PBEM v93 n05 (10Sep93)

======================================================================
Atlantis 1.0: A Retrospective                             Greg Lindahl
======================================================================

Russell Wallace recently sent out the rules for Atlantis 2.0, his
second generation version of Atlantis. Version 1.0 was a very simple
implementation of a strategic fantasy wargame: no economy other than
trade between players, movement limited to one map square per turn,
and no specialization of player factions.  I'd like to take a few
moments to look back at version 1, and talk about a feature of the
game that I don't think was intended.

It has less to do with the rules, actually, or the game mechanics, than
with the world on which the game took place.  I call the feature I'm
going to talk about "instability". A little instability is good for
the soul, but a lot is not.

The boardgame Risk suffers from a similar instability. In Risk, the
goal is to conquer the world. Every turn, players receive income
depending on how much territory they hold. In addition, every turn
each player (usually) receives a card, and sets of 3 cards can be
turned in for bonus armies.

So far, it's a very simple game. The interesting situation arises
because the number of bonus armies rises with each set turned in.
Since the other forms of income in the game are roughly constant,
eventually the cards become the most important factor. In addition,
when a player is wiped out, the conquerer receives their cards, and
can immediately turn in a set for additional armies. These rules
combine to mean that the game generally ends with one player turning
in cards, wiping out the weakest player, turning in more cards, wiping out
the next weakest player, and so on until they are the only player
left.

Is this good? Well, in the case of Risk, it is a good thing, because
it means that the game ends after a limited amount of play, and the
players don't have to stay up all night to finish a game. Most
boardgames have similar mechanisms that force an end to play.

So what does this have to do with Atlantis? Well, in Atlantis, the
players who controlled the forests and mountains were the only people
who could produce, among other things, weapons. Weapons were necessary
if you wanted to tax the peasants, and taxing the peasants was a very,
very lucrative pastime. It was so lucrative that it created a similar
unstable situation. The Atlantis world was divided up into small
islands, each of which had 5-10 mountains and forests. Once a group of
players had control of one of these mountain or forest squares, then
interesting things began to happen.

Adopting a "first strike" strategy was very important. There was no
advantage for the defender in combat, other than that the attacker had
to move to the defender was, wasting possible study time, and the
spread-out nature of defensive operations thus meant that attacking
had a large advantage.  The combat system was a "winner take all"
system when the sides were asymmetrical -- and attacking armies tend
to be more concentrated than defensive armies.

Second, some the game mechanics were quite useless in the starting
game. You could use wood to build bows, or boats. Bows meant you could
tax and raise a huge army. A boat was, well, a boat. You could sail to
other islands and run into huge armies. Thus, building boats was a
waste of time until you controlled your entire island.

Finally, the game was turned into an instant kill-or-be-killed contest.
If a group of players suffered a setback, it was likely that they were
totally hosed and were best off dropping out and restarting elsewhere.
This process would have continued until one group was in charge of
each island.

What would have happened next? Well, we may never know since Russell
isn't adding any new players and it seems that quite a few of the old
players have dropped out, but odds are that the various islands would
start invading each other -- and it's possible that the older islands
might have such a superiority in number of weapons that they could
take out a second island. And a third. And so on, producing a global
runaway.

What could have made this situation more stable? Most open-ended games
strive to set up a situation where the system tends toward
equilibrium, not towards total victory after a few battles. Putting
limitations on taxation income would have done that in Atlantis. Not
only was the money produced per turn per man too large, but the total
amount of taxable money available was so large that even at this
stage, where the first island is mostly held by one group, over 90% of
the tax money is still waiting to be collected. The rules do allow for
an equilibrium situation as the population balances against the food
supply, but the initial populations were so low that this equilibrium
was never reached.

Second, the terrain could be more mixed, such that each island
contains many possible production centers instead of just a few groups.
This might create a situation with balance-of-power dynamics... or
not. Finally, the combat system could be revamped.

Version 2 is very different from Version 1. Among other changes,
weapons are no longer needed to tax, so control of weapons production
will no longer be the only way to success. Russell has stated that the
terrain will be much more mixed than Version 1. But the combat system
has not changed much. My prediction is that Version 2 will turn into a
runaway, much like Version 1, and I'll probably end up dead, much like
Version 1. However, just like Version 1, I'll have fun dying, and
that's what gaming is all about, after all.