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

pi rule instead of komi for go?
http://lifein19x19.com/viewtopic.php?f=45&t=12507
Page 1 of 1

Author:  phillip1882 [ Fri Dec 11, 2015 4:21 pm ]
Post subject:  pi rule instead of komi for go?

if we implemented the pi rule instead of the komi rule for go, what would be your first move as black?

Author:  Kirby [ Fri Dec 11, 2015 4:37 pm ]
Post subject:  Re: pi rule instead of komi for go?

Click Here To Show Diagram Code
[go]$$c
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . B . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

Author:  DrStraw [ Fri Dec 11, 2015 4:43 pm ]
Post subject:  Re: pi rule instead of komi for go?

Perhaps it would be a good idea if you told us what the pi rule is.

Author:  schawipp [ Fri Dec 11, 2015 5:04 pm ]
Post subject:  Re: pi rule instead of komi for go?

DrStraw wrote:
Perhaps it would be a good idea if you told us what the pi rule is.
I would guess, he means the "pie rule" instead: http://en.wikipedia.org/wiki/pie_rule. The "pi rule" (w gets pi komi) doesn't seem to make much sense... ;-)

Author:  MinjaeKim [ Fri Dec 11, 2015 5:17 pm ]
Post subject:  Re: pi rule instead of komi for go?

Kirby wrote:
Click Here To Show Diagram Code
[go]$$c
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . B . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

I'm sure any pro player as the second one to move will choose to swap. It even feels like a possible move to play as black with usual komi. In my feeling, the first move on the second line will be slightly better for black, and on the first line will be better for white.

Author:  hyperpape [ Fri Dec 11, 2015 5:48 pm ]
Post subject:  Re: pi rule instead of komi for go?

I have no idea. The last time I brought up the topic most people wouldn't make suggestions: viewtopic.php?f=45&t=5501&hilit=Pie

If you forced me, I guess I'd pick 2-8 or something.

Author:  Jhyn [ Fri Dec 11, 2015 9:12 pm ]
Post subject:  Re: pi rule instead of komi for go?

schawipp wrote:
DrStraw wrote:
Perhaps it would be a good idea if you told us what the pi rule is.
I would guess, he means the "pie rule" instead: http://en.wikipedia.org/wiki/pie_rule. The "pi rule" (w gets pi komi) doesn't seem to make much sense... ;-)


The only difference with komi 3.5 would be that you win by 0.14159... instead of 0.5, but the .5 doesn't make more sense. Personally I prefer saying "3 point komi, white wins ties"... Or just leave the ties be.
Imaginary komi, now...

Author:  DrStraw [ Sat Dec 12, 2015 8:19 am ]
Post subject:  Re: pi rule instead of komi for go?

If that is what is meant then I would think that e is a much more widely used number then pi and so we should have the e-rule. It would also me more apt for e-mail games, I guess.

Author:  Bill Spight [ Sat Dec 12, 2015 9:37 am ]
Post subject:  Re: pi rule instead of komi for go?

Jhyn wrote:
schawipp wrote:
DrStraw wrote:
Perhaps it would be a good idea if you told us what the pi rule is.
I would guess, he means the "pie rule" instead: http://en.wikipedia.org/wiki/pie_rule. The "pi rule" (w gets pi komi) doesn't seem to make much sense... ;-)


The only difference with komi 3.5 would be that you win by 0.14159... instead of 0.5, but the .5 doesn't make more sense. Personally I prefer saying "3 point komi, white wins ties"... Or just leave the ties be.
Imaginary komi, now...


Right. The i rule! :mrgreen:

Edit: I win jigo. ;)

Author:  Jhyn [ Sat Dec 12, 2015 11:32 am ]
Post subject:  Re: pi rule instead of komi for go?

Bill Spight wrote:
Jhyn wrote:
The only difference with komi 3.5 would be that you win by 0.14159... instead of 0.5, but the .5 doesn't make more sense. Personally I prefer saying "3 point komi, white wins ties"... Or just leave the ties be.
Imaginary komi, now...


Right. The i rule! :mrgreen:

Edit: I win jigo. ;)


And since both players can say it... Both players win?
That's an interesting interpretation. (Shame it only works in English)

Author:  Bill Spight [ Sat Dec 12, 2015 12:12 pm ]
Post subject:  Re: pi rule instead of komi for go?

Jhyn wrote:
Bill Spight wrote:
Jhyn wrote:
The only difference with komi 3.5 would be that you win by 0.14159... instead of 0.5, but the .5 doesn't make more sense. Personally I prefer saying "3 point komi, white wins ties"... Or just leave the ties be.
Imaginary komi, now...


Right. The i rule! :mrgreen:

Edit: I win jigo. ;)


And since both players can say it... Both players win?
That's an interesting interpretation. (Shame it only works in English)


"White won, but Black did not lose." :mrgreen:

(See http://senseis.xmp.net/?TenThousandYearKo%2FRulesCrisis )

Author:  Matti [ Tue Dec 15, 2015 2:57 am ]
Post subject:  Re: pi rule instead of komi for go?

If one is looking for ideas to replace komi, here is one. Play two games with alternate colours. Add the results of both games. The palyer who has the higher total score wins. The first game is a bit different from traditional game, because one might settle to lose a little instead of taking a big risk. But the second is not. A similar idea is been used in football, ice hockey, etc. matches.

Author:  Matti [ Tue Dec 15, 2015 4:52 am ]
Post subject:  Re: pi rule instead of komi for go?

Marcel GrĂ¼nauer wrote:
First move 3-14.

Seems good for black.

Author:  skydyr [ Tue Dec 15, 2015 7:43 am ]
Post subject:  Re: pi rule instead of komi for go?

Matti wrote:
If one is looking for ideas to replace komi, here is one. Play two games with alternate colours. Add the results of both games. The palyer who has the higher total score wins. The first game is a bit different from traditional game, because one might settle to lose a little instead of taking a big risk. But the second is not. A similar idea is been used in football, ice hockey, etc. matches.


How do you score resignations?

Author:  Jhyn [ Tue Dec 15, 2015 6:22 pm ]
Post subject:  Re: pi rule instead of komi for go?

skydyr wrote:
Matti wrote:
If one is looking for ideas to replace komi, here is one. Play two games with alternate colours. Add the results of both games. The player who has the higher total score wins. The first game is a bit different from traditional game, because one might settle to lose a little instead of taking a big risk. But the second is not. A similar idea is been used in football, ice hockey, etc. matches.


How do you score resignations?


I assume a resignation would be for the entire match - that is, the first game is played until the end whatever happens.

It might also make sense to equal a, say, 50-point defeat with a resignation. This way, you might catch a draw by crushing your opponent in the second game even after resigning. But since players would never resign the first game anyway, I don't think it would change much.

Author:  Matti [ Wed Dec 16, 2015 3:16 am ]
Post subject:  Re: pi rule instead of komi for go?

Jhyn wrote:
skydyr wrote:
Matti wrote:
If one is looking for ideas to replace komi, here is one. Play two games with alternate colours. Add the results of both games. The player who has the higher total score wins. The first game is a bit different from traditional game, because one might settle to lose a little instead of taking a big risk. But the second is not. A similar idea is been used in football, ice hockey, etc. matches.


How do you score resignations?


I assume a resignation would be for the entire match - that is, the first game is played until the end whatever happens.

It might also make sense to equal a, say, 50-point defeat with a resignation. This way, you might catch a draw by crushing your opponent in the second game even after resigning. But since players would never resign the first game anyway, I don't think it would change much.


Both are viable options. Another one is, that a player may for example state:"I offer a resignation as 40 points win for you".

Author:  hgj11 [ Fri May 06, 2016 3:32 am ]
Post subject:  Re: pi rule instead of komi for go?

Matti wrote:
If one is looking for ideas to replace komi, here is one. Play two games with alternate colours. Add the results of both games. The palyer who has the higher total score wins. The first game is a bit different from traditional game, because one might settle to lose a little instead of taking a big risk. But the second is not. A similar idea is been used in football, ice hockey, etc. matches.


Using two boards could be another interesting variation.

Black places his/her first move to one of the boards.
Next, the two player alternately play one stone to both boards.
When they agree finishing on one of the boards, they continue the game on the other one with "normal" play.
Finally, they count the sum of the scores on the two boards.

This version eliminates the "using" of the knowledge of the result of the first game in the fuseki of the second one.

Author:  Mike Novack [ Fri May 06, 2016 5:01 am ]
Post subject:  Re: pi rule instead of komi for go?

The problem with schemes like that (any that take into account the point count margin of victory) is that they fundamentally change the game.

It is precisely the fact that the margin of victory DOESN'T matter that leads to the exciting fights we see in high level games. That's because it "pays" (is the correct strategy for) the player who is behind even slightly to take desperate risks to pull the game out.

Suppose instead "score matters". Let's say we had a match, with the winner of the match being the player with the highest score total of all the games added up. The play would not be exciting. It would NOT pay for the player who was a couple points behind in one of the games to take the risk of losing by 40 points in exchange for a couple percent likelihood of winning by a point or two.

Remember, we shouldn't use the strategy of kyu players to think about what is good for go. At the levels of play closer to beginner proper strategy if behind might well be "just keep as close as you can" because BOTH players will likely be making mistakes worth 10-20 points before the game is over.

Author:  Sennahoj [ Fri May 06, 2016 5:14 am ]
Post subject:  Re: pi rule instead of komi for go?

Mike Novack wrote:
It is precisely the fact that the margin of victory DOESN'T matter that leads to the exciting fights we see in high level games.


I'm not sure this is true --- I remember reading somewhere that in settings where the margin of victory matters, games tend to involve a lot of risk, because both players have an incentive to try to get "more" all the time, no matter who is leading

Author:  Mike Novack [ Fri May 06, 2016 1:54 pm ]
Post subject:  Re: pi rule instead of komi for go?

Sennahoj wrote:
Mike Novack wrote:
It is precisely the fact that the margin of victory DOESN'T matter that leads to the exciting fights we see in high level games.


I'm not sure this is true --- I remember reading somewhere that in settings where the margin of victory matters, games tend to involve a lot of risk, because both players have an incentive to try to get "more" all the time, no matter who is leading


In a game where margin of victory matters, it is correct to accept SOME more risk to increase the result. One compares the likelihood of increasing the margin to decreasing it.

In a game where the margin of victory doesn't matter, it is correct to accept ANY level of risk to possibly pull the game out. In other words, the cost of losing by an even larger margin is zero.

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