When should I stop building Civs? A: Don't build them.

[blahmaster6k]

Yeah, "nonsense". You know, difference between us: I spot nonsense and add information, you feel attacked and get Defensive.

Your rebuttal underscores the fundamental issue of your approach: dumbed down to worthlessness.
If your math cannot handle the complexitiy, well, then the math is lacking. Try again. Come up with a better model.

For all your scientific Method thumping, you're really bad at taking criticism. I didn't argue that your thing is inconsistent or the math wrong, I attacked the very premises and your reply is: "but then my theory won't work"
Yeah. That's the point. There is a variety in methodology because you cannot easily handle problems like this in 2 pages of equations.
Think about that.

He's never going to acknowledge this. He just ignores every reply that tells him such and dismisses us. This whole thread is pointless. He's also attacking every counterexample as a "special case" when his model is itself a very special, dare I say unique in it's simplicity, case that he, in his own words in the title and in his conclusion in the pdf, is trying to generalize.

 

[DeadEyeTucker]

You're very dismissal of anyone claiming or defending that opening with mass CIC production for a few years is the optimal meta for most cases, but you haven't given evidence for the opposite either. It seems the only thing you proved, assuming your math is correct, is that MIC from day 1 is superior than CIC for x years then MIC for putting out raw MIC output capacity. Not irrelevant information, but as far as I am aware, the prevailing school of thought is NOT "Build CIC for x amount of years for its superior MIC output capacity", but because it is generally the best strategy from an overall perspective. You've admitted yourself that this 'study' does not take into account the need to trade for resources or to build other vital infrastructure such radar, refineries, silos, etc. But these are tasks that must be done in almost every game to some degree or another.

 

[DeadEyeTucker]

@personwithhat, below is one of the threads that tried to address the question of CICs or MICs. The thread will not teach someone with your math skills anything new in math, but it does a good job bringing up questions that need answers or assumptions when building a production model for any individual country.

German build up: A simple comparison between a CIC build and pure MIC.

I haven't run this test since LaR came out, so in light of some of the tweaks the game has had over the past year, I thought I would run this test again. But first, let's ask Conan what is is best in life. Conan has played HOI4 with me in MP...

forum.paradoxplaza.com

Below is another thread that gives many observations of an optimized German build.

1.9 German Economy Benchmark Thread

There were some changes to economy in Husky and I'd like to optimize my Germany Buildup 1936-1939 again. To see where I stand I would like to compare my builds with others. Therefore I'd suggest the following benchmark: Date: 1.6.1939 (Short...

forum.paradoxplaza.com

Both threads have many more observations than one person could possibly arrive at by themselves. I hope they can help you with your model.

Reading Secret Master's thread is a waaay better experience than this one. And his MP group sounds awesome. It's interesting that he ran almost the exact same test but in game as OP, got roughly the same disparity in equipment by 41, but has a much more nuanced take away.

 

[blahmaster6k]

Reading Secret Master's thread is a waaay better experience than this one. And his MP group sounds awesome. It's interesting that he ran almost the exact same test but in game as OP, got roughly the same disparity in equipment by 41, but has a much more nuanced take away.

The important part is that Secret Master put his case study in context of an actual game, gave credit to scenarios where it's better to do one or the other, acknowledged the inherent shortcomings of not accounting for trade, and took that into consideration for real-game applications of his tests. That thread was a meaningful discussion where everyone was willing to hear each other out, and a lot of people learned something from it. This thread is not.

 

[foamingjetty]

The below are simply outside constraints that are unrelated to the economic strategy that is taken:
- Civs traded or building Infra: For resources
- Civs built to maximize future military infrastructure: Fuel silo's, synthetics, airports, repair, etc.
As noted they are buildings of a fixed value that are independent of when you build them.
Unlike military factories, the loss of which introduces a substantial opportunity cost.
- Civs built to maximize Mils at a later point in the game: A trade off. You lose total military IC but gain more of whatever you're aiming for. SITUATIONAL DECISION

In your search for a quick rebuttal to all dissenting views at once, you've just accidentally closed this thread. I went from mildly disappointed that this wasn't going anywhere to entertained in just two words.

I know I can't read, but reading this was fun. Cheers.

 

[Simon_9732495]

Read the previous 4 pages. Or my post immediately before your post.

That is your answer to my statement, that it would help YOU to write down your claim in a precise manner.
It's not very polite and not very helpful. Why do you answer in that manner?
If your exact claim is in the post before quote it and you are finished.

Yet you complain about:

Cherrypicking portions of my arguments and then pretending as if I've said something which I didn't, repeatedly.....

That is because your claim isn't precise.
The "good" thing on a vague claim is you can easily shut down critics, by saying: "I didn't mean THAT."
But the bad thing is the discussion about it will be unprecise arguing that gets everybody angry and impolite.
Sadly that seems to have happend to you.
Please do yourself a favor and write a posting: THIS IS PRECISELY MY CLAIM, to get the discussion back on the rails.
If you have done that already in other postings: Good for you. Just copy paste it together and make it clear.
Don't leave the task of collecting and assembling what you said to others. You will not be happy with the result.


Your answer to my text about "overall IC produced" "critical equipment" and "wasted IC is:

"I think all equipment is critical if you properly manage your production lines and technology, with various garrison templates to maximize the use"
"In 1936, for most nations, there is typically lots of useful equipment. Why, you can read any of the past 8 snippets I've written regarding this"
"Therefore I can build Mils early"

I'll try to explain it again, with a bit more detail:
1) In HOI4 you start with a certain number of MILs producing certain equipment, typically at highest efficiency. I think before you start the game you can reshuffle MILs and don't lose efficiency.

2) If you change a production line later you LOSE the efficiency. That is the most important example where you SACRIFICE "overall IC produced". You can call this a "waste of IC" because you'd have a lot more "overall IC produced" if you never change a production line.

3) That's why optimization on "overall IC produced" alone isn't leading anywhere.

4) Another example is building CIVs to get more MILs later. You SACRIFICE the production of the MILs you'd build in 1936 to build CIVs to have more MILs later on.
It's not that the support equipment you produce in 1936 is not useful at all. But sacrificing 1936 produced support equipment for more Medium3 Tanks in 1941 might be the optimal thing to to.

5) And you cannot deal with this with "proper management of production lines" as you claim. You cannot get more "modern equipment built" by managing your production in 1936. But you can by building CIVs to have more MILs later on.

6) Optimizing you economy on "modern equipment built" is what you do in competitive games. If you optimize on "overall IC produced" you are not allowed to change a production line and will lose for sure.

 

[Algarde]

I've already defended it. None of you are bothering to defend yours, which is my only point.

Nah, we play the game, the REAL game, instead you spend time with useless math, Play more and you'll see the results. Bye.

 

[blahmaster6k]

2) If you change a production line you LOSE the efficiency. That is the most important example where you SACRIFICE "overall IC produced". You can call this a "waste of IC" because you'd have a lot more "overall IC produced" if you never change a production line.
3) That's why optimization on "overall IC produced" alone isn't leading anywhere.

I once saw a screenshot in an AAR of someone's production window. The player in question was still producing Interwar Fighters despite the pop-out where you choose the equipment for a new production line showing that he already had 1940 fighters researched. The reasoning he gave for this was that if you change your production line to the new fighters, you lose efficiency and losing efficiency is bad, because you will end up producing fewer fighters. I believe that player learned a lot from critiques of his strategy in that AAR thread, but I'm going to use it as an example here anyway. This thread demonstrates the same logic as that original screenshot.

The math in the OP is correct, given the constraints. I'm going to ignore the unrealistic constraints for the sake of this argument though, and pretend they make perfect sense.

The conclusion that can be drawn from the math is NOT that you should never build civilian factories. The conclusion that can be drawn from the math is ONLY that you will have more total equipment produced until date X if you build only MIC from day 1. This is different from the OP's stated conclusion in his PDF and in the thread title.

Even if all his assumptions didn't take away from the relevancy and were 100% applicable to a real game, the goal of the game isn't to have the most total equipment produced by date X. The goal of the game is to WIN, and winning is rarely as simple as having the most total equipment produced by date X.

To go back to the example of the AAR thread, upgraded 1940 fighters will shoot down interwar fighters at a rate upwards of 30 to 1. It's not helpful if you have 30,000 interwar fighters and your opponent was only able to produce 4,000 1940 fighters by 1940. The 4,000 1940 fighters will still win in the end even if the country producing them had less overall IC produced throughout the game. I realize that this is just one example of a rather extreme case in tech disparity, but the logic applies to the rest of the game as well. As @Zauberelefant mentioned in his post about German strategy, Germany doesn't need to produce many guns because of all the equipment they gain from captured stockpiles. But those free guns aren't counted in IC production, and most guns built by Germany in the interwar period will be excess when these free guns are taken into account. OP has agreed that building up more factories to increase production later of critical technologies is a "special case" where it makes sense to build some CIC first. However, he also claims that these critical techs are only important on a nation-specific basis, and can be discounted from a general model. The true fact of the matter is that every single nation needs many of the same critical techs if they want to win the war. FIghter 2+, 1941/43 tanks of some variety, amtracs, naval variants, CAS 2, take your pick, every country needs at least some of these to be successful. It's impossible to discount as a special case something that is relevant to literally every country.

A common thread I see in the OP's replies to others is that he sees every counterexample as a special case, and that his generalized conclusion in the OP is applicable for all but some special cases. I would like to postulate that there is never a one-size-fits-all solution for every country, every single country is its own special case, and there exists no way to make a coherent claim that building either CIC or MIC is always the better option for winning the game. I've already mentioned earlier in this thread, out of the seven majors in the game, I would build MIC first on two of them, CIC/MIC on one of them depending on game plan(and still at least some CIC, just not as much as for other countries), and CIC first on the other four. My most common build, CIC for at least two years, is only done 100% of the time on 4/7 of the major nations, which is barely a majority and hardly enough of one to be able to draw sweeping generalizations from.

 

[Broe]

CIV for spies anyone? covert actions? it is used for a lot more now than before.
the usefulness or value of CIV have increased.

 

[Guffrus]

I didnt read the whole thread so I dont know if this has been covered already but you need to factor in the advantages of tech advancement as well;

Because higher tier tech is so much more powerful than lower tiers of tech by building civ first and mil later you may find that the value of what has been produced could be far higher with later mil factories even if the overall mil production is actually lower.

But the issue is further muddied by things like motorised and support equipment which is unaffected by tech advancement so optimal production would require you to have a goal in mind, e.g. i would like to complete the building of my starting army at X point in time and then you work backwards from there and say well i cant research and produce any of these items in that timeframe so they cannot be part of the starting army, i could probably have Y amounts of guys at that time and they would need Z amounts of motorised and support equipment so initially i need A amount of military factories so that i can build all of that basic equipment before i have to switch to the more advanced stuff that i can research and build in time by the time i have unlocked the research and need to start building it etc.

So optimal might actually be something like building 11 military factories to start with and then building civilian factories for a time and perhaps then building a few more military factories to build the new tech you just got and then building civs a while longer until you reach the point where you think you wont benefit from building any more.

But there could be outside forces which dictate when you must have an army ready by if you are not the initiator of war, perhaps an AI can be predicted to 100% accuracy or close to and you could fine tune your plans to what you know it will do but a human could do anything so you have to respond to their actions as they respond to your actions which would if not completely nullify the question of civ vs mil it certainly does in large part.

All you can really be sure of is that the earlier you build a civilian factory the more value you will extract from it but building even 1 more than you 'need' to (whatever that means) is extremely wasteful because it doesnt do anything for you but without having a crystal ball to know exactly what is going to happen you cant really know how many you will need.

I think i would suggest that whether or not you build civilian factories would depend on how long you think the game is going to last for and or whether you think you will need to build any at all and or if you think you can get away with delaying building additional military factories or not.

In any case its probably not a good idea to build too many of them so perhaps a rule of thumb might be something like dont build them after the first year or after the first 2 years etc

Unless of course you can see that you have a need for them, you might gain or lose them by taking ground or being bombed etc so that again is another variable which in itself is more significant than the original calculations presented.

Or if you wanted to take a more balanced approach and perhaps not be worried about hitting a specific timing attack or being optimal and just want to keep things simple you could say, I will always keep one set of 15 factories producing more civilian factories. Obviously this guarantees you are working suboptimally but perhaps in multiplayer where you cant pause, assuming you arent allowed to pause, reducing your cognitive load by establishing this sort of algorithum could be more valuable than the production costs etc.

There are most certainly more than enough variables to exclude any simple blanket statement / answer to the question but thank you for posting it, I found your data useful for considering the question.

 

[Voigt]

Things that happen in a real game:
Because you build Civs you don't need ressources for production, which means you can go to Free Trade and reap the benefit of construction, production and research.
You don't start with Dispersed IV, meaning your Mils in the beginning are much worse than later on in 1939.
There is only so much Motorized, Support EQ and Weapons I you need, you don't win the war by garrision everything with divisions which have lots of support companies. You use very cheap infantry and produce enough Inf EQ for those, but then produce as many planes and tanks as possible. LSPART are decent, aswell as 1941 MARM2. 1940 Fighter II can be used aswell.

If you just build mils and you are on 1939 dispersed, you can just switch 30 factories at once from another production line to the new researched equipment, which can help if you get the tech in 1939 so you have more stuff in 1940. Because you don't assigned new constructed Mils, but from your pool of existing Mils.

Still for various reasons (non factory buildings, trade, spystuff) I prefer to build Civs till mid to late 1937 as Germany, and to mid 1938 as Japan on Vanilla. Even on France I build some Civs in Algier, but only in MP and when I expect no Vichy forming, so I have some Civs to use in Africa for trade. (Core Algier with 10 Infrastructure and as new French Capitals provides lots of supply, steel and building speed)

 

[Gyrvendal]

To the OP: Do you actually play the game? Why don't you try scenario 1 and scenario 2 with a major nation of your choice (not US it's really a special case) and try to actually win the war?

 

[DrinkingHeavily]

Now it's your turn, if you feel like defending your points.

I think the OP is a practitioner of The Secret. If he keeps repeating it, putting it out there, eventually it will magic itself into truth.

Spoiler alert: It won't.

This thread is precisely the scientific method. OP took a shot, and peer review shows that life doesn't work that way. Peers don't "have" to submit a paper to have a substantive refutation -- that's not how this [or life] works. Is it annoying that he took the time to do something and it didn't properly account for all relevant variables? Sure. But pushing harder isn't going to lead to a different outcome.

Note, of course, that the community didn't initially make this a "win" or a "loss." But since the OP wants to frame it confrontationally, OK.


Just take the L, dude.

 

[kettyo]

personwithhat is the type of guy who drives up to a motorway in the opposite direction and then argues with all the ignorant fools who come against him

 

[Louella]

What might be an interesting experiment would be the case of Canada. Historically Canada produced a huge amount of the motor vehicles used by the Allies. The CMP truck is the most famous product of the Canadian motor industry during the war, but there were others. Not sure if it makes sense in MP for Canada to do something similar, could be worth investigating.

 

[Secret Master]

Reading Secret Master's thread is a waaay better experience than this one.

I'm surprised no one summoned me to this thread. But I would have just linked the previous test anyway.

 

[Zauberelefant]

I'm surprised no one summoned me to this thread. But I would have just linked the previous test anyway.

You we're mentioned, don't you worry.
And as you can see, would have seen your time wasted.