New game: Teime
luigi
by Luis Bolaños Mures
Teime is a new connection game which is played on a regular hexagonal
board with 10 cells per side.
The game starts on an empty board. Each player has stones of his
allocated color: white or black. In their turn, starting with plack,
players put one stone on a vacant cell.
The game is won by the player who creates a "teime". A "teime" is a
closed connection of like-colored stones among three board sides of
which at least one is not adjacent to the other two. Corners are
considered to belong to the sides they touch. Please note that any
closed connection of like-colored stones among more than three sides
is always a teime.
Since the first player has a distinct advantage, the pie rule is used.
Draws are not possible in Teime. Also, as in Hex, any random filling
of the board always produces one, and only one, winner. In fact, my
objective with Teime was to create a game which corrected Hex's few
faults, which are IMHO the irregular board and the fact that the sides
every player has to connect are fixed from the start.
I am pretty sure of Teime's drawlessness, but I cannot find the way to
formally prove it. I will thank you if you help me on this point.
Enjoy!
Harald Korneliussen
This sounds a lot like Christian Freeling's Havannah, or Cameron
Browne's Cross. The difference is that both those games have
additional win conditions/counter-conditions, in addition to the
"fork". I don't remember the discussion (if there was one), but I
think there was some reason that the simpler, more elegant game that
you propose was rejected.
I would talk with Nick Bentley, he's experimented more than most with
these sorts of games (hex-hex board connection games).
Disclosure: I like Hex, and I didn't much like Freeling's attempt to
"fix" it :-)
Nick Bentley
good game. But it has it's own flaws. Like Y, cells near the center
are extremely important, more important than in Hex, and this limits
the depth a little, IMO.
You can play a version that is quite close to the one presented by
luigi, in real time at the ig game center, by opening a game of Mind
Ninja an selecting the pattern "Almost Y". The differences are: in
the online version corners don't belong to sides, and the first mover
is trying to form the pattern described by Luigi, and the other is
trying to prevent it. But since this can only be done by forming the
same pattern, it amounts to the same thing. These changes were
introduced in order to balance the game more to the second player's
favor, but honestly, I don't think it's really an improvement over the
version Luigi describes. (although there are some unusual tactics
having to do with the corners. Specifically, in the online version,
player 2 can now try to "funnel" his opponent towards corners in order
to prevent the Y pattern from forming.
I'm in the middle of playing some bizarre variations on Cameron
Browne's Cross by email at the moment. I hope one of them works,
because they are really strange.
About Havannah: yeah, it's a bit lame. But the tactics are very rich
and interesting, which I believe is mainly due to the presence of the
loop win condition.
Bill Taylor
This allegedly new game is not greatly different from other attempts
to make a decent connection game on a hexagon, that uses
the alternate-edges-Y condition as the chief win criterion.
In particular, this claim...
> as in Hex, any random filling
> of the board always produces one, and only one, winner.
...is complete nonsense!
The board is almost certain to fill up with both players
getting to create a "Teime", namely one connecting edges A,B,D
and the other edges D,F,A. (Marking edges cyclically, as obvious.)
That is, both create a winning position, and the one who does it first
wins the game. It becomes a mere race. Which we hate. And in
particular, it essentially becomes a race to sprint across from
one side to the opposite one, in the above case A to D,
(then, in this game, join up to a third).
What on earth is everyone thinking of!!
Of course it is playable, but it's a mere race, and so of no interest
to us true connection-game afficionados, who regard the essence
of a connection game as being its win/loss complementarity,
described in the excerpted comment above.
See my parallel post, "The HEXA-Y problem", for a full account.
-- Bruising Bill
** Q: What is a consultant?
** A: A person who does a con job on you then lives like a sultan.
Bill Taylor
> About Havannah: yeah, it's a bit lame.
Definitely lame, as it stands.
In particular, the special status of corner cells is ridiculous.
As I have written before, the essential nature of Havannah is
the tension between the global goal of making a Y (or cross)
connection, competing with the local goal of making a loop.
This tension is a marvellous frissant, and leads to all sorts
of new tactics and strategies that do not appear in the usual
connection games.
Of course, it re-introduces a "race" element into the game,
which is unfortunate, but seemingly unavoidable, except by
re-writing the Y goal, and changing the board shape.
Even so there is a race to make a loop, so this element
is unavoidable.
But in any event, corners should *always* be regular cells,
belonging to both adjacent sides. That aspect of Havannah
is lunacy, and Joao Neto and I have been playing "Lisbon",
which is essentially Havannah without special corners,
for ages. It's a great game!
> But the tactics are very rich and interesting, which I believe is
> mainly due to the presence of the loop win condition.
Quite so! :-)
-- Beaming Bill
Q: What's the difference between a preacher and a used car salesman?
A: The used car salesman knows when he's lying.
Nick Bentley
wrote:
In the version I play, one person is trying to form a Y of the type
luigi describes, and the other is trying to block its formation. This
eliminates the race aspect, but still leaves the blocker with a
similar connection goal.
But for those for whom anything but isomorphic or identical connection
goals is distasteful, this version of the game will not satisfy. I
love isomorphic/identical goals where they are possible, but I also
wholeheartedly embrace asymmetry in those cases where it leads to an
interesting and balance-able game.
luigi
> In particular, this claim...
>
> > as in Hex, any random filling
> > of the board always produces one, and only one, winner.
>
> ...is complete nonsense!
>
> The board is almost certain to fill up with both players
> getting to create a "Teime", namely one connecting edges A,B,D
> and the other edges D,F,A. (Marking edges cyclically, as obvious.)
>
> That is, both create a winning position, and the one who does it first
> wins the game. It becomes a mere race. Which we hate. And in
> particular, it essentially becomes a race to sprint across from
> one side to the opposite one, in the above case A to D,
> (then, in this game, join up to a third).
Oops! Seems like I should have thought a bit more before claiming my
game's win/loss complementarity.
Anyway, as this game gets rid of that weird
corners/sides distinction, I still prefer it over Havannah, specially
if you also consider loops as winning formations. Moreover,
"teimes" (or however you want to call them) are harder to make than
both Havannah's forks and bridges, making this hypothetical game a bit
(just a bit, unfortunately) less race-like than Freeling's popular one.
Bill Taylor
> In the version I play, one person is trying to form a Y of the type
> luigi describes, and the other is trying to block its formation. This
> eliminates the race aspect, but still leaves the blocker with a
> similar connection goal.
We play a similar version, but where the Y-forming player has to make
a genuine three-alternate-sides connection, a "symmetric" Y.
(OC it may connect other sides as well, irrelevantly.)
We find this asymmetry is not actually a bug, but a FEATURE,
in "chicken" games (like Unlur) where both players play the weaker
side's stones, until someone (is "chicken" and) elects to take
the weaker side (which he deems has now become strong enough
to win with all those early moves), and (necessarily) passes,
and then regular game alternation begins.
There are many game types where one side is "naturally"
stronger than the other, such as Hexagonal Y with only one
player reqyuired to make a Y, and the other to stop him,
(equivalently: make his own Y or an opposites-crossing).
Others are small-board moku games, where the board is
too small and the moku-size (typically 5) is too large for
the moku-former to win ordinarily. Another would be
a square connection game where the same person
has to connect all four sides; and so on.
The typical case is:- one player has to perform some
quite hard task, and the other only has to prevent him.
No-one knows who is which, until the "chicken" declaration
has been made. That person then takes the difficult task.
> But for those for whom anything but isomorphic or identical connection
> goals is distasteful, this version of the game will not satisfy.
Of course the true afficionado will love both types!
> I love isomorphic/identical goals where they are possible, but I also
> wholeheartedly embrace asymmetry in those cases where it leads to an
> interesting and balance-able game.
How can one say better than that!! :-)
-- Bantam Bill