Re: Galaxy: More on communication and alliances)

From: hall@kronos.arc.nasa.gov (Michael Hall)

Date: Thu, 23 Jul 1992 17:24:04 +0000

Now crossposted to rec.games.design...

In article <1992Jul20.102311.11903@uni-paderborn.de> jerry@uni-paderborn.de (Gerald Siek) writes:
>rwallace@unix1.tcd.ie (russell wallace) writes:
>>kanderso@reed.edu (Karl Anderson) writes:
>
>>>                   [ ... ]                      The best solution
>>>that I can think of [to the problem of massive alliance victories]
>>>is to eliminate communication between players.
>>>Alternately, messages could only be sent via ship, that is, a player
>>>would attach a message to a ship and the recipient would only
>>>recieve it when that ship reached one of its worlds, or perhaps a
>>>group of it's ships.
>
>Too complicated!   In fact I'd like so have *more* communication
>facilities in Galaxy, for example anonymous broadcast messages.
>
>>My favoured solution to the problem of huge alliances steamrollering
>>everyone is,
>>*** NOT TO ALLOW ALLIANCE VICTORIES WITH FOURTEEN *!@!#? PLAYERS!!! ***
>
>That's it.   Once a strong alliance is established the other players
>have but three choices:
>	* Fight alone (and lose the game)
>	* Join the alliance (and "win" together with 13 other players)
>	* Try to form a counteralliance (which is very difficult and
>	  mostly fails because of the lazyness of the other non-organized
>	  players)
>All three situations happend to me and I found none of them very exicting.
>So the maximum alliance size that could claim a victory should be limited
>on 1-3 races, depending on the game size.

As one of the 14 victorious players, I faced the above choices, and
certainly it would have been suicide for me to choose anything but
to join the alliance.

If victory were permitted for only one race, it would in theory pretty
much fix this problem by causing the alliance to fragment and fight itself.
(You might still get steamrollered by the alliance early on, but the
alliance would be at least weakened because of paranoia about the
inevitable backstabbing.)

In practice, the game would have gone on for a *long* time to get down
to one victorious player and probably would have died under its own weight.
I had one of the two largest empires in the game, which was but a small
fraction of the galaxy, and it was already hell to manage, and I had access
to a lot of empire management software that other players did not.  I was
ready to end the game, as I think most were, simply because I couldn't bear
the thought of managing an even bigger empire.  I had 3 dozen worlds, 6 dozen
ship types, and the killer: 300+ ship groups, which had to be sorted through
when writing orders.  Now some of that is my "fault": I should have designed
fewer ship types, which would have decreased the number of ship groups,
sacrificing efficiency for sanity.  But mostly it is Galaxy's inherent
problem.

To solve this problem, one could increase the cost of ships.  Russell
already did this once.  Nearly equivalently, one could decrease the size
of the worlds.  If this is done in the extreme, then maybe, just maybe,
large numbers of worlds could be manageable.

Abstraction would be an alternate solution.  If the player could
group groups into fleets, this would help;  I often had to issue 20+
orders just to get a war fleet to move from one world to the next, and I
had to sort through which of the groups at the first world were actually
defense fleets intended to remain in place.  Another possibility is that
one could simply abstract away ships altogether: players could produce
"goo" of a certain amount and tech level.  This goo could be sent to
other worlds, where it would mix with goo belonging to the same player
there.  This might simply average out the tech levels, or you could
use the mean and variance (keeping track of the summation of the goo
techs and the goo techs squared).  Well, this might be pretty weird,
and might not work, but I'm just trying to avoid having hundreds of ship
groups.

I understand that Russell is working on another space empire game, so I'm
throwing out these ideas with that in mind.

--
Michael Hall            \
NASA Ames Research Center\
hall@ptolemy.arc.nasa.gov \

References

galaxy communication and alliances
kanderso@reed.edu (Karl Anderson)
Fri, 17 Jul 1992 00:25:21 +0000
Re: galaxy communication and alliances
rwallace@unix1.tcd.ie (russell wallace)
Fri, 17 Jul 1992 10:23:23 +0000
Galaxy: More on communication and alliances)
jerry@uni-paderborn.de (Gerald Siek)
Mon, 20 Jul 1992 10:23:11 +0000

Referenced By

galaxy communication and alliances
kanderso@reed.edu (Karl Anderson)
Fri, 17 Jul 1992 00:25:21 +0000
Re: galaxy communication and alliances
rwallace@unix1.tcd.ie (russell wallace)
Fri, 17 Jul 1992 10:23:23 +0000
Galaxy: More on communication and alliances)
jerry@uni-paderborn.de (Gerald Siek)
Mon, 20 Jul 1992 10:23:11 +0000