Atari-Go with gomoku?

[Elom]

Would it be useful to only use atari-go as a method of teaching when in tandem with gomoku? Perhaps they may offset each other's possible negative side effects for go.

[Bill Spight]

Gomoku is its own game, with no relation to go except placing Black and White stones on intersection of the go board.

Atari go is a variant of go with similar strategy. It's easier to live, however. It is best to play atari go with no passing. If a player has no play, they lose. You can move up to Capture-Two, Capture-Four, and so on. The more captures you need, the closer the game is to regular go.

[EdLee]

Hi Bill,

Atari go is a variant of go with similar strategy.

Is Atari-go the same as Capture-go ( = Capture-1 ) ?

It is best to play atari go with no passing. If a player has no play, they lose

Are there some objective reasons no passing is better ?
Since I never knew about this no-passing variant ( until reading about it on this forum some years ago ), I've never tried it.
This way ( Capture-N with passing ), if the beginner can get a tie with the more experienced person, I congratulate them ( as opposed to losing ).

Elom, this is the first time I hear about gomoku in relation to the teaching of Go... other than this chat with curious bystanders:

Code: Select all

Oh, we used to play this all the time!
Great! Then I don't have to explain the rules!
Well, maybe a refresher wouldn't hurt...
Sure! ( time passes )
Oh! We played this differently...
Could you show me how you played it ? ( time passes )
Oh, I see. :) You were playing Pente, not Go. :)

I don't have any strong anecdotal evidence one way or the other, but my feeling is mixing Pente with Go for raw beginners is not helpful for teaching Go ?

[Bill Spight]

Hi Bill,

Atari go is a variant of go with similar strategy.

Is Atari-go the same as Capture-go ( = Capture-1 ) ?

Yes.

It is best to play atari go with no passing. If a player has no play, they lose

Are there some objective reasons no passing is better ?
Since I never knew about this no-passing variant ( until reading about it on this forum some years ago ), I've never tried it.
This way ( Capture-N with passing ), if the beginner can get a tie with the more experienced person, I congratulate them ( as opposed to losing ).

Almost every form of no pass go has a kind of territory. I. e., points where one player can play with impunity but the other player cannot. That leads to the correspondence between moves and points of territory. No pass capture go, then, provides a natural way for a beginner to learn about territory. Also, the frequent question about why isn't it advantageous to play inside the opponent's territory, forcing him to fill his own territory to capture the stone or stones, does not arise, because with no passes there is no such gain. (OC, I am talking about Capture-N.)

How does a tie happen in Capture-One? What I have observed is that the teacher simply counts territory. Why not just keep playing and teach territory that way?

All you need to live in Capture-One is a two point eye. That means that on, say, a 7x7 board tactics matter less and strategy matters more. In a way, the Capture Game is even more strategical than regular go.

[Bonobo]

[…] In a way, the Capture Game is even more strategical than regular go.

When teaching beginners, I have lost SO many games of Atari-Go because at some point I forgot it was Atari-Go and I played some throw-in

[EdLee]

Hi Bill,

How does a tie happen in Capture-One? What I have observed is that the teacher simply counts territory. Why not just keep playing and teach territory that way?

A tie happens when neither side can reach N captures in Capture-N (passes optional and OK). ( Because the stated goal from the beginning is that we play this introductory verion, where whoever captures first wins, and not whoever controls the most territory wins. Thus a tie is possible. ) Sometimes that happens as early as game 4 with a beginner (they lose all first 3), at which point happy congrats. But usually they never get to a tie on day 1 against a much more experienced person (which is understandable).

I have nothing for or against the no-pass-and-continue-to-count variant; it's unclear to me why one version has to be any better than another. So far, from my experience, Capture-1 with pass has worked very nicely. Only until the recent post here, I never knew about Capture-N where N>1 ; also, I wasn't clear what's N ( N stones or N instances of capture ? ). We went with the latter in our most recent try.

[EdLee]

Hi Tom,

I played some throw-in

Nice teaching moments. (They win and feel happy, and both of you learn something new. )

[Bill Spight]

I have nothing for or against the no-pass-and-continue-to-count variant; it's unclear to me why one version has to be any better than another. So far, from my experience, Capture-1 with pass has worked very nicely. Only until the recent post here, I never knew about Capture-N where N>1 ; also, I wasn't clear what's N ( N stones or N occurences of capture ? ). We went with the latter in our most recent try.

Well, as I said, No pass capture go has an inherent concept of territory which is like that of regular go. (It's really territory scoring with a group tax. ) Which means that once only territory is left you can stop and count the score. Just like with regular go.

I found out about Capture-N on this site: http://www.vimage.co.jp/~iehiro/cgi-bin/gocaffe/ . N stands for the number of stones, if you want to keep the correspondence to territory.

[Bonobo]

Hi Tom,

I played some throw-in

Nice teaching moments. (They win and feel happy, and both of you learn something new. )

And a big laugh for all of us each time it happened. And of course I explained to them why I played there, so they really learned something new. Whether I learned something new … I don't know, especially b/c it happened to me so often …

[Elom]

Hi Bill,

Atari go is a variant of go with similar strategy.

Is Atari-go the same as Capture-go ( = Capture-1 ) ?

Now you've given me the idea for Atari-state go (stone in atari? You lose.)

It is best to play atari go with no passing. If a player has no play, they lose

Are there some objective reasons no passing is better ?
Since I never knew about this no-passing variant ( until reading about it on this forum some years ago ), I've never tried it.
This way ( Capture-N with passing ), if the beginner can get a tie with the more experienced person, I congratulate them ( as opposed to losing ).

Elom, this is the first time I hear about gomoku in relation to the teaching of Go... other than this chat with curious bystanders:

Code: Select all

Oh, we used to play this all the time!
Great! Then I don't have to explain the rules!
Well, maybe a refresher wouldn't hurt...
Sure! ( time passes )
Oh! We played this differently...
Could you show me how you played it ? ( time passes )
Oh, I see. :) You were playing Pente, not Go. :)

I don't have any strong anecdotal evidence one way or the other, but my feeling is mixing Pente with Go for raw beginners is not helpful for teaching Go ?

Nice story .

My feeling with pente I share with yours— I played a few games a while back, and thought it might reduce my awareness for stones in atari in go. Which is, as far as I know, the opposite of the disputed effect some worry for when teaching via atari-go, namely, an over-focus on capturing stones (as Bill Spight mentioned, capture-N go's similarity to territorial go increases with N, which should alay such a fear at least to some extent).

So on their own as introductions, they may negatively sway your beginner go when you eventually play, but alternated beforehand and I wonder if these opposites cancel out. Time for some field testing, perhaps .

[Bill Spight]

My feeling with pente I share with yours— I played a few games a while back, and thought it might reduce my awareness for stones in atari in go. Which is, as far as I know, the opposite of the disputed effect some worry for when teaching via atari-go, namely, an over-focus on capturing stones (as Bill Spight mentioned, capture-N go's similarity to territorial go increases with N, which should alay such a fear at least to some extent)

I was one of those who thought that starting raw beginners off on Capture Go was not good because it encouraged the bad habit of trying to capture stones. Two things changed my mind. The first was hearing about someone who taught Capture Go to a class of school kids one day and then let them play among themselves. One little girl became invincible by simply keeping her stones connected. The second was learning about the no pass version. It immediately became apparent to me that with the no pass version, once the beginner learns to avoid capture, Capture Go becomes a game of territory.

But it is still not regular go. For one thing, a two point eye is alive. And sacrifices do not work. This is where Capture-N comes in. Capture-Two means that a two point eye dies by the sacrifice of one stone. It also means that snapbacks work. BTW, Ed, that's a good reason to count stones, not captures. A snapback no longer loses, per se, because afterwards each player has one capture.

In Capture-Two a three point eye with an opposing stone in the center is seki. In Capture-Four it dies when the opponent sacrifices three stones. (That's why I recommend skipping Capture-Three. )

With large enough N Capture-N is almost indistiguishable from regular go with territory scoring and a group tax.

[jlt]

You may also want to try the Strasbourg rules. Passing is allowed. The winner is the player who has the largest number of stones on the board. This is more or less equivalent to regular go with territory scoring and a group tax.

[Bill Spight]

You may also want to try the Strasbourg rules. Passing is allowed. The winner is the player who has the largest number of stones on the board. This is more or less equivalent to regular go with territory scoring and a group tax.

It's even more like Stone Counting.

Which persisted into the 20th century in parts of China.

[EdLee]

Hi Elom,

Thanks.

That episode has happened countless times over the past 15 years;
the most recent one was probably two weeks ago.
( And one can substitute Pente with Othello; similar outcome. )