Dev Diary #36 – Construction

[bendkok]

Will infrastructure be expanded the same way as buildings?

 

[Llama_Guy]

Looks great!

Do construction worker wages always come directly from the state?

Would it make sense that wages to an extent comes from the investment pool instead, dependent on factors such as laws, production methods, and what's being built. So building government structures = state pays more of the wages, building industries = capitalists pay more wages via the investment pool, dependent on your economic laws etc.

 

[Jamaican Castle]

Will infrastructure be expanded the same way as buildings?

IIRC Infrastructure is created by buildings, chiefly railroads and ports.

Would it make sense that wages to an extent comes from the investment pool instead

Apparently, this is exactly how it works:

The investment pool can be used to cover the costs of all allowed constructions, so how much you can draw from it each week depends on where your construction allocation is going, ie whether you are currently building things that can be funded by it.

 

[Kyoumen]

I haven't had a chance to read the whole thread, so my apologies if this has already been suggested, but one thing that occurs to me is that infrastructure miantenance could be a function of the construction sector, and that if construction workers aren't assigned to it, the efficacy of infrastructure slowly degrades. Normally this would be bottom of the queue priority, but the player could move it up (and the AI presumably would when it started degrading to unacceptable levels). This would give some baseline employment to the sector and a small amount of goods use even when you weren't building things.

(Realistically, it would also cost more and take more time to fix the longer it's been neglected, but that might be a bit too realistic.)

 

[Kondio_patrona]

Can't construction be POPs need that's also used by the state to build things?

 

[grommile]

Can't construction be POPs need that's also used by the state to build things?

Several people have asked that question, and Paradox have explained their choice.

"Show dev responses" is your friend

 

[Xeiki]

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?

That actually sounds rather nice & realistic. Let's say I'm playing France and decide "we will build a barracks in every state, starting immediately!", that would drive prices up because there's simply not enough people to build all these buildings at once. This leads to the next point:

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?

That'd be on the player - just like governments in real life, you'd have to space out your construction projects appropriately, trying to maintain a relatively constant trickle of orders to keep up capacity so that in an emergency, it can be quickly expanded. I'm reminded of the British (naval) armaments industry after WW1, which expanded hugely during the war, only to then shrink afterwards. When civiclian demand for ships didn't arise, the government stepped in and supplied certain firms with a limited number of orders to keep them afloat and maintain skilled labour & expertise in the country. So it would be a strategic element of planning to say: "Okay, we just invented next level railroads - let's try to modernize our country over the next decade, and lets see what we'll do with our unused construction capacity afterwards" instead of "next level railroad invented - purchase new railways for the whole country at once!" Granted, it's more micro-heavy than just queueing construction in every state as soon as it is available, but it would also be an element of player planning & agency which could more or less involve into a gameplay element itself.

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?

If you want to build a huge, modern port in Nauru in 1837, then ... it should take a long time and/or be hugely expensive, since you'd have to ship everything there. Besides, timber, sand and bricks should be available almost everywhere. Most construction at the time wasn't high-tech at all, and later on in the game, when things start to become high-tech, you should have the ability as a developed country to ship advanced machinery to far away places with your by then steam-driven merchant navy anyway. This advanced machinery would then be used to set up local production plants for cement/whatever, and a few years down the line, you can build up some capacity locally if it is really worth the huge up-front investment.

I realize though that I'm probably in a minority here, actually liking the very limited player control (under certain political systems) and inefficient investment by capitalists in Vicky2. It's a balancing factor for player nations. As far as I can see & understand it, the current system means that you'll always be able to do the most optimal thing for your country, and I currently don't get how the system is interlinked with the socio-economic model, since, well ... it's entirely player controlled. This: "with the player's job being to appropriately size that capacity while minimizing its cost" sounds like a really straightforward "build up capacity to the maximum you can use and only scale down if you run out of money." There's nothing to throw a wrench in the player's plans. I spend 40% of my budget on military, 20% on other state expenses and 30% on construction, with 10% used to build up my reserves, maintaining a constant fraction of my pops as soldiers, state employees and construction workers who'll never be out of a job & will never have any reason to grow unhappy or rebell against my incompetent rule. Excluding wars for a moment, it seems like you'd have to go out of your way (like cutting funding entirely) to cause mass-unemployment and wreck your country.

As for the "there's no capitalist AI!" post earlier in the thread: all AI nations run on "capitalist" AI, since they all manage their own economies. There is an economic AI in place. So that seems like a non-starter of an argument.

But again, I don't think I understand the system at all, so my objections are most likely wrong on multiple levels.

 

[Revan F]

I'm clearly disppointed by contreuction being one another mandatory governement-run thing.
Remember : we talk about the pinnacle historical period of laissez-faire economics

 

[Bearmarshal]

This is such a minor nitpick that I feel bad about bringing it up, but the cut red ribbon in the artwork is way too long. Based on the height and distance between the two poles, it should barely reach the ground on either side if it was stretched before they cut it. Instead, the gap between the cut ends is really small, and the ribbon has crumpled up a bit, meaning the distance between the poles would have to be twice as wide. The only explanation would be that the ribbon was arranged after they cut it to make it look better on camera, but that feels a bit far-fetched.

Also, it doesn't look like it was cut in the middle, and there's no scissor in the picture (as far as we can see, but it seems to continue off-screen to the left).

As I said, really minor, but since I assume this is going to become one of the loading screens and thus will annoy me for all eternity, I wanted to mention it.
(To the people who hadn't noticed it before but will be as unable to unsee it as me, I'm sorry.)

 

[Eddie121]

The queue screenshots show 2 buttons to move a construction up or down the queue, why not just make it drag&drop instead?

Obviously we need both drag&drop for small moves, and Shift+Up/Down to move to begin/end when it's getting too long

 

[Tamwin5]

That'd be on the player - just like governments in real life, you'd have to space out your construction projects appropriately, trying to maintain a relatively constant trickle of orders to keep up capacity so that in an emergency, it can be quickly expanded. I'm reminded of the British (naval) armaments industry after WW1, which expanded hugely during the war, only to then shrink afterwards. When civiclian demand for ships didn't arise, the government stepped in and supplied certain firms with a limited number of orders to keep them afloat and maintain skilled labour & expertise in the country. So it would be a strategic element of planning to say: "Okay, we just invented next level railroads - let's try to modernize our country over the next decade, and lets see what we'll do with our unused construction capacity afterwards" instead of "next level railroad invented - purchase new railways for the whole country at once!" Granted, it's more micro-heavy than just queueing construction in every state as soon as it is available, but it would also be an element of player planning & agency which could more or less involve into a gameplay element itself.

Having to remember to go and add another building every 2 weeks or however long it takes sounds like a horrible experience and needless micro. The current system doesn't have that problem: If you queue up a bunch of stuff all at once, it just sits around until your construction industries get around to it. You still need to make sure you have roughly the right amount of construction industries relative to the amount of construction going on, and large country-wide building projects might require a temporary expansion of it, but it doesn't need to be babysat by the player. Remember, players will optimize the fun out of a game. Don't design systems which facilitate that.

Furthermore, I'd argue that for a large country wide infrastructure initiative, it makes sense to have it planned out entirely at the start, even if it takes a while for some of them to start construction. Eisenhower didn't say "We'll fully modernize these 4 highways! The rest? don't worry, I'll tell you about them once these have finished construction". So needing to manually trickle feed orders would be decreasing the immersion and accuracy.

There are ways around both of these problems I'm sure, but they are things to keep in mind.

 

[dunka2]

Having to remember to go and add another building every 2 weeks or however long it takes sounds like a horrible experience and needless micro. The current system doesn't have that problem: If you queue up a bunch of stuff all at once, it just sits around until your construction industries get around to it. You still need to make sure you have roughly the right amount of construction industries relative to the amount of construction going on, and large country-wide building projects might require a temporary expansion of it, but it doesn't need to be babysat by the player. Remember, players will optimize the fun out of a game. Don't design systems which facilitate that.

Furthermore, I'd argue that it makes sense that for a large country wide infrastructure initiative, it makes sense to have it planned out entirely at the start, even if it takes a while for some of them to start construction. Eisenhower didn't say "We'll fully modernize these 4 highways! The rest? don't worry, I'll tell you about them once these have finished construction". So needing to manually trickle feed orders would be decreasing the immersion and accuracy.

There are ways around both of these problems I'm sure, but they are things to keep in mind.

Oh my, frequently what makes sense is not in fact what happens when it comes to government procurement and investment. As for your Eisenhower example, stops, setbacks, poor planning, and false starts is really how it happened. The Federal Highway Administration set out the standards following the Federal Aid Highway Act of 1956, and highways were built piecemeal across the country for the next 70 years. It's still unfinished, and there are numerous gaps and various stretches that will simply never be built. The interstate system under Eisenhower wasn't even the first attempt, with two separate federal laws in 1916 and 1921. Large infrastructure projects are in fact routinely plagued with poor planning and underestimating, though the underestimating is rather paradoxically planned.

Of course the solution to what you imagine to be excessive micro is simple and already exists in another paradox game, HoI4. Have construction buildings create construction units which are assigned automatically to projects and can be reassigned as the player sees fit, perhaps for a cost.

 

[Tamwin5]

Oh my, frequently what makes sense is not in fact what happens when it comes to government procurement and investment. As for your Eisenhower example, stops, setbacks, poor planning, and false starts is really how it happened. The Federal Highway Administration set out the standards following the Federal Aid Highway Act of 1956, and highways were built piecemeal across the country for the next 70 years. It's still unfinished, and there are numerous gaps and various stretches that will simply never be built. The interstate system under Eisenhower wasn't even the first attempt, with two separate federal laws in 1916 and 1921. Large infrastructure projects are in fact routinely plagued with poor planning and underestimating, though the underestimating is rather paradoxically planned.

Of course the solution to what you imagine to be excessive micro is simple and already exists in another paradox game, HoI4. Have construction buildings create construction units which are assigned automatically to projects and can be reassigned as the player sees fit, perhaps for a cost.

I mean the main concern here is gameplay, not accuracy. But I'm not saying there wouldn't be slowdowns and that everything would proceed smoothly according to plan. What I'm saying is that the plan would be "build all the highways", and then the construction industry handles all the details (slowdowns and piecemeal construction included).

And the solution already exists in the current design of construction for Vic3: Construction buildings create construction which are assigned automatically to projects and can be reassigned as the player sees fit. I'm just poking some holes in a suggestion for how things could be different.

 

[dunka2]

I mean the main concern here is gameplay, not accuracy. But I'm not saying there wouldn't be slowdowns and that everything would proceed smoothly according to plan. What I'm saying is that the plan would be "build all the highways", and then the construction industry handles all the details (slowdowns and piecemeal construction included).

And the solution already exists in the current design of construction for Vic3: Construction buildings create construction which are assigned automatically to projects and can be reassigned as the player sees fit. I'm just poking some holes in a suggestion for how things could be different.

Yes I understood what you were doing. The issue for myself is that the construction system doesn't mesh with the entire rest of the game. If you're trying to create an economic simulator with supply and demand and variable wages, then the construction industry should also be part of that.

 

[Rinmun]

Having to remember to go and add another building every 2 weeks or however long it takes sounds like a horrible experience and needless micro. The current system doesn't have that problem: If you queue up a bunch of stuff all at once, it just sits around until your construction industries get around to it. You still need to make sure you have roughly the right amount of construction industries relative to the amount of construction going on, and large country-wide building projects might require a temporary expansion of it, but it doesn't need to be babysat by the player. Remember, players will optimize the fun out of a game. Don't design systems which facilitate that.

Furthermore, I'd argue that for a large country wide infrastructure initiative, it makes sense to have it planned out entirely at the start, even if it takes a while for some of them to start construction. Eisenhower didn't say "We'll fully modernize these 4 highways! The rest? don't worry, I'll tell you about them once these have finished construction". So needing to manually trickle feed orders would be decreasing the immersion and accuracy.

There are ways around both of these problems I'm sure, but they are things to keep in mind.

Easy solution to that. Queue up as much as you want, but have an option to manually decrease the rate at which you want to build stuff. Additionally to the the order of the queue I should be able to say "use X amount of capacity for that building max" or "use X max amount of capacity for buildings globally".
That way you could, in a private building sector, either go for maximum speed to build something up fast, or smooth it out over a longer time for a more stable building industry sector or in order to not compete with your pops too much.

 

[Sir Tornado]

I like the whole concept in theory, but making this into a centrally operated system is a bit off. Ideally, this construction capacity will increase and reduce based on market demand.

 

[Xeiki]

Yes I understood what you were doing. The issue for myself is that the construction system doesn't mesh with the entire rest of the game. If you're trying to create an economic simulator with supply and demand and variable wages, then the construction industry should also be part of that.

This is also my main concern here. Having a production queue and abstract production capacity is fine for a game; most games don't even go into this much detail. Except that Vicky 3 is first and foremost meant to be a society simulator; it breaks the illusion of being more of a guiding hand than a micromanager - and personally ordering and queuing the construction and meddling in what gets built where and how even if you don't order the construction yourself (I assume that people can build stuff on their own initiative, because if they can't, it'd be rather ... ) I've got no problems building state road railroads, for example, or state owned factories, but the construction system seems to cover all ongoing construction, with the only difference being that sometimes you don't pay for what's being build.

@Tamwin5: A few solutions for decreasing the mental capacity the player needs to spend on stretching out construction have already been mentioned - another method would actually be to use the queue feature already in the game, but *only* for government construction, and make it so that you adjust the rate at which stuff is build, and how many things can be built at the same time. Then you can order highways for your whole country, but decree that only X amount of money is spent each month, with only Y number of production sites being active at any time. A semi-automated system.

That of course also means that other construction (like factories, farms, etc.,) would be build by the private sector independently, but that seems not to be the case (?).

 

[Prince Ire]

This is also my main concern here. Having a production queue and abstract production capacity is fine for a game; most games don't even go into this much detail. Except that Vicky 3 is first and foremost meant to be a society simulator; it breaks the illusion of being more of a guiding hand than a micromanager - and personally ordering and queuing the construction and meddling in what gets built where and how even if you don't order the construction yourself (I assume that people can build stuff on their own initiative, because if they can't, it'd be rather ... ) I've got no problems building state road railroads, for example, or state owned factories, but the construction system seems to cover all ongoing construction, with the only difference being that sometimes you don't pay for what's being build.

@Tamwin5: A few solutions for decreasing the mental capacity the player needs to spend on stretching out construction have already been mentioned - another method would actually be to use the queue feature already in the game, but *only* for government construction, and make it so that you adjust the rate at which stuff is build, and how many things can be built at the same time. Then you can order highways for your whole country, but decree that only X amount of money is spent each month, with only Y number of production sites being active at any time. A semi-automated system.

That of course also means that other construction (like factories, farms, etc.,) would be build by the private sector independently, but that seems not to be the case (?).

Your pops can't build things on their own initiative, everything is built by the player. I'm extremely disappointed by it too, but that economic development would be 100% player directed was established months ago.

 

[amberambitions]

Why is everyone so concerned about the fact that pops can't build things now of all times? Capitalist AI in Vic2 was one of the most commonly complained about things about Vic2's economy. It made command economies just factually better. And it was never a fixable problem. Because trying to make every single pop have an AI potentially advanced enough to know how to build profitable buildings means every AI is either incredibly stupid and bad at it, or the game runs at 1 tick per hour.

So the solution was to get rid of it entirely and replace with a mechanism for encouraging you to build to pop demands. That's what the Investment Pool is for.

This solution definitely isn't perfect. I don't like how global it is, it needs to be a little more localized. On top of market access, I think there should be some sort of penalty based on the distance from the closest construction sector building. But it's certainly a lot better than the EU/CK/Stellaris model, and it's leagues better than Vic2's Capitalist AI.

 

[Xeiki]

@Prince Ire : Thanks for the clarification!

Why is everyone so concerned about the fact that pops can't build things now of all times?

For me, because I just now realized that this was the case, so I'm talking about it only now.

I also just don't get it, frankly. The economic AI also governs what AI nations do. The arguments on why player-controlled construction is a good thing seem to focus on "the AI is horrible and would make stupid decisions". But, uh ... wasn't that exactly the argument for taking player control away in the case of military micro? "The player has such a huge advantage over the AI that they could always win by outsmarting it." So, why doesn't it apply in this case? Doesn't it only mean that contrary to Vicky 2, where the player could micro-cheese the AI by military means, we can now cheese it via economics? If the AI is indeed so horrible that it would frustrate players to no end and makes horrible counterproductive decisions all the time (just a hypothetical), doesn't that automatically mean that AI nations will be at a huge disatvantage because they never build what they actually need? Which in turn means that the player is hugely advantaged by default? Which in turn means the AI must be able to cheat to hide its defficiencies and allow for challenging gameplay, if it truly is so horrible? Because if it were made to be better, the game would run at 1 tick per hour?

In other words: why is it fine for the player to be penalized by having his military be controlled by "stupid" and/or hugely resource intensive AI (which would slow down the game immensely, apparently, even if the player isn't themselves in a war since it has to run for all AI countries, too), but not fine for the player to have to suffer the same "horrible" AI the AI-controlled countries have to run with, levelling the playing field there as well?

Anyway, since I'm late, I guess this point has been discussed ad nauseam at this point, so feel free to ignore me. It's just baffeling to see that a game focused on socio-economic simulation doesn't actually simulate the "economic" bit, and makes no meaningful distinction (in practical terms, as far as I can see) between free market economies and command economies. Doubly baffling since the era simulated was basically defined by the rise of capitalism and industrialism, which in turn drove a lot of the societal changes/unrest. On a related note, one of the main reasons I never opted for state capitalism/communism over interventionism in Vicky2 was precisely because I wouldn't have to micromanage everything, but would still be able to protect key industries. Now, whichever system I opt for, I'll always have to micromanage everything. I already dread playing a large country lategame ... since it seems that we'll only be able to build factories in the same increments (add 5k/10k (??) workers each step), it'll be a lot of fun trying to build factories fast enough to keep up with population growth. That was one of the main reasons why Vicky2 got screwy lategame - even if you had unlimited budget, a huge economy and a population with extremely high literacy/education, shedding unprofitable industries and building new ones was almost impossible, because deleting one factory to have space for a new one would leave you with 500k unemployed pops and you couldn't build fast enough to get them employed again. So subsidizing huge factories was more or less a must to keep unemployment down. Or just ignore the issue, I guess. Anyway, tangential tangent is tangential, and switching to new goods is partly handled by production methods, so this issue is most likely mute.