Placid/discussion

Sub-page of Placid
[Diagram]
Placid Ko 1 (outer stones alive)  


Willemien 11 kyu: I don't understand it. I thought a ko is always worth atleast one point (a point for the captured stone? (or under area counting the option to fill it) or are placid ko's ko's within dead groups only?

fractic: This is not a term that is easily explained. It's more a mathematical thing than very relevant to play. First you seem to misunderstand what is meant by the local count. See that page for an explanation. Secondly you need to know stuff like miai value and CGT. See the komaster page for an actual computation on a non placid ko.

Bill: I have added an example. Is that better, Willemien?


[Diagram]
Placid ko 2 (outer stones alive)  


Willemien Thanks Bill getting better

I understand: If Black wins this ko the local score will be +3. (3 points of black territory) and If White wins it it will be -18. (10 points of white territory and 8 black prisoners)

But I do not understand:

The local count is -4, Is the local count the difference (3 - (-18) = 21) divided by 3 equals 7 and because the Black has taken the ko in diagram Placid ko 1 the local count is -4, If White has taken the ko (diagram Placid ko 2 ) the local count would be -11? But then the count is dependent on to who 's turn it is to make a ko threat ?

Bill: No assumption is made about who has taken the ko or not or whose turn it is. The position is what it is.

willemien I think I expessed myself badly. only the position is important but the positions in placid ko 1 and placid ko 2 do differ. And is the count of placid ko 2 -11? please explain if I am seeing it all wrong.

fractic: Yeah the count in the second diagram is indeed -11.

Willemien cont. and it is placid because it has no influence on other parts of the board? Or because each play in the ko gains 7 points?

Bill: Placid is just the name Prof. Berlekamp came up with to describe this kind of ordinary ko position. It is placid because who is komaster does not affect the count, by definition.

Willemien I don't understand this yet. but will comment on it later

Willemien cont. The 21 is divided by 3 because it takes 1 (local) play to end the ko for one player and 2 (local) plays for the other player.(in either diagram) (move this explanation up if it is correct or comment here if it is not) Also add examples to hyperactive position and active position because I don't underatand a lot of it (yet).

Bill: Active positions are quite rare. I do not have an example readily at hand. As for hyperactive positions, I think that hyperlinking to examples is good enough. SL is hypertext, after all. :)

Willemien I think giving (at least one) example directly on the page is better. especially because you then can add al CGT ins and outs about the position (and use the CGT terms) without disrupting people who want to read it as a Go theory article, Also this way it is safe from editing by people who want to edit it to go theoretic? perspective only.

fractic: I'll just put a hyperactive position here temporarely to explain the difference. I hope this is right though please correct me if I'm wrong Bill.

[Diagram]
corner  

fractic: The count of this position depends on who is komaster. It's allways sente for Black and we know that sente plays don't affect the count so it's not hard to find out what the count is.

[Diagram]
Black komaster  

fractic: If Black is komaster B1-W6 is sente. So the local count here is +3 counting the marked points

[Diagram]
White komaster  

fractic: If White is komaster she can block with W2 since if Black cuts, White wins the ko anyway. Now the count is only +1. So who is komaster makes a difference.


Willemien Thanks fractic Maybe it is an idea to move your example to the hyperactive if it is correct. (sorry for mispelling your name the first time, corrected)

Still a bit confused What is the difference between an active position and an hyperactive position Also Is the "count" not the summation of all "local counts"?

Bill: In ordinary parlance, that might be the case, but here count is a technical term, meaning the local territorial (or area) value.

Willemien cont. Also I was wondering about komasters /komonsters is Black a komonster in a game if part of is the komonster diagram? (except in the diagram itself and only because Black will win any ko until White plays square

Bill: Well, komonster is a komaster who can gain by waiting to win a ko (because of extra ko threats). Until Black captures the four White stones, White has an indefinite supply of ko threats, and so might well be komonster for a ko somewhere else on the board.

Willemien Oops again. My example is was not a one sided double ko as I intended. (only usable a ko threat source for one player) need to study double ko more or maybe it is all relative.

Bill: It is one-sided, in the sense that White has an indefinite supply of threats (while it lasts), but not Black.


Placid/discussion last edited by fractic on January 7, 2009 - 23:00
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
RecentChanges
StartingPoints
About
RandomPage
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Goproblems.com
Login / Prefs
Tools
Sensei's Library