Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics

Dynamic vs static territories

A topic by Calandiel created Apr 30, 2020 Views: 1,105 Replies: 11
Viewing posts 1 to 9
(1 edit) (+1)

Demian and I were rediscussing the best approach to territories. As some of you may know, we want borders to be more fluid than those in most strategy games (especially at the beginning of the game). As such, we talked about dynamic provinces that can grow and shrink. There are, however, multiple ways to implement them that we are considering. We'd like to hear your thoughts on the subject (and possibly your own ideas as to how the system should be approached).

To introduce both models, we'll need some shared assumptions. The map is divided into tiles (hexes). Those hexes are then aggregated into territories. Each territory contains at most a single settlement (you can think of settlements as cities in Civilization. They're the main hub of population in any given territory). Settlements can then exert control over other territories forming more complex structures. Here, we'll call these structures "provinces". These aggregations will be dynamic in either model.

Our two models are what I'll call a "dynamic" and a "static" model. In the latter, territories are predefined, like they are in, say, Endless Legends or Humankind. Whereas in the "dynamic" model territories would gain and drop tiles depending on the hold they have on them. 

Now, before you immediately say "Well, that's easy, I want more dynamic stuff!", the choice isn't as simple as that. Static regions have a lot more going for them than may meet the eye. 

In no particular order, they simplify pathfinding a lot, allowing for much more efficient path caching and direct usage of algorithms superior to A* (such as HPA*). They would also allow us to store animal pops directly on them, removing the need to recalculate accessible animal populations for each territory with each monthly tick. 

They also prevent issues with possibly excessive border gore in situations where a decaying coastal settlement could get squished into a thin line by two neighboring coastal settlements. There are some countermeasures we could take against that but they generally increase code complexity (which is *not* a good thing in an already complex simulation).

And, lastly, they're much easier to implement and maintain. Though, that's a bit less of a consideration (but it may open using them to bootstrap the simulation with the goal of replacing them by a dynamic model later on).

I think their issues are much easier to spot(rigidity, lack of adaptation to evolving terrain so they'd need to be based on terrain profile more than biomes, etc). If you can think of other issues or advantages (for either model), or a different model entirely, let us know. Otherwise, share your thoughts on the subject below.

With the static model, could there be an option to still be able to edit the borders of a province. Say there are 3 provinces that share a border, but it would make more sense to organize the borders of the provinces differently. Could this be possible with the static model? The idea is that 90% of provinces would not change and remain as they were when they were made, but in cases where you want to tweak the borders you can and the game will know to update the pathfinding with the new "static" layout.

It's not just pathfinding but also any other type of cache we may come up with (as well as them being the same as regions used to store animal data). If they could be redrawn, we'd lose out on these optimizations.

I still personally think that the benefit of dynamic territories outweigh potential costs. Subdivision of land is a thought invention, and subdivisions need to be able to change in order to reflect the ways sapients view their society and the value of local regions.  If not, how will situations like the arbitrary borders of European colonies in Africa versus the geography-based borders of Europe be represented?

(1 edit)

What I want to know is if if territories will be associated with geographical regions, or if they are randomly made. (Think rivers and mountains define borders for the first, and the post-WWI borders that still trouble Africa and the middle east today). Are the regions limited to one city? Are other factions able to build a city in that territory or region and cause conflict or cooperation based on Culture? If it is forced to one city, then its static and easy to manage. If other cities  can be settled in that region from other cultures/factions then it becomes dynamic, which is more elaborate but more realistic. 

I like the Endless legends large territories for easier management of areas, but Civilizations is more immersive to me.

In a mix of the two, a faction would have an influence over a region and the more influence the more control over people and resources from that region. While other civilizations can settle or grow their territories (like in Civilization) into that region causing power struggles over resources. But if that region is foreign (like Egyptians in Norway), then they get negatives on them. It also hinders empires from just taking everything around them. Their influence will be too foreign or savage to the natives thus causing it to be weaker in farther away regions and recently conquered regions.  

I think Influence is a major part of territories as well. You can't control where people go, and so you receive people from other places, maybe unsettled wild-land nomads in the earlier start, but large amounts trade bringing in people from around the globe talking highly of their home in the later game. 

Are there similarities between the mountain people and the plains people for them to work together? Maybe they may be too different, and the mountain people  don't accept the ways of others in your empire and so it becomes more difficult to expand in that region. All these would have an effect of how easy it is for your empire to grow.  Do the surroundings like you and want to be part of you? Or do they resist you like the Gauls with Rome? 

I see total war games as static as they are unmoving territories, but are restricted by the fact that reference existing places, they don't create new maps like Civilization does, even in their Total War: Warhammer games they stick to what was already built.  I think their games make a good reference with public order and religious/cultural influence, but I don't think much else be applied to Songs of Eons d/t that SoE's philosophy is much closer to Civilizations style. 

In summary I like the static Territories for geographical reasons and managing simpler areas with multiple mechanics. Like city builders, or Total War games, Endless Legends.  But I prefer dynamic for territories and constantly changing worlds with influence and expansion mechanics, like Civilization games, Endless Space.

I love extra small coastal(sea/lake or river) city states, I like how they make a map a lot more alive and interesting. One of the largest reasons I'm waiting for Sote is being able to play as mayor of small city in large dynamic world with a lot of events that directly or indirectly influence you. And if territories will be quite large, states like Frankfurt, Lubeck, Hamburg, Bremen will lose their chance to be in game.

(6 edits)

I would say it depends on the economic system.

  1. If we have a model based of off the tiles, where each tiles contains resources that can be gathered through a resource gathering operation (RGO), which uses POPs for labour, who in turn consume resource both for sustenance and for production, we would expect a tile to both produce and consume. In accordance with the principles outlined in the famous article on how a game might model a realistic economy, it would be reasonable that the different RGOs or POPs would trade their respective produce with the others of the same tile, through a clearing house in each tile, so that the farmer sells directly to the carpenter next door. Intra-tile trade. However, no tile is an island, wherefore trade would be conducted with other tiles too. Inter-tile trade. Here, it seems reasonable that the tile as a whole would trade with another tile as a whole, clearing house to clearing house, with a POP of specialised traders living of any profit. This POP might be tiny and insignificant, or large and rich, depending on the level of trade. Since long distance trading is dangerous and expensive, trade links between tiles would only be established if it were profitable, which of course varies depending on supply and demand. A poor tile might only have a single trade link, to a local hub. The local hub might have a couple, including one to the local capital. That in turn will have a very strong trade link to the capital of the entire empire, which is the centre of the web that covers the region.
  2. Another solution is to forgo the principle of intra-tile trade, and make all tiles within a certain distance (a day's trek, so maybe six tiles away, depending on terrain?) from a population centre share a single clearing house, located in the population centre, which is treated as instantly available to all the tiles. One might call it cheating for the purpose of simplification. Only from this central clearing house can trade take place, which then is treated as "between province A and province B", rather than between tile a and b. This system mean losing the realistic intra-tile trade of the former solution, but on the other hand goes well with the principles of a gatekeeper state, and staple cities and staple ports of yore.
  3. Here, we see how an economic system might come into play with regards to provinces/territories. If tiles a), b), c), and d), are all sparsely populated, with little to trade but low-investment primary goods, they might not afford or want or imagine trading with anyone else than their closest town, A. The other towns, B and C, are too far away, or aren't really in the same economic zone. And so, it makes sense for tiles/villages a), b), c), and d), to be part of the same province as A, and to be separated from B and C. Likewise, it seems reasonable that that isolated mining colony in the middle of the desert would be part of the same province as the only town nearby, which is the only trade partner of the colony, its fundamental gateway to the rest of the world.
  4. On the other hand, it depends on what we want the territories to be. Military divisions? Administrative divisions? Legal divisions? Economic divisions? Natural divisions? The wiki isn't very clear about this, wherefore I feel a bit lost. While it seems reasonable that a mountain valley be a single province, a single "area", "region", or whatever one might call it, it also seems reasonable to assume that it will continue to be so, as long as people live there why would it grow or shrink, if the valley stays constant over centuries? There isn't really much need for dynamic provinces in such a case. On the other hand, if a horrible war devastates a once prosperous river valley, razing cities and sowing the fields with salt, one might very well imagine several small provinces merging to one or two poor and depopulated ones. If the territories are more of an administrative division, for example for taxation, the divisions would be based of off central planning, and change over time, like those of Rome in ancient times. In some cases borders between provinces might seem arbitrary, like the lines that separate Hoboken from Manhattan, while keeping Manhattan with Queens. There are a myriad similar cases around the world.
  5. I think in the end I would opt for a system that is dynamic, but still quite rigid. Population patterns, cultures, and economic relations change over time, but only very gradually in pre-industrial times.  Here in Sweden, the well-populated and powerful regions that formed the nucleus of our kingdom around the year 1000, are still today the centre of population and power, even though Sweden has grown, and new population centres have appeared. If the system in SotE allowed change, but only at certain intervals, and after a certain critical threshold had been achieved, it might be good compromise between realism and computing efficiency.

I must say that I would have greatly enjoyed the ability to change the constitution of certain duchies in CK2, depending on how the world changes...

As a word of clarification.

The article you linked is actually also the one on which our currently preferred economic model is based (in a modified form to allow good quality to be taken into account).

However, keeping a market on each tile is impossible due to RAM constraints. I mentioned cities/settlements in my post above for a reason. We can have about 20k entities storing pops per million tiles. These cities then have territories from which they can extract resources (through farming, mining, fishing, woodcutting, foraging, hunting and other means). Rural population is stored on a separate list on those entities.

I'm under impression some people see tiles as our equivalent of provinces from Paradox' games. They're not. Settlements are. Tiles act as dynamic pixels to define environment for settlements so that things like available resources and farming rates can be determined.

That is exactly the reason why I linked to it, it's a great model!

Thank you for that clarification on tiles and RAM constraints, I've been confused about that. Regardless of what we should dream of, the game needs to be playable. It seems then that point 2 is the valid one: groups of inhabited (only?) tiles will combine into a single economic unit, a province, in which production, consumption, and trade, is summed up and managed, while outside trade will be handled by the province as a whole.

The question remains though: how will the size and borders of these economic units be determined? Natural borders, like seas, rivers, mountain ranges and the likes, seems easy enough, but apart from that? Will it be travel distance, like the 100 départements of France? Cultural preference? Maybe economic needs, so that each province tries to include certain basic resources to be somewhat self-efficient? Or rather the opposite, so that they tend to specialise, "the province of the verdant fields", "the province of the deep woods", "the province of the great mines"?

If it is simply geography that defines the province, then it makes sense that it would be quite permanent. If on the other hand it is economy and culture, or even administration and other factors, certain events might change the provinces over time, behooving a more dynamic system. For example, to this day, Sweden's capital of Stockholm is divided in the middle by the prehistoric traditional provinces of Södermanland (south) and Uppland (north). But since the 17th century, there has a been an independent Stockholm county, and the two former provinces, now counties, have "lost" their common border, and been pushed south and north. 

Maybe a system in which "settlements" are static, but "territories" are not?

There would be set locations for settlements (representing the "ideal" location for settlement in that area), but the territory they control can change. So when you colonize new land, you get a new territory with only 1 tile which holds the settlement.

Example/Comparison picture for uncolonised land: (grey dots represent settlement locations)

Example for newly colonised land:

As it builds up, the settlement starts exerting its influence over neighboring tiles and further. It could even control tiles, that under static territories would belong to another territory.

Tiles that would belong to another territory are behind gray line(could, or could not be shown during actual gameplay):

When it reaches next "settlement location", it creates new settlement under new territory(separated with thinner line):

Now, the agents system could come into play, because first settlement agent may want to keep all the territory it previously had for itself, while the new settlement agent will consider tiles that are closer to it (but were claimed by first settlement earlier) rightfully its own, and on top of claiming new tiles for itself, it will also try to "retake"(purchase, ask overlord for intervention) some of the tiles from the first settlement.

There could be laws for that so player can decide whether he wishes for territory borders to have the same distance to both settlements, use "static" territory borders as reference or use the "first come, first serve" rule.

As a sidenote, I also wanted to ask: will movement be tile- or territory-based?

Sounds like a good compromise, especially since most great cities, or even settlements, were built on certain strategic locations, that did not change much over time, other than with technological advancements and the discovery of certain resources, like minerals, or sometimes with great environmental changes. Paris and London were settlements long before they became great capitals. Stockholm, the capital of Sweden, is a fairly new city, less than 800 years old. Before it was founded as a proper settlement, "its province" was divided by the "provinces" of other settlements. The same is true with many other cities around the world, such as Warsaw in Poland, which was a insignificant town before the royal court was transferred there from Kraków, or Madrid, which shares a similar history. 

The only question is how to do with those myriad cases where different settlements in the same region, "province", have switched places over the years, in terms of being the leading one. The history of ancient Mesopotamia and Egypt have endless cases of this, as has many other regions that have been civilised for many years. Should it be possible to change the "capital" of a province? 

In any case, I think we can all agree that it would be best with a system in which the simulation itself changes matters depending on endogenous variables, such as culture and economy. New trade routes, new technology, new influences, leading to an insignificant town to become the leader of the entire region.

To give an update, we went with static territories/provinces.

They'll be on average the size of irl Luxemburg (the country, not the city).