Dev Diary #36 – Construction

[grommile]

In a worst case scenario a shortage of construction good would occur thus making construction prices skyrocket.

The problem being, that could easily become an uncorrectable situation, because trying to correct it (by building more construction buildings) will make it worse (by increasing demand for something that is already in shortage).

A capacity model seems much easier to numerically tune here.

 

[Adimon6]

The problem being, that could easily become an uncorrectable situation, because trying to correct it (by building more construction buildings) will make it worse (by increasing demand for something that is already in shortage).

A capacity model seems much easier to numerically tune here.

Remember that construction is more of a service than a physical good. It wouldn't serve as an input good for any factory (unless building maintenance isn't designed this way) so in this scenario the government would bid only against pops. Such competition wouldn't be equal and I can imagine it ending up with some pops not being able to withstand higher prices and losing some SoL in effect. Prices going up drive demand down and there is no shortage anymore, problem solved.

 

[Rinmun]

The problem being, that could easily become an uncorrectable situation, because trying to correct it (by building more construction buildings) will make it worse (by increasing demand for something that is already in shortage).

A capacity model seems much easier to numerically tune here.

Hmm, would be cool if some consumer goods had a buffer, but I don't know how complicated that would make things. Like people don't need to build houses every day, even if they become became rich overnight they would maybe wait some years before building a bigger house, waiting for the right circumstances. So pops should really only buy housing when the price is relatively low.

 

[Brynjar]

Bear in mind that "buildings" are not really individual buildings, but are instead entire industrial sectors. So you're not just building one building in one specific place, you're building all the necessities to support that sector of the industry. Which can very much end up being statewide and be affected by the terrain of an entire region.

I'm really not following the logic there. Why should it be so much slower to build a factory town along an already existing railway on a plain in a valley than on some plains in the middle of nowehere with no infrastructure? Mountainous terrain and forests already seems to have a malus to infrastructure, so it shouldn't be infrastructure that is the reasoning. The only reasoning I can imagine being done by the devs here is that clearing forests or mountains takes more time.

The problem with that is that it is generally not how settlements and industries are built. Usually they are built in a suitable spot. If you want to utilise a waterfall to produce power for a smeltery you don't blow a whole in the mountain right next to the waterfall (which is what the current system seems to be modelling), you find a suitable location nearby and use power lines to transport the power there. In Victoria 3 however, you will ideally want to build that smeltery far, far away from the waterfall since somehow building 500 km of power lines is easier than building the smeltery on that plain 10 km away from the waterfall...

The effect you are describing seems like it would be much better modelled by making these states support lower amounts of population (and therefore limiting how much you can expand an industry), combined with existing infrastructure. As it is now, a terrain seems to be giving both a flat reduction to construction efficiency and a construction efficiency malus from the infrastructure modifier (market access). The real world effect of this seems to be that you are better off establishing a new industry on a plain with no existing infrastructure than you are establishing it on a plain in a valley right next to a railroad. To me that makes absolutely no sense, and it seems to me like the devs have just slapped on another state modifier just to have something to fill the enormous UI elements with.

I really think the system would be better off with scrapping the terrain modifier to construction efficiency and make sure market access/infrastructure has a meaningful impact on construction efficiency. There doesn't seem to be any good reason to make the construction efficiency double dip the terrain modifiers.

 

[Grubnessul]

Why is this done through another counter? Shouldn't this be done through workers/construction worker pops?

 

[grommile]

Why is this done through another counter? Shouldn't this be done through workers/construction worker pops?

If you select "Show dev responses only", you will see some statements from Paradox regarding the motives of this design.

 

[Lord de Seis 85]

Thank you for the dev diary Wiz.

One thing I can't help but notice is that concrete isn't a construction good in Victoria 3? I feel like this is an oversight if it is the case. You are aspiring for Victoria 3 to be a society simulator where you are building up your nation but leaving out quite possibly the most important building material that those nations were built upon. This is taken from Wikipedia:

"Concrete is the second-most-used substance in the world after water,[2] and is the most widely used building material.[3] Its usage worldwide, ton for ton, is twice that of steel, wood, plastics, and aluminum combined."

Just my thoughts on it.
Keep up the good work.

Cheers,
Dan

 

[dunka2]

Finding a way to make the construction industry private is an aspiration we have, actually! Like many of you have pointed out, it does sound really cool and more realistic.

As it turns out though, after quite a bit of experimentation, this is harder than you might expect at first blush. For example, would you really want a gameplay dynamic where the fewer buildings you construct at once, the cheaper it is to construct those buildings, making it optimal (but very inconvenient) to only construct one thing at a time? What about if you do build a large number of buildings at once, creating a need for a large number of construction workers, who then get fired as soon as the construction completes because there's no projects left to work on? If construction industry is all very local in nature, how do you build ports to connect your overseas markets when local access to construction materials is non-existent? Do you have to set (and potentially constantly adjust) a construction budget that determines how much resources the construction industry has to operate with?

What's interesting is the process you have described is how the construction industry functions in the real world. As economic activity increases, usually more construction is needed in line with increased investment. This requires more workers and material, but hiring those workers can take time, potentially driving up wages if unemployment is low and thus construction costs in the short term. When construction orders start to fall due to decreased economic activity or opportunity for growth, this causes a retraction in the construction sector. It's an industry very prone to boom and bust cycles of a much more precipitous nature than the wider economy. And of course it would be optimal to build below your capacity, but then you would also be missing out on potential earnings if you put off building. Your opponents probably won't be holding off either. This would be an unnecessary worry in any case if the capitalist AI built things as well, but that's a non-starter.

A way to help cushion this boom and bust, at least until you guys figure out housing or some other more constant construction need, would be to have construction buildings produce a certain amount of services when their entire capacity isn't used. This wouldn't be enough to keep a large construction sector going, forcing the building to shed workers, but it would keep some around while other construction continues.

A benefit to this is that having local construction provides a bit of services for cheap using mostly labourers in newly developing areas. Local construction also means you can't just suddenly build a huge port and every production building you need halfway around the world with your construction sectors back in Europe. You want that huge military harbour in Papua New Guinea, Mr. Kaiser? You better get the workers there to build it, and its gonna cost you dearly. It, I think, makes for better and more interesting gameplay which also models how areas developed historically.

 

[Gilbert95]

It's an interesting thought, but that wouldn't actually create a choice for players. You don't decide where a Good goes, it gets consumed automatically by pops and buildings that want it. If there was a Construction Good that was consumed by both building buildings and by wealthy pops, the pops would always claim their share whether you like it or not.

Now, if there was an interesting Goods Substitution option (maybe wealthy pops can displace some of their housing/Construction demand into Services, or poorer pops might be satisfied with raw building materials?) that would open up a bit more flexibility for the player.

This is a good point. Wiz said that he was considering adding some bonus for unused construction capacity (similar to the boni given for unused bureaucracy capacity, etc.). Perhaps that bonus could affect SoL in other ways, maybe as a modifier to SoL growth. Because construction capacity is generated primarily by low strata laborers, I'd like to see an effect that increases SoL for upper strata pops, thereby making this a way to increase stratification in your society (for good or ill).

 

[Jamaican Castle]

But if construction was a good rather than a capacity, couldn't the state just take priority, so that pops get nothing for a while, again lowering their SoL?

That isn't how goods and standard of living work: all goods that are demanded in a week will be supplied unless you go into shortage, in which case everyone takes the hit equally. There isn't a system of priorities. (They could make one for one specific good, of course, but at that point it would work drastically different from every other good.)

Pop SoL also isn't really designed around the assumption that a whole good category will just not be there, especially transiently. Luxury goods might be so ruinously expensive that you don't get pops above, say, Wealth 20 because they can't support it, but it's not designed for a scenario where a pop can get everything else it needs to be Wealth 30 but this one item sometimes isn't available regardless of price.

 

[BeauNiddle]

"There are ways round this but it then risks breaking up the order things are calculated in (including possibly breaking siloing which will break multi-threading)"

I don't exactly know what this means, so maybe you already answered my question. But if construction was a good rather than a capacity, couldn't the state just take priority, so that pops get nothing for a while, again lowering their SoL?

My comment was about if you tried to make it a half and half system.

You are correct, if they altered Construction to be only a good, the system would just work. But it would throw up other edge cases. As Jamaican Castle explains goods are created and destroyed per tick so you can bend/break the cost of construction. Also goods seem to be either state specific (train tickets) or world tradable which can lead to other issues (I believe there was an AAR where Canada sold electricity to the UK). If you enter into another country's market then you can end up buying their construction goods (for good or ill), etc.

From the sounds of it Paradox will keep their focus on moddability for the game so I'm sure the modders will try a version where construction is a good. It will be interesting to see if it works.

 

[BrotherJonathan]

Will construction of buildings still require resources (steel, cement, wood, etc.)? I remember in past Vicky games it was almost impossible for non-Great or Secondary Powers to build much of anything for the first 20 years or so of the game because of constant cement shortages, among other things.

 

[Jamaican Castle]

Will construction of buildings still require resources (steel, cement, wood, etc.)?

Construction Sector buildings turn resources (and labor) into Construction, which then gets distributed to the buildings in progress. The exact resources depend on the production method(s) in your Construction Sectors.

 

[Revolution 11]

Will construction of buildings still require resources (steel, cement, wood, etc.)? I remember in past Vicky games it was almost impossible for non-Great or Secondary Powers to build much of anything for the first 20 years or so of the game because of constant cement shortages, among other things.

This will no longer be a (worldwide, all pervasive, unending) supply problem because the world is divided up into national markets of all the major powers and a few minor powers. Any resource shortage will be limited to each individual national market. Additionally, because resource shortages are no longer explicitly present but represented abstractly and indirectly by price increases of the good in shortage, there will be no hard limit or wall to your industries but rather a stifling and crippling pricing problem. This allows the player to continue their intended activities unrestricted by anything besides the cash reserves of the affected Building or the cash reserves of the nation (if you choose to subsidize the Building).

Yet this system gives clearly marked incentives to still fix the supply problems for the sake of affordability, industry competitiveness and profits, employment, and efficiency.

It is a win win win scenario for all players.

 

[Perdikkas76]

View attachment 813307

Hello and welcome to another Victoria 3 development diary! Today we’ll be returning to more mechanics-oriented dev diaries, starting out with a very important mechanic for the economic development of your 19th century nation - the construction of new Buildings.

Construction in Strategy games tends to follow a pretty typical formula: you save up money, order a construction and pay a lump-sum cost, wait some time, and the new building pops into existence. As mentioned in Dev Diary 12, however, the vast majority of expenses in Victoria 3 are not lump-sum costs but applied over time as part of your national budget. So how does it work instead? To answer that, there’s a few concepts we need to cover, namely Construction Capacity, the Construction Sector and the Construction Queue.

Let’s start then with Construction Capacity - which is actually just named Construction in-game, but we’re calling it Construction Capacity here to differentiate it from the overall concept of building things. This is a country-wide value of your nation’s overall ability to make progress on new buildings in a single week. For example, if your country produces a total of 100 Construction and a new Textile Mill costs 300 Construction, you’d expect to be able to build that Textile Mill in a total of 3 weeks. However, it’s a little more complicated than that, as we’ll see below when we explain the Construction Queue.


With Construction Sectors present in Lower Egypt, Matruh, Sinah and Palestine, the Egypt in this screenshot generates a respectable amount of Construction for the early game, though their finances may struggle a bit to fund it all.

View attachment 813308​

So, how do you produce Construction? This is where the Construction Sector comes in. All countries get a tiny amount of ‘free’ Construction Capacity to ensure that you never get stuck in a situation where you need Construction Capacity to expand your Construction Sector but need a Construction Sector to get Construction Capacity. This amount is woefully small though, and wholly insufficient even for a small nation, so if you’re not planning to run a subsistence economy long-term you will definitely need to invest in a proper Construction Sector by building more Construction Sector buildings in your states.

Mechanically speaking, the Construction Sector is a type of government building which employs people and uses goods to output Construction Capacity with a variety of different Production Methods, ranging from simple Wooden Buildings to modern arc-welded Steel and Glass structures. It does work a little bit differently though, in that the amount of Goods used by the Construction Sector each week depends on the actual need for Construction Capacity - if your Country is producing a total of 500 Construction Capacity, but will only need 250 for ongoing projects that week, the total usage of Goods in the Construction Sector is cut by half - though you still have to pay the wages of all the Pops employed there.

More advanced methods of construction are expensive and require complex goods - but you will find it difficult to build up a true industrialized economy without them.
View attachment 813309​

Ultimately, what this means is that how fast you can build things depends entirely on how much money, goods and research you’re willing to throw into your Construction Sector - having only a handful of Construction Sector buildings using only Wood and Fabric will certainly be cheaper and easier than building up a sprawling Construction Sector using Steel-Frame Buildings, but will naturally limit your ability to industrialize your nation.

So then, how does Construction Capacity actually turn into finished buildings? This is where the Construction Queue comes in. Each country has a nation-wide Construction Queue, with each project in the Queue corresponding to building a single level of a Building in a specific State. For example, a Construction Queue in Sweden might look like this (all numbers are examples):

1. Expand Government Administration in Svealand (250/300 Construction Capacity remaining)
2. Expand Fishing Wharves in Norrland (155/180 Construction Capacity remaining)
3. Expand Fishing Wharves in Norrland (180/180 Construction Capacity remaining)
4. Expand Rye Farms in Svealand (180/180 Construction Capacity remaining)
5. Expand Port in Götaland (240/240 Construction Capacity remaining)

Each week, your produced Construction Capacity is allocated to projects in the Queue in order of priority, with a maximum speed at which projects can proceed (so it’s never possible to, say, build the Panama Canal in a single week). Using the above construction queue as an example, let’s say the maximum progress that can be made each week is 50, and Sweden is producing 112 Construction Capacity.

This would mean that projects 1 and 2 would both be allocated 50 Construction Capacity, while project 3 would get the left-over 12 and projects 4 and 5 would not progress at all in that week. It would take 5 weeks for entry 1 to finish at that pace, but after only 3 weeks, project 2 will be down to only 5 progress needed, and so most of the Construction Capacity allocated to it will be freed up for other projects. This also means that project 2 will actually finish before project 1, which is perfectly normal, as different buildings require different amounts of Construction Capacity to complete - it’s easier to build a Rye Farm than a Shipyard.

With just above 40 construction output and the help of some local Construction Efficiency bonuses, this country is able to make rapid progress on the Wheat Farms and Iron Mines at the top of the queue and even get a bit of weekly extra progress on the Logging Camps.
View attachment 813310​

If all this seems confusing, don’t worry! All you really need to understand is that the more Construction Capacity you have, the faster things go - but a large Construction Sector will need to be kept busy with multiple projects at once if you want to use its entire output.

There is one more important factor to Construction, which is a modifier called State Construction Efficiency that governs how effective each point of Construction Capacity you put into building Buildings in a State is. For example, a state with a +50% bonus to State Construction Efficiency means that every Construction Capacity allocated to projects in that State actually results in 1.5 progress on said projects, while a malus of -50% would reduce it to 0.5 actual progress.

A few factors that will increase or decrease State Construction Efficiency are:

• Terrain-based State Traits, such as mountains or jungle, tends to reduce State Construction Efficiency
• Building a Construction Sector in a State increases the local State Construction Efficiency
• Low Market Access reduces State Construction Efficiency

Industrializing the Amazon Rainforest is neither easy nor cheap.
View attachment 813311​

That’s it for today! Join us again next week as we continue talking mechanics, on the topic of Market Expansion!

sorry,but 'the state' is 'sozialistisch' per default??

 

[Prince Ire]

As other posters have already said, this limits construction to a government directed sector. Why should we pay for the construction workers wages if they are not employed if we were in the private sector?

Construction sector can be modeled as the other Buildings and provide construction capacity without being state directed.

Also, construction queues? Where is the market? Supply and Demand? where are the private sector actors? Is the player the only one deciding what is going to be constructed?!

It's been clear since very early on that they want the economy in Vicky3 to be almost entirely player driven, unfortunately.

 

[Daurien]

I like it. This is a very elegant solution to the gameplay problem that other Paradox games have when it comes to construction. Why I need to provide a Dyson Sphere site with an enormous amount of materials just to get them to start using a small amount of materials has bugged me for a while.

 

[Te. Kenzo]

Finding a way to make the construction industry private is an aspiration we have, actually! Like many of you have pointed out, it does sound really cool and more realistic.

As it turns out though, after quite a bit of experimentation, this is harder than you might expect at first blush. For example, would you really want a gameplay dynamic where the fewer buildings you construct at once, the cheaper it is to construct those buildings, making it optimal (but very inconvenient) to only construct one thing at a time? What about if you do build a large number of buildings at once, creating a need for a large number of construction workers, who then get fired as soon as the construction completes because there's no projects left to work on? If construction industry is all very local in nature, how do you build ports to connect your overseas markets when local access to construction materials is non-existent? Do you have to set (and potentially constantly adjust) a construction budget that determines how much resources the construction industry has to operate with?

All these quite tricky questions, each of which add a number of gameplay concerns that need solutions, fade away when you treat Construction not as a variable-cost good bought and sold on the open market but as a capacity, with the player's job being to appropriately size that capacity while minimizing its cost. While I'd love to continue experimenting with Construction Sectors in the future to see if we can find a model where they can operate privately without damaging gameplay, I find it provides just the right level of player decision-making at the moment, while still being quite a bit more involved and interconnected with our socioeconomic simulation than construction usually is in strategy games.

I agree a private constructor sector shouldn't have only the state as a costumer, but also pops and other industries.

 

[Idle America]

Will there be some highly advanced buildings that are impossible to build without advanced construction methods? I'm finding it hard to imagine building Canals or modern factories with just wood and fabric.

That's quitter's talk! If I can build my first canal in Minecraft using wooden sticks, so can you!

 

[fightkirito]

As urbanization progresses, people moving from other places naturally and surely need more houses to live in. These construction workers usually build houses and then hire them to build government facilities when the government needs them.