If any one topic is ‘the core’ of this System Hack, this would be it. Base building is the underlying gameplay loop of RimWorld and it is also a topic du jour in RPG circles, with the (admittedly poorly detailed) stronghold building from the D&D days of yore coming back into focus as more gamers want broader storyline opportunities. For our purposes, of course, if we’re making a colony sim we need to make a colony. But what exactly is the best way to do that?

Base building from my perspective is sandwiched between two examples which effectively bracket the space we have to work in. On the heavy end is RimWorld itself, a computer-assisted colony manager where everything is measured out in five foot squares and the player has complete power to place elements as they want them, as long as everything fits. On the light end is the new generation of stronghold building rules, most effectively typified by Free League’s games, notably Forbidden Lands and Twilight:2000. These games add a strong layer atop their roleplay frameworks, but the actual mechanical existence of a Forbidden Lands stronghold is merely a list of buildings with requirements and effects. We know the first item is too much, but we know the second is not enough. So what will base building in our System Hack actually look like?

Let’s define the bounds we’ve set forth a bit more clearly. At a minimum we must include the features of the base and what they do. At the maximum we can outline an actual physical layout of the base. The middle, where I’m aiming for, still defines layout without necessarily charting it precisely. Even if we don’t want the system to involve graph paper, we still need to define the base in terms of what physical spaces it contains, what those spaces themselves contain, and the relationships between these spaces. Let’s once again define a unit we can work with. We don’t want the unit to be exact space, per se, but it will represent space. Let’s call it a room. The important spatial facts about a room are that it’s rectangular and therefore has four exits, and that it comes in a few sizes. By defining rooms as ‘small’ or ‘large’ we can give them parameters, like how many stations they hold.

So what’s a ‘station’, then? Yes, I just picked this word, but it’s a good approximation of what most of the interactive furniture in RimWorld represents, a place where a single pawn (in RimWorld parlance) can work on a task. While beds don’t exactly operate like stations they are about the same size, so for the sake of our System Hack we can treat a bed as a single station as well. This also lets us, basically as soon as we introduced it, move away from room ‘size’ as a parameter; while small, medium, and large rooms may have 1, 2, and 4 stations respectively, there can also be halls which can scale upward to fit as many elements in one ‘room’ as you want, with scaling construction and energy costs as you expand further.

Stations are used for production, and we can roughly follow RimWorld on this, though specifics may be dependent on how we want to write our tech tree. Overall though, the basic idea will be the same, with raw materials being turned into finished goods. Each task can have its own difficulty (as detailed in the previous article), and we can even get into things like Effect Die limits depending on where the colony is in the tech tree. The idea there would be not to introduce complexity for the sake of complexity, but potentially limit the number of stations that are needed. The question that can be answered now is one of resources. RimWorld has a fair amount of granularity when it comes to resources: At a basic level for building materials there are wood, stone, and metal, but both stone and metal come in several different types, and metal has much more important properties than just its structural uses. RimWorld has already chosen to simplify; digging ‘compacted steel’ out of the ground sidesteps the fact that smelting is a hugely important and hugely complicated facet of moving from digging something out of the ground to getting useful metal out of it.

Resource extraction broadly is going to have to be abstracted to a significant level; in RimWorld much of the challenge inherent in gathering resources from the map is measured in distance. As your colony expands, you will use up the closest metal resources and chop down all the closest trees. If our game is not going to have a map, or at least not have a non-abstract map, it means the entire resource gathering cycle needs to be reimagined. There are ways to do this, to be certain, but it must take a completely different approach from RimWorld instead of abstracting or adopting the current map system.

There are, roughly speaking, two essential types of resources in RimWorld, renewable and non-renewable. Wood, and most foodstuffs (be they cultivated or hunted) are renewable, they will regrow or reappear on the map after a certain amount of time. Mineral deposits are non-renewable; once you mine all the steel and ‘compacted machinery’ on your map, it’s gone. In RimWorld, this ends up creating distinct strategies for getting important resources like steel and components in the early, middle, and late game stages. In the early game, mine what you can and buy the rest; in the middle game, focus on making renewable trade goods so you can buy enough steel and components, and maybe use caravans if you aren’t playing a nomadic settlement (the gravship in Odyssey does a lot to short-circuit this mid-game challenge). In the late game, you can make components and deep drill for steel. In thinking through this System Hack, we have, ironically, the late game elements set out, because fabrication benches and deep drills will be ‘stations’ that can be used to do tasks, like many other elements in the game. Resource gathering in the early game will take some more thought. We’re once again going to be using an abstracted map, but now that map is talking about what’s going on outside the base. I will fully concede that I want to work a hex map into this, but at least for early game, most resource gathering will be happening inside of what would be a hex. In the spirit of supporting that later hexploration, we can define six zones around the base: North, South, Northwest, Southwest, Northeast, and Southeast. Beyond directions there will be distance; I think two or three radii within a hex makes sense, in addition to a ‘periphery’ zone that will be immediately around the base. Of course, if we’re more concerned about distance, we should be more concerned about hauling as well. How long does it take to walk out to the other side of a hex? How much can you carry back? With these questions existing in the design, it suddenly introduces considerations about vehicles and pack animals, which have a much lighter touch in RimWorld.

While I will be covering the overworld and other planetary inhabitants in another article, it’s likely that your base’s connection to others will be dependent on geographic elements like roads. That could add a fascinating dimension to trade; having no road connection may reduce the appearance of hostile neighbors, but it also reduces the appearance of friendly ones, all but forcing you to venture away from the colony to trade. This also means the purpose of different entities on the overworld could vary; you could have other settlements, but also trading posts and maybe even towns and cities. RimWorld’s planetary economy is relatively simplistic, that is something we could add some more detail to if it makes the underlying gameplay loop more interesting. For now, though, considerations of the overworld impact base building because transportation will be part of the base. A road would be a large project, which implies that like many early civilizations, water travel could unlock accessibility another way. And this is of course on the earlier side of the tech tree, before air (and space?) travel becomes available as well.

---

Let’s summarize. The base buildings will have varying sizes, defined by ‘stations’ that can fit within. These can be standalone buildings or rooms within a building, made out of differing structural materials (wood, stone, metal). There will also be ‘outdoor’ stations as well; as we discussed in the article on tasks and work there should be space for agriculture in addition to other outdoor activities. These are going to be secondary, though modeled in a similar way and potentially keyed into other activities which are more related to the outdoor environment. That environment will extend beyond the actual space of the base, where resources like wood, stone, and metal will have to be extracted. Resource pressure is likely to be more of a present challenge than in RimWorld, where even basic resources like wood will require journeying further and further from the base and make considerations like vehicles and draft animals more relevant. These environments will tie into the rest of the overworld in a hex-based layout, but for the base itself access to either roads or water will be important for early game transportation.

In looking into how this will all play out, resources may benefit from a little simplification, although some segments may get more complexity. Wood and stone can likely just be wood and stone; the differences between, say, limestone, granite, and slate may be more than is necessary for this game. Instead, it might be more interesting to look into stone and masonry, separating out aggregate stone, flagstone, and more artificial items like brick and concrete. Food will likely have a lot more distinction as well, though at the back end may condense like in RimWorld, at least to meat, vegetable, and grain if not just meat and vegetable. What’s going to make the farming and hunting decisions more interesting though will be the overworld

Similarly, actually defining what the stations are is going to be heavily dependent on the tech tree. We likely don’t want as much differentiation as RimWorld; a few different workshop types is likely enough for most purposes, with some specificity getting baked in at the higher tech levels for things like transportation, robotics, and genetic engineering. Even then, we want to keep the list of stations from getting to sprawling or creating optionality that won’t itself lead to interesting decisions.

The hope is that base building will be full of interesting decisions. Even if the rules aren’t going to cover square by square space allocations, we will be creating a building layout and forcing choices about, say, separate buildings in a complex versus one large facility. Co-location of stations and the people working at them will invite social interactions, while considering the resource availability of the surrounding area will induce interesting decisions about time management and trade. As the extent of what’s contained in the base comes into focus, it’s clear at least to me that there is a strong possibility to make this game about tough decisions and strategic resource management, even if we don’t aim for the same level of specification as its digital predecessors.

Now it’s time to start putting more stakes in the ground. We can talk about stations and production, but to discuss what the stations are and what they do, we need to talk about the tech tree. Similarly, we can talk about foodstuffs, wood, stone, and metal, but we need to specify elements of the overworld to discuss what crops and animals are common where, or how much detail we want to go into beyond key metals like iron, copper, silver, and gold. The good news is that we can start sketching out ‘the lists’ which cover what you can build, grow, hunt, and trade that will eventually make up the backbone of the game itself.

Next time, we’ll look into overworld mechanics and start thinking about how the base relates to other settlements around it and how many of these relationships we want to mechanize in the game. After that, we’ll start with the lists, beginning with the tech tree and moving on to base elements, resources, and skills. Still a lot of specification to do, but there is a light at the end of the tunnel. When the lists are drafted, it’s time to carry all this back to our initial scaffold and maybe even begin playtesting! Let’s carry this energy through the end of the year and see if we can get a finished System Hack sometime in 2026.

Like what Cannibal Halfling Gaming is doing and want to help us bring games and gamers together? First, you can follow me @levelonewonk.bsky.social for RPG commentary, relevant retweets, and maybe some rambling. You can also find our Discord channel and drop in to chat with our authors and get every new post as it comes out. You can travel to DriveThruRPG through one of our fine and elegantly-crafted links, which generates credit that lets us get more games to work with (which is eactly what we did here)! Finally, you can support us directly on Patreon, which lets us cover costs, pay our contributors, and save up for projects. Thanks for reading!