Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

The IOM. The apocryphal claim that the Emperor is the master of "a million worlds" sets a good benchmark for estimates of the possible Imperial population, but what is a million worlds? And has GW attempted to get solid figures on census data? Was Cmdr Kubrik "send in the next wave" Chenkov onto something when he expended conscripts faster than lasbolts? These are fun questions
In constructing some fluff, I came up with the idea of an embattled subsector with a settled population of 300 billion humans. From that I wanted to know how large a local military it could support, what casualty rates were sustainable, and how significant an asset to the IOM 300 billion people actually is. So strap in, it's going to get numerical!


.
The Total Imperial Population: A Strawman Census

You got 1,000,000 worlds. Some of these are Forge Worlds, others Agri, some Hive. All in all, if you count Earth as an "average" human-inhabitable planet, Forge and Agri worlds as just an outsourcing system that trades industry for sustainable population with the other worlds, you can expect each world to be able to sustain a mean population of 5,000,000,000 (5 billion) humans. This gives us a total population of 5,000,000,000,000,000 (5 million billion, or 5 quadrillion). That's a lot. A lot a lot.
The average rate of population growth (calculated as birthrate minus deathrate) on Earth (M.3) is 10.7 more people per 1,000 population, as of 2016. Using this as a non-crisis population growth across the IOM, that gives an extra 53,500,000,000,000 (fifty-three trillion, five-hundred billion) more people per year. If the IOM wanted to maintain a stable population, it needs to expend 53,500,000,000,000 lives in war per year. That's like expending the current world population 7,500 times per year.

But that's just a bunch of zeroes, how silly large is this exactly? A subsector of average population distribution with a total population of 300,000,000,000 would comprise just 60 worlds. That's such a small fraction of the original million that you'd barely spot it in a margin of error when estimating the IOM. That population leaves a spare 999,940 worlds to go. If the whole IOM took a 10% population reduction, it would consume the subsector's population over 1,600 times.

TL;DR:
IOM Worldcount: 1,000,000
IOM Population: 5,000,000,000,000,000
IOM Population Growth Rate (excl. war): 53,500,000,000,000
IOM Maximum Sustainable Wartime Fatalities: 53,500,000,000,000 (or 7,500 M.3 Earths)
=======================
Subsector Worldcount: 60
Subsector Population: 300,000,000,000
Subsector Population (as a proportion of the IOM population): 0.006%
Subsector Population Growth Rate (excl. war): 3,210,000,000
Subsector Maximum Sustainable Wartime Fatalities: 3,210,000,000 (or 0.5 M.3 Earths)



.
Military Eye-Spy: Redshirt Army Edition

The above numbers are impressive, sure, but you can't have the entire population act as active military personnel - there would be nothing left for basic industry or administration. It has been said somewhere, I do not recall where, that a country could keep 10% of its population in the military indefinitely. Most modern countries will never try this, but some come close. North Korea at this time has the largest standing army in proportion to its population, clocking in at 47.4 members per 1,000 citizens, and keeps another 305.7 in "reserve". The IOM functions a lot like North Korea in many ways, but has a much more vibrant economy and far more military branches to maintain. Taking North Korea as a good proof-of-concept, it would not be unreasonable to run the thought experiment of inducting 25 citizens per 1,000 into the Imperial Guard. Just how enormous is this?

The resulting figure is 100,000,000,000,000 (100 trillion) personnel. That's a lot of cannon fodder. 100 trillion personnel is roughly equivalent to inducting the entire M.3 Earth 14,000 times. Taking the same process across our beloved subsector, that gives us 7,500,000,000 (7.5 billion) men. That's world-conquering figures!

Going back to the population growth rate, that figure is not 100-percent recruitable. It may be expendable, but it supplies the same distribution of professions as the preexisting population. If not all of it goes into the military or gets killed, the IOM has some new planets to populate - to the tune of up to some thousands of worlds each year. But that new population does get drafted into the military, and from it we can determine a yearly recruitment uptake. Going from the IOM's 53,500,000,000,000 and recruiting 25/1000, we get well over 1,300,000,000,000 (1.3 trillion) fresh-faced recruits. Applying the same process to the subsector, we get 80,250,000 (80 million, 250 thousand) per year.

For comparison's sake, WW2 saw somewhere around 21,000,000 to 25,000,000 military deaths due to war (including 8-11 million Red Army). Assuming a median total of 23,000,000, and a timeframe of six years (the approx duration of WW2), that gives us an average of over 3,800,000 military deaths per year. Comparing that to the recruitment rates, we can find that the subsector can maintain 21 WW2-level scenarios without reinforcement, and the IOM can take on over 340,000 WW2-level scenarios. This tells us that the IOM can suffer global war on a third of its planets and not care. Grimdank, amirite?

TL;DR:

Assumed Military Induction Rate: 0.025 (25 per 1,000)
=======================
IOM AM Population: 100,000,000,000,000
IOM Recruitment Rate: 1,300,000,000,000
IOM WW2-Level Engagements: 340,000
=======================
Subsector AM Population: 7,500,000,000
Subsector Recruitment Rate: 3,800,000
Subsector WW2-Level Engagements: 21


.
In Closing

The IOM is big. So big that numbers that don't end in at least six zeroes usually aren't worth counting. The IOM is so powerful that it could hypothetically maintain 340,000 planetary scale engagements indefinitely, and still have the manpower to deal with a good hundred thousand other short-term crises. A subsector of 300 billion people sounds like a lot. Like it would be noticeable on a map. It is not. If, in the blink of an eye, all 300 billion souls were enslaved by Dark Eldar, murdered by Orks, or eaten by Tyranids, nobody would notice.

And this was the conservative estimate.

This reinforce the idea that the Imperium may lose a thousand world without even noticing it really. On a side note though, I don't think that the Imperium need a military close to the size of those of Earth during WWII. With orbital bombardment and assault and the fact that the Imperium doesn't occupy territory (it mearly ask for a tithe) armies can be ridiculously smaller in proportion compared to those of Napoleon or Hitler. I would also suspect most Imperial planet to have a median population inferior to 2 billions. The Imperium seems to be structured around a series of massive hive world, forges and fortresses surrounded by agri, feral, feudal and civilised worlds that have a much, much lower importance. These are the panets that can be taken and destroyed without much impact.

epronovost wrote:
This reinforce the idea that the Imperium may lose a thousand world without even noticing it really. On a side note though, I don't think that the Imperium need a military close to the size of those of Earth during WWII. With orbital bombardment and assault and the fact that the Imperium doesn't occupy territory (it mearly ask for a tithe) armies can be ridiculously smaller in proportion compared to those of Napoleon or Hitler. I would also suspect most Imperial planet to have a median population inferior to 2 billions. The Imperium seems to be structured around a series of massive hive world, forges and fortresses surrounded by agri, feral, feudal and civilised worlds that have a much, much lower importance. These are the planets that can be taken and destroyed without much impact.

Regarding WW2 sized armies, I only used WW2 casualty rates to illustrate how bloody rediculous the IOM is - 340,000 WW2-level crises would only balance out to a sustainable recruitment rate. The standing army figure is abstracted down from North Korea's "we expect total invasion any day now" rate of 47.4 active soldiers per 1,000 citizens, to a more moderate 25. That makes it somewhere between Eritrea and Israel, or about double the proportion of soldiers in Singapore or South Korea, and way above the rate of a peacetime nation like the USA.

I would justify this rate for a few reasons. First, the IOM is constantly at war inside and out, and is never sure when the next sector-wide catastrophe will occur. It not only has to export soldiers to frontline areas, which leaves a good deal of soldiers stuck in transit all the time, but also has to police each and every planet to prevent or subdue uprisings. These uprisings could come from Feudal rulers vying for greater power, from ancient (or not so ancient) national identities being mobilised, from political units that wish for a measure of self-determination, or from religious conflicts. These, combined with the constant threat of raids or invasion, and the unending number of military branches that have to be capable of working alone if necessary (coordinated effort being something of a luxury in the grimdark future), results in the need for both large external and large internal forces. Second, the IOM's access to supermassive weaponry is only there to keep it on a level with its external threats - at any time a significant force could be eradicated in just a moment. Due to this the IOM needs to diversify and spread out its forces, each again needing to be capable of launching and seeing out operations on their own. This results in the further need to a vast multitude of large forces in combat or transit or simply stationed somewhere.

Regarding lower median population, that's certainly possible. It depends heavily on the proportions involved, however. If you take a model of the Imperium that sees it in constant and dire threat, the IOM would have a significant imperative to both keep its populations large and keep them well-armed. This would develop some drive to keep the proportion of under-developed worlds quite low. Feudal worlds refer largely to a system of governance, where the issue of local sovereignty has not been consolidated into a single planet-wide political actor (in the same way that feudal Britain eventually developed into an absolute monarchy, with a centralised government, military, and legal system). On the route of agri, forge, and hive worlds, we have some logistical balancing to do. Earth, with a mix of industry, agriculture, comparatively low technology, and non-total planetary domination (nature still wins out in huge chunks of the world) is able to sustain a good 7 billion people. Less would be easier, and more is debatable. A hive world has none of those drawbacks to habitation, resulting in multiple vast "hive" cities that act more like individual nations than our conception of a city. This allows them to have populations well in excess of 5 billion, and could possibly develop habitation for a good 20 billion with polluted wastelands to spare. The questions this leaves us with are: How many forge worlds exist per hive? How many agri worlds is needed to feed a hive of 20 billion (I suspect just one could do it, possibly two small ones)? Does the IOM count resource or death worlds (and if they do count, how do they count)? Are the populations of non-hive worlds relevant?

If you put those together with some degree of common sense I don't think it would be unreasonable to estimate as an example of proportion:

1x Forge World (population statistically irrelevant)
5x Hive World (20 billion population model)
6x Agri World (population statistically irrelevant)
Misc. In-System Resource Bodies (uninhabited masses used for obtaining raw material, aren't counted as part of the "million worlds")
Leaving uncivilised, death, and undocumented worlds as largely irrelevant to the calculation due to being outlier worlds.

That gives us an accounted population of over 8.3 billion people as an average population across the 12 "relevant" worlds. Dubious as a calculation, but it's there.

Yeah, but what about those 1,000 Space Marines!? They're doin' the real work...

 Elbows wrote:
Yeah, but what about those 1,000 Space Marines!? They're doin' the real work...

The million Space Marines would fall within the military population of the IOM but, let's face it, a million Marines to cover a potential 340,000 WW2-level engagements leaves a lot to be desired.

 Selym wrote:
If you put those together with some degree of common sense I don't think it would be unreasonable to estimate as an example of proportion:

1x Forge World (population statistically irrelevant)
5x Hive World (20 billion population model)
6x Agri World (population statistically irrelevant)
Misc. In-System Resource Bodies (uninhabited masses used for obtaining raw material, aren't counted as part of the "million worlds")
Leaving uncivilised, death, and undocumented worlds as largely irrelevant to the calculation due to being outlier worlds.

That gives us an accounted population of over 8.3 billion people as an average population across the 12 "relevant" worlds. Dubious as a calculation, but it's there.

The Necromunda Underhive book throws common sense out of the window.
We're told that Hive worlds often have populations "exceeding a thousand billion people". For simplicity, let's use a thousand billion as an average.
We're told that there are "thousands" of hive worlds. For simplicity, let's assume that "thousands" of hive worlds means five thousand, e.g. 0.5% of worlds.

Using just pages 7 and 9 of Underhive, we get (with an absolutely huge margin of error):
5,000 hive worlds
1,000,000,000,000 people per hive world (1 trillion, or 1,000 billion, or 1 x 10^12).
5,000,000,000,000,000 (or 5 x 10^15) people for just the hive worlds.
Using your assumption that "Forge and Agri worlds as just an outsourcing system that trades industry for sustainable population with the other worlds" (and therefore have statistically irrelevant populations compared to hive worlds), we're actually within a few orders of magnitude of your initial 5 x10^15 estimate of people in the Imperium. Not a bad variation considering how wildly different our sources are.

Nice work! The variables in play then are simply "how many people can we fit into a hive world" and "logistics and resources used to be a problem, until we realised harvesting entire planet systems is actually quite easy".
The Imperium is so far beyond an M.3-Earth-Comparable scale that it just throws reason out the window and gives us statements like:

- "340,000 WW2s wouldn't even reduce the military population by a relevant number"

- "Capturing and colonising 1,000 worlds in a year is only a problem when a Xenos invasion happens"

- "I can't see the bodycount for the population"

- "Chenkov got those medals because he solved an overpopulation problem"

#GrimDank

The Hive Worlds of the Imperium are classed as having upper limit population of 500 billion (3rd edition 40K rulebook, p. 114). The 5th edition rulebook on p. 115 estimates there to be 3.238 * 10^4 or 32,380 hive worlds in the Imperium. If all 1,000,000 worlds of the Imperium were hive worlds with 500 billion people that is 5 * 10^17. That is an UPPER limit for the Imperium's population because we know that not all worlds are maximum population hive worlds.

Minea from that page in the 5th edition rulebook is described as a typical example of a hive world, and it has 154 billion population, well under the 500 billion maximum. Assume we are still generous and give all hive worlds a population of 250 billion to account for less populated worlds elsewhere in the Imperium. 32,380 hive worlds of 250 billion population each is 8.095 * 10^15.

That is more than "500 trillion" for the hive worlds alone.

5 * 10^17 as an upper limit is 5 million billion people, or 500 quadrillion people.

I would take those Necromunda quotes as hyperbole because there are no hard numbers given, and a thousand billion exceeds GW's own population classifications for hive worlds.

The other flow-on implication from the hive world population numbers is that warp travel cannot possibly be as dangerous or unreliable as it is sometimes made out to be, for the simple fact that if it were, then these hive worlds that are reliant on regular food imports would have starved to death a long time ago. They would not have endured for centuries, maintained their population, let alone grow them.

Therefore the amount of interstellar traffic has to increase proportionally.

GW is wisely vague with any hard numbers. So that in an Imperium of a million worlds your subsector can pretty much look like anything you want. Have a agricultural world of just 30 million farmers, working hard to reduce the massive amounts of imports the nearby hive world of 125 billion imperial souls needs (as an example).

Aside from the cornerstones that Iracundus already has pointed out I can give you this calculation I did this very week. To determine what regimental number I want to give my revamped Steel Legion representing one being raised in the latter stages of the ongoing war on Armageddon I did the following:

- 5th Edition Codex Imperial Guard states that a Cadian Infantry regiment has something like 4000-10000 soldiers (4000 soldiers being present at a parade, being less than half of the regiment).

- Minea with 154 billion has a imperial presence of 3,3 (PDF and annual recruitment - assuming that that is for the tithe regiments).

So, in Minea's case they have 0,002% in the military service (with a much larger part employed by the Departmento Munitiorum to actually enable all that).
If that number about Korea is correct, they have 4,7%. The Imperium would have unbelievable amounts of regiments if that was the case (easily by a factor of 1000 more).

If most worlds follow Cadia's military structure (something GW still seems to push even in this edition) I then went on to check some WW2 data.
During WW2 Germany deployed 31% of its population in its military (roughly (22 out of 70) mio) and Russia 18% (roughly (35 out of 197) mio).
Given that Armageddon survived first the largest Waaagh in imperial history and now is a playground for both Khorne and Tzeentch I decided to throw out the usual very low recruitment numbers.

Given the "we are doomed" situation of Armageddon and its strategic value I can imagine that the Imperium issues a "not one step back" order and raises its recruitment like crazy.
Assuming that 25% of Armageddon's population will be drafted for the war effort we get the following:

At a population of 1 billion people we have about 250 mio for military service.

Now for each frontline soldier there are a certain amount of people working in the rear. Mechanics, drivers, cooks, etc.
For a US Soldier in WW2 that was 9, for Germany and Russia I need to draw from memory as I do not have the book with me where I read it. But I think Germany had 5 per soldier and Russia 3 or 4.

Now let's assume the situation on Armageddon is truly dire and the Imperium can only afford 4 rear line guys for each soldier and we get a potential number of 5000 regiments per billion hive world inhabitants.

(At this point I had to laugh at GW inability with numbers, given that Armageddon was supposed to be super duper ready for Ghazghkull's second attempt and fielded a whopping amount of 155 regiments. Yeah... truely all-out war.)

Now given that Armageddon is one of the Imperium's more important Hive Worlds and named alongside Minea and had the highest tribute rate prior to the third war for Armageddon it is safe to assume it has something between 100-300 billion inhabitants. And the numbers can be adjusted accordingly.

If you believe that GW's number are correct as stated in Codex Armageddon however (155 regiments at the start... so 1550000 soldiers...) that'd be half of what Minea raises every year... without being threatened by the largest Ork Waaagh known to mankind.

TL;DR GW has some serious issue with numbers, which is both a curse and a blessing.

Very interesting thread. The numbers given above for Armageddon are very interesting, and suggest that 750,000 regiments at least could be raised from the populaiton, given a regimental size of 10,000 troops (I suspect the Militia regiments are probably more in line with the Valhallan regiments in size, so more like 100,000 per regiment, but that doesn't matter too much).

I have assumed a population of 150 billion equivalent to Minea, but frankly Armageddon is probably on the upper limit of population for hive worlds, being a very important world in the Segmentum Solar, where the most densely populated planets are. Considering the likely huge losses in life from the Orks opening salvoes upon the planet from orbit, downgrading the numbers somewhat doesn't seem unreasonable.

That gives a staggering 7.5 billion guardsmen-under-arms that could be supported in combat by Armageddon, more than the entire population of Earth today. The Imperium in disaster mode should be truly terrifying.


As a side note, there is no possible way to reconcile this with 155 regiments, unless Armageddon militia regiments are ridiculously oversized on the order of several million troops each. The only other way I could see this working is if most of the militia regiments are not formed up prior to the war beginning, and only the regiments on active garrison are listed. Once the attack commenced, they could then have mobilised thousands more militia regiments to respond to attacks as needed, but prior to that they kept them working in the manufactorums and so on to prepare for the invasion.


Automatically Appended Next Post:

Iracundus wrote:
The other flow-on implication from the hive world population numbers is that warp travel cannot possibly be as dangerous or unreliable as it is sometimes made out to be, for the simple fact that if it were, then these hive worlds that are reliant on regular food imports would have starved to death a long time ago. They would not have endured for centuries, maintained their population, let alone grow them.

Therefore the amount of interstellar traffic has to increase proportionally.

I think this is definitely true in general, and things like how the Inquisitors move around in the Dan Abnett series would be much more risky and less routine if interstellar travel was always risky.

However, the Warp is not a fixed thing, and we know it responds to emotions. So a large warzone, with lots of anger and fear and pain and death, likely churns the Warp up something nasty and probably impacts the safety of Warp travel to that system. I reckon the way Warp travel is described as being so unreliable largely only applies to trying to reach large warzones (and the odd other patches of unusually disrupted Warpspace).

Of course, this probably doesn't change that communications and timings are difficult to predict, so Hive worlds probably have to keep fairly large stockpiles of resources in case the next few jumps all arrive late due to some weird time delay.

Otherwise we would never get Imperial Navy battleships that are still in commission after 10,000 years or more- the odds of that happening over thousands of Warp jumps is ridiculous as described, yet the Navy doesn't have a crazy attrition of capital ships.

 Haighus wrote:


I think this is definitely true in general, and things like how the Inquisitors move around in the Dan Abnett series would be much more risky and less routine if interstellar travel was always risky.

However, the Warp is not a fixed thing, and we know it responds to emotions. So a large warzone, with lots of anger and fear and pain and death, likely churns the Warp up something nasty and probably impacts the safety of Warp travel to that system. I reckon the way Warp travel is described as being so unreliable largely only applies to trying to reach large warzones (and the odd other patches of unusually disrupted Warpspace).

Of course, this probably doesn't change that communications and timings are difficult to predict, so Hive worlds probably have to keep fairly large stockpiles of resources in case the next few jumps all arrive late due to some weird time delay.

Otherwise we would never get Imperial Navy battleships that are still in commission after 10,000 years or more- the odds of that happening over thousands of Warp jumps is ridiculous as described, yet the Navy doesn't have a crazy attrition of capital ships.

Tfw you get redpilled on the warp travel narrative by a probable heretic

Ships is the big issue. All well and good having a billion people trained as soldiers, but that's just a problem waiting to happen if you're keeping them as PDF because you can't move them off-world.

Hive worlds will need to colonise most of their home system, they need Agriworlds to bring in goods (and probably to take waste) but they're probably also going to have to be able to feed at least half their world via orbital in-system facilities less you tear the warp the shreds at the edge of your system - (not to mention the demand this places on Navigators).

So if an Imperial capital system is strong dependant on local, off-world facilities to keep it functioning, you don't need an army to take it - just a few pirates and enough system craft to mess with their logistics.

A hive world's fighting force is going to focus on this and civil containment. Their PDF know that they're never going to have to fight aliens, their job is to put down apocalyptic uprisings and cull the population in times of famine. Their elite PDF/IG regiments specialising in holding/retaking orbital facilities with the minimum of disruption to the infrastructure (at the expense of lives).

WW2 level engagements would be rare, it's unlikely you ever need to hold that much ground (unless fighting against Tyranids). Easy to see how small armies with massive stakes would be the norm.

Tastyfish wrote:
Ships is the big issue. All well and good having a billion people trained as soldiers, but that's just a problem waiting to happen if you're keeping them as PDF because you can't move them off-world.

Hive worlds will need to colonise most of their home system, they need Agriworlds to bring in goods (and probably to take waste) but they're probably also going to have to be able to feed at least half their world via orbital in-system facilities less you tear the warp the shreds at the edge of your system - (not to mention the demand this places on Navigators).

So if an Imperial capital system is strong dependant on local, off-world facilities to keep it functioning, you don't need an army to take it - just a few pirates and enough system craft to mess with their logistics.

A hive world's fighting force is going to focus on this and civil containment. Their PDF know that they're never going to have to fight aliens, their job is to put down apocalyptic uprisings and cull the population in times of famine. Their elite PDF/IG regiments specialising in holding/retaking orbital facilities with the minimum of disruption to the infrastructure (at the expense of lives).

WW2 level engagements would be rare, it's unlikely you ever need to hold that much ground (unless fighting against Tyranids). Easy to see how small armies with massive stakes would be the norm.

I am in the process of working out how much needs to be shipped in to support ground forces logistically, and I am then going to extend that to supporting the Hive world as a whole. Suffice to say, it is going to be a lot! Interestingly, food supplies are a pretty small component of modern military logistics, with ammo and fuel being the two primary concerns. This is in stark contrast to ancient armies, who had food supply as their primary logistical concern.

I would also contend that Ork and some Chaos invasions also require very large forces to oppose them, as they will engage in large scale engagements on the ground. Against the other main factions, having a lot of troops around probably just makes it more likely you have some ability to respond to the far greater mobility of the enemy, so I would still contend it is useful. You cannot easily predict where a Dark Eldar force is going to attack, so defending strategic locations across the planet with a large number of troops is a must. Granted, it is not likely to be as large as the numbers needed to counter Orks, Tyranids and large uprisings.

Imperial freighters are like ships prior to the last century and a half- they carry armaments and unboard weaponry, and are armoured for combat. I think it would take more than a few additional pirates to seriously threaten a Hive world, because the merchant shipping already has to deal with piracy as default.

Where does the stuff about too many ships translating into a system causing Warp destabilisation come from? I've not seen anything on that before, and there is a hell of a lot of space that a ship could arrive into on the edge of a solar system. The Sol system, for example, is pretty much fully developed, and I imagine has minimal in-system agri resources, as well as huge military, pilgrim, industrial and administrative inter-system traffic, yet Warp destabilisation is never mentioned as an issue for the Sol system.

As far as I know ships have pretty much no effect on the warp other than at their entry/exit point, where they open a hole in reality. Given that you could happily have a good few dozen of those opening in Earth's orbit and not even be able to see them, the presumably hundreds of thousands of supply ships translating into and out of Sol each year wouldn't really matter.
Also gotta bear in mind that Sol has the following warp phenomena while still being stable:

- Hundreds of thousands of supply vessels to Terra each year
- Endless convoys of rotating soldiers
- *Daily* arrival of Black Ships carrying untrained and possibly quite potent psykers
- Pan-galactic naval orienting from the Golden Throne / Astronomicon
- A greater population than probably any other system in the IOM
- Endless coming and going from the Mechanicus on Mars
- Imperial Fist mission launches (and possibly a few other chapters, too)
- And naval patrols to/from nearby systems, in all likelihood

That's a lot of warp that din du nuffink

Hive worlds do produce some food themselves, but it is just woefully inadequate to maintain their populations. The bodies of the dead are recycled producing the ubiquitous "corpse-starch" rations. Necromunda Underhive makes mention of liquified (human) waste being used to feed fungal farms. The nobility of the upper spires might have greenhouses etc... However the hive world as a whole is still dependent on food imports. If it is actual recognizable food then it is for the upper classes. If it is processed rations or raw material to make processed rations, then it is for the general populace.

Chris Wraight's Custodes books described Terra as being utterly dependent on imports, to the point where an Inquisitorial imposed quarantine barring ships from landing for even a short period started to cause starvation. Of course it would have been the poorest segments of society starving first.

A Hive World system would be absolutely buzzing with ships. These would count as System Ships or Monitors in BFG rules, being ships without warp capacity. They would suffice however for intra-system escort duties or transport between any intra-system colonies or facilities.

That all makes sense to me. In the information below, I have basically assumed Armageddon is receiving all of it's necessary food from off-planet.



So I have done some further research on the Armageddon situation (with the assumed population of 150 billion). Here are some thoughts and rough estimates on the basic supplies needed to maintain a total war footing of 150 billion people.

In my research, it seems that modern units (from WWI onwards) seem to need a bare minimum of 100 tons per day per division (about 10,000-15,000 soldiers, so roughly analogous to an IG regiment. I will assume they are the same from henceforth). That basically includes their food requirements and basic ammunition for defensive operations. Any movement demands more resources, and operating in areas without easily available water increases the supply need greatly too. Intensive operations, particularly artillery use, also massively increases ammunition expenditure. So basically, a division simply holding it's ground, with light firepower in a wet climate, can be sustained for 100 tons per day.

It seems average for divisions in WWII was somewhere more around 300-400 tons per day, with offensive operations extending up to 800 if a lot of artillery was used, or if the units were operating far from railheads/ports. In fact, distance between the railhead/port and the front seems to generally be the greatest limiting factor in logistics- the further you move from the off-loading point, the more fuel needs to be used for the trucks transporting the equipment, and the less there is available on arriving at the destination. This is what halted the advance of Patton in Normandy- he outran his supply lines as the trucks simply couldn't supply enough fuel from Cherbourg with how far they were travelling.

As a side note, distance from off-load point to the front is not the only limiting factor. Rommel in North Africa basically never really stood a chance of capturing Alexandria in Egypt, both because of the distance he had to move across the desert from the ports in Libya, but also due to the absolute maximum amount of tonnage that could be unloaded at those ports. Simply put, with both issues combined, he could never reasonably transport enough resources into Libya to support a sufficiently large army across the desert into Egypt once the British reinforced enough. By contrast, as he pushed the 8th Army back, thy got closer to their supply point, and could reinforce and resupply much easier. They also has a much large shipping capacity from Egypt, so could sustain a much larger force across the same desert in the opposite direction.

I looked at modern forces- a modern US armoured division consumes up to about 2300 tons during intensive offensive operations, and about 1500 in intensive defensive operations. A US armoured division being about the most resource-intensive unit I can think of, I think this probably serves as about the upper limit for modern forces. The possible exception is the logistical support needed for an amphibious assault.

This shows a general trend of increasing logistical needs as time goes on and unit firepower and motorisation/mechanisation generally increases. Hypothetically, this could be projected to consistently increase, which would leave us with astronomical resource needs for 40k forces. However, I don't think this is the case. I think the maximum resources needed by a unit will peak. Why?

Well, first of all, the fuel needs have increased as units became more motorised/mechanised. Modern armoured division units are fully mechanised, which makes them the most fuel-hungry troops... but they can't give them more armoured vehicles, so their fuel needs cannot increase for that reason. There could be a greater involvement of helicopters maybe, which adds more fuel, but US divisions are pretty saturated already. So what else can change? Fuel efficiency. Now, this could go either way. The hummer has got less efficient in recent years, despite the engines ebing unchangd, because they have added more armour in light of ambushes and IEDs. So up-armouring and adding more weight to vehicles could decrease efficiency. On the other hand, engines and fuels have generally got more efficient (the exception being the M1 Abrams, which has a right guzzler of an engine just so they can have a multi-fuel, powerful plant). I think that we are close to peak inefficiency, and form this point vehicles are only going to get more efficient. Obviously, over the service life of an individual vehicle, like the hummer, it is going to get less efficient, but the next hummer replacement should be more efficient for the same capability. I think that this means military vehicles in the future are only going to becmoe more efficient, and fuel needs will drop from the present requirement. The only other factor that could impact this is faster vehicles, but that is hard to quantify.

The area that is more difficult to judge is ammunition. The weight of ammo used has consistently increased since ancient times, and could be projected to increase. The main limiters on this nowadays are basically what a soldier can reasonably carry on themselves for an operation, and what can be reasonably supplied to artillery as they fire it at the enemy. Artillery fire rates have increased, but I don't think massively so on the whole. Infantry has obviously become much more ammo-hungry, but the ammunition is generally lighter, so they can carry more. They probably still need to be resupplied more. I think it is reasonable to assume this will continue to increase as logistical capacity increases, and perhaps as robotic squad equpiment mules become a thing, but it probably won't be able to increase hugely.

The difficulties of applying this to 40k:

1) Fuel:
Imperial Guard vehicles seem to run on anything, so their adaptability is good, but their efficiency is hard to predict. On the whole, it seems Imperial vehicles are much more efficient than modern ones, in part perhaps due to promethium. I would say fuel needs are no more than they are for the modern armoured division, and quite possibly less. In addition, Armageddon is a highly developed world, and the vast majority of the fighting will be in Hive cities and industrial complexes, where transport is good. Therefore the resupply distance is likely short, further reducing fuel costs. Most of the units in the Hives are also likely to be Hive Militia, and probably not mechanised. This further reduces fuel needs. They are likely still motorised, especially with civilian equipment, but frankly the Hive likely has a public transport system which is sufficient to allow a great degree of strategic mobility to defending militia units, and they don't need offensive mobility for defending from Ork attacks. That is what the Steel Legion units are for. On this note, resupplying units out in the Ash Wastes is likely to be considerably harder. Furthermore, these units are much more likely to be mechanised and carrying out offensive operations to strike Ork staging areas. These are the units most likely to approach the US figures for fuel. They are also almost certainly a minority of the overall forces. Therefore, I think I will assume a lower fuel cost than the US division overall. Fuel constituted about 1000 tons of the total weight per day for the division, so I am going to half this to 500 tons.

2) Ammunition:
Missiles, heavy bolters, anything else firing a solid munition, are all roughly analogous to their counterparts today, just supposedly more effective 39,000 years in the future. In particular, artillery is basically the same, and artillery accounts for most the ammunition expenditure of units. We know the Imperium luuuuuuvs artillery, but then most the units on Armageddon will be militia units, likely without too much access to heavy artillery. Armageddon is a heavily industrialised world with a large production of military equipment, so it can likely still provide a fiarly heavily armed militia, but lower than you'd expect for regular Steel Legion units. Therefore we can assume the ammunition needs to be a bit lower than you would expect due to the limitations of reserve troops, but not massively as if it was a world with lower industrial output. Now, the big changes- energy weapons, in particular lasguns, have much lower ammunition requirements. You basically don't need to resupply lasguns if they are in low-intensity fighting, and even against Orks there will be low intensity fighting. This significantly reduces the ammunition needs. However, as artillery still forms the bulk of the ammunition expenditure, lasguns won't make that much impact. Take the two factors together, and I think we can safely give quite a low ammunition expenditure compared to todays units, probably happily at WWII levels despite the relatively high firepower of 40k units. The modern US armoured division is projected to use up to 1200 tons of ammunition in maximum intensity fighting. WWII divisions were expected to use somewhere closer to 100-200 tons per day, maybe 300. I think 300 tons is entirely acceptable with the considerations above.

3) Other supplies:
This includes food, medical supplies, maintenance parts and equipment, replacement gear and so on. Some of this is non-military stuff like cigarettes and things (or lho sticks for our 40k counterparts). Being the Imperium, I am not going to devote any dedicated resources to these- they will have to be bartered/pinched/scavenged from civilians. The food is basically going to be the same, I can't be bothered to try to work out if they have some kind of super energy-dense rations, and the medical supplies and replacement gear are likewise going to be similar. Maintenance parts are hypothetically less needed, due to the increased reliablility of Imperial equipment. I am going to leave it the same though, because this is a small part of the overall total, and I think the differences are small. 50 tons per day for the lot, about 25 tons maximum for the food alone. Soldiers need more energy than people not fighting, so they have a bit more in terms of food than civilians (often around 4,000 calories per day). Note this is for the modern regiment- WWII regiments seemed to eat 35 tons per division, rather than 25. I suspect this is primarily due to improvements in packaging- much WWII food was canned, rather than vacuum packed in plastic, so that would increase the weight a lot. 35 tons vs 25 tons is a fairly huge increase though! I am going to use the lower, modern values.


This gives us a total of 850 tons per day per regiment (~10,000 troops) in active combat. I believe this includes the resources of all the units involved in supplying this, but not of inactive troops in reserve. Just to be sure, I will assume the logistics troops are not being fed in this total, and will count them later.

Sources for the above:
This article from the US army: www.dtic.mil/dtic/tr/fulltext/u2/a241430.pdf
Lots of info from this thread: forum.axishistory.com/viewtopic.php?f=46&t=90657


Now, we can work out the requirements from our 37.5 billion troops we have raised on Armageddon for the Ork invasion. There are 7.5 billion frontline soldiers. I have done some reading, and it seems Allied units in WWII operated a 2/3rds frontline, 1/3rd reserve policy, at every level, although many of these levels would be entirely within enemy range. So two squads would be forward with one behind, two platoons would be forward, with one acting as the reserve (but still within combat range), two companies forward, one in reserve etc., all the way up to corp level. This works out as an individual soldier spending about 50% of the time on the frontline (when counting company reserves and below as in the line, due to being within firing range of the enemy still). This basically allowed a sufficient level of reserves to respond to unexpected enemy action- for example, the Battle of the Bulge was entriely contained using reserves deployed in this manner. Applying this reserve structure to Armageddon, we have 375 million guardsmen on frontline duty at any one time.

This means we need 318,750,000 tons of supplies per day to keep these units operating in the frontline. The reserve troops mainly only need the food. This being a total war situation, I am going to count them as civilians in terms of expenditure needs when calculating the food imports needed to sustain Armageddon. As Armageddon is basically a giant military factory, I am going to assume the civilian population is reasonably important and continuing to produce weapons and gear for the frontlines. However, this gear is broadly analogous to the ammunition and maintenance of the frontline regiments, so I am mainly going to include that within the supplies needed for the troops. Therefore food is the primary civilian concern. This probably doesn't include vehicle production, which we know Armageddon is famous for, but I don't really know how to reasonable calculate the raw material imports needed to sustain the industry. In large part because I don't know how many tanks Armageddon produces, and because I don't have an easy way to project the battlefield losses of troops when fighting Orks, because this is likely to be significantly higher than casualties sustained when fighting humans.


What food does Armageddon need on the whole? In WWII, troops used about 6Ib (2.7kg) of food per day. This is the basis of the 35 tons per day per division above. For the modern figure, it would be about 1.9kg per person. British civilians during rationing had about 1.1kg of food per week. Despite this, they had ~3000 calories per day, and the lower weight likely comes down to significantly less packaging with fresh foodstuffs. Hive food is unlikely to be fresh. Civilians likely wouldn't need so many calories as soldiers, but then they are probably working heavy, manual labour too- therefore 1kg per day is probably reasonable for them (I am assuming less packaging than needed for soldier ration packs, but more than wartime Britain).

So, this gives us a projected food need of 1.9kg per day for soldiers, and 1kg per day for civilians. That gives us 176,625,000 tons per day of food for non frontline soldiers and civilians, and a total minimum resource requirement for Armageddon of 495,375,000 tons of supplies per day. Armageddon seems to only have domestic fuel production, and that is likely used for domestic industry, so we can assume that 500 million tons of supplies need to be shipped to the planet each day for it to survive. Wow! That is a lot of stuff.

Note that this does not include resources for building more gear- weapons and vehicles and other equipment, for which Armageddon would need further millions of tons I am sure. This is much, much harder to estimate I believe. In addition, additional regiments being transported to the planet are not included (this covers some of the material and soldier losses). However, both of these factors will take up available transportation and convoy protection services.

I also have no idea of the stockpiled resources on Armageddon prior to the war, so it may be that it can sustain itself for a reasonable time before requiring external supplies, as the war was expected. The fleet will probably have to be considerably larger than the minimum level I estimate, due to likely combat losses on traversing the system with a heavy Ork fleet presence.

I will probably have a look at what kind of fleet is needed to transport these supplies when I get chance.





Automatically Appended Next Post:
Hang on, I've made an error in the calculations half-way through. Editing them to be correct.

Done. If anyone was wondering, I forgot to divide the active troops by 10,000, which basically meant I assumed each individual guardsmen was using 850 tons per day! This increased the reasources needed to over 3 trillion tons per day...

 Haighus wrote:
I would also contend that Ork and some Chaos invasions also require very large forces to oppose them, as they will engage in large scale engagements on the ground.

If 40K had anthing to do with realism those Chaos and Ork forces would be so badly outnumbered and woefully undersupplied that there's no need to call for reinforcements. If the hive world needs X amount of crap per day for their PDF and militia an attacker needs more, as they're attacking and all. Sure, they can capture some stuff - but they'd still have to have some pretty hefty stockpiles with them, cutting down the number of troops they can bring. Which ofc are already pitiful because GW and numbers. Armageddon mustering 7.5 billion troops? Are there enough CSM alive to march through them all?

The IoM has about a million marines, take or give a bit, so if all attacked Armageddon they'd still lose if they didn't get their average kill/death ratio up to 7500-1. I'm sure it doesn't take 7500 men to kill a marine (or ork, or CSM), more like ten - or let's say 50 because they're just militia and might lack heavier support. So 7500/50=150 - an attacker going in with less than 150 million troops isn't winning. But ofc, in the fluff a squad of marines or a regiment of heroic troops can conquer planets...

Can I just question a fundamental assumption here? I don’t think that Forge Worlds count in the “million worlds” figure–the Adeptus Mechanicus is, technically, an independent allied nation, not a part of the Imperium proper.
Yes they get a sweet deal (how many modern nations would allow an ally to nave a permanent place on its ruling council/parliament?) but then humanity in the form of the Imperium is basically their only hope for extended survival. Regardless of that though, they still maintain home rule and don’t tithe but rather negotiate contracts with Imperial bodies for resources in exchange for technology etc. so I’m pretty sure the Administratum doesn’t include them in its headcount. Or worldcount. This may also extend to Mechanicus-aligned Knight Worlds.

Spetulhu wrote:

 Haighus wrote:
I would also contend that Ork and some Chaos invasions also require very large forces to oppose them, as they will engage in large scale engagements on the ground.

If 40K had anthing to do with realism those Chaos and Ork forces would be so badly outnumbered and woefully undersupplied that there's no need to call for reinforcements. If the hive world needs X amount of crap per day for their PDF and militia an attacker needs more, as they're attacking and all. Sure, they can capture some stuff - but they'd still have to have some pretty hefty stockpiles with them, cutting down the number of troops they can bring. Which ofc are already pitiful because GW and numbers. Armageddon mustering 7.5 billion troops? Are there enough CSM alive to march through them all?

The IoM has about a million marines, take or give a bit, so if all attacked Armageddon they'd still lose if they didn't get their average kill/death ratio up to 7500-1. I'm sure it doesn't take 7500 men to kill a marine (or ork, or CSM), more like ten - or let's say 50 because they're just militia and might lack heavier support. So 7500/50=150 - an attacker going in with less than 150 million troops isn't winning. But ofc, in the fluff a squad of marines or a regiment of heroic troops can conquer planets...

First of all- Orks clearly can- they are listed as deploying forces in the billions. How their logistics cope with this is a mystery, but they are capable of transporting the sorts of numbers to threaten a Hive or Forge world, which means a LOT of Orks.

Secondly, when I said Chaos, I didn't just mean Chaos Marines. Obviosuly Daemons have the potential to swarm a planet, but even aside form that large Chaos forces generally have a huge component of Lost and the Damned cannon fodder who do most of the dying and fighting, with the Marines acting as elite shock troops in support.

If Marines fight without such support, they don't engage in massive slug-fests. They target key areas- decapitation strikes on planetary governors, sabotaging supply depots, knocking out transport links and so on. They use their massively greater mobility to avoid most of those troops- Marines are all about bringing a locally superior force to bear against a small, key section of the enemies' defences. Is having the huge number of troops therefore useless against Marines? I would say they are still useful because more troops makes it easier to guard every strategic point against an enemy where you don't know where they will next strike. If you have troops everywhere, finding the poorly defended weakpoint becomes more difficult. The same is true when fighting Eldar of all stripes, and to a lesser extent Tau. I am not so sure about the numbers Necrons can bring to bear, but they generally seem to operate on the strategic strikes principle regardless.

This means that apocalyptically scaled battles are only likely against Orks, Tyranids, Chaos (Daemons and Lost and the Damned), and renegade humans (and certain other xenos, like Hrud), whereas the numbers are mostly used just to reduce the strategic mobility of elite forces like the Tau, Marines, Necrons and Eldar. The numbers are still useful eitherway, but they don't stop either from being possible.

Mr_Rose wrote:Can I just question a fundamental assumption here? I don’t think that Forge Worlds count in the “million worlds” figure–the Adeptus Mechanicus is, technically, an independent allied nation, not a part of the Imperium proper.
Yes they get a sweet deal (how many modern nations would allow an ally to nave a permanent place on its ruling council/parliament?) but then humanity in the form of the Imperium is basically their only hope for extended survival. Regardless of that though, they still maintain home rule and don’t tithe but rather negotiate contracts with Imperial bodies for resources in exchange for technology etc. so I’m pretty sure the Administratum doesn’t include them in its headcount. Or worldcount. This may also extend to Mechanicus-aligned Knight Worlds.

This is a very good point, although I was contend that the Adeptus Mechanicus and Imperium are more like a political union than close allies- something more like England and Scotland in the UK. Technically separate countries, with separate legal systems, but with a unified political government.

That doesn't mean the Forge worlds and other Mechanicus fiefs are going to be included in the Imperium tally though, and I think you are probably right that they are not. Forge worlds are also generally listed as having huge populations too, so that is probably not a totally insignificant addition to the million worlds.

Spetulhu wrote:

 Haighus wrote:
I would also contend that Ork and some Chaos invasions also require very large forces to oppose them, as they will engage in large scale engagements on the ground.

If 40K had anthing to do with realism those Chaos and Ork forces would be so badly outnumbered and woefully undersupplied that there's no need to call for reinforcements. If the hive world needs X amount of crap per day for their PDF and militia an attacker needs more, as they're attacking and all. Sure, they can capture some stuff - but they'd still have to have some pretty hefty stockpiles with them, cutting down the number of troops they can bring. Which ofc are already pitiful because GW and numbers. Armageddon mustering 7.5 billion troops? Are there enough CSM alive to march through them all?

The IoM has about a million marines, take or give a bit, so if all attacked Armageddon they'd still lose if they didn't get their average kill/death ratio up to 7500-1. I'm sure it doesn't take 7500 men to kill a marine (or ork, or CSM), more like ten - or let's say 50 because they're just militia and might lack heavier support. So 7500/50=150 - an attacker going in with less than 150 million troops isn't winning. But ofc, in the fluff a squad of marines or a regiment of heroic troops can conquer planets...

Don't worry, almost all sci-fi writers have no sense of scale. In terms of selling plastic minis, GW might be right about pushing marine lines (or not, but whatever), but the Marines on both sides are a statistical irrelevancy as far as the IOM is concerned. Until someone converts a whole world to chaos, ofc.

 Selym wrote:

Spetulhu wrote:

 Haighus wrote:
I would also contend that Ork and some Chaos invasions also require very large forces to oppose them, as they will engage in large scale engagements on the ground.

If 40K had anthing to do with realism those Chaos and Ork forces would be so badly outnumbered and woefully undersupplied that there's no need to call for reinforcements. If the hive world needs X amount of crap per day for their PDF and militia an attacker needs more, as they're attacking and all. Sure, they can capture some stuff - but they'd still have to have some pretty hefty stockpiles with them, cutting down the number of troops they can bring. Which ofc are already pitiful because GW and numbers. Armageddon mustering 7.5 billion troops? Are there enough CSM alive to march through them all?

The IoM has about a million marines, take or give a bit, so if all attacked Armageddon they'd still lose if they didn't get their average kill/death ratio up to 7500-1. I'm sure it doesn't take 7500 men to kill a marine (or ork, or CSM), more like ten - or let's say 50 because they're just militia and might lack heavier support. So 7500/50=150 - an attacker going in with less than 150 million troops isn't winning. But ofc, in the fluff a squad of marines or a regiment of heroic troops can conquer planets...

Don't worry, almost all sci-fi writers have no sense of scale. In terms of selling plastic minis, GW might be right about pushing marine lines (or not, but whatever), but the Marines on both sides are a statistical irrelevancy as far as the IOM is concerned. Until someone converts a whole world to chaos, ofc.

This is true, although I would contend that Marines have a strategic impact far beyond what their numbers would suggest, or even their tactical capabilities. The ability to basically show up to any planet, and insert a team locally superior in combat power to the vast majority of forces in the galaxy pretty much anywhere on that planet, and then extract them again with generally minimal casualties? That is a huge level of strategic flexibility. I mean, imagine if you could teleport an SAS company wherever you wanted? Iraq war? Basically ended in the first engagement when a squad teleports into Hussein's office. This is the kind of strategic mobility Marines have against the majority of foes, and it allows them to achieve huge force multipliers. Not to mention if they need wholesale destruction, they can bombard from orbit, with enough potential firepower to destroy entire planets if sanctioned. I think the Space Marine fleets are the real strength of the Chapters.

Don't forget orks reproduce at a ludicrously fast rate and might as well have infinite resources considering the amount of scrap produced and them just eating whatevers around, including their own young etc. so they can probably match imperial guard in numbers if you look at all the worlds they control in their various empires and all the various Waaaghs.

 john27 wrote:
Don't forget orks reproduce at a ludicrously fast rate and might as well have infinite resources considering the amount of scrap produced and them just eating whatevers around, including their own young etc. so they can probably match imperial guard in numbers if you look at all the worlds they control in their various empires and all the various Waaaghs.

Orks are always said to outnumber all other sentient species of the galaxy combined. Fluff's just usually from imperial perspective so we never really get shown how things are beyond Imperium or its points of interest, but the galaxy is mostly Ork space. Other species are crammed on strings of archipelagos on the edges of vast seas of orkoid space

Morgasm the Powerfull wrote:

 john27 wrote:
Don't forget orks reproduce at a ludicrously fast rate and might as well have infinite resources considering the amount of scrap produced and them just eating whatevers around, including their own young etc. so they can probably match imperial guard in numbers if you look at all the worlds they control in their various empires and all the various Waaaghs.

Orks are always said to outnumber all other sentient species of the galaxy combined. Fluff's just usually from imperial perspective so we never really get shown how things are beyond Imperium or its points of interest, but the galaxy is mostly Ork space. Other species are crammed on strings of archipelagos on the edges of vast seas of orkoid space

That's pretty grimdark

Utterly enjoyed the supply tonnage calculation. Thank you for that Haighus. And from what I can gather from the GW source materials, the ratio of militia to Steel Legions (mechanized infantry) is about 5:1.

A few more points to ponder for you (and honestly myself). And just so that is has been said: I honestly doubt that anyone at GW has spent too much thought on all this. True enough Sci-Fi generally suffers from a lack of understanding of the true scales of space. GW so far keeps on being vague about specific numbers. Saves them from spending serious thought, but at the same time robs themselves of any sense of scale (those ridiculous 155 regiments).

The hive world of Armageddon, I assume, will be ranking up among the more populated worlds. Likely around 300 billion. If not more. And while that would instantly grant us, more or less, twice our world's entire population in arms, actual frontlines troops, with the supply handled by the Departmento Munitiorum, the hive world won't raise its max potential number in an instant. Here comes the next problem of writers, besides basic math and that is time.

Often writers throw around long periods of time to make it sound impressive. Completely neglecting both human adaptability and the strain put upon a system by a running conflict. One truth of armed conflict that will be valid in the future, as it does now, always has been and always will be is, that it is expensive. With the estimated amounts of supply by Haighus and apparent catastrophic losses, it will take dozens of equally industrious worlds to keep the conflict on Armageddon running (and no, Voss Prime won't be enough on its own).

And even then, they can only maintain it for so long. The best examples of total war economies are Germany and Russia and there, over the course of the entire war, have been 31% and 18% of the population been employed in military service. (Sadly, am lacking data about Britain) Even America, without having portions of its industry bombed and/or occupied was having serious issues towards the end of the war. So, while I doubt we'll see a million+ regiments raised on a hive world with a sufficient population, it is likely to be in the hundreds of thousands. With the capacity to replenish and refit said units over the course of a war.

We are to believe that the conflict between the IoM and the Orks lasted for decades, before both of them apparently had to team up against demons. With the hindrance in warp travel it is hard to accept that the world is still "doing okay-ish".

And that warp travel to Armageddon had to be quite good in the first place. I really like the points made for the scope of imperial space travel. And have to think that for hive worlds the same geopolitical factors apply as for large cities these days, namely (besides others):
- access to trade routes
- access to fertile lands/natural ressources
- reliable shipping/land/warp routes

Thus, like today for a metropolis, there is a need for hightened imports. So fightings will have to be along spaceports, highways, pipe lines. The life-lines of a hive. But all we get about the capacity of landing craft is along the lines of "a lot".
As precise as the Ork number son Armageddon. What we do know that it was only due to their, rather limited, teleportation technology that the Waaagh could maintain numerical superiority. And once that specific space hulk was gone favor started to swing towards the Imperium not only in space but the planet itself. But that is like the last bit of info before: Demons!

About the assessment of forge worlds... it is my understanding that, like Haighus points out, they are a somewhat autonomous force within the IoM. But nevertheless a part of it. Unless the Codex Adeptus Mechanics states something differently (which I haven't read yet, as I am still trying to catch up with all the new 8th edition fluff), I am not aware that anything states that AM worlds (forge worlds) operate entirely outside of Imperial rule. Thus, I'd have assumed that they are a part of the "millions of worlds" of the IoM. And not extra.

I've been thinking, (dangerous, I know), if the 40k universe never specifies the size of regiments (and we know it is cannon that they vary greatly), and if the IOM necessarily has a Hive World population count in the dozens or hundreds of billions, and if Armageddon is only summoning 115 regiments, and if one or a handful of regiments is all that's needed to conquer a planet (the defences of which vary from none to Fortress World), then is it possible that an Imperial Regiment actually has way more than 10,000 - 100,000 men?

What if the Administratum put some zeroes in the wrong place and started organising regiments on the scale of 1,000,000 - 100,000,000 men?

115,000,000 to 11,500,000,000 men defending Armageddon or 10,000,000 - 500,000,000 attacking a low- to mid-tech world seems much more reasonable than current numerical estimates.

Ignoring, of course, places where fluffbooks state stupid things like "only 10 million or so men were needed at Vraks".

Well, I think we can savely assume that the number of regiments stated in the 3rd Ed. Codex Armageddon are... well, bollocks.

The militarum regimentos are a purely organisational structure and most likely (as stated) vary greatly from planet to planet. Militarum regimentos from Catachan with its population of 12 million citizens are going to be much smaller than one from a hive world.

And even there the differences might be quite significant. As was pointed out before a regiment of mechanized infantry (in Armageddon's case: the Steel Legion) might have way less soldiers than a regiment of ash waste militia. Which makes sense, given the high maintenance cost of a fully motorized combat regiment.

We know only for sure a few things. Like the 5th edition codex gives us the number of a Cadian infantry regiment, numbering something between 8000-10000. More or less. And that size is likely going to be copied by every world who can do so, given how everyone seems to have the hots for Cadia.

And then we know that everything is thrown out the window anyway, once the Imperium has to form battle/army groups to respond to threads. Combining multiple regiments of different nature into an effective combat force.

Which doesn't have a defined organisational structure, rank of command, etc. Probably due to the vast variety within the Imperium.

We also know the Tundra Wolves, a light infantry regiment from Valhalla (which has hive cities) had a strength of 120,000 guardsmen when Chenkov commanded them. This seems like a broadly comparable kind of regiment to something like a hive militia unit.

 Haighus wrote:
We also know the Tundra Wolves, a light infantry regiment from Valhalla (which has hive cities) had a strength of 120,000 guardsmen when Chenkov commanded them. This seems like a broadly comparable kind of regiment to something like a hive militia unit.

I would agree, but it's Chenkov so the literary analogy is too obvious. Normal regiments compared to Chenkov are US/UK/GER WW2 military units in comparison to the Red Army. Which brings the figure back down to 10,000 or less.

And oddly consistent number to be returning to...

It's not that odd. The sources you've been looking at are almost certainly the same ones that the fluff writers were skimming when they were doing their writing. WW2 is also the only recorded modern intensive industrial war situation to refer to, so it makes sense that it is the background.