Games of Interest at the 1000-Year Game Design Challenge
Nick Bentley
through all the entries at the 1000-Year Game Design Challenge and
pick out the ones that caught my eye. Links to the games and comments
at the link below.
http://nickbentley.posterous.com/games-of-interest-from-the-1000-year-game-des
Check em out and comment for the sake of discussion.
christian
> Though I *should* be working or something, I couldn't help but wade
> through all the entries at the 1000-Year Game Design Challenge and
> pick out the ones that caught my eye. Links to the games and comments
> at the link below.
>
>
> Check em out and comment for the sake of discussion.
Cartography looks great. Can't say much about Tricala, but somehow I
don't see it last a thousand years. The same for Charing Cross. The
rest isn't my cup of tea, so I can't comment.
You didn't mention Flume. I fear Mark will notice ;-) But fun aside, I
feel it is a serious contender for a lasting game.
Nick Bentley
> feel it is a serious contender for a lasting game.
I agree, except I don't like dots-n-boxes style games much. No offense
Mark: Flume is clearly excellent for what it is.
Nick Bentley
Mark Steere
> I'm curious Mark: why Flume, and not, say, Oust?
Both Oust and Flume will certainly thrive 1000 years from now. But
the Flume rule set is easier and its gameplay more accessible. You
have to play Oust 100 times before the counterintuitive tactics take
hold.
I like the millennium theme. It takes time for a body of strategic
knowledge to grow. Oust is the greatest game in the world, but the
world doesn't know it yet. The world still thinks it's Go, lol
Nick Bentley
Hmmm. So, I feel that if you're trying to win the contest, as opposed
to just submitting your best game, Oust may *still* be the better
choice.
My reasoning: Flume is like other dots-n-boxes games in that the
opening is obscure to newbies. The vast majority of people don't see
the depth in these kinds of games are because the openings are so hard
to comprehend.
The contest administrators aren't going to be able play any of the
entries in depth enough to get past opening bafflement. In other
words, the games which give a good first impression will win even if
they're not the most worthy of victory. Ergo, Flume is maybe not the
most strategic choice. Rebuttal?
Nick Bentley
I guess you already have in claiming that Oust tactics are too
obscure. But for what it's worth, I found Oust more accessible on
first play than Flume.
christian
> My reasoning: Flume is like other dots-n-boxes games in that the
> opening is obscure to newbies. The vast majority of people don't see
> the depth in these kinds of games are because the openings are so hard
> to comprehend.
That's true to an equal degree, if not more so, for Oust. For lack of
playing experience I'm not qualified to give a detailed comparison but
I feel Flume stands a better chance.
>
> The contest administrators aren't going to be able play any of the
> entries in depth enough to get past opening bafflement. In other
> words, the games which give a good first impression will win even if
> they're not the most worthy of victory. Ergo, Flume is maybe not the
> most strategic choice. Rebuttal?
Of course meeting the condition isn't dependent on the nature of the
game alone. Anyone who claims to be able to predict what the world
will be like for humans in a thousand years from now, if there are any
that is, or what 'being human' even means by that time, is fooling at
least one person.
marks...@gmail.com
On 22-Aug-2011, Nick Bentley <nicko...@gmail.com> wrote:
> My reasoning: Flume is like other dots-n-boxes games in that the
> opening is obscure to newbies.
Flume is similar to Dots and Boxes in this respect, though I personally
don't know of any other "dots-n-boxes games" - not that I doubt their
existence.
From the 1000 year contest: "Elegance: Entries should be easy to learn,
teach and play. We favor entries that can be explained succinctly and played
in under an hour."
The choice between Flume and Oust as contest entries was a tough one. Oust
is harder to learn though. If you look at the two rule sheets, Flume can be
explained in half a page, with the bottom half being used for example moves.
Flume:
http://www.marksteeregames.com/Flume_Go_rules.pdf
Oust:
http://www.marksteeregames.com/Oust_rules.pdf
More people had trouble understanding Oust than Flume, though a lot of
people had trouble with the Flume rule sheet too.
Yes, the opening play in Flume is "random" in a sense. You're just laying
down stones, setting up the board in preparation for the "thinking" phase,
which only lasts for a few turns.
A "randomized" opening followed by a short but intense thought period makes
for a short game *with negligible turn order advantage*. You've got a game
of (essentially) 30-35 turns with no more turn order advantage than many 60+
turn games.
Nick Bentley
>don't know of any other "dots-n-boxes games" - not that I doubt their
>existence.
I know of several. There's even another dots-n-boxes-derived entry in
the 1000-Year-Contest, called Zuniq:
(I'll be surprised if that one doesn't suck terribly, but you never
know)
One that I like is just like dots-n-boxes, except that when any
enclosed area is formed, the player claims *all the squares* in that
area. That makes the opening less obscure, simply because the mid-
game arrives faster.
I could name a few others as well, but you get the idea.
> If you look at the two rule sheets, Flume can be
> explained in half a page, with the bottom half being used for example moves.
Good point. This probably matters.
Nick Bentley
> playing experience I'm not qualified to give a detailed comparison but
> I feel Flume stands a better chance.
You may be right. I can only speak from personal experience and for me
Oust was the better experience for me out of the gate (I also like it
more over-all). If I'm not like the administrators, my opinion is
meaningless.
> Of course meeting the condition isn't dependent on the nature of the
> game alone. Anyone who claims to be able to predict what the world
> will be like for humans in a thousand years from now, if there are any
> that is, or what 'being human' even means by that time, is fooling at
> least one person.
THIS is the really interesting part of the contest. I wish I'd known
about this contest earlier so I could design a game around my wild
speculations about what we'll be in 1000 years. I have doubts about
whether there will even be humans in a thousand years. I could have
designed a game for cockroaches.
Mark Steere
>
> Anyone who claims to be able to predict what the world
> will be like for humans in a thousand years from now, if there are any
> that is, or what 'being human' even means by that time, is fooling at
> least one person.
Who? George W. Bush?
christian
Ok, three.
christian
>
> I know of several. There's even another dots-n-boxes-derived entry in
> the 1000-Year-Contest, called Zuniq:
>
>
> (I'll be surprised if that one doesn't suck terribly, but you never
> know)
>
Doesn't look so bad. Fencing off playable areas speeds up the game,
effectively 'cutting out the middle game'.
The reason for the restriction on same sized areas (I take that to be
the number of enclosed squares) is unclear to me.
There's also a rather traditional looking abstract named Push:
http://danielsolisblog.blogspot.com/2011/07/thousand-year-game-design-challenge_31.html
I'm not so taken by a YouTube presentation though.
Nick Bentley
> I'm not so taken by a YouTube presentation though.
Yes, there were several abstracts I didn't include, including one on a
Go board that I thought you might be interested in.
Anyone want to wager a guess about which games will win? I fear "our"
kind of games don't have a chance. One of the 2 judges is female and
few women seem to enjoy perfect-info, geometrical abstracts.
My money is maybe on Librarium.
christian
> Yes, there were several abstracts I didn't include, including one on a
> Go board that I thought you might be interested in.
Can you give a link, I don't feel like wading through the whole bunch.
> Anyone want to wager a guess about which games will win? I fear "our"
> kind of games don't have a chance. One of the 2 judges is female and
> few women seem to enjoy perfect-info, geometrical abstracts.
>
Yes, but you can't generalize up to one person, so who knows.
Cartography might be a game women enjoy. But that's the same
generalization.
I don't quite agree 'our' games don't stand a chance. The only known
games that are over a thousand years old, or their ancestors at least,
are Go, Chess, Mancalas, Morris ... and Draughts isn't too young
either.
Great that Luis entered Yodd!
marks...@gmail.com
On 23-Aug-2011, christian <chri...@mindsports.nl> wrote:
> Great that Luis entered Yodd!
It's unanimous. Yodd, the greatest game that nobody ever actually played.
christian
>
> It's unanimous. Yodd, the greatest game that nobody ever actually played.
We're in the process of remedying that :)
christian
http://mindsports.nl/index.php/the-pit#territory
Bottom of the list (for alphabetical reasons only).
I'll have to add a few diagrams and Ed's working on an applet (the hex
one for starters). That way Yodd can be played along with Ketchup and
Sygo.
I'll copy hexamples from Luis' rulesheet, but a couple of square
examples would be welcome.
christian
> I'll copy hexamples from Luis' rulesheet.
christian
> ... but a couple of square examples would be welcome.
Luis, what I need is a couple of similar exaples as the ones in Yodd,
on say a 5x5 board.
Everything is different of course, for instance you can split a
diagonal pair with two stones _unless_ you contact one own group.
It's the other way around really :)
I'd rather see you figure out a couple of representative tactics
because you're deeper into the game.
Please send the stuff to my email address 'christian at mindsports dot
nl'.
Nick Bentley
similar to *Star, just with totally different tactics. The more I
think about Yodd, the more I love the idea. I hope, hope, hope it
plays well.
christian
I have no doubts about that. I forgot to include the Xodd link:
http://mindsports.nl/index.php/the-pit/xodd-624
Despite Luis' saying that joining groups becomes harder, and that the
game is more tactical than the hexversion, I have rather high
expectations of it.
This is all 'intuition' of course, and since mentioning that concept
will trigger a response by Mark anyway, it would be nice to hear
something more insightful about Yodd than "it sounds like a Kris Burm
game" and "the greatest game that nobody ever actually played".
Nick Bentley
> will trigger a response by Mark anyway, it would be nice to hear
> something more insightful about Yodd than "it sounds like a Kris Burm
> game" and "the greatest game that nobody ever actually played".
Another apparent feature is that there is some clarity in both tactics
and strategy *despite* what appears to be a colossal branch factor.
There aren't many games like that. The few that are, like Amazons,
make for heady company.
I wonder what is lost if you drop friendly stones only. I guess the
friendly-only version would be colder because your opponent would
block off territories where you you'd want to avoid putting stones.
I'm starting to see the tactics by playing through mini-situations in
my head and they look tantalizing.
christian
> I wonder what is lost if you drop friendly stones only.
About everything I'm afraid. White on his first move would have to
drop to singles or pass (and if he keeps passing, does he win with 0
groups?). You don't have that 'passing' problem in the current
version.
If that were settled (extra rule), black could keep clogging and white
would have to stop him with 2 or 4 groups to force black to drop
another two singles ... doesn't feel good.
>
> I'm starting to see the tactics by playing through mini-situations in
> my head and they look tantalizing.
I'm trusting they will be but I'll wait till the applet is ready :)
Nick Bentley
My one fear is that it will turn out that the player who wins the
center wins the game. It certainly is true on small boards. It may be
that tactical play in the center dominates everything else at high
level play.
luigi
> http://mindsports.nl/index.php/the-pit#territory
> Bottom of the list (for alphabetical reasons only).
>
> I'll have to add a few diagrams and Ed's working on an applet (the hex
> one for starters). That way Yodd can be played along with Ketchup and
> Sygo.
Wow, that's wonderful news! Thank you very much, Christian!
On 24 ago, 19:30, christian <christ...@mindsports.nl> wrote:
> On Aug 24, 3:44 pm, christian <christ...@mindsports.nl> wrote:
>
> > ... but a couple of square examples would be welcome.
>
> Luis, what I need is a couple of similar exaples as the ones in Yodd,
> on say a 5x5 board.
>
> Please send the stuff to my email address 'christian at mindsports dot
> nl'.
Done!
I'm also planning to write a standalone Xodd ruleset, in case it turns
out to be better than Yodd.
luigi
> My one fear is that it will turn out that the player who wins the
> center wins the game.
Well, it's the same for chess ;-)
luigi
> diagonal pair with two stones _unless_ you contact one own group.
> It's the other way around really :)
That's right. So, when you form a diagonal pair, you can play an
opponent stone nearby...
. . .
O . X
. O .
... to not only increase your opponent's group count, but actually
ensure the connection of your own stones!
There should be more interesting tactical patterns like this to be
found in Xodd. It may well turn out to be better than Yodd, after
all...
> Despite Luis' saying that joining groups becomes harder, and that the
> game is more tactical than the hexversion, I have rather high
> expectations of it.
Well, those are facts. The point is whether they favor the game or
not. Let's see:
- As there will be more groups at the end of the game, differences in
score can be higher, and so slight differences in skill are more
likely to become apparent.
- More tactics could be a good thing in a game where stones are never
removed from the board.
Just speculating. It's easier to predict a good gameplay in Yodd,
because you can interpolate from a fair bunch of connection games on a
hex board. For Xodd, more experience will be needed.
If thick chains are too difficult to make, and are needed for
territories to pay off, then Xodd won't likely work. On the contrary,
if diagonally connected chains ensure the creation of more opponent
groups than are in the chain, or it's likely that you'll be able to
solidly connect a fair amount of them, then Xodd could be great. Of
course, it's the same for both players, so this objections may well be
irrelevant (?!?).
A large part of the answer must lay on the specific tactical tricks
provided by the square grid.
luigi
> and strategy *despite* what appears to be a colossal branch factor.
There is an upper bound of
2m(m-1) + 2m + 1
moves available on any given turn, m being the number of empty spaces
(the '1' is a pass). So, on average, there are less than
(2n^2/3) + 1
moves available on each turn, n being the size of the board (total
number of cells). This is the integral of the previous function for m
= n, divided into n.
So, in plain English, on a reasonable base-8 hex board or base-13
square board (169 cells), there are, on average, less than
19041.667
moves available on each turn. 15000, maybe? That's more or less like
arimaa.
christian
> Done!
Me too: http://mindsports.nl/index.php/the-pit/xodd-624
>
> I'm also planning to write a standalone Xodd ruleset, in case it turns
> out to be better than Yodd.
Which may very well be the case, in fact I'm almost sure :)
Of course the applet will keep up the group count and indicate legal
cells to play after the player has selected a color to play. Ideally
it will exclude cells on the first move that have no legal follow up
move. I'll take that up with Ed because he's much better at the kind
of deductive reasoning needed for that. I'm sure he'll come up with a
smart implementation.
christian
> I'm sure he'll come up with a smart implementation.
Ok, I've made the square to point switch in Xodd.
Mirroring is a problem with hexhex bords and odd-sized square boards.
Black can start with a white stone in the center and next mirror
white's moves with reversed colors.
I know of another quintessential game with a similar (though not the
same) problem, namely Emergo.
Here the rule is simply that the first player may not start in the
center.
Unless there's a smarter solution, I suggest XYodd does the same.
christian
> On Aug 25, 3:15 pm, christian <christ...@mindsports.nl> wrote:
>
> > I'm sure he'll come up with a smart implementation.
>
consisting of a white and a black stone. It will double as a hot
button for selecting the color of the next move. The board then will
show all legal moves for that color. If a move is made, the board will
show all legal follow-up moves for _either_ color, indicating legal-
white, legal-black and legal-both using a white, black or grey marker.
How's that sound?
christian
> Mirroring is a problem with hexhex bords and odd-sized square boards.
> Black can start with a white stone in the center and next mirror
> white's moves with reversed colors.
>
Or maybe it's not a problem because white can answer with two black
stones, one at either side of the first one.
Or any two black adjacent stones for that matter: if perpendicular
black would have to mirror two white stones and thereby create one
white group, putting himself in a two aginst one disadvantage.
So maybe my previous post is bull ;-)
Nick Bentley
> Or maybe it's not a problem because white can answer with two black
> stones, one at either side of the first one.
> Or any two black adjacent stones for that matter: if perpendicular
> black would have to mirror two white stones and thereby create one
> white group, putting himself in a two aginst one disadvantage.
>
> So maybe my previous post is bull ;-)
not only that, but if the opponent connects another like-colored group
to the center group, Symmetry is broken. This game seems highly
resistant to mirroring for several reasons.
christian
Black would start with a white stone. This saddles white up with 'more
groups' and prevents a white pass, so white must play for one or three
groups. Adding two whites would be silly. Adding one white adjacent to
the first stone would require another withe one to the same group(or a
pass for the second stone), so that too would be silly.
So adding two black stones adjacent would appear to be the sensible
way to break symmetry: either the two stones are already symmetrically
placed, or black would have to mirror the move by placing two white
stones, leaving him with two groups and his opponent with one.
Nick Bentley
> way to break symmetry: either the two stones are already symmetrically
> placed, or black would have to mirror the move by placing two white
> stones, leaving him with two groups and his opponent with one.
I may be imagining things wrong, but if Black opens with a white stone
in the center, why wouldn't white play two white stones, both adjacent
to it, on his next turn? Having strength in the center seems super-
important for keeping your opponent's groups divided later in the game
(on small boards it's a certain win, in fact), and that move would
give White center dominance.
christian
> I may be imagining things wrong, but if Black opens with a white stone
> in the center, why wouldn't white play two white stones, both adjacent
> to it, on his next turn?
For one thing, black can add more white groups. You're playing for the
_least_ number of groups so your opponent tries to saddle you up with
more.
Nick Bentley
Right, but if you dominate the center with a group, all your groups
will, in the end, be much more connect-able than black's groups. I
wager that a center group makes them so much more connect-able that
Black's attempts to start more will be fruitless. In the the end you
only have to connect two of your white groups to the center group to
make up for it, and in the meantime Black will be seriously segmented.
When I play through the scenario on small boards, it's not contest.
White crushes Black. So if you're right, as the board size grows, not
only will this advantage have to attenuate, but reverse polarity. How
would that happen?
luigi
> Ok, I've made the square to point switch in Xodd.
Thank you!
> Mirroring is a problem with hexhex bords and odd-sized square boards.
> Black can start with a white stone in the center and next mirror
> white's moves with reversed colors.
Before I read any of the subsequent posts, this is the reply I got
ready when you mailed me on the matter:
"It occurs to me that the seed of asymmetry every Yodd/Xodd position
must have in the form of at least one odd group should be enough for
the losing player to prevent continuous mirroring play from his
opponent:
- Strategically, the losing player will try to use his extra group to
create a faster frame than the mirrored one from his opponent. When he
finally joins his groups, his opponent will still be one stone away
from doing so. In the scenario you suggest, And, from the point of
view of the non-mirroring player, there couldn't be a better cell for
the extra group to be placed than the center cell.
- Tactically, there are at least two kinds of moves which can't be
mirrored:
a) Occupying one cell and its symmetrical complement on a single
turn, with stones of the same color.
b) Playing next to the odd group, if it's in the center cell. If you
change parity by doing so, your opponent won't, and viceversa. This
way, the second move of the turn can't be mirrored.
This should be enough to discourage/break persistent mirroring play.
Let's see an example 3x3 Xodd game:
(coordinates as in chess; notation of moves is preceded by the Color
of the played stone)
1. Wb2, Wb1 Wc2;
2. Bb3 Ba2, Bc1;
3. Wa3, pass;
4. pass, and White wins with 2 groups to 3.
Or, even better:
1. Wb2, Wb1, Wb3;
and Black can't mirror."
luigi
I agree with Nick here. In Yodd, a group in the center is not likely
to become a burden at all. It's the ideal place from which to spread
influence across the board and connect your groups in the future. In
Xodd it could be a bit different, because isolating groups is easier,
but anyway.
christian
> I agree with Nick here. In Yodd, a group in the center is not likely
> to become a burden at all. It's the ideal place from which to spread
> influence across the board and connect your groups in the future. In
> Xodd it could be a bit different, because isolating groups is easier,
> but anyway.
Boy is this game difficult to grasp. It really requires an applet and
a fair couple of games to get any idea at all. No telling what comes
out of it in the end in terms of strategy and tactics, except that it
will be inticate. I remember coming across a very early havannah
position once, some 25 years later, and the most striking thing was
that it didn't look like a havannah position at all, despite the fact
that havannah can have widely different types of positions. It's the
same as a Chess player recognizing a position as silly on sight, you
immediately _see_ something is wrong. It illustrated how thinking
about strategy had solidified over the years.
luigi
Marcos D
> effectively 'cutting out the middle game'.
Thanks :)
> The reason for the restriction on same sized areas (I take that to be
> the number of enclosed squares) is unclear to me.
There is no a priori reason (as in many abstract games). The restriction came up during the design process.
Regards,
Marcos