I wonder if anyone ever played havannah on seriously big board? Like there is go 37x37 (though that is a bit overkill).

I wonder how much it would change game, due to shifing costs of different win conditions. Would seem that the smaller the board the less relevant rings are, and the reverse for larger boards, so possibly on large boards rings could even become more prominent win structure while forks and bridges maybe become threats? Does such thinking make sens?

It is possible to play a variant of Havannah on a tetrahedron. I called it Tetrahex, but maybe TetraHavannah would be a beter name. This 3D-board is comparable to an infinite Havannah board. I would like to hear your opinion about it.

Here are the rules and a gameboard for Tetrahex.

I think size 12 is still a playable format for Havannah in 2D.

This touches on the fact that Havannah is not so much the implementation of an underlying idea, but rather a lucky merger of win conditions. Because the ring is invariant to board size the game is not. The right balance is around base-10, meaning that 8 and 12 are not too far off to be playable.

Being only slightly off-topic, I hope this may be of interest to some readers.

If I remember correctly, Christian invented Havannah (or at least introduced it to the world) in around 1980.  In the mid-1990s, unaware of its existence, I came up with this suite of games, which, although ultimately quite distinct from the Freeling brainchild (faster and nastier, for a start), nevertheless probably has enough in common with it to be considered a not-too-distant cousin.

The boards for all three games in the series were quite deliberately designed in such a way as to maintain parity between 'link' and 'loop' lengths, thereby avoiding the variation in balance which Christian has referred to above.  As you will see from the rules, however, the move up from Trig to the larger boards of Trigger and Triggest brings a commensurate increase in the number of ways to win (or lose!) - to be precise, from 2 to 5 to 9. This inevitably makes for steeply growing difficulty in keeping positional focus.  The bigger the board, the more easily the game spirals out of control, so that a typical win at Triggest leaves a considerably higher proportion of cells unoccupied than would be the case at Trig.  Some might regard this as a blemish, but I can only say that games in all three formats tend to be exciting affairs, the need to monitor threats from so many angles being a thoroughly stressful experience.

Anyway, I draw all this to your attention for what it's worth.  At the time of writing, I'm trying to interest Richard in implementing at least the root game Trig.  Any support in that direction would be most welcome.

christian, what happens if board is too large or too small? are you *sure* that something “bad” happens?

Hi Lazy,

I only play base 8 and 10 but this is what I assume will happen on larger boards (small boards have been figured out quite extensively). Besides being invariant to board size, the ring is mainly a tactical weapon and the fork is its strategical counterpart. The bridge is the oil in the machinery. Fork and bridge are not invariant to board size, so the relations between these two on the one side and the ring on the other, change with board size. Obviously base-2 is less than interesting, and so is base-20 because tactics and strategy by then have become 'disconnected'. So there is an optimum somewhere.

I'm completely satisfied with the way play has panned out over the years, but Havannah imo. is not a really great game precisely because of this dependence on board size .

The ring still plays a critical role in Size 4, which I find an interesting game.

I would also like to see a hexhex12 or bigger board. There are boards of all even sizes between 4-10, but I believe the 10 board does not feel like a big game because it still feels a bit rushed and overly tactical. Even if 10 is a balanced size, it does not make sense to use that as the excuse to not add a bigger board because there are already small sizes which are also unbalanced.

Havannah being one of my fav games, so for me it would also be interesting to see how it plays on something more extended. But not sure if 12 would give sufficiently distinct play from 10, as a first post in thread said, on large boards rings might be more dominant structures, so something that would have better chance of showing that would be nice, maybe something like size 18. Though I presume games would be looooong :). Also I kinda slightly suspect that possibly rings might still not become all that dominant, as to me it seems that main problem with rings is how easy (well comparatively, at least it seems so to me) it is to disrupt them.

I would also say 18 would be better than 12 to add. But I fear most players would think 18 to be too big, thats why I said 12 or bigger.

In 2012 I played against Timo Ewalds' program 'Castro' in the context of a Man vs Machine challenge and Castro, more than the other programs, was extremely, not to mention annoyingly, ring-happy. For me it felt as if all plans were constantly disrupted and I couldn't get into my way of playing at all. So in my opinion larger boards would show a considerable ring dominance that would affect the internal balance such as it is known in the base-10 game. Up to base-12 that may be a matter of taste, but on larger boards the whole nature of the game might change and, so far as I'm concerned, not for the better. But anyone who wants to find out for him or herself has my blessing. :)