Life In 19x19

Side note: the idea that stones are the simplest elementary particle and have two different "frequencies" (black and white), roughly corresponds to string theory. So, if string theory is ever proven, it could be said that Go is such a good game because it has the same structural basis as the universe.

And if it turned out that truth of the (old?) standard model was a brute fact, in the sense that it couldn't be further explained, then we'd prefer a game with a dozen or more different pieces, each of which had seemingly arbitrary powers and rules?

hyperpape wrote:And if it turned out that truth of the (old?) standard model was a brute fact, in the sense that it couldn't be further explained, then we'd prefer a game with a dozen or more different pieces, each of which had seemingly arbitrary powers and rules?

Uh... no.

I agree with a lot of your principles for good board game design. In fact I'd say most of your other rules can boil down to applications of #5 (All rules must be simple as possible) and #6 (Enough variation to challenge humans).

1) One type of unit, of two opposite "charges"
Certainly the fewer unit types is simpler (#5). There are actually games with only one unit type for both players, i.e. nim, dots & boxes, but those tend to be too easy. So we have two types, maybe the smallest number with a good challenge (#6).

2) 2 dimensions
Again, 3D or higher seems very complicated for board games (and even other games), so it's out (#5). 1D is out by (#6) (although maybe a 1D game could be challenging?)

3) No Movement
Obvious application of (#5). Actually hex has less movement than go; in go stones are added AND removed!

4) komi
If you mean komi specifically, I'm not sure that's actually a design principle. I think maybe a better rule is something like "equally challenging for both players" (an application of #6). There was some discussion about draws, and maybe too many draws are bad, but komi doesn't have to prevent draws, only if you add 0.5. Some people would argue that perfect play SHOULD give a draw. If Black plays perfectly, should he lose by 0.5? Should white? As a side note, tic-tac-toe with pie rule is a draw, just like normal tic-tac-toe. And hex can never be a draw, even with pie-rule.

To me, the absolute simplest game that has playable complexity is actually hex. But I still prefer go, because I think overall the challenge and enjoyment is greater. But as palapiku says, that could be a result of the community, not the game itself (although I doubt it).

nagano wrote:So the game needs to have:
1. One type of unit, of two opposite "charges".
2. Two dimensions.
3. No movement.
4. Komi.
5. All rules must be simple as possible to avoid unnecessary complication.
6. Enough variation to challenge humans.

Here is a go like game, Gone (See http://senseis.xmp.net/?AntiAtariGo ), that meets all of the requirements except #4,

Gone is played like go, but with these rules:

1) No pass allowed.
2) No capture allowed.
3) No suicide allowed.
4) If you have no move, you lose.

It is surprisingly difficult.

emeraldemon wrote:I agree with a lot of your principles for good board game design. In fact I'd say most of your other rules can boil down to applications of #5 (All rules must be simple as possible) and #6 (Enough variation to challenge humans).

1) One type of unit, of two opposite "charges"
Certainly the fewer unit types is simpler (#5). There are actually games with only one unit type for both players, i.e. nim, dots & boxes, but those tend to be too easy. So we have two types, maybe the smallest number with a good challenge (#6).

That does not matter, as long as all game pieces have equal ability.

2) 2 dimensions
Again, 3D or higher seems very complicated for board games (and even other games), so it's out (#5). 1D is out by (#6) (although maybe a 1D game could be challenging?)

You have no idea how complicated 3D is. Try this. As far as 1D is concerned, it would have to be an awfully long line to allow for much variation.

3) No Movement
Obvious application of (#5). Actually hex has less movement than go; in go stones are added AND removed!

Yes, but no stone is ever moved from point a to point b, so that is irrelevant.

4) komi
If you mean komi specifically, I'm not sure that's actually a design principle. I think maybe a better rule is something like "equally challenging for both players" (an application of #6). There was some discussion about draws, and maybe too many draws are bad, but komi doesn't have to prevent draws, only if you add 0.5. Some people would argue that perfect play SHOULD give a draw. If Black plays perfectly, should he lose by 0.5? Should white? As a side note, tic-tac-toe with pie rule is a draw, just like normal tic-tac-toe. And hex can never be a draw, even with pie-rule.

No, komi isn't a design principle in itself. It's just the only method that currently exists that is sufficient. Uh... about those Tic-Tak-Toe comments. I believe that I wrote Tic-Tak-Toe, but was thinking of Hex. If you look at my analysis and apply it to Hex, it is valid.

To me, the absolute simplest game that has playable complexity is actually hex. But I still prefer go, because I think overall the challenge and enjoyment is greater. But as palapiku says, that could be a result of the community, not the game itself (although I doubt it).

I would agree, if it were not for the fact that I'm dissatisfied with the effectiveness of the pie rule and don't know of any usable method other than basing the game on scoring.

I think that Amazons ( http://en.wikipedia.org/wiki/Game_of_the_Amazons ) is the kind of game you are looking for, even though it involves movement.

Bill Spight wrote:I think that Amazons ( http://en.wikipedia.org/wiki/Game_of_the_Amazons ) is the kind of game you are looking for, even though it involves movement.

Hmm... yeah, if you add komi to it, it looks fine. Though it does have a minor "flaw" in allowing diagonal movement. So it doesn't *quite* meet my criteria.

You might find arimaa interesting too. It seems to be the closest to your principles I've seen so far for a game that's still using moving pieces in a chess like manner.

While there are different pieces they only differ in strength, and not in movement type. So it comes close to the ideal of a chess type game with only one piece.

The first move advantage is reduces by the second player choosing his starting formations after the first one. This is obviously not as fine tunable as komi. On the other hand the variable piece placement makes a variation using the rule that one player places them, and the other chooses the side viable and interesting.

Something else you didn't adress when you said you concluded that go was the best game was the board size. your criteria dont discuss board size at all. what makes 19x19 perfect?

We do not know if 19x19 is perfect but, from a human view, it is about the most demanding size still suitable in duration for human play.

Square grid allows for a strategically dynamic game. Maybe other grids would fit, too. We do not know any better fitting grid structure yet though.

Square board is regular enough and convex so that there are no boring remote parts, which implicitly would simplify global strategy. Due to the square grid, a rectangular board minimizes the number of corners in the board shape. Therefore a rectangular board is optimal for a square grid. As a finer consideration, a square board is optimal for a square grid so that none of the edges is less interesting than other edges. Board symmetry maximizes global strategic interaction consistently.

Odd square board sizes have the advantage of the existence of a single center intersection. Maybe this feature is over-estimated though. In the following, I presume a preference for odd sizes.

19x19 is known by every advanced player to be more interesting than 17x17. The next question is if 21x21 is more interesting than 19x19. Since 21x21 has been tried only rarely so far, we cannot be sure. One theory says that bigger boards lead to too long and boring endgames but this could simply be a consequence of our missing better understanding of bigger boards. Therefore currently the game duration is the decisive limiting factor.

19x19 suffices to require a lifetime to master it, as they say.

Li Kao wrote:You might find arimaa interesting too.

It is in some ways simpler than chess, but really checkers is closer to meeting my requirements than it is (and that's not very close).

ramanujan wrote:Something else you didn't adress when you said you concluded that go was the best game was the board size. your criteria dont discuss board size at all. what makes 19x19 perfect?

I mostly agree with Robert on the reasoning for 19x19. See the Sensei's article "Different Sized Boards" for more information. Why did you say "Something else you didn't address..."? What else do you believe I have left out? Do you agree or disagree with my reasoning?

i think just about any connection game can be applied to a 3d board easily and with interesting results.

for example, lines of action; each piece moves like a queen, but the number of squares it can move is determined by the number of pieces along that line. a 6x6 board with a border row where the pieces start off, and the first player to connect all thier pieces win.

in the 3d version, i imagine a 3 player game, one player going top to bottom, another from left to right, and the third from forward to back. perhaps 5x5x5 rather than 6x6x6.

Yes, but again, this violates rule 4.

not sure i agree komi is nessicary to make a game interesting.
any game with movement generally has no point scoring, giving the first player a slight advantage.
however, having a slight advantage is no garentee to win, not by a long shot. in the 3d version, where you have 3 players, the other two players could easily gang up on player 1, making the game unfair for him.

even in go where you have komi, black wins 52% of the time. is that fair?