Interesting Go Variation

John Tromp

未讀,

1996年10月11日凌晨3:00:001996/10/11

收件者：

Dear Go friends,

Pondering on the simplicity of current Go rules, I came up with the
following simple variation, that might be called "simultaneous capture":

After placing a stone, remove ALL (ie. both black and white) stones that
have no liberties.

This has some weird effects, a major one being that kos differ radically.
In a "double tiger mouth", the one not starting the ko is now the one who
has to make the first ko threat!
(I'm assuming a superko rule is in effect to forbid whole board repetition)

. O # . . O # . . O # . . O # . . O # . . O # .
O . . # O # . # O # . # O # . # O . . # O . . #
. O # . . O # . . O # . . O # . . O # . . O # .
. . . . . . . . . . O . . . O # . . O # # . O #

start black white black white b makes
starts makes answers 'takes a threat
a ko ko threat threat back' etc...

It's like every simple ko becomes a two step ko; it can be fought on
the white or on the black side.
The new rules thus give the first player a complete victory
on a 2 x 1 board, whereas under normal rules both players should pass:)

Another major effect is that most sekis disappear: the side with less
points to lose will just fill in one of the shared eyes, leading to
the capture of both groups.

I'm sure there are many other interesting side effects I have yet
to contemplate.

(one is that beginners learning these rules will likely be much less
confused about which side gets captured by a board play, as they often
are with current rules:)

It may be fun playing a few games under these rules to get a feeling
for just how much they affect play as we know it.

Btw, I can very well imagine someone else having considered
this variation before, and would be delighted
to hear about any of their findings.

regards,

%!PS % -John Tromp (http://www.cwi.nl/~tromp/)
42 42 scale 7 9 translate .07 setlinewidth .5 setgray/c{arc clip fill
setgray}def 1 0 0 42 1 0 c 0 1 1{0 3 3 90 270 arc 0 0 6 0 -3 3 90 270
arcn 270 90 c -2 2 4{-6 moveto 0 12 rlineto}for -5 2 5{-3 exch moveto
9 0 rlineto}for stroke 0 0 3 1 1 0 c 180 rotate initclip}for showpage

Bill Taylor

未讀,

1996年10月17日凌晨3:00:001996/10/17

收件者：John Tromp

tr...@cwi.nl (John Tromp) writes:

|> Pondering on the simplicity of current Go rules, I came up with the
|> following simple variation, that might be called "simultaneous capture":
|>
|> After placing a stone, remove ALL (ie. both black and white) stones that
|> have no liberties.

Well now, this is amazing!

Such an obvious thing, and no-one seems to have thought of it before!

And, as John says, it seems to change the game a *little*, but leave it
very largely the same game we know and love so well. Indeed, I'd guess
90% of games would be identical, and most of the rest virtualy so.

Even the ko situation is not so *very* different; simple ko fights still
exist as before, with the biggest change likely to be that a different
player has to start them.

So we have here a GENUINE VARIATION, that is STILL ESSENTIALLY GO.

Nice one John!
=============

John's remarks about kos seem to be precisely correct; and I did a
quick check to find all results of 2x2 to 6x6 are the same; (though not
all sequences are exactly the same!) More interesting still, 3x2 go,
which is a bizarre NO-MOVE DRAW (!) at superko normal go, is now a much
plainer 6-win to black! Fascinating.

However, on the big board, the chief difference, as John guesses,
seems to be in sekis. The most common sort that ever actually occur
in real games, (or rather, threaten to occur), are unchanged...
______________
|. O X . X O .
|O . X X X O .
|X X X O O O .
|O O O O . . .

|. . . . . .
|

However, as John partly observed, no-eye sekis are not necessariy the same.

______________
. X O . X O .
. X O . X O . This little situation isn't even a seki, (yet)!
. X O O X O .
. X X X O O . At area scoring, it will still gain a point to play
. . . . . . . a dame in here. Let's say black plays...

______________ Now we get this. This *is* a seki, a 1-liberty seki!
. X O X X O . It's really more like the "Jun Kan Ko", that could
. X O . X O . occur at the standard game; in effect a seki.
. X O O X O .
. X X X O O . If both pass, the 9 points are shared: 0 score.
. . . . . . . If X takes, it finishes up O5-X4: 1 to O.
If O takes, it finishes up O3-X6: 3 to X.
So: neither can take!

HOWEVER...
______________
. X O O X O . But if O had've moved first in the original; now...
. X O . X O .
. X O O X O . If both pass, it's O5-X3: 2 to O.
. X X X O O . If X takes, it finishes up O5-X4: 1 to O.
. . . . . . . If O takes, it finishes up O3-X6: 3 to X.

So: X should take, and win back one point!

Well no, there's not a lot in it! But as John implied, if one has greatly
fewer stones in the seki than the other, he should capture it away. If the
seki is close to evenly balanced (like the third one above), it could make
for some very delicate "under-the-stones" style calculations!

All in all, it looks like a fun variant; and, as most go games, and all
strategy would be identical, it might be worth taking seriously.

Thanks John. And if anyone else wants to lend a hand with some fun
analysis, please jump right in!

-------------------------------------------------------------------------------
----------- ______________________________________________ -----------
| * * * | Bill Taylor W.Ta...@math.canterbury.ac.nz | * * |
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | * * |
| * * * | ============================ | * |
| | FREE THE HANDICAP NINE !! | * |
| * * * | ============================ | * * * |
----------- -----------
-------------------------------------------------------------------------------
In the land of the blind, the one-eyed group is king!
-------------------------------------------------------------------------------

Bill Taylor

未讀,

1996年10月18日凌晨3:00:001996/10/18

收件者：

Looking again at John Tromp's SIM-KILL variation...

> After placing a stone, remove ALL (ie. both black and white) stones that
> have no liberties.

...here's another little oddity.

Similarly (which I should've mentioned yesterday), asymmetric seki of the
"over-nadake" type, are DEAD for the walling group. e.g...

|. . . . . . .
|

.---------------------------------------. ####################################
| _|__ | _|__ _|__ |
| / \ | / \ /xxxx\ | Bill Taylor
|----( )---+----( )(xxxxxx)---|
| \____/ _|__ \____/ \xxxx/ | W.Ta...@math.canterbury.ac.nz
| | / \ /xxxx\ | | ================================
|-------+----( )(xxxxxx)---+-------|
| _|__ \____/ \xxxx/ __|_ |
| / \ /xxxx\ | /xxxx\ | An eye for an eye,
|----( )(xxxxxx)---+---(xxxxxx)----| a group for a group.
| \____/ \xxxx/ | \xxxx/ |
| | | | | |
`---------------------------------------' ####################################

Bill Taylor

未讀,

1996年10月21日凌晨3:00:001996/10/21

收件者：

This is the same post as earlier, with the boo-boo corrected.
=============================================================

Looking again at John Tromp's SIM-KILL variation...

> After placing a stone, remove ALL (ie. both black and white) stones that
> have no liberties.

...here's another little oddity.

Similarly (which I should've mentioned yesterday), asymmetric seki of the
"over-nadake" type, are DEAD for the walling group. e.g...

______________
|X X . O . . . White is DEAD.
|X X X O . X .

|X X X O X . .
|. . . O X . X

Bill Taylor

未讀,

1996年10月21日凌晨3:00:001996/10/21

收件者：

Following myself up... sigh... but I must correct this braino, kindly
pointed out to me by John Rickard.

|> Similarly (which I should've mentioned yesterday), asymmetric seki of the
|> "over-nadake" type, are DEAD for the walling group. e.g...
|>
|> ______________
|> |X X . O . . . White is DEAD.

|> |X X X O . X .
|> |O X X O . X .
|> |O . X O . X .

|> |O O O O . X .

|> |X X X X X . .

|> |. . . . . . .

With so many liberties, white can live, of course, by capturing. But in this...

______________
|X X . O . . . White is DEAD.

|X X X O . X .
|X X X O X . .
|. . . O X . .

|O O O O . X .

|X X X X X . .

|. . . . . . .

Sorreeeeeee!

-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
How to measure size:
===================
In China: measure the height and the width.
In Japan: measure the interior, and add on any previously excreted bits.
-------------------------------------------------------------------------------
.===========================. .- - - - - - - - - - - - - -.
|| || : nese :
|| CHINESE CHINESE || : Japa Japane***
|| CHINESE CHINESE || : ese : **
|| CHINESE CHINESE || : J n : *
|| || : apa Jap : *
`===========================' `- - - - - - - - - - -** - -'

-------------------------------------------------------------------------------