Perseus Patrol devlog #7: Federation control and alliances
Previously talked about the major concepts I am thinking to merge for a compelling, lived in world:
- game theory and crisis bargaining,
- escalation ladder.
Theory is nice and needed, but practice makes it enjoyable for the players. To bring it even further into reality, let’s simulate logically what happens with the first decision each colony can make:
- stay/join the federation,
- leave to be independent.
To complete the mix, what is leading up to a decision like that?
- Motivation differences create friction,
- friction escalates to thresholds,
- thresholds trigger change in state,
- enough state change and it results in alliance broken or strengthened.
The colonies
To entice a colony to be part of something larger, the “larger” has to sooth the colony’s needs. In a world with four major action area (military, engineering, science and diplomacy) the needs are corresponding to the same areas too:
- to be safe, e.g. its or its allies military power outweights the dangers in the local area, to use force projection to secure interests,
- to grow, e.g. able to feed their citizens, house them, economical prosperity,
- to learn, e.g. new techniques and scientific achievements in collaboration,
- and to influence, e.g. culturally or politically exert concessions, change minds and hearths, and get early warnings of adversaries.
If any of these are fulfilled by an alliance, the colony is warming up to that faction or getting more supportive of it even.
The Federation
Treating an alliance as a living being, it grows and shrinks in a natural way by their elements interacting with each other. This means that the “larger” entity is only a self-determinism of each actor independently and exists until at least one colony is supporting the entity.
Let’s simplify into positive and negative factors how a single colony’s relationship to the alliance changes. The actions are not exhaustive, just representative:
- Military angle:
- positive:
- policing without civilian losses
- eliminating outlaws
- Steeldome visits delivering supplies
- negative:
- mobilizing military forces without reason
- attacking its interests
- blockade of its trade
- positive:
- Engineering:
- positive:
- deploy repair crews
- build new facilities
- construct civilian ships
- negative:
- remove repair crews without finishing
- deconstruct facilities for their raw materials
- deconstruct civilian ships for their raw materials
- positive:
- Science:
- positive:
- share technology
- improve facilities
- negative:
- scan owned facilities
- scan force deployments
- positive:
- Diplomacy:
- positive:
- share culture
- networking and schmoozing
- share intelligence
- negative:
- bribing
- spying
- stealing
- assassination
- positive:
The main idea is that these actions can be taken by any colony or core world against any other, influencing their relationships. The core world of an alliance holds it together and when its influence is failing to reach a certain threshold, the influenced will secede from the alliance. If another alliance builds up enough influence, it may join them and this cycle can be reversed by the player with effort put into it.
The key will be to design a system that allows for a sandbox like behavior yet supports the narrative of a failing but originally good alliance’s last “hours”.
Could allow even the player to secede from the Federation and join another faction or start his/her own!
Case studies
Learning from the greats will definitely help in cracking the alliance model.
Civilizations
https://civilization.fandom.com/wiki/Alliance_(Civ6)
In this series, alliance mechanics changed a lot over the episodes, yet the underlying phylosophy I think was always the same: accrue points based on trade and similarity, lose with military movements. The game is centered around the “can your civilization stand the test of time” theme which pushes the alliance system to a secondary support role in my eyes. Most playthroughs do not care about maintaining any alliance, more like pushing down the confrontations until you are strong enough to wither any war scenarios and find win conditions. Still, using benefits from this system could allow for the carrot part of the carrot and stick based relationship building.
In my game, the parallel would be if the civilization you control is composed of an alliance of the settlements.
Imperium Galactica II
Even earlier model, very similar to how Civilizations handle the relationships between civs, including the spying and diplomacy. Most probably it was inspired by earlier Civ instalments.
While great for its time, and still stands to have fun playing with it nowadays, the systems are too simplistic for my game. When the theme is centered around complex group behaviors, comparing powers and giving a basic behavior of the sides won’t cut it.
Hearts of Iron
Haven’t played it, but read about it only. Seems very player decision centric approach, where the AI mimics similarly the player decisions. It is a great game but not the direction I am looking for.
Maybe I am going wrong about this research approach. Most RTS are designed like a god-game, the player is the ultimate decision maker where one city deciding allegiance is very simple: it is only for challenging the player to handle some resource constraint:
- Civ’s morale and growth rate,
- IG II’s empire goals,
- HoI’s military powers and intelligence,
etc.
In Perseus Patrol, the player is the ultimate decision maker for the ship only as its captain. These mechanics will be interesting for ship morale and internal turmoils to give depth to dilemmas and player decisions. However too shallow to model the inter-galactic politics after this. A new approach is needed.
Game of Life
A simple model we learn early in programming. Only two rules and you can create a non-stable system where for humans, it seems like a living noise.
Alliance dynamics
Learning of how scholars think of modeling alliance and coalition behaviors could be the unlock I am looking for.
https://www.sciencedirect.com/science/article/pii/S0972860016301621
https://webs-deim.urv.cat/~juanalberto.rodriguez/Tesis-Ismael-Gonzalez-Yero.pdf
http://slantchev.ucsd.edu/courses/pdf/ramsay-fang-alliances6.5.pdf
Quick scan of these documents show one insight for me: alliances can be modelled as cluster of points in a graph.
This matches perfectly with the celestial model I built, each colony is a vertex of that graph, the relationship between the colonies are the adjacency strengths. Using these, I could build an algorithm that defines the alliances forming, with some threshold crossed, a new alliance can be formed or an existing one dissolved.
A* route calculation, water flow algorithm
Thinking of an alliance as a concept radiating from its core world, each colony that belongs to the alliance should have a strong enough route. This route can be calculated like water flowing on terrain. Start from the core world and expand the wave, weakening it with each relationship it has to “flow” through. If there are more than one route, like core world is close enough to multiple colonies, then a single colony’s allegiance can be calculated as a sum of core world direct effect and the nearby other colony’s effect (indirectly from core world).
This can help model a dynamic effect of an alliance growing or shrinking with threshold values deciding. If a colony shrinks below set value to a e.g. the Federation, it may secede into independent status. If its allegiance grows to any alliance above a threshold, it may join to that alliance. Or in the case of independent colonies growing allegiance to each other, form a new alliance themselves.
Simplest possible model: solidarity
https://hackaday.com/2025/06/25/simulating-empires-with-procedurally-generated-history/
An interesting idea presented by Fractal Philosophy, who found Peter Turchin’s concept:
In my crude words, high solidarity between members creates the empire/alliance, over time the solidarity wanes leading to a breakdown of the empire. And the cycle starts again.
This concept worth more deep dive as could give a simple enough to understand yet feeling living enough to be interesting backdrop for the alliance system in Perseus Patrol.
Summary
In this article I try to expand on the concept of alliances in Perseus Patrol and got to the conclusion:
- Previous games are not focusing enough on alliance dynamics to get inspiration or model my solution over theirs.
- Instead foundational research needs to be learned and discovered the alliance dynamics from game theory as a good direction.
- Modeling expansion and shrinking of alliances can be compared to water flowing on terrain and use similar routing and wave algorithms to simulate.
- Introducing threshold values enable the simulation to dissolve or create alliances dynamically.
- Using the solidarity concept can simplify and present the player with a number to drive with decisions.
Next steps will be very complicated for this:
- Read in depth the found research material,
- Design the concept with representation (UI/UX) in mind,
- Work out a simulation algorithm for alliances,
- And procedurally generate the starting point of the story.
Next, I will dive into the research system implemented and being tested for Steeldome’s improvements.
