Borders and Territory in Civilization-style Games

This post is the beginning in a series about borders and territory in Civilization-style games.  It is meant to familiarize the reader with the various borders-systems of the Civilization series and sum up how borders are generally integrated into the modern turn-based strategy game.

• Civilization II (CIV2, 1996)
• Sid Meier's Alpha Centauri (SMAC, 1999)
• Civilization III (CIV3, 2001)
• Civilization IV (CIV4, 2005)
• Civilization V (CIV5, 2010)

Civilization II

There was no borders system for Civilization II.  The only restriction on city placement were the minimum-distance constraint, and the only way to control AI movement through your land was via Zones of Control from your own units.  Cities could work any tile not being worked by any other city.

Alpha Centauri

Borders were first introduced to turn-based civilization-style games in Sid Meier's Alpha Centauri (SMAC).  Before that time, clear map-level delineations between players did not exist.  In SMAC, borders extended from a faction's bases to a certain, large, radius when the base was first founded.  When borders conflicted, the game used a "who's base is closer" style of adjudication to decide which faction controlled the territory.

Borders could be violated without going to war.  If you discovered another faction's units within your borders, you could demand that they leave, though you faced the risk of them declaring war over the request.  The AI was programmed to not ask for you to leave its borders unless it could see an offending unit.  Importantly, bases could not work tiles not within your borders, and I recall that bases could not be founded in other faction's territory as well.  However, I'm not completely certain about that.

Civilization III

Civilization III introduced the concept of culture related to borders.  Newly founded cities only expanded your borders 1 tile from the city square, and as differing culture levels were reached, the border from the city expanded to the "fat cross" familiar to fans of Civilization 4.  Border conflicts were adjudicated by checking the total value applied to each tile every turn, which each city in range applying its culture value to the tile each turn.  A side effect of this system was that borders were not particularly "sticky" and could change quickly based on founding new cities.

Borders could still be violated without war.  Right of passage agreements would let players use each other's roads and railways, though it also lead to the concept of "RoP Rape" where a player could sign an RoP with an AI, move their units into position around each AI city, and then declare war.  Their units would then be in position to gut the AI empire before any effective response could be mobilized.

The idea of culture-flipping was also introduced.  A city could convert from one civilization to another if it was sufficiently pressured by opposing culture.

Finally, unlike Alpha Centauri, a player could see everything that happened within his borders.

Civilization IV

Civilization IV iterated on the CIV3 design, adding more culture buildings and filling some of the holes.  A player could only enter another player's borders in a state of war or with an open-borders agreement.  Additionally, when war was declared, units were automatically removed to prevent players sneak-attacking the AI.  Tiles and cities also accumulated culture points over the course of the game.  As a consequence, borders became more "sticky" and harder to change as the game progressed.

Culture flips remained, as did the big-fat-cross (BFG).  Below, the city of Aachen has expanded to its 2nd ring borders.  Great Artists were also introduced, and could be consumed to add a large amount of culture in a city to dramatically expand borders.  Early in the game, a properly used GA could overwhelm the established borders of another player.  However, later in the game, due to the accumulation of culture on tiles, the Great Artist bomb became less effective.

2nd ring borders in CIV4

Civilization V

The culture/borders/territory design was mostly rewritten for Civilization 5.  Cities would still gain culture and expand borders based on that culture but tiles would be gained individually.  Tiles could also be gained by using gold to "rush buy" the tile.  However, tiles gained through culture would be gained based on an algorithm outside of the player's control.  My own experience with the tile-picker is that it works reasonably well, but there are certainly instances where I'd prefer a different tile from the one it picked.

City view in Civ5 with predicted tile expansion.  The picker prioritizes luxury resources, but here is torn between choosing the dye or the incense tile.

In CIV5, a player cannot flip tiles from another player without either conquering the offending city or using a Great Artist to culture bomb the tile.  Tiles also did not accumulate culture.  Instead they are assigned to a particular city (either the city whose culture claimed it, or the city working the tile).  Tiles and borders then shift with possession of the city.  This was a dramatic departure from CIV4, where newly conquered territory often needed an influx of culture just to establish a workable border, but was presaged in a CIV III WW2 scenario.  In that scenario, all culture added to a city was given to the current owner.

The same city after the 1st border expansion.

Today

Borders come from culture, generated in cities.  We expect borders to be respected by the AI and other players.  We expect borders to be "sticky": once they're established only dramatic changes can move them.  Effectively, we expect the behavior of borders to mimic those of a real life, sovereign nation state.

Most interesting to me, however, are that borders do not reflect the map terrain. Borders reflect the position of cities and their culture.  Cultural borders do not care whether or not a tile is a plains, mountain, or ocean.  For games which as their basis ask players to plan and grow around the terrain of the map, this seems a strange omission.

In the next post, I'll talk about the differences in game-play between dynamic and static borders.  I plan to take examples from the Civilization series as a whole and a number of Paradox and Creative Assembly titles.

EitB v9 Released

I got everything packed up and ready to go last night.  It's now sitting in moderation at ModDB, but once it's cleared it will be available here:



Until that clears you can grab the build from my dropbox at this link.

The cumulative changelog for EitB v9 can be found in PDF form from my dropbox here.

What Makes a Good Trait?

Every civilization is lead by a leader.  The leader has some number of traits, which change the base rules for that player.  Traits vary in power. Some civilizations have access to more than one leader, and a civilization's various leaders will have differing traits.  Most commonly, a leader will have 2 traits.

Leader traits in Civilization are so effective because they can completely change how you approach the game.  They help diversify the number of competitive strategies, and add a healthy variance to the game.  Therefore, when I am designing traits, those are the sorts of effects I want to create. However, not all traits are created equal.  The financial trait is causing me no end of headaches in developing Erebus in the Balance.

The financial trait doubles the speed of markets (an early building that gives gold) and moneychangers (a mid-game building which boosts gold production with a modifier).  Additionally, in every tile, if the tile produces 2 or more commerce, it will produce an additional commerce.  2 commerce tiles are fairly common in CIV, so the end result is that Financial produces a 50% increase in commerce for many tiles.

Ironically though, outside of the balance concerns, financial would otherwise be a model trait.  It clearly distinguishes itself from other traits by providing a type of bonus that no other trait does (a direct tile yield change).  It changes a player's game plan dramatically, and encourages a player to plan around leveraging the trait to maximize its effect.

Organized is a good counterexample to financial.  Organized doubles the speed of courthouses (reduce city maintenance by 50%) and lighthouses (+1 food from water tiles), and reduces civic upkeep by 50%.

It is a solid trait with a reasonable level of power, but isn't very exciting.  Its effects are primarily under the hood.  The player knows that his empire is more efficient, but he's not doing anything much different from his standard gameplan.  In other words, the player does not need to behave differently to leverage his advantage from being organized, beyond simply stepping up the time-frame of all of his plans.

Financial poses the larger problem to me.  It is a good trait, encourages a different style of play, but is so strong that it distorts the leader trait choices in multiplayer.  In general, the games on RBCIV can be loosely classified into two groups: games where financial leaders are banned and games where they are allowed.  In games where they are allowed, financial leaders dominate the picks.  While organized isn't a model trait, at least it doesn't destabilize the leader-pick meta-game.  Considering this, I realize that there was an important point I left off my initial list.

A "good" trait:

• Creates space for new and competitive strategies
• Encourages players to play to their trait's strength to unlock its full potential
• Is not a must-have for every possible competitive strategy

EitB Dev Journal 3: Where we are and where we're headed

I'm going to publish my 5th build of EitB later today as soon as it finishes uploading to dropbox.

There are always competing pressures on release dates.  I always feel that there's more that I can do, and I want the project to be perfect.  At a certain point however, I need to just publish and let the feedback roll in.  That feedback is critical for further improvements.

Particular rough spots that I feel need to be looked at:

1. Nox Noctis
2. Invisibility in General
3. Council of Esus spells
4. Religions unit pricing and strength

Then of course, there are the usual rough spots that need to be ironed out as we go along:

1. Interface quirks
2. Documentation
3. How does the AI handle it?

There's another point that I need to address in the near future.  How do I describe all the changes that have been made to the mod concisely and in such a way that players are not lost at sea in an ocean of changes?  I think that will be the hardest thing for me.

For now though, I wait with a bit of nervousness to see what people think of it.

EitB Dev Journal 2: Eras

Fall From Heaven was originally split into three eras: Ancient, Classical, and Medieval.  Eras are used to set the starting techs for non-ancient starts and also delineate the technology tree into distinct sections. These eras, however, are too broad.  

EitB then, will have 5 eras, tentatively named:

1. Thaw (Worker Techs including Bronze Working and Archery)
2. Discovery (All techs that enable T2 units and Smelting)
3. Expansion (All between T2 and T3)
4. Innovation (All techs that enable T3 units and Arcane Lore)
5. Mastery (Everything else)

Start Era Options; System Use Eras are used for state religions.

I plan to use these new eras to modify the coloring on the tech tree to make a visible distinction between the technology tiers, building on the Heat Map mod I did last year (that colored techs in the tree based on their cost to research).

EitB Dev Journal 1: The Council of Esus

The council of Esus presents something of a conundrum to the player and designer in the original Fall from Heaven (FfH).  It function as a religion, but has no temples nor religious heroes.  The council is concepted as the religion of the shadows, of what is hidden.  So how do we represent that in game?

When you follow a state religion, you gain certain benefits from the religion's base attributes, and from your civics.  If you're not following any state religion, you gain all the base attribute bonuses, but none of the civic bonuses.  When you're running the council of Esus in EitB, you'll be able to get the best of both worlds.  All religions will continue to contribute their base bonus, and your civics bonuses will be tied to the Council of Esus.

The Order, like all the other religions, turns off the base bonuses from having CoE in all of your cities.

The Council of Esus, however, can gain the benefits of having the order in your cities!

Theme in Fall From Heaven 2

As part of the design work for EitB, I found myself asking the question: what is FfH about?

If you had asked me that question originally, before I had started working on the tech tree, I would have told you that Fall From Heaven was about specialization; that there are many different paths to victory, but they each require a commitment from the player to a particular path.  Additionally, I would have told you that the areas in which you could specialize were:

• Infantry
• Archers
• Cavalry
• Recon Units
• Wizards
• Priests and Disciple Units

I was wrong.

I would say that FfH is about Magic and Religion.  A large portion of the technologies on the tech tree are about either Magic or Religion.  The economic and mundane part of the tech tree feels scatterbrained and somewhat unfocused, but the Magic and Religion parts feel well designed and thoughtful.  Each technology in those areas feels like it fits.

This observation has played out again and again in the FfH PBEMs that I've spectated.  Conflicts between players have never been decided by whether or not a player decided to go archers or axemen, but rather by spell choice and usage, priests, and summons.  All but the very best mundane units are ignored, and most units do not ever have a reasonable opportunity to see play in any particular game.

What does this mean for design in EitB?

I was running into a lot of difficulty trying to differentiate the various Mundane specializations from each other.  Now I realize that what will probably be best, is if I redesign the mundane parts of the tree to emphasize the religious and magic lines.  Mundane units in EitB will be the supporting units behind your religious and magic specializations, as they are in FfH, but they won't be tied to a tech tree that pretends that the Mundanes are the main show.