dirty hack if you would not like the game to be annulled, and would like to resolve things without falsifying the score in a close game: you could also continue real playing but with a condition that each player must place the same number of stones into territory. This is not an official solution though and obviously only works between honest players.
If analysis mode is enabled, can you share the variation using the analysis mode? Just thinking.
Not in scoring. Things don’t really work properly while scoring is active. You’d have to cancel, return to the game, maybe pause because otherwise clocks are running, and then share etc.
Might work even less well if your opponent doesn’t understand you.
For scoring, after resolving the life and death disputes, the game position should be restored to what is was after the passes. Resolving life and death (under Japanese rules) is done by considering what would happen under hypothetical continuations of play. Note: these hypothetical analysis should also apply some technical rules that are likely unfamiliar to many beginners (such as special rules dealing with ko and consideration of whether any hypothetical capture might enable the opponent to place another living stone). In many cases, unawareness of these technicalities won’t affect the life and death status, but in some rare cases, such rules can make a huge difference.
Not reverting (i.e., changing) the final board position could lead to incorrect scoring. Actually playing out further moves to capture dead stones also reduce one’s territory. However, under Japanese rules, dead stones are ruled to be dead (and removed) without actually having to play out those additional territory-reducing moves.
As I mentioned earlier, beginners are usually taught simplified approximations of the Japanese rules, since dealing with the full complexity is beyond the abilities of even many intermediate players.
For example, this procedure given by @jannn only works in most, but not all, situations:
First, requiring players to place stones into territory may lead to some pathological situations that could be exploited by a player acting in bad faith. If players were forced to make such territory filling moves, eventually they would run out of eyes and otherwise secure groups may be captured. This would not be the same as simply a matter of who had more territory, but also group tax and eye shape would have an effect (and komi would not be in the picture), so the outcome of this self-filling procedure could be quite different than what should be the proper territory score. However, the ultimate purpose of this suggestion is to essentially use something like pass stones (where a player must either place a stone on the board or give their opponent one point). So, these issues could be dealt with by using pass stones instead.
Second, the above procedure would result in life-and-death resolution in accordance to various area scoring rules. In most cases, this agrees with the Japanese rules, but there are rare cases where the outcome would be quite different.
Ah, the players need to agree when the game ends. But this agreement never happens when one of the players thinks his group of stones can live. Then the player who thinks a group of stones is dead has the power to end the game, because the other player can’t say “let’s continue”, since from now on it’s all hypothetical (even if the invasion is successful, as I’ve just learned), just to check out of curiosity.
Yeah, I was wondering why it’s a problem to continue the game and it’s just in the scenario when one player passes while the other tries uselessly to invade. If there’s an answer balancing every move, there isn’t even a problem continuing the game. Why wouldn’t this happen irl, or even this can generate a 1-point-difference problem?
Edit:
@yebellz 's comment posted 1 minute after mine deals with my second paragraph.
I think there is a misunderstanding here. A game resumption is not hypothetical. It is normal play and thus it is not reverted.
Hypothetical play only starts after the game has stopped (both players passed) and they disagree about some group status, but neither player wants to resume the game. The hypothetical play only serves to confirm which player is right: if a group can’t be ultimately captured in hypothetical play, it is alive.
No, you seem to have misunderstood something. As @gennan said, in a stopped position, either player can ask to continue the game for real (for whatever reason or whim). If nobody wants to continue - which is usually why both passed in the first place - AND they cannot agree which is alive and which is not, only then comes hypothetical play (like “this is alive since you could not capture it if we were to continue” etc.).
Oh ok, I was thinking about some simple situation, when one player places a stone inside a big empty territory and the opponent passes, like “Don’t even start, it’s dead”. The player who wants to invade has no reason to pass, because he thinks he can make a territory inside. But in this case I see now there’s no disagreement and he’s just committing suicide, while the other player is happily getting points. The game is not over yet.
The agreement question only comes up after both players passed.
Only then the scoring process starts and the agreement question comes up.
As long as one or both players keep playing, they are not asked which groups they think are dead because the game hasn’t stopped yet. So there is no agreement or disagreement.
Also, when player A invades, player B passes and player A continues, this says nothing about who is right in their decision. We can’t say that player A will lose their invasion (and thus lose points). We’d need a board position to determine which decisions were correct.
Also, putting a lone-stone into another player’s apparent territory might not need further follow-up. For example, you might have a position like this:
This could even possibly lead to a dispute (among beginners), say if Black claims to be alive and that the single White stone is dead, and White claims the opposite.
Btw, there is an ironic situation here. Because of its creator’s whim, go is a game that cannot be modeled by a purely territorial view, nor by a purely area view. It only works correctly when the two are - at least theoretically - combined.
If the main game phase were played under an area view, then dame parities and B’s ability to play both first and last could disrupt the score by one point. If the second phase - possible disputes and cleanups - were played under a territory view, then unanswered capturing moves could disrupt the score by a few points.
Hence the various solutions. The button makes the game more territory-like while staying under area scoring. Pass stones make the game more area-like while staying under territory scoring. Area komi 7.5 sets the win threshold in a way that is a bit high but prevents parities to influence the winner. And hypothetical play makes disputes neutral to territory or area view, since the position is reverted afterwards anyway.
