Developer Diary 12

This developer diary was first posted in the M&T forum here, and team members answered some questions in the thread. It was further posted on the M&T subreddit here, and team members also commented there.

Dev Diary #12 - Buildings
A wise man once said that buildings are nothing more than glorified modifiers. Glorified modifiers they may be, they are nevertheless the main method of investing wealth into the economy. This diary will go into how we tried to improve upon that by fitting it into the 3.0 systems, created as a collab of myself but mainly KJH. Enjoy.

Let’s start from the basics. At its core, buildings serve two main functions. One is to be a place where the state can dump money into, and the other is to provide some return as a result of that investment. In that sense, key changes of 3.0 can be summarized into, first, making clear distinction between two types of buildable objects, and second, adding 2 new functions on top of the old ones. Let’s start from the first one, two types of buildable objects.

During the development of 3.0, one of the things that we wanted to add was a representation of property, meaning things like arable lands, forestry's, fisheries, etc. Their role would be to decide who gets what share of the economic surplus that was generated by the industries, as well as putting a tangible cap on industrial development. It shouldn’t be possible to feed the world from some small remote island, even if that island has all the workers it needs to do so.

At first, we didn’t consider those properties to be tied to construction and building. We gave a purely political property share to each class, and made property size be something that is set in stone based on provincial stats. Property size was not supposed to change over the course of the game.

This, however, quickly led to many issues. The first problem was that we had no idea how to model change in property share in a way that is fun and engaging. We attempted tying it to the total wealth share of a class, but since it’s not possible to interact with class wealth in a direct and immediate way, it wasn’t very understandable or enjoyable. We also faced issues in regions like the Ukraine, where there is an abundance of arable land compared to its starting population, making them super wealthy without the effort of investment or construction. All in all, such an abstract model of property didn’t really work in practice.

To remedy these problems, we made key conceptual changes to property. First, we decided on a clear distinction between current property size and maximum property size. Maximum property size represents physical, spatial limitations of how much property can be built for a given type, while current property size represents how much of that is actually constructed/developed and in use. This fixed the Ukraine problem, representing its potential while also restricting its initial, current size.



In the end we made Property work as fixed ‘units’. Ownership is per 1 unit, for example the state can own 2 farmlands and 5 fisheries. One unit represents the same size in all provinces, with smaller provinces simply having a lower maximum. Each unit in Property is a buildable object that the state pays ducats to build, which is then used to purchase input materials and acquire labor for construction. When the project is done, the built units become owned by the state, existing as state property. The state is responsible for paying maintenance for the units of Property they own, but can also gain income if it’s profitable. Property is the first type of ‘buildings’ in 3.0, focused on supply and demand.



Classes gain wealth from property through its fundamental role in the economy. Property provides space for industry to expand onto. If an industry has the labour and goods to expand, almost all industries need property to achieve full output and thus actual expansion. Industry as such is upheld by property. Given the importance of property in industry, it also forms the basis of profit and wealth distribution. All income and losses are distributed to property holders based on their share of the property in question (say the Peasantry’s share of Farmland). If the yearly revenue is income, additionally all labour receives 10% of all profit, with the other 90% of income distributed to property holders. All loss is absorbed entirely by property holders. Classes use their revenue and gain on average to decide whether to expand or shrink their holdings and thus space for industry, providing them with an essential role in economic growth.



The second type, infrastructure, is something more traditional, akin to roads and ports in 2.52, yet expanded beyond a communications efficiency role. While Property is closely tied to the economic systems and the role they serve within it, infrastructure is more focused on existing as an option the state can spend money on for a concrete, definitive benefit. Infrastructure in its effects is represented as a series of ‘ranks’, where increasing from Rank 1 -> 2 for example has a definitive gain. However underneath a rank is a number of units representing size, which is different between provinces. A large province may require 20 units of roads to achieve rank 2, while a small province gets rank 2 at just 4 units.



When constructing, the UI directly facilitates a simple method of constructing to a certain rank. Since units of infrastructure require a fixed amount of upkeep to maintain, a large province with 20 units for rank 2 has higher maintenance cost than a small one with 4 units for rank 2. Unlike property, each individual infrastructure unit isn’t owned by anyone, and doesn't have an innate effect. Only a collection of units does something, when enough of them exist to raise the rank of the infrastructure in a province.



What makes infrastructure and property much more interesting than plain old buildings are two special functions that we added. First, they require goods and labor for both initial construction and ongoing maintenance. This means that the money provided for a construction project does not get removed from the economy, instead simply changing hands to the industries selling goods and classes providing labor. And this economic stimulation through increased wages and additional consumption will persist over time, as upkeep needs to be afforded continuously.

Second, classes can invest into infrastructure and property. They can build any property should they have the wealth and will to do so, and Elites (Nobles, Burghers and Clergy) can invest into the province's infrastructure fund which maintains and builds a bunch of infrastructure units. This means that provincial economies persist and improve on their own even without direct state involvement. Especially Elites will increase their wealth by constructing new property units, from which they gain wealth and thus power.

Based on those two functions, infrastructure and property becomes heavily tied into politics and the economy. This leads to many interesting scenarios, such as a focus on disempowering elites and expropriating their property (and thus wealth base), leaving elites unable to fund infrastructure, resulting in underfunding and falling into decay, which reduces jobs devoted to maintaining them, forcing the state to either accept economic decline or increase state spending to fill in the vacancy, which can be extremely costly. It can also lead to a scenario where you rely on powerful elites to finance your provincial infrastructure, focusing state wealth on other ventures.

Fundamentally, the benefits of properties such as farmlands or fisheries are economic. Their benefits may be more difficult to measure on the surface, yet they are the key to economic expansion and political balance. Infrastructure is more akin to the buildings of 2.5, and the primary means of provincial development. Their construction provides concrete benefits, improving the province and industry within. While Buildings may be modifiers in 2.5, they are something else entirely in 3.0.

That’s about it for today. Also dropping a kinda mystery screenshot, but more of a reveal. Peace.