Massively multiplayer games, and the tron dolequeue grows [was Games reflecting life ]

From: Emlyn (onetel) (
Date: Mon Jul 17 2000 - 09:01:56 MDT

----- Original Message -----
From: Robert J. Bradbury <>
To: <>
Sent: Monday, July 17, 2000 6:47 PM
Subject: Re: Games reflecting life [was Re: No AI for Nano/No Nano

> On Mon, 17 Jul 2000, Emlyn (onetel) wrote:
> > It sounds really cool; similiar to Utopia. Looks like they stole a time
> > machine, came forward and saw Utopia, then went back in time and made
> > own copy. Or else Utopia is a rippoff.
> I suspect that Utopia is a son-of-son-of-son-of-Empire.


> > Utopia is not a great game. It's fairly mediocre in my reckoning; it's
> > of clunky fixes to try to retain game balance, each such resulting in
> > changes so that each round, the rules have been changed to stop an
> > unbalanced tactic from the round before, and which invariably introduce
> > their own new problems.
> Empire went through the same process as people discovered loopholes
> and they were blocked which then made more complex loopholes.

I'm sure the more rabid dynamists among us would find this an appealing
metaphor; I agree.

> > But it is 60,000 players, and an entirely simplistic abstract
> > model (if you want to think in terms of a map, it is as if you were
> > to every other player), that makes the game incredible. It is just soooo
> > many people; such a game is qualitatively different from any small scale
> > game. Strategies that work for 2 adjacent opponents, or 4, or 20, do not
> > work for 60,000; new strategies emerge which are innapropriate in games
> > smaller numbers of players.
> If you are adjacent to every other player, then it is N-dimensional
> (which is an interesting concept). Empire was entirely reality based
> (2-dimensional), which made the formation and keeping of treaties an
> interesting exercise.

I have found in multiplayer wargames with more than a few players (5 plus?),
that this kind of N-dimensionality bogs the game down mightily. Usually
there is some concept of "attack", which allows you to damage an opponent
and maybe improve your own position directly, but leaves you open to
"attack" from others. With many players, everyone is afraid to "go in",
because they are almost guaranteed to cop more than one attack on them in

This effect seemed to be fairly linear; the more players, the more
conservative (and endlessly boring) the play. Many strategy games employ
special rules to limit the scope of each player (2d maps are a good
example), to allow scalability.

Except that with 60,000 players, the effect almost dissapears. There are so
many people, that hardly anyone has their eye on you; you often fight an
opponent in splendid insignificance. Except someone does notice your
temporary weak state sometimes, and sometimes that someone is a member of an
online alliance that could be hundreds of people strong, and you are in
deep, deep trouble. Or maybe that happens to your enemy.

I think, as I said in a previous post, that there is tremendous scope for
social experiment by testing ideas in online games. You can attract
thousands of volunteers free! These games usually include toy
economic/political models. Why not try real ones? We might even learn

Something else that's been bugging me is a comment Robert made recently
about games driving electronic/neural interface development. I'm wondering
whether AI research has been boosted in at least some limited ways by the
demand for good game AIs, and whether we are seeing significant decrease in
that driver now that isolated people can find other actual humans (always
more interesting opponents) to play against. Anyone know whether there
effect is happening?


This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:34:48 MDT