Should the construction be a good?

[magistairs]

The idea is not new but as I see only benefits (more player agency, a lot more realism and cool interaction with MAPI/trade/foreign investment) to it, I would like to discuss it.

The main consequence would be the use of MAPI, which is way more realistic (constructing far away is more expensive than constructing where the construction companies are based and get their resources)

This also simulates the transportation of the workers to the construction site.

This leaves 2 details to design:

- Is MAPI of the input goods based on the state of the construction sectors or the construction site?
By just changing construction to a good, it would be the former, which is not totally realistic if, for instance, constructions sectors are in state A, iron in state B and the current construction in state B too: the construction would be impacted by MAPI 2 times while the resources are actually extracted and used in the same state.

- Can the good be traded?
As is, it shouldn't because the players would be unable to keep their construction for themselves.
But it would be a very interesting feature going along the new trade system and importance of foreign investments.
It requires some mechanic, I think about simply allowing the player to precisely decide how much of the construction can be exported.

Last note, I remember the devs said it was designed how it is for the sake of simplicity, but this was before MAPI, foreign investment and automated trade, this is why I think it's worth reconsidering today.

What do you think?

 

[TheFlemishDuck]

There is a lot of things that could be changed/improved that might come on the agenda. The simple system kinda works here, i'm not entirely convinced that the benifits of altering the system would bring as much added value as the cost of having to implement it. The caveats you describe are reasonable and just seem to make it harder. That said i get your considerations and why you think in this route, you thought about the matter a fair bit it seems so it perhaps merrit some discussion.

-leasing out construction capacity: there are things here that sounds somewhat reasonable (although perhaps less so by 19th century logistics) as it happens that country's provide the materials and manpower to build in some other country. And then there is also a degree that such a thing seems "potentially exploitable" for the creative min maxers. ill give credit to the consideration because its something that kinda exist afaik.

When you talk about "construction good x made in province x where it would by used by a construction sector in province Y" .... and then we would even have to what consider the volume of locally supplied building materials compared to the ones that come from outside the province .... nay on the computational overload and complexity there, the game already cant calculate transport distances and prices to begin with anyway.

To me this seems more mod territory, though then you get this isue that since construction currently isnt a good you would need to make cosntruction a good that construction sectors consume .... that i guess will have its own caveats.

 

[FranklyJustNess]

I think that's not really a good idea. The issue is, that maybe government would buy construction-points, but construction sectors are government run buildings. You'd be hit with MAPI costs twice as a single economic actor.

And if they were simply privatized, then you'd have problems with responsiveness - building sectors in advance where you wanna build, but then they aren't privatized, and you privatize them, build a bunch of things, and move onto a different state and those stop being profitable.

It would also mean you could drain private owners of their money. No need to centralize construction sectors in states with good resources, because you can just put them anywhere you need them, enjoy low construction costs, while MAPI hits the construction sector owners, who are importing all goods.

And I don't even wanna consider the horror scenerio where Construction is a local good.

 

[magistairs]

Indeed the idea is to add depth without costing too much work, this is why I tried to stick as much as possible to what we have:

1. Make it a good, trading disabled
2. Set the demand for this good based on where and how much construction points are spent (there is no additional computation since it just uses MAPI)

This is simple enough and brings something to the players with almost 0 work

Then we enter in the realm of refactoring the thing with design pros and cons, potential bugs and of course development time.

Trading construction and change construction sectors to use their input goods where the construction is happening are simple enough in my opinion but they can't use the existing systems without additional work, so they can be considered a bonus

 

[TheFlemishDuck]

Indeed the idea is to add depth without costing too much work, this is why I tried to stick as much as possible to what we have:

1. Make it a good, trading disabled
2. Set the demand for this good based on where and how much construction points are spent (there is no additional computation since it just uses MAPI)

Lets consider how simple or complex this is.

As you now, currently you have a capacity set by how many sectors you have, leading to a tab that can calculate how much it can construct according to how much capacity it has, and prices will be determined how the used capacity relates to the availabillety of construction goods in the country, with mapi calculated regionally for where the construction sectors are at.

If we consider our capacity determined by the availability of a construction good .... well how do we do that in relation to prices? i assume that in such a scenario i have either construction sectors or "industry" providing the good at a price based in part how cheap (an perhaps regional) the can get their goods and when we add orders to the queue the system has to calculate its "capacity" from that as to how much it might parallel build. Which will be? Will it be that the AI decides its ok to build more to the tune that your construction good has a 50% markup in the country because your overusing it? Or is the construction tab going to limit itself capacity wise so to run an amount of parallel constructions that keeps the price of the construction good around +0% markup? Perhaps on certain occasions the player will want to limit the construction tab to build at a rate where input prices and costs are cheap, and in other circumstances he might not give a damn about costs as long as the buildings he needs are pumped out faster, but then you dont have that control anymore as the player? As otherwise the player could overbuild or underbuild construction sectors in relation to resource prices in the economy and both imho have valid applications.

I guess point 2 is on that matter? That basically when you have say 100 construction goods in your market you get to set the amount to which the AI might overuse that or must underuse that to get to a certain price for the construction good? Like some buttons where you say "build as if you had 150 capacity" even if its only 100?

 

[magistairs]

I think that's not really a good idea. The issue is, that maybe government would buy construction-points, but construction sectors are government run buildings. You'd be hit with MAPI costs twice as a single economic actor.

And if they were simply privatized, then you'd have problems with responsiveness - building sectors in advance where you wanna build, but then they aren't privatized, and you privatize them, build a bunch of things, and move onto a different state and those stop being profitable.

It would also mean you could drain private owners of their money. No need to centralize construction sectors in states with good resources, because you can just put them anywhere you need them, enjoy low construction costs, while MAPI hits the construction sector owners, who are importing all goods.

And I don't even wanna consider the horror scenerio where Construction is a local good.

Yes, simple MAPI is not perfect, we would need actual supply cost based on distance to be realistic, but this is not worth developping just to change construction, I totally agree with that.


Let's not consider ownership for now, realistically construction companies should be able to be privately owned, this is too far from the current design so let's say everything remains state owned.
(We can discuss it but it opens many topics like buildings ownership, private investment buying construction to the government or the opposite, the government buying from private companies, or even compliance with existing features like companies, HQs, foreign investment...)


About the double MAPI impact, I think it's not too complicated to move the input goods demand to the state where construction site is, and it's by far the most realistic, for instance if the UK is constructing in India, it should be cheap next to an iron mine but expensive of they have to bring all the materials there.
However this requires a new mechanic, either:

-A: allow buildings to produce where they are not, consuming and outputting everything there

-B: make the construction company fixed but the construction sectors dynamic and automatically built and destroyed. For instance we would build the company to level 10 like we build the construction sectors currently, it then would be allowed to sustain 10 construction sectors and distribute them proportionally based on the current construction queue (automated, with build time or not)


About construction being a local good... yes this would be a nightmare,l with the current system, transport and electricity are already huge pain points in the game, we don't want to add it to the very basic gameplay loop of a day 1 game.

However if the above solution B is chosen, they are de facto a local good and can use the existing system to avoid being traded.


But again I want to make clear that I don't think construction absolutely needs a rework and I think there are many more topics to spend substantial development time on.
It's just that, if done wisely using the existing systems to cost a negligible amount of development time, the benefits could easily outweight the cost imho

 

[monksimus]

This has come up a bunch before - personally I am in favor but I think it is more important to have internal trade fixed first. I would just copy the world market system over national markets and replace merchant marine with transportation and that gets you close.

One way that having construction be a good could be really helpful especially for new players is governments could set the amount they want to spend in their budget directly and it would automatically adjust the amount of construction that buys based on market conditions. Right now you can get these wild swings in spending based on whether the investment pool is covering their allotted construction and it takes some practice to get a sense of what your average budget looks like. Just saying ‘spend 50k each week’ would eliminate that problem and smooth things out.

 

[magistairs]

Lets consider how simple or complex this is.

As you now, currently you have a capacity set by how many sectors you have, leading to a tab that can calculate how much it can construct according to how much capacity it has, and prices will be determined how the used capacity relates to the availabillety of construction goods in the country, with mapi calculated regionally for where the construction sectors are at.

If we consider our capacity determined by the availability of a construction good .... well how do we do that in relation to prices? i assume that in such a scenario i have either construction sectors or "industry" providing the good at a price based in part how cheap (an perhaps regional) the can get their goods and when we add orders to the queue the system has to calculate its "capacity" from that as to how much it might parallel build. Which will be? Will it be that the AI decides its ok to build more to the tune that your construction good has a 50% markup in the country because your overusing it? Or is the construction tab going to limit itself capacity wise so to run an amount of parallel constructions that keeps the price of the construction good around +0% markup? Perhaps on certain occasions the player will want to limit the construction tab to build at a rate where input prices and costs are cheap, and in other circumstances he might not give a damn about costs as long as the buildings he needs are pumped out faster, but then you dont have that control anymore as the player? As otherwise the player could overbuild or underbuild construction sectors in relation to resource prices in the economy and both imho have valid applications.

I guess point 2 is on that matter? That basically when you have say 100 construction goods in your market you get to set the amount to which the AI might overuse that or must underuse that to get to a certain price for the construction good? Like some buttons where you say "build as if you had 150 capacity" even if its only 100?

Good point, I omitted this consequence of making it a good, the prices would fluctuate based on sell and buy orders.
For now I was just thinking about staying close to the current system with the nation wide price being fixed, like gold, and the buy/sell orders always be 1:1.
The price would fluctuate only from MAPI.
I think it's good enough for a simple implementation (the investment pool already takes into account the predicted cost of construction before starting one).

Of course it could also be extended like you say, if the AI is able to combine the predicted productivity with the predicted cost of construction.
In a country with very profitable buildings, the construction good would be overused and cost up to +75%, while in a country with barely profitable buildings and/or expensive wood/iron/etc.. it would be underused and the price would go down until constructing a new level becomes worth it again.

But honestly I think this makes it too complicated and even annoying to the player, so I would really just stick with the current fixed sell/buy orders.

 

[magistairs]

This has come up a bunch before - personally I am in favor but I think it is more important to have internal trade fixed first. I would just copy the world market system over national markets and replace merchant marine with transportation and that gets you close.

One way that having construction be a good could be really helpful especially for new players is governments could set the amount they want to spend in their budget directly and it would automatically adjust the amount of construction that buys based on market conditions. Right now you can get these wild swings in spending based on whether the investment pool is covering their allotted construction and it takes some practice to get a sense of what your average budget looks like. Just saying ‘spend 50k each week’ would eliminate that problem and smooth things out.

Yes, this is about logistics costs and it would be way more realistic to have it based on the distance so changed along the MAPI system.
But this is a huge rework the devs already consider so there is no benefit in proposing such change today, this is why I'm trying to stick to the existing mechanics

Controlling construction based on its cost would be great imo! It would solve the problem mentioned in the post before (actual sell/buy orders instead of fixed) as well as allowing trading (export only the construction which is not used by the government or the private sector, import only if the government spending + investment pool willingness to spend create more demand than supply)

 

[TheFlemishDuck]

Yes, this is about logistics costs and it would be way more realistic to have it based on the distance so changed along the MAPI system.

Lets consider the limitations from Mapi, from there we can perhaps draw a better strategy for the intended result than to have the transport cost be represented trough mapi with needing a good.

For what regards mapi, if i'm in St petersburg and i want to send something to Moskou and something to vladivostok, the price of the transport cost that Mapi would impose is the same for both destinations. The transport cost hence can be the same between a province just next door and one more than 10.000km away. The same logic would apply when you have construction as a good between source and destination. So with the current systems existing regarding logistic (being very bare at that) you dont necessarily get far in terms of precision eitherhow. However if you can live with the abstraction, the thing you then kinda seemingly want is that there is atleast a markup between building in a province where construction is produced and where it is consumed. Such things can be conceptualized in a different way, for example a differentiation in price can be differently modelled in a differentiation of construction efficiency. Afterall in a province where construction sectors are present there is a bonus to construction speed given by the construction sectors. You could choose to make this effect stronger, so that in a province with construction sectors (or one that is abstractly producing construction) construction is cheaper not only because of its local input goods but because its local construction modifier. The thing here is that if mapi across the board is a 15% price penalty for example lets say the presence of local construction sectors on average gives about 15 to 20% added construction efficiency, that the lack of construction efficiency provided by construction sectors in another province for the tune of giving it that +15 to 20% construction efficiency will mean that functionally its construction is 15% more expensive, as if it was thus Mapi affected. The point here is that this implementation does not require to raise the cost of construction in a "Mapi affected" province trough a good if you can make it so that "it does not enjoy the same construction efficiency thus its affected in cost to a similar tune as Mapi would". As a practical implementation that simply means that you amplify the construction efficiency provided by construction sector levels while potentially somewhat decreasing their overal output.

 

[magistairs]

Lets consider the limitations from Mapi, from there we can perhaps draw a better strategy for the intended result than to have the transport cost be represented trough mapi with needing a good.

For what regards mapi, if i'm in St petersburg and i want to send something to Moskou and something to vladivostok, the price of the transport cost that Mapi would impose is the same for both destinations. The transport cost hence can be the same between a province just next door and one more than 10.000km away. The same logic would apply when you have construction as a good between source and destination. So with the current systems existing regarding logistic (being very bare at that) you dont necessarily get far in terms of precision eitherhow. However if you can live with the abstraction, the thing you then kinda seemingly want is that there is atleast a markup between building in a province where construction is produced and where it is consumed. Such things can be conceptualized in a different way, for example a differentiation in price can be differently modelled in a differentiation of construction efficiency. Afterall in a province where construction sectors are present there is a bonus to construction speed given by the construction sectors. You could choose to make this effect stronger, so that in a province with construction sectors (or one that is abstractly producing construction) construction is cheaper not only because of its local input goods but because its local construction modifier. The thing here is that if mapi across the board is a 15% price penalty for example lets say the presence of local construction sectors on average gives about 15 to 20% added construction efficiency, that the lack of construction efficiency provided by construction sectors in another province for the tune of giving it that +15 to 20% construction efficiency will mean that functionally its construction is 15% more expensive, as if it was thus Mapi affected. The point here is that this implementation does not require to raise the cost of construction in a "Mapi affected" province trough a good if you can make it so that "it does not enjoy the same construction efficiency thus its affected in cost to a similar tune as Mapi would". As a practical implementation that simply means that you amplify the construction efficiency provided by construction sector levels while potentially somewhat decreasing their overal output.

Fair! As MAPI is just a 0 to 40% price change, 10-20 for most of the game, abstracting it with construction efficiency where the construction sectors are is good enough or can be slightly tweaked.

But I have to admit I didn't even know about this construction efficiency, and I have several hundreds of hours... so I would still see a benefit to change it to a good so it inserts into an existing system, is more expectable for new players and eases the learning curve (this includes everything related to MAPI like changes from technology and laws, lack of convoys, devastation, occupation, manipulation of the feature by the player...)

The second benefit of being already aligned with other systems is that, the day we have a thinner transport cost system, the construction will profit from it without needing a specific "construction distance cost" work/design

 

[realCuervo]

I doubt it.

Construction can be traded globally: Ridiculous if construction team sat in a boat for 2 month to do their work overseas in 18th century.
Traded nationally: not very different from current system - construction sectors already provide local efficiency.
Local goods: NOT another headache unbalancable good please!

Glass and steel act as a buffer layer as pops and global market absorb the surplus when there is less construction ongoing (when player is busy or not enough private fund). Removing the abstract layer will only expose that problem imo.

 

[OxfordNik]

I'd actually make it two new goods: construction (could be local, or usable as now, produced by construction sectors), and housing (a local good, with different production methods for urban centres evolving with technology, requiring construction to switch from one to the other).

I'd also require construction as an ongoing 'maintenance' good, and potentially tie it in with switching from one PM to another (e.g. to reflect the cost of new furnaces when switching to Bessemer steel; new lines being built when switching to railway-assisted plantations/lumber camps; new equipment being installed with better mining sectors), etc.

That should hopefully resolve the problem of not having an ongoing demand for construction once you've developed a particular state, while allowing you to continue to develop others.

 

[SpiritRiven]

Indeed the idea is to add depth without costing too much work, this is why I tried to stick as much as possible to what we have:

1. Make it a good, trading disabled
2. Set the demand for this good based on where and how much construction points are spent (there is no additional computation since it just uses MAPI)

This is simple enough and brings something to the players with almost 0 work

Then we enter in the realm of refactoring the thing with design pros and cons, potential bugs and of course development time.

Trading construction and change construction sectors to use their input goods where the construction is happening are simple enough in my opinion but they can't use the existing systems without additional work, so they can be considered a bonus

There is a mod that does exactly this!

Steam Workshop::Private Sector Construction

steamcommunity.com

Construction goods are added as local goods, and consumed locally to contribute to the construction balance nation-wide. There is one construction good for each tier of construction.

...

Construction is consumed by the Construction Regulator, which is automatically set to the same level as the construction sector in the state. This employs 100 bureaucrats per level.

 

[Froonk]

Construction should be a good but this should come with a change to construction sectors getting the port treatment. I assume this will be a difficult thing to implement in general so I'd hope it'd come with some sort of logistics rework rather, one that makes things like rail, ports and waterways important to movement of goods and people.

 

[Ari Rahikkala]

The construction-as-a-good idea is weird, because on the one hand it really does feel like a perfect match for the direction the game has been developing, but then on the other hand, the current system really doesn't feel wrong or bad to play, either: It's hard to find an actual problem that needs solving here. And if you think about the actual details, there turns out to be surprisingly many moving parts, and if you get it wrong, you make the whole game terrible. So the devs really aren't kidding about that it's a concept that requires a lot of prototyping.

By just changing construction to a good, it would be the former, which is not totally realistic if, for instance, constructions sectors are in state A, iron in state B and the current construction in state B too: the construction would be impacted by MAPI 2 times while the resources are actually extracted and used in the same state.

This seems entirely fine to me, though? I think construction-as-a-good would represent things like bricks, sheet metal, finished lumber, windows, etc.. Maybe also engineering and architecture services, and even migrant labor a bit, but mostly stuff where it's 100% sensible that stuff really is getting hauled from the steel plants and lumber mills to the construction industries, and then from the construction industries to the construction sites, and both moves have a cost. And also that you could take the construction, put the construction on a ship, and haul it to the other side of the world.

As is, it shouldn't because the players would be unable to keep their construction for themselves.
But it would be a very interesting feature going along the new trade system and importance of foreign investments.
It requires some mechanic, I think about simply allowing the player to precisely decide how much of the construction can be exported.

This... also seems entirely fine to me (that is, fine to have construction be tradable, with just the same tariff/subvention controls that other goods get). You'll note the lack of player complaints that they got screwed by the world market buying all their grain and starving their population, for instance. The world market seems to work quite well to help balance out over- or underproduction rather than be a cause of trouble, and I expect it would be just the same with tradable construction, too.

The actually scary part to me is the UI and game mechanics for controlling how much your government budget and investment pool buy construction. The way that it works now is only slightly fiddly - you sometimes have to pause construction because your fraction of reinvestment to total construction budget is bigger than your economic policy's private construction allocation, is all. I can think of several ways to "improve" on that that actually end up being way fiddlier or just less fun to play. Yet if the private economy can build construction industries, you do have to come up with some answer there.

 

[magistairs]

There is a mod that does exactly this!

Steam Workshop::Private Sector Construction

steamcommunity.com

Oh great! I will directly try it

 

[Froonk]

The construction-as-a-good idea is weird, because on the one hand it really does feel like a perfect match for the direction the game has been developing, but then on the other hand, the current system really doesn't feel wrong or bad to play, either: It's hard to find an actual problem that needs solving here. And if you think about the actual details, there turns out to be surprisingly many moving parts, and if you get it wrong, you make the whole game terrible. So the devs really aren't kidding about that it's a concept that requires a lot of prototyping.

I think there are a lot of issues with the current manifestation of centralized construction queue and it has been there since release of the game. It's just things that made the game better on the whole since release both conceals and exacerbates the issues with it. Most important are how a state is able to mobilize and focus whole of its construction at any target, regardless of any priori in regards to distance, logistics or available infrastructure and how private sector benefits are generalized to all private actors collectively which makes companies be way too powerful and regional HQs leech off private investment of its host country (when it should come from company coffers).

It is however a lot of things they need to do at the same time. For example logistics of construction (I feel like making it just a local good with subsistence buildings and urban centers providing some baseline construction could work, but it just adds another thing to rework in future as transport and electricity being localized feels temporary too), as well as how capital ownership works (so that individual manor houses, financial centers and company HQs would all need their own money supply and management, which could better come with some DLC representing stocks, banks and financial instruments in general).

So it feels like it is something that can be improved greatly but also at the same time too centrally connected to everything to rework without having to rework it again multiple times to accommodate more changes.

As it is, it is "adequate" in being a framework for rest of the game but since the initial 1.0 release it feels more and more stretched out and more irrational the more (and better) systems are added into the game.

 

[magistairs]

Oh great! I will directly try it

Well I tried and to me this is unplayable, the spendings fluctuate by thousands each week even when set to 1%

But it's very good to see the caveats of construction as a local good+"construction centers" being built dynamically where the current construction is happening

It doesn't mean it's bad, just needs a lot of tweaking

But to go back to my first posts, I still think construction as a global good+MAPI would be less tedious

 

[SpiritRiven]

Well I tried and to me this is unplayable, the spendings fluctuate by thousands each week even when set to 1%

But it's very good to see the caveats of construction as a local good+"construction centers" being built dynamically where the current construction is happening

It doesn't mean it's bad, just needs a lot of tweaking

But to go back to my first posts, I still think construction as a global good+MAPI would be less tedious

Steam Workshop::Free Market Construction

steamcommunity.com

Here was the other one I was using until it stopped updating, but it seems recently it was just updated. Maybe give that a try.