Life In 19x19
http://www.lifein19x19.com/

An interesting new rule idea.
http://www.lifein19x19.com/viewtopic.php?f=45&t=12514
Page 1 of 3

Author:  Joelnelsonb [ Mon Dec 14, 2015 8:45 am ]
Post subject:  An interesting new rule idea.

So this may have been thought of and talked about already but I have never heard of it. I suggest that a player ought to lose 2 points for each independent group he/she has on the board. Maybe you already know where I'm coming from with this and maybe you don't. Let me explain. It comes from what I consider to be the most simple and pure (quite subjective of course) scoring system: No-pass Go being one way where the game doesn't end until one player has no legal move. The other way is to say that the object of the game is to get the most stones on the board. Either way, think about this(and forget komi for now) If the game comes to what we generally consider to be the "end" (which is technically just the point at which no more productive moves are available and thus the game is counted up to see who's ahead) and Black has control of 181 intersections while white has the other 180, black would appear to have won without question. However, imagine that black has four separate living groups while white has one big connected group, if you play the game out completely, both players will fill in their entire territories until only the two eyes necessary for life are left vacant. However, black will reach this point having filled in 173 intersections (181-8) where as white would be able to fill in 178 (180-2). At this point, white has more stones on the board than black and black will not make another move or he will self-atari and be killed, losing by an even larger margin. Under no-pass rules, black will be forced to self-atari giving white the victory one again. I don't expect anyone to do anything with this or ever play this way, I just found it interesting and wondered what you all would think.

Author:  Uberdude [ Mon Dec 14, 2015 9:03 am ]
Post subject:  Re: An interesting new rule idea.

People have indeed thought and played with this rule, the Chinese did that for thousands of years. It's called group tax: http://senseis.xmp.net/?GroupTax .

Author:  Bill Spight [ Mon Dec 14, 2015 9:55 am ]
Post subject:  Re: An interesting new rule idea.

Uberdude wrote:
People have indeed thought and played with this rule, the Chinese did that for thousands of years. It's called group tax: http://senseis.xmp.net/?GroupTax .


People sometimes think that the group tax was a feature of area scoring. And area scoring with a group tax persisted into the 20th century in China. However, the oldest known game records with scores appear to have used territory scoring with a group tax. And, as I have indicated, the natural scoring method for no pass go with prisoner return is territory scoring with a group tax. :) See this post, viewtopic.php?p=179724#p179724, and this rather longish post, viewtopic.php?p=179798#p179798 . See also viewtopic.php?p=179786#p179786 and viewtopic.php?p=195133#p195133 .

Edit: Somehow my links ended up being the same. I have corrected that. :)

Author:  Bill Spight [ Mon Dec 14, 2015 11:24 am ]
Post subject:  Re: An interesting new rule idea.

Joelnelsonb wrote:
So this may have been thought of and talked about already but I have never heard of it. I suggest that a player ought to lose 2 points for each independent group he/she has on the board. Maybe you already know where I'm coming from with this and maybe you don't. Let me explain. It comes from what I consider to be the most simple and pure (quite subjective of course) scoring system: No-pass Go being one way where the game doesn't end until one player has no legal move. The other way is to say that the object of the game is to get the most stones on the board. Either way, think about this(and forget komi for now) If the game comes to what we generally consider to be the "end" (which is technically just the point at which no more productive moves are available and thus the game is counted up to see who's ahead) and Black has control of 181 intersections while white has the other 180, black would appear to have won without question. However, imagine that black has four separate living groups while white has one big connected group, if you play the game out completely, both players will fill in their entire territories until only the two eyes necessary for life are left vacant. However, black will reach this point having filled in 173 intersections (181-8) where as white would be able to fill in 178 (180-2). At this point, white has more stones on the board than black and black will not make another move or he will self-atari and be killed, losing by an even larger margin. Under no-pass rules, black will be forced to self-atari giving white the victory one again. I don't expect anyone to do anything with this or ever play this way, I just found it interesting and wondered what you all would think.


Emphasis added.

This seems to suggest that no pass go and stone scoring are the same, or almost so. That is not the case. For instance,

Click Here To Show Diagram Code
[go]$$Wc No pass go
$$ -----------------
$$ | . X X . X W W . |
$$ | X X X X X X X X |
$$ | O O O O O O O O |
$$ | . O O B . O B . |
$$ -----------------[/go]


By stone scoring the net score is 0. In No Pass Go White wins even if he plays first, as he is one move ahead. :)

See http://senseis.xmp.net/?NoPassGo and http://senseis.xmp.net/?NoPassGoProblem1

Author:  Joelnelsonb [ Wed Dec 30, 2015 8:48 am ]
Post subject:  Re: An interesting new rule idea.

I thought this was interesting and decided to share it. It's a fully completed game ending in jigo. Essentially, its an ultimate seki position where both sides finish with 173 stones on the board and are both unable to make another move due to self-Atari. I played this game with a buddy and I've been told that this is extremely uncommon and so we decided to fill it in completely just to check it out. Note that white is collecting no komi and group tax is obviously applied under the basic principle that the winner is the one to get the most stones on the board. So I guess my real question stemming from the original post is: Why on earth is this not the standard way to play Go?

Attachments:
Jigo.JPG
Jigo.JPG [ 151.86 KiB | Viewed 17946 times ]

Author:  HermanHiddema [ Wed Dec 30, 2015 10:20 am ]
Post subject:  Re: An interesting new rule idea.

Joelnelsonb wrote:
Why on earth is this not the standard way to play Go?


Because filling the board is extremely tedious.

Author:  Bill Spight [ Wed Dec 30, 2015 11:10 am ]
Post subject:  Re: An interesting new rule idea.

HermanHiddema wrote:
Joelnelsonb wrote:
Why on earth is this not the standard way to play Go?


Because filling the board is extremely tedious.


Well, the group tax persisted into the 20th century. Go with a group tax, whether by territory or area scoring, was played longer than any other form of go, as far as we know. And the players did not bother to fill the board.

Author:  Joelnelsonb [ Wed Dec 30, 2015 11:32 am ]
Post subject:  Re: An interesting new rule idea.

Right, filling the board out will never be necessary given that any seasoned player can tell who will win far ahead of time and resign if need be. Nonetheless, I don't quite understand three things: why it's the norm to play without group tax, why people like to keep score in such a way that doesn't account for points left in dame, and why Komi includes a half point...

Author:  HermanHiddema [ Wed Dec 30, 2015 11:39 am ]
Post subject:  Re: An interesting new rule idea.

Quote:
Nonetheless, I don't quite understand three things: why it's the norm to play without group tax, why people like to keep score in such a way that doesn't account for points left in dame, and why Komi includes a half point...


Removing group tax and counting territory (and hence not dame) both simplify the game slightly. As they say: perfection is not when there is nothing left to add, but when there is nothing left to remove.

The purpose of the half point is to prevents ties (Note that komi may or may not include half a point, depending on the event).

Author:  hyperpape [ Wed Dec 30, 2015 2:47 pm ]
Post subject:  Re: An interesting new rule idea.

Joelnelsonb wrote:
Why on earth is this not the standard way to play Go?
Ask yourself why on earth it should be and you can find the answer.

Author:  Joelnelsonb [ Wed Dec 30, 2015 4:15 pm ]
Post subject:  Re: An interesting new rule idea.

I'm not meaning to sound like "the way I think it should be played is the only right way and anyone who thinks differently is stupid". After all, I was on this forum a short while ago bashing things like including dame in the scoring and what not. It's just that after gaining a better understanding of the finer points and origins of the game we call Go, I want to better understand where these other sorts of counting methods and rule sets come into play. So territory scoring is said to simplify the game, how so? To a new player learning the game, telling them that you lose two points for every separate group would sound like some additional rule that's added onto the game. However, with a better understanding of the core concept of the game, you realize it's just the proper way to deduce who will win the game without actually having to fill in the entire board as in the picture above. But that kinda' brings me to my question: Is there another alternative core objective to the game? For instance, it seems to me to make the most sense to state the rules of Go as follows:

- Any stones on the board of the same color that lay directly adjacent to each other are considered to be a single unit and therefore share liberties (A liberty being a vacant point directly adjacent to a stone).

- Every stone or group of stones must maintain one vacant liberty at all times in order to remain on the board (a player cannot end their turn while one of their own stones/groups remains without a vacant liberty). Once a player places his stone for that turn, he then removes and stones/groups of his opponent that no longer enjoy at least one liberty.

- Never at any time in the game may an exact board position be repeated (enough said on that I believe).

- The player who is able to get the most stones on the board is deemed the winner (Like in the picture above, the game never actually ends, it just reaches a point where neither player wished to play another stone, and therefore the player who stands in the deficit is at that point declaring resignation.

You may call this a rule or just the nature of gameplay:
- Players alternate placing a maximum of one stone at a time, on any vacant intersection on the board with the black player placing the first stone. The only stipulations as to where stones can be placed are found in the liberty and repetition rules stated above.The game ends when one player resigns, deeming his opponent the victor.

So what I'm getting at is: What are other ways of laying out the confines of the game such that the more popular methods of scoring the board make sense (no profit from points left in dame, no group tax, requiring pass stones etc)?

Author:  Jhyn [ Wed Dec 30, 2015 4:32 pm ]
Post subject:  Re: An interesting new rule idea.

I'll let the rules experts sort out your proposition but what you describe correspond to my understanding of the Chinese rules. At the risk of being pedantic,

Quote:
- Every stone or group of stones must maintain one vacant liberty at all times in order to remain on the board (a player cannot end their turn while one of their own stones/groups remains without a vacant liberty).


you must be more precise: you can play a move removing the last liberty of one of your groups if by doing so you remove the last liberty of an opponent's group (in other words, you have to say in which order groups are removed).

This being said, I am pretty sure there will be some borderline cases (the last play on the board is a ko in which the active player cannot play) where this will differ from Chinese rules.

Author:  Joelnelsonb [ Wed Dec 30, 2015 4:39 pm ]
Post subject:  Re: An interesting new rule idea.

Jhyn wrote:
you must be more precise: you can play a move removing the last liberty of one of your groups if by doing so you remove the last liberty of an opponent's group (in other words, you have to say in which order groups are removed).



Good point, adjustment made.

Author:  Bill Spight [ Wed Dec 30, 2015 7:28 pm ]
Post subject:  Re: An interesting new rule idea.

Joelnelsonb wrote:
It's just that after gaining a better understanding of the finer points and origins of the game we call Go, I want to better understand where these other sorts of counting methods and rule sets come into play. So territory scoring is said to simplify the game, how so?


There is something of the eye of the beholder here. However, territory scoring is not a simplification of area scoring.

Quote:
To a new player learning the game, telling them that you lose two points for every separate group would sound like some additional rule that's added onto the game. However, with a better understanding of the core concept of the game, you realize it's just the proper way to deduce who will win the game without actually having to fill in the entire board as in the picture above.


Indeed, it would seem like a strange additional rule. :)

Quote:
- The player who is able to get the most stones on the board is deemed the winner (Like in the picture above, the game never actually ends, it just reaches a point where neither player wished to play another stone, and therefore the player who stands in the deficit is at that point declaring resignation.


From this it is not clear to me what you think the core concept of the game is. Usually the game does not reach a point where neither player wishes to play another stone; instead one player does not wish to continue play, while the other one does. This point does not depend upon who has more stones on the board at that point, although that is one factor.

Click Here To Show Diagram Code
[go]$$Bc
$$ -------------------
$$ | . X 7 X 9 X W O 3 |
$$ | X X X X X X X X X |
$$ | X O X O 1 O X O X |
$$ | O O O O O O O O O |
$$ | . O 8 O B 2 O X 6 |
$$ -----------------[/go]


:w4: at :wc:, :b5: captures :w4:, :w10: at :bc:

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


At this point Black has the move but does not wish to play, even though he has one more stone on the board than White. (As it happens, White does not want to play, either; but I could have easily contrived a position where White would have had fewer stones on the board than Black but more plays.)

Author:  Jhyn [ Wed Dec 30, 2015 7:56 pm ]
Post subject:  Re: An interesting new rule idea.

Bill Spight wrote:
At this point Black has the move but does not wish to play, even though he has one more stone on the board than White. (As it happens, White does not want to play, either; but I could have easily contrived a position where White would have had fewer stones on the board than Black but more plays.)


Your remark is very clear and I realise I misunderstood the previous post. Today I had another idea. If we allow another legal move which is giving back a prisoner to your opponent, does no-pass go becomes the same as most-stones go with group tax? (maybe disregarding some final ko situations)

Was there ever a ruleset allowing this? Of course it is inspired by the idea of pass stone.

Author:  Bill Spight [ Thu Dec 31, 2015 12:39 am ]
Post subject:  Re: An interesting new rule idea.

Jhyn wrote:
Bill Spight wrote:
At this point Black has the move but does not wish to play, even though he has one more stone on the board than White. (As it happens, White does not want to play, either; but I could have easily contrived a position where White would have had fewer stones on the board than Black but more plays.)


Your remark is very clear and I realise I misunderstood the previous post. Today I had another idea. If we allow another legal move which is giving back a prisoner to your opponent, does no-pass go becomes the same as most-stones go with group tax? (maybe disregarding some final ko situations)

Was there ever a ruleset allowing this? Of course it is inspired by the idea of pass stone.


No pass go with prisoner return yields territory scoring with a group tax. Professor Berlekamp demonstrated that over 20 years ago. As it turns out, the oldest existing game records that are scored seem to use territory scoring with a group tax. :)

Author:  Joelnelsonb [ Thu Dec 31, 2015 3:37 am ]
Post subject:  Re: An interesting new rule idea.

Bill Spight wrote:

No pass go with prisoner return yields territory scoring with a group tax. Professor Berlekamp demonstrated that over 20 years ago. As it turns out, the oldest existing game records that are scored seem to use territory scoring with a group tax. :)


So even under that idea, wouldn't it be relevant whether the amount of dame left over at the end of the game is an even or odd number?

Author:  Matti [ Thu Dec 31, 2015 8:39 am ]
Post subject:  Re: An interesting new rule idea.

Dames are small.

Author:  Bill Spight [ Thu Dec 31, 2015 10:33 am ]
Post subject:  Re: An interesting new rule idea.

Matti wrote:
Dames are small.


Yes, but there are smaller plays. ;)

Author:  Bill Spight [ Thu Dec 31, 2015 12:20 pm ]
Post subject:  Re: An interesting new rule idea.

Joelnelsonb wrote:
Bill Spight wrote:

No pass go with prisoner return yields territory scoring with a group tax. Professor Berlekamp demonstrated that over 20 years ago. As it turns out, the oldest existing game records that are scored seem to use territory scoring with a group tax. :)


So even under that idea, wouldn't it be relevant whether the amount of dame left over at the end of the game is an even or odd number?


First, let me clarify what I said about Professor Berlekamp. He did not, I think, mean to produce a set of rules with a group tax. In fact, he added an encore with immortal stones to eliminate it.

Now, on to dame in no pass go with prisoner return.

Click Here To Show Diagram Code
[go]$$W Odd or even
$$ --------------------
$$ . O . . X . O . . X .
$$ . O X X X . O O O X .
$$ . O O X . O O X X X .
$$ . . O X X X X O X . .
$$ . . O O O O O O X . .[/go]


No pass go with prisoner return. 19x19 board. Net territory plus prisoners and dead stones = 0. Only dame left otherwise. No other seki.

White to play.

The key is to realize that the number of dame elsewhere is odd.

Click Here To Show Diagram Code
[go]$$W White gets the last dame
$$ --------------------
$$ . O 2 . X 3 O . 1 X .
$$ . O X X X 4 O O O X .
$$ . O O X . O O X X X .
$$ . . O X X X X O X . .
$$ . . O O O O O O X . .[/go]


:w5: fills a dame elsewhere.

The remaining dame after :w5: are even, so White gets the last dame. Zero is a second player win, and Black will play first. White wins.

Note that the local eyes are not territory, as they are necessary for life. Also note that the remaining local neutral point is not dame in no pass go, since neither player can afford to fill it. It is also necessary for life. The fact that it remains unfilled changes the parity of the board, which is why White gets the last dame when the territory score is even.

Click Here To Show Diagram Code
[go]$$W Failure
$$ --------------------
$$ . O 1 3 X 5 O 4 2 X .
$$ . O X X X . O O O X .
$$ . O O X . O O X X X .
$$ . . O X X X X O X . .
$$ . . O O O O O O X . .[/go]


After :w5: there are two neutral points left that are necessary for life. Black gets the last dame elsewhere and wins.

So yes, there are cases in which getting the last dame affects best play. :)

Page 1 of 3 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/