Victoria 3 - Dev Diary #4 - Goods

[Florda421]

I just want some level of ridiculous to be allowed. I think I understand civilians can't buy tanks, but what's the compromise?

 

[kozlovsky]

At the current version of the game the economy isn't fully initialized with all values until a few weeks in (very much like in Victoria 2), though this is something we're working on and absolutely plan to have fixed for the eventual release version. It has to do with how many circular and interconnected effects there are in the game (for example, a drop in grain price due can lead to pops increasing their wealth level, which results in them buying more grain, which raises the price of grain again) which all needs to be properly accounted for during game initialization.

You're the dev here, but I always wondered why don't Paradox's games just simulate a month or two of gameplay (obviously, without war declarations and such) before game start. You usually have to sit idle for some time waiting for all the values to properly initialize anyway, so why not do it behind the scene? Maybe it's what is happening in Vic 3 already though.

 

[lachek]

You're the dev here, but I always wondered why don't Paradox's games just simulate a month or two of gameplay (obviously, without war declarations and such) before game start. You usually have to sit idle for some time waiting for all the values to properly initialize anyway, so why not do it behind the scene? Maybe it's what is happening in Vic 3 already though.

Vicky 2 did actually do this to some extent, and it was the approach we used in early Vicky 3 development as well. It has the downside that since everything is interrelated with the economy, anything you change in history scripts would be out-of-date as the game starts - for example, if I tell the game to make some Wealth 12 Clerks in Missouri at game start, but then the economy runs a few months in the interim, those Clerks might end up Wealth 10 with half of them turning into Farmers before the game starts.

So at this point in development it's preferable to us to ensure the scripted setup is adhered to at game start, even if that makes for an economically unstable early game. At release we will likely do some limited behind-the-scene iterative pre-ticking (instead of trying to compute an economic equilibrium between this many agents which would be, well I don't want to say impossible, but probably a bit overkill for a video game) but we need to limit what can happen as a result of that - for example we wouldn't want Pops to engage in economic migration prior to game start.

 

[GrounchoVilla]

so if im reading this correctly, in the British market you can buy an eighth for 47 pounds? sounds kind of pricey tbh i guess colonizing the hindu kush didn't help push prices down

 

[mladjo176]

View attachment 732592
Anybody want to guess what these are? I thought vegetables for a second but they seem too expensive relative to other goods.

I'm pretty sure that's opium bro

 

[Lorehead]

At release we will likely do some limited behind-the-scene iterative pre-ticking (instead of trying to compute an economic equilibrium between this many agents which would be, well I don't want to say impossible, but probably a bit overkill for a video game) but we need to limit what can happen as a result of that - for example we wouldn't want Pops to engage in economic migration prior to game start.

Then I was underestimating the number of agents by a lot. Sounds good!

 

[Woifee]

At the moment, Services is a single good that is non-tradable (meaning it can only be bought and sold locally) but it would be easy to split it up into different kinds of intangible goods if we wanted to.

There comes to mind, will Vic3 have Tourism, so countries profiting from tourism might blow up their service sector to get revenue from pops from other states as will. So kinda exporting services/improrting money.

 

[Mr. Wiggles]

The biggest issue with railroading precious metal and oil discoveries is that it really does remove the real strategic thinking behind it because the player already knows where deposits are going to be ahead of time. For instance, in Victoria 2 the player knew that Arabia would become an oil producer so it was always beneficial to conquer the Arabia Peninsula decades before it actually became strategically valuable, when the AI would have no knowledge of that. It robs any real decision making or scramble for such resources when the player knows well ahead of time where they're going to pop up just because of bringing in outside knowledge. Since you can't stop the player from bringing in that outside knowledge, what you can do to make the decision making more impactful and necessary is make the discovery of the resources more randomized and give the discoveries a chance of not happening at all or happening elsewhere, within enough reason. That way the player doesn't actually know with 100% certainty what's going to happen in any particular game.

If that's case, why not making the iron and coal deposits discovered after 1836 appear randomly out of UK and Germany? If I want to play Araby I want to have the oil, if I want to play as a south east nation I want to have rubber near, if I play as South Africa i want diamonds and coal.

The player will always have "unfair" advantages over the AI. If you are playing as the UK or France you will probably take a good chunk of the world for yourself anyway. If you are playing a minor the other countries should get there before you do, leaving you the scraps.

 

[SignedName]

Is there any possibility for Armes Blanches (i.e. melee weapons) to be a type of Military Good? Swords, spears/lances, combat knives, bayonets, etc. were used throughout the period as weapons alongside firearms, and were produced in different factories that specialized in blade making. Melee weapons were especially prominent in colonial conflicts among peoples with limited access to firearms or who preferred to fight in close-quarters engagements, and European powers sometimes found themselves at a disadvantage if they were not themselves equipped with their own melee weapons for self-defense. Not to mention cavalry's heavy usage of cold weapons throughout all armies, even the modern European ones, of the Victorian era. It wouldn't make much sense for cuirassiers or hussars to require small arms for maintenance, yet the need to supply them with appropriate equipment ought to still be represented.

 

[safferli]

Any kind of insubstantial posts that do not

(instead of trying to compute an economic equilibrium between this many agents which would be, well I don't want to say impossible, but probably a bit overkill for a video game)

My General Equilibrium Models course is years in the past and all I remember was that it was fiendishly difficult... but I'm not entirely sure you could even calculate a GEM on Victoria, at least not in the standard Arrow-Debreu sense. Some of the assumptions compared to the V3 reality are what we economists like to call "heroic" assumptions One thing immediately springs to mind that the preference functions are not monotonous non-decreasing. In fact, they're not even continuous!

A quick googleing tells me that a thing called "Computationable General Equilibrium Models" exist, and this is what is used these days by the World Bank, etc. But it seems to me that they don't use GEM any more, but simultaneous equations based off Keynesian Macro models. Which would make them easier to compute I guess (I know less of computer science than I do of economics), but would definitely fit the V3 economy better (from what I can see from these DD and playing V2 so far).

 

[Make Victoria 3]

So not directly a "Tractor goods" but abstracted into Engines&Coal/Oil.

What will happen if one can only supply either engines or coal/oil but not both?

And generally what will happen to unsold goods? I hope there will at least be a limited factory stockpile that prevents situations where one day you have a surplus then soon after due to job changes you have a small shortage, then a surplus again etc...

 

[Antediluvian Monster]

If that's case, why not making the iron and coal deposits discovered after 1836 appear randomly out of UK and Germany? If I want to play Araby I want to have the oil, if I want to play as a south east nation I want to have rubber near, if I play as South Africa i want diamonds and coal.

The player will always have "unfair" advantages over the AI. If you are playing as the UK or France you will probably take a good chunk of the world for yourself anyway. If you are playing a minor the other countries should get there before you do, leaving you the scraps.

If they just put every real world deposit there that would not be very historical either since some deposits would be beyond reasonable for technology of the era. The end result might not be too different from having random deposits with weighted chances of discovery for those locations that have exploited the resource. And on the other hand if they limit the deposits to just the ones that were actually exploited in the era that would end up as unfair for e.g. South America.

Rubber is rather different case than rest of those resources. It's not actually native to Southeast Asia but was imported there by the British during the period of this game. The SEA plantations then put the initial market leaders like Brazil out of the market. Can't say if Southeast Asia had some kind of intrinsic advantage for this or if it was more incidental, perhaps a thriving rubber plantation economy could have also been set up in Africa.

Is there any possibility for Armes Blanches (i.e. melee weapons) to be a type of Military Good? Swords, spears/lances, combat knives, bayonets, etc. were used throughout the period as weapons alongside firearms, and were produced in different factories that specialized in blade making. Melee weapons were especially prominent in colonial conflicts among peoples with limited access to firearms or who preferred to fight in close-quarters engagements, and European powers sometimes found themselves at a disadvantage if they were not themselves equipped with their own melee weapons for self-defense. Not to mention cavalry's heavy usage of cold weapons throughout all armies, even the modern European ones, of the Victorian era. It wouldn't make much sense for cuirassiers or hussars to require small arms for maintenance, yet the need to supply them with appropriate equipment ought to still be represented.

I'd say using steel as a proxy would be sufficient. It's rare and valuable (presumably we will be getting some kind of primitive subsistence mines as well?) in the less developed countries (which in a way is everyone before Bessemer) and if you can mass produce steel you can probably mass produce bayonets.

...Or who am I kidding? We don't just need swords for cavalry, we need the ability to choose between cutting and thrusting swords. And a HoI-like bayonet designer for the infantry (spoiler: longer is better as guy with short bayonet puts himself into immediate disadvantage in a gunfight).

 

[mergele]

On the danger of jumping to conclusions to early: The interface for checking up on goods looks like it could actually be good.

 

[lachek]

With regards to discoverable resources such as Gold and Oil, they are not totally randomized but rather distributed throughout the world in potential quantities based on our research of actual deposits discovered today. Discovery chances are based on technologies, so it's also influenced by who controls what part of the world.

From a design perspective we don't actually care if a player wants to pre-emptively capture a part of the world before they would realistically know that part of the world will contain a valuable late-game resource. You wouldn't realistically be able to know in 1836 what technologies would eventually lead you to the capacity to develop Automobiles by the turn of the century either, but we're not hiding or randomizing technology progression for realism's sake.

That does not mean all regions with potential for Gold or Oil will be guaranteed to manifest that potential by end-game, even if they did so in actual history. If you want to try to pre-empt historical development you do so at your own risk. Personally I think it's more fun to see and react to the economic, political, and diplomatic shake-up of one of these discoveries than to use my knowledge of history strategically to win more, but for those who feel differently it's very easy to mod the removal of potentials from regions that didn't manifest it before 1936 - even impose historical discovery dates if you so wish.

 

[Amorenkaire]

It's cultural. So if French people become obsessed with Wine, then all French people will demand relatively more Wine regardless of which country they live in, though this will be moderated somewhat by its availability.

Here's hoping that Civilian Firearms can be modelled and be an Obsession for full 'Merica mode!

 

[joe9594]

That does not mean all regions with potential for Gold or Oil will be guaranteed to manifest that potential by end-game, even if they did so in actual history. If you want to try to pre-empt historical development you do so at your own risk. Personally I think it's more fun to see and react to the economic, political, and diplomatic shake-up of one of these discoveries than to use my knowledge of history strategically to win more, but for those who feel differently it's very easy to mod the removal of potentials from regions that didn't manifest it before 1936 - even impose historical discovery dates if you so wish.

Does the demand for a resource influence the chance that deposits will be discovered. Also could you show us one of the production potential heatmaps to get an idea of what it looks like?

 

[osondoar]

Vicky 2 did actually do this to some extent, and it was the approach we used in early Vicky 3 development as well. It has the downside that since everything is interrelated with the economy, anything you change in history scripts would be out-of-date as the game starts - for example, if I tell the game to make some Wealth 12 Clerks in Missouri at game start, but then the economy runs a few months in the interim, those Clerks might end up Wealth 10 with half of them turning into Farmers before the game starts.

So at this point in development it's preferable to us to ensure the scripted setup is adhered to at game start, even if that makes for an economically unstable early game. At release we will likely do some limited behind-the-scene iterative pre-ticking (instead of trying to compute an economic equilibrium between this many agents which would be, well I don't want to say impossible, but probably a bit overkill for a video game) but we need to limit what can happen as a result of that - for example we wouldn't want Pops to engage in economic migration prior to game start.

There is a similar problem in Climate Models when trying to initialize them to a specific condition. What we do is to run a very long simulation that forces the values we want at every step of it. For example, to initialize a historical run on 1900, we may run a long experiment of 1000 years were we keep the 1900 conditions fixed (radiative forcings, atmospheric composition...). We wait for the simulation to stabilize and then use that as the starting point for the historical ones

A similar approach may be an overkill for you depending on the ability of the game code to keep this scripted values fixed without a major rework, but bear in mind that you will just need to do this once for each game version.

About the equilibrium part, that is not even desirable as the state you are trying to simulate is not in equilibrium. And that's good because you may take an approach similar of what we do in climate. In our example, there is no reason to prefer year 1000 to 999 of the simulation, so we usually run several members initialized from different years of the simulation, with that count ranging usually from 10 to 100 depending on the experiment. If you do something similar with the game, for example randomly selecting between 5 different states for initialization, you can have slightly different starting setups that could lead to a more compelling early game due to it being played slightly different each time

 

[SignedName]

I'd say using steel as a proxy would be sufficient. It's rare and valuable (presumably we will be getting some kind of primitive subsistence mines as well?) in the less developed countries (which in a way is everyone before Bessemer) and if you can mass produce steel you can probably mass produce bayonets.

...Or who am I kidding? We don't just need swords for cavalry, we need the ability to choose between cutting and thrusting swords. And a HoI-like bayonet designer for the infantry (spoiler: longer is better as guy with short bayonet puts himself into immediate disadvantage in a gunfight).

Why stop at bayonets? Customizing the armament of troops from the type of primary weapon they're using to their sidearms to the degree of armor they're wearing (mostly helmets reducing deaths via artillery, but cuirasses and even chainmail were still in use in this period- early ballistic vests were developed as well, though they saw limited military use- but hey, if you've got cash to splash, why not?), with tradeoffs in terms of cost, availability, troops stats, etc. would be amazing. Even just changing what type of gun your soldiers use would be a big step up from the way Victoria 2 does it (simple stat bonuses from tech/inventions).

 

[Xain]

With regards to discoverable resources such as Gold and Oil, they are not totally randomized but rather distributed throughout the world in potential quantities based on our research of actual deposits discovered today. Discovery chances are based on technologies, so it's also influenced by who controls what part of the world.

That is a super elegant solution. This way, we don't really know which resources will end up being tapped during the playthrough, while at the same time avoiding weird outcomes like discovering gold in Oxfordshire.

That said, I really think that a completely randomised mode should be added just for the fun!

 

[Antediluvian Monster]

Why stop at bayonets?

Because people were seriously worried that length of your rifle-bayonet combination was some kind of significant advantage in decades before WW1. In retrospect this is bit amusing.

(Not that length of your weapon wasn't of concern earlier, e.g. with minie ball there was still preference for the longer rifle-musket over the shorter rifle (which tended to be reserved for light infantry) but that had the more practical consideration that troops needed to still fire over each others' shoulders which needed a longer barrel)

Even just changing what type of gun your soldiers use would be a big step up from the way Victoria 2 does it (simple stat bonuses from tech/inventions).

There should be at least need to re-outfit your troops with another bunch of (generic) small arms/artillery to take advantage of new weapons. With raised regular troops that's not even problematic, they just keep the old stats until you re-outfit them. Reserves are more complicated though.