JRaup, I think you might be on to something important.
The initial IC figures for the majors are pretty much taken straight out of Madison's 1938 GDP, and one IC roughly equals 1 billion dollars, divided by 360 days of the HoI year.
The world total IC also adds up to Madison's world total in GDP, which is 4,019.9 billion 1938 US$.
Basically, that is our starting point for IC.
In fact, I'd argue that we should take those figures, and reduce them by another 6% or so, to account for the two years of growth from 1936 to 1938.
On resources, in 1936 we should have enough of all resources for IC production to cover world's needs, plus another small amount as buffer. I'd say adding 10% to world needs should cover it nicely.
Here are the values from province.bak, HoI 1.05c:
2201 provinces
Average infrastructure = 25.36
Total IC = 4420
Total Manpower = 3150
Total Oil = 2592
Total Steel = 8203
Total Coal = 14826
Total Rubber = 2597
As you can see, total IC is a little over Madison's figures, about 10%. This is mostly because many minors have their IC production vastly overstated. Case in point is Yugoslavia. According to Madison, in 1938 it has 22 billion GDP, while in HoI it has 53 IC starting, more then double that.
Coal is overabundant, by 67%. Steel is even more overabundant, by 85%. Rubber is overabundant, by 18%.
My world strategic material production report states that global oil production in 1937 was 272 million tons. Looking at the HoI value, I'd assume that this figure, or a similar one, was used to derive total global oil production value.
It seems fairly obvious to me that the limiting growth factor here is rubber production. However, if we take into account the fact that oil and excess coal can both be converted to rubber, we can easily guesstimate that maximum supportable IC based on initial HoI resource production is about 50% more then the initial IC. This looks fairly well, considering that our rough estimates show that world's GDP in 1947 being about 50% more then in 1936.
In fact, looking at all these figures, it seems Paradox took a very easy road: taking Madison's GDP values straight for majors, divided by 10 and distributed over 360 days. For minors, they took Madison' values where available, and tweaked them mostly upwards for playability. Coal, oil, and steel (iron ore), seem to have roughly been taken from World Strategic Material production report for 1937, with minor changes. Rubber I guess was simply adjusted to these existing values, not seeming to have any relation to any of the other values I can see.
Various unit costs, supply costs and ratios, conversion techs, and fuel usage by units seem to have been mostly worked into these numbers backwards, to come up with roughly the right number of units and historical shortages (more or less). In fact, most of the fiddling with shortages seems to have been done in the area of resource distribution, rather then working with totals, as production is not correctly divided among the historically producing areas.
While this is a valid model, it is by no means ideal. We need to figure out actual world's NEEDS of main resources in 1938. Since IC production ratios for coal, steel, and rubber are fixed, we do not have the luxury of picking and choosing numbers for those 4. In fact, working with total IC, we need to work out total production of resources needed to support that IC.
Regardless of actual ratios of coal, steel, and "rubber" used in production, we are stuck with hardcoded ratio of 2:1:0.5. Basically, most discussion of how valuable coal is compared to steel compared to rubber is pointless, because we already know. Steel is twice as valuable as coal, and half as valuable as rubber. WE CANNOT CHANGE THESE RATIOS.
At this point, lacking other data, I'd assume that at any given point, global resource production matches global needs for that resource, given global GDP produced at that point. In other words, we should use GDP to figure out how much of resources we need for that GDP. Using Madison's 1938 global GDP, this is what I got:
Global GDP (1938): 4,019.9 billion US$
Global IC: 4,020
Coal Total: 8,040
Steel Total: 4,020
Rubber Total: 2,010
Given an uneven distribution of these resources, we'll have shortages in some countries and excess in others. We also need some way to support world GDP growth up to 1947. Assuming this growth will equal 50% over the 11 year period, we come up with these values for the end of game scenario:
Global GDP (1947): 6,029.85 billion US$
Global IC: 6,030
Coal Total: 12,060
Steel Total: 6,030
Rubber Total: 3,015
These are the tools we have at our disposal to create that global IC:
1. Initial overabundance of three key resources. I'd argue that a small excess, of about 10%, is the way to go here.
2. Oil conversion into rubber. Obviously, this is a limited help, beucase oil is needed for units, and because you still need extra coal and steel, in much larger quantities, to match the extra rubber.
3. Changing efficiency for countries, to give them an increase in production without changing their IC.
I think solution #3 is the way to go, with various industrial efficiency technologies. As I understand it, it is possible to increase the total production of IC produced with the same amount of resources, which is EXACTLY what we need in this case.
The small initial excess of key resources should support some growth of IC through normal means (growing it in provinces). It should also be there to allow World Market to function (so most nations would have some excess to trade for what they need). Oil conversion into rubber is not a terribly relevant factor. Majors will have a better use for their oil (units) while minors can only convert excess oil up to their rubber needs, which are minor anyway.
If we do this, we mostly avoid the problem of converting coal to oil and rubber to avoid shortages. Since, globally speaking, there isn't all that much extra to be had, only 10%, even at best conversion ratios, you can only squeeze 10% more IC out of those resources. All we need to do is to tweak conversion ratios to make it unattractive.
Basically, here is the solution I am outlining:
1. Take 1938 GDP figures, adjust down for 2 years of growth (from 1936), give same IC to the world in billions of 1938 dollars.
2. Give world enough coal, steel, and rubber to support that IC, plus 10% extra in each resource.
3. Create a set of production efficiency technologies, giving an average of 50% increase in production, with same resources. Maximum efficiency gain should be about 100% (for U.S., for example).
This leaves oil out, as it is a separate issue. It also leaves conversion factors out, as they are not going to be terribly important. We can set those as we choose. Oil can be handled as follows:
1. Take maximum war global oil consumption for military needs. Use that value to create oil point total.
2. Work with whatever ratio was used to convert military oil usage into points, and work out individual unit consumption figures, based on real usage numbers.
3. Up the global oil production total by some factor, to represent extra oil usable for military needs and/or oil-to-rubber conversion, and trading. My suggested value here is also 10%, to be in line with other resources.
Note that I try and skip the issue of how much of actual oil production went into military vs. civilian needs. I work backwards from maximum level of military wartime consumption, and take that to be the production total, plus another 10%. This organically limits us to maximum historical fuel usages, plus 10%, and another possibly another 10% if all excess coal is converted to oil at 1:1 ratio. So, theoretically, it is possible to fuel a global mechanized war machine about 20% bigger then the biggest historical war machine. I think this number is fairly reasonable.
Of course, the thorny issue is the distribution of resources, which is the actual problem. This being done, we come back to the first step of the resource distruction, which is defining each resource. The second step is defining relative ratios of materials within a resource. The third step is using the production figures to determine resource production, and making sure totals add up to our predetermined world totals.
By necessity, step two is related to the final testing, where we see if things actually work out to produce desired economic results. I have no magic method to offer here, other then to try out different material ratios and see what we get.
Feedback?
Zerli
APPENDIX
Using Madison's world GDP for 1938, and assuming 1 billion US$ equals 1 IC over 1 year, we get:
1,000,000,000 US$ / 360 days = 2,777,777 US$
Therefore 1 IC = 2,777 million US$ (in 1938 dollars)
I'd like to take that figure and set it down as the correct IC-to-real-money ratio.
