Is it worthwhile to build up Infrastructure for the build speed modifier?

[Simon_9732495]

(This is from another thread, but I think it should have it's own.)

Short answer:
No.

Long answer:

The building speed for construction of industry like CIVs, MILs, Docks or Refineries (not Infrastructure itself) is calculated as:

Building speed = 5 * Number of assigned factories * (1 + sum of modifiers) * (1 + Infrastructure percentage)​

5 is the base building speed of 1 CIV​

"Number of assigned factories" is 0 to 15 in the construction queue​

Modifiers are e.g. (Construction1, -2, -3, Free trade, Captain of Industry, Autarky, Undisturbed Isolation); Example: (1 +10% +10%) = 1.2​

With 90% Infrastructure you have a multiplier 1.9 at the end and with 100% you have 2.0.​

That is a increase in building speed of about 5%.​

The duration of the construction is:

Duration = Building Cost / Building speed ​

Therefore if it is worthwhile to build up Infrastructure first depends on:

• How expensive is the building (More expensive, infrastucture helps more)
• How many building slots has the state (More slots, infrastucture helps more)
• Which production buffs are on the building (Less Buffs, infrastucture helps more, because it takes longer)
• Which production buffs are on infrastucture
• What is the infrastucture before
• How much do you upgrade the infrastucture

Let's do an example in favor of building up infrastucture:

Germany 1936​

Building CIVs​

In a state with a lot of slots (7)​

CIVS only have a 5% buff (closed economy) at that time​

And are expensive​

Lets take a slot with 70% infrastucture​

Infrastructure has a 5% buff​

Updating infrastructure to 10 and then building 7 CIVs takes 594 days.
Just building 7 CIVs takes 564 days.

It's seems it's almost worth it, and you will get more slots later, BUT:

• You need 10 slots to only get breakeven
• You will get several buffs on CIVs (construction1,2,3, free trade, Advisor Schacht, Autarky) They turn the page again.
• When building CIVs the already built CIVs help building the next ones. Thats in favor of not building infrastructure first.
• Later you build MILs in the slots you get late. They are cheaper and for them it's only worth with 19 slots.

IF you want to build up the infrastructure in a state anyway because of resources, then it could be different. I'm not sure if building before is better than building afterwards.
Could be before because early building is so slow. Or it could be later, because early CIVs count more because of snowballing.

But if it's only about the buildings, building infrastructure before is not worth it, because the states don't have enough building slots.


One exception, the USA:
The USA have the modifier "Undisturbed Isolation" that gives -50% on construction.
There it is breakeven with ~6 Slots for CIVs.
So I think in the USA in states with a lot of Slots and resources it's better to build infrasturcture first until you can remove the debuffs on CIVs and MILs.

 

[sekelsenmat]

Your post is valid for western Europe, but as it has already been calculated before, you forget to consider that the viability of infra is greater if you have states with low starting infra and lots of slots.

Let me give you an extreme example. In two states, one with starting infra 1 and another 7, we build 3 infra in each. How much will the building speed be increased in each?

0.3/1.1 = 27.3% speed increase for the state with starting infra 1
0.3/1.7=17.6% speed increase for the state with starting infra 7

So yes, building infra for speed is useless in Western Europe, but not necessarily elsewhere. At least the Soviet Union has some states which can benefit from this.

Also, in some states you might want to build infra anyway for resources.

 

[Mousetick]

One exception, the USA:
The USA have the modifier "Undisturbed Isolation" that gives -50% on construction.
There it is breakeven with ~6 Slots for CIVs.
So I think in the USA in states with a lot of Slots and resources it's better to build infrasturcture first until you can remove the debuffs on CIVs and MILs.

Also as the USA, you can get the New Deal national spirit which grants +20% to infrastucture construction speed. It's not there at the start, but you get it early by going with FD Roosevelt in the November 1936 presidential elections - if I remember correctly. And it remains there even after FDR dies.

Otherwise I'm not sure the answer is as clear cut "No" as you present it based on one example. There are many factors to consider as you point out. One of them is the resources gain. Those increased resources can save you CIVs that you don't trade away for importing resources, or give you more CIVs by exporting more resources.

 

[Telenil]

Several nations start in Civilian Economy. That's better than the US but still an additionnal -30% to factory construction speed.

 

[Simon_9732495]

Let me give you an extreme example. In two states, one with starting infra 1 and another 7, we build 3 infra in each. How much will the building speed be increased in each?

0.3/1.1 = 27.3% speed increase for the state with starting infra 1
0.3/1.7=17.6% speed increase for the state with starting infra 7

In your example it still is not worth it with less than 5 Building slots.

Building up 3 Infra = 114 days
Building 4 CIVs in Level 4 Infra = 392 days
Total 506 days
Building 4 CIVs in Level 1 Infra = 499 days

There are no states with Infrastructure Level 1 and 5 Building Slots.

 

[Simon_9732495]

One of them is the resources gain. Those increased resources can save you CIVs that you don't trade away for importing resources, or give you more CIVs by exporting more resources.

I adress that in that paragraph:

IF you want to build up the infrastructure in a state anyway because of resources, then it could be different. I'm not sure if building before is better than building afterwards.
Could be before because early building is so slow. Or it could be later, because early CIVs count more because of snowballing.

But if it's only about the buildings, building infrastructure before is not worth it, because the states don't have enough building slots.

 

[Simon_9732495]

Several nations start in Civilian Economy. That's better than the US but still an additionnal -30% to factory construction speed.

I did an Excel to quickly calculate the numbers.
If you provide an example (Which nation, which state, which building) I can do the math.

I can upload the excel if someone wants it. I didn't so far because my excel is not in english...

 

[Telenil]

I did an Excel to quickly calculate the numbers.
If you provide an example (Which nation, which state, which building) I can do the math.

I can upload the excel if someone wants it. I didn't so far because my excel is not in english...

The maths might be a little more complicated if we switch from Civilian Economy to an other mobilisation law at 150pp.
But let's take a state like Coahulia in Mexico, it starts at 4/10 infrastructure and 2/5 slots. We are in Civilian economy (-30% factory construction speed). What are the numbers if I build, say, one or two infrastructures before switching to civ factories?

 

[Simon_9732495]

The maths might be a little more complicated if we switch from Civilian Economy to an other mobilisation law at 150pp.
But let's take a state like Coahulia in Mexico, it starts at 4/10 infrastructure and 2/5 slots. We are in Civilian economy (-30% factory construction speed). What are the numbers if I build, say, one or two infrastructures before switching to civ factories?

Building Speed of Infra is modified by Export Focus + 10% and Callistas -10%
That makes it 25 with 5 factorys (5*5*(1+10%-10%))
Building 2 Infa costs 6000 and will be done in 6000/25 = 240 days

Building Speed of CIV is modified by Export Focus + 10% and Callistas -10% and Civilian Economy -30% and Infra (1.4)
That makes it 24.5 with 5 factories (5*5*(1+10%-10%-30%) * (1 + 40%))
29,4 with 6 factories
34,4 with 7 factories
Building 3 CIVs will take 440 + 367 + 314 = 1121 days

With Infra 6 it will be:
28 with 5
33,6 with 6
39,2 with 7
And will take 386 + 321 + 275 = 982 days

Together Infra + CIVs it's 982+240 = 1222 days, which is about 100 days longer.



If you build the CIVs with only the -10% from early mobilisation and infra 6 it will be:
36 with 5
43,2 with 6
50,4 with 7
And will take 300 + 250 + 214 = 764 days

Together 1004 days.
But now you cant compare to the 1121 anymore because the CIVs you built in Infra 4 also profited from early mobilisation.

Puh...
I'd say it will be a bit worse with Infra first, but you will have more Oil...

 

[Telenil]

All right, so infrastructure doesn't seem to really pay for itself in construction time, though there may be some edge cases with law changes or high consumer goods %.

You would put infrastructure in rich resource areas and perhaps build there because it's faster, but thumb rule is that you shouldn't build infrastructure just to build faster.

 

[porta80]

Also the complex answer is no. Not for building speed, for resources and supply yes, the building speed is the additional benefit then.
Problem even on civilian economy the offset is already to high without any tech.
And most low infra states also do not have enough building slots, even later with tech not to benefit from it. Also the later the game the more construction techs you have and later you switch building mils and there its nor worth it anymore.
Getting to early or partial mob, what you wanna do anyway just makes it worse.
And building a civ earlier helps xou building more civs. And the first 2 building techs are done in less then a year if you focus on that what you also schould do.
And on most states you cant get enough building slot to even get even.
Even not as USA.

 

[Yanotoshi]

I did an Excel to quickly calculate the numbers.
If you provide an example (Which nation, which state, which building) I can do the math.

I can upload the excel if someone wants it. I didn't so far because my excel is not in english...

Yes please. If it is in German, it wouldn't be a big dealto use.

 

[Simon_9732495]

Yes please. If it is in German, it wouldn't be a big dealto use.

See my Excel-Sheet attached.

You can edit the yellow fields:

Free Building Slots = free slots in the tile
Infrastructure Start = infra before building
Infrastructure End = infra after infra building
Modifiers MIL = all modifiers for MILs (construction, war economy, MEFO, advisor, ...)
Modifier Infra = all modifiers for Infra
Modifiers CIV = all modifiers for CIVs

 

[Yanotoshi]

Thanks a lot

 

[noobermenschen]

(This is from another thread, but I think it should have it's own.)

Short answer:
No.

Long answer:

The building speed for construction of industry like CIVs, MILs, Docks or Refineries (not Infrastructure itself) is calculated as:

Building speed = 5 * Number of assigned factories * (1 + sum of modifiers) * (1 + Infrastructure percentage)​

5 is the base building speed of 1 CIV​

"Number of assigned factories" is 0 to 15 in the construction queue​

Modifiers are e.g. (Construction1, -2, -3, Free trade, Captain of Industry, Autarky, Undisturbed Isolation); Example: (1 +10% +10%) = 1.2​

With 90% Infrastructure you have a multiplier 1.9 at the end and with 100% you have 2.0.​

That is a increase in building speed of about 5%.​

The duration of the construction is:

Duration = Building Cost / Building speed ​

Therefore if it is worthwhile to build up Infrastructure first depends on:

• How expensive is the building (More expensive, infrastucture helps more)
• How many building slots has the state (More slots, infrastucture helps more)
• Which production buffs are on the building (Less Buffs, infrastucture helps more, because it takes longer)
• Which production buffs are on infrastucture
• What is the infrastucture before
• How much do you upgrade the infrastucture

Let's do an example in favor of building up infrastucture:

Germany 1936​

Building CIVs​

In a state with a lot of slots (7)​

CIVS only have a 5% buff (closed economy) at that time​

And are expensive​

Lets take a slot with 70% infrastucture​

Infrastructure has a 5% buff​

Updating infrastructure to 10 and then building 7 CIVs takes 594 days.
Just building 7 CIVs takes 564 days.

It's seems it's almost worth it, and you will get more slots later, BUT:

• You need 10 slots to only get breakeven
• You will get several buffs on CIVs (construction1,2,3, free trade, Advisor Schacht, Autarky) They turn the page again.
• When building CIVs the already built CIVs help building the next ones. Thats in favor of not building infrastructure first.
• Later you build MILs in the slots you get late. They are cheaper and for them it's only worth with 19 slots.

IF you want to build up the infrastructure in a state anyway because of resources, then it could be different. I'm not sure if building before is better than building afterwards.
Could be before because early building is so slow. Or it could be later, because early CIVs count more because of snowballing.

But if it's only about the buildings, building infrastructure before is not worth it, because the states don't have enough building slots.


One exception, the USA:
The USA have the modifier "Undisturbed Isolation" that gives -50% on construction.
There it is breakeven with ~6 Slots for CIVs.
So I think in the USA in states with a lot of Slots and resources it's better to build infrasturcture first until you can remove the debuffs on CIVs and MILs.

Another point to consider is a lot of players will build mostly or all Military factories. @Secret Master recently posted a thread showing an all Mil buildup is better for Germany through 1940, after which you are raking in conquered factories with the military your Mils produced. As USA I build most states' infrastructure to at least 8 even though I am building largely Mils (I like to play USA -vs- Fascist World so am building a couple dozen of refineries early on though). Playing France for the first time I have realized (even though the war is still six months away) that my country would have been a lot better off building Mils instead of Civs during 1936.

In addition to gaining resources another specific reason to build infrastructure is building up supply lines. Once I start invading most of my top construction projects are ports and infrastructure in newly conquered lands.

 

[jpd]

In addition to gaining resources another specific reason to build infrastructure is building up supply lines. Once I start invading most of my top construction projects are ports and infrastructure in newly conquered lands.

Exactly. Building infra is about more than construction speed of factories. Supply capacity plays a huge role, as does resource extraction.

Play as germany. Conquer poland (and don't share it with the USSR). Now check up on the supply capacity in the zones on the russian border. It will be around the 70-80 mark. Now build up infra to level 10 in selected states that give you a continuous supply line from berlin to the front zones. Lo and behold, your supply capacity is now a whopping 210 per zone. That makes quite a difference

 

[Simon_9732495]

Another point to consider is a lot of players will build mostly or all Military factories. @Secret Master recently posted a thread showing an all Mil buildup is better for Germany through 1940, after which you are raking in conquered factories with the military your Mils produced. As USA I build most states' infrastructure to at least 8 even though I am building largely Mils (I like to play USA -vs- Fascist World so am building a couple dozen of refineries early on though). Playing France for the first time I have realized (even though the war is still six months away) that my country would have been a lot better off building Mils instead of Civs during 1936.

I'm not sure if I get your point...
When builiding MILs, building Infrastructrure before only for the speed modifier is even less worth it, because MILs are cheaper than CIVs.


If you build infrstructure for ressources or supply lines the return of investment calculation has to count that in, which I didn't, because how?
I concentrated on aswering the question: "I want to build up my economy as fast a s possible. Will my factory count be higher at date x, when I build up Infrastructure first?", because that can be calculated.


Another thing to consider: (wrote already a bit about it, but to put it clearer)
If building CIVs and MILs is debuffed now and the debuff will go away later, it's better to not build the debuffed things now if you can build something useful. As USA this can be Infrasturcture in places with ressources AND/OR a lot of building slots.

 

[Androbey]

While this specific discussion is valid.

The objective of Infrastructure as a state building is complex compared to the shared buildings. Each state building (slotted) have a single purpose.
When state level construction is viewed, IMO Infrastructure is the only one that has multiple objectives.

I am of the opinion that the Wiki accurately lists the objectives of Infrastructure and in the order of importance:

"Increases supply and supply throughput. Increases the speed of land units. Resources output is increased by 10% per level. Increases construction speed of  shared buildings."

So, you should build Infrastructure to:

First: Increase your supply to your front line units.

Second: Increase the strategic speed of your land units.

Third: Increase the strategic resources that are in a state. This will depend on what is causing a production bottleneck and the availability of that resource in trade.

Fourth: Increase the construction speed of "Shared buildings". This applies to ALL Shared Buildings, not just CIVs & MILs.

So IMO, the increase in construction speed is the least important of the abilities of infrastructure.

The question that I couldn't answer is:

What is the construction speed Increase?
Is it a flat one-time increase or is it a percentage increase per level of infrastructure?
Then what is the amount of increase?

 

[Simon_9732495]

What is the construction speed Increase?
Is it a flat one-time increase or is it a percentage increase per level of infrastructure?
Then what is the amount of increase?

The formula is in the first posting.

The building speed for construction of industry like CIVs, MILs, Docks or Refineries (not Infrastructure itself) is calculated as:

Building speed = 5 * Number of assigned factories * (1 + sum of modifiers) * (1 + Infrastructure percentage)
5 is the base building speed of 1 CIV"Number of assigned factories" is 0 to 15 in the construction queueModifiers are e.g. (Construction1, -2, -3, Free trade, Captain of Industry, Autarky, Undisturbed Isolation); Example: (1 +10% +10%) = 1.2With 90% Infrastructure you have a multiplier 1.9 at the end and with 100% you have 2.0.That is a increase in building speed of about 5%.
The duration of the construction is:

Duration = Building Cost / Building speed

Building 1 more level of Infantry speeds up the building from 5% to 9%
Level 1 -> 2: 9%
Level 9 ->10: 5%

 

[noobermenschen]

I'm not sure if I get your point...
When builiding MILs, building Infrastructrure before only for the speed modifier is even less worth it, because MILs are cheaper than CIVs.

I was posting in support of your point that infrastructure normally does not cut down total construction time, since players will normally put Mils in most of those new slots.