Introduction



I've been a lurker on these forums for awhile and have gotten a lot of helpful information from the guides and tips here, so wanted to give something back. I used to get stomped by the AI on Normal, but recently won an Elimination Victory with the Broken Lords on Serious due to what I've learned here. This is hardly brag-worthy, but it's to give some context as to where I am with the game and what perspective I'm coming from.



Like any 4X game, your opening has a huge impact on how later turns play out due to the "snowball" effect. Basically, the more FIDS you can squeeze out of every turn, the bigger and faster your "snowball" will grow as the turns begin to pile up. Thinking about this today, I've never quite min-maxed my opening, and often wonder which is better: food first or industry first? Should you focus on food first and grow your population as fast as you can to assign it other places later? Or is it better to focus on industry first, get building, and let your population grow with help from tiles, improvements and luxury resources? I'll share what I've learned when attempting to answer this question in this guide.



Food costs of population growth



The first step to answering this question is to analyze food costs in regards to population growth. There's a great thread here in regards to Science costs, but I haven't seen anything like it for population and food costs. Below is a table breaking down how much food it costs to reach a certain number of population (only the first 10), how much food it costs to hit that mark, and how much of an increase that is over the previous amount required.







If you plot the food increase on a graph, it's not quite exponential, but it's hardly linear. Courtesy of Wolfram Alpha:



Fit to Curve: The actual curve.

Fit to Linear: Does not fit a line very well.

Fit to Exponential: As an exponential curve it's close.



The point is, population growth costs more and more food the more population you have. This creates a "soft cap" of sorts, which means there's no set limit, but around a certain range your population growth will slow down and will take substantial food to push it further. For me on Normal speed in games where I win somewhere between turn 150-250, my cities tend to top out at around 10-12 population. Looking at the table above and the graphs in the links, you can see the curve get steeper around 6-7 population. This becomes relevant in the tests I show you below.



The next step to answering the question of "food first" vs "industry first" is to run a simple test to see how each approach performs.



Test Parameters



The goal of these tests was to create a baseline comparison, so many factors are intentionally left out to control as many variables as possible and keep random chance out of the comparison. While not “realistic” for how an actual play through might go, it could still be helpful to see the difference between “all-in” food versus “all-in” industry.



Map: Default as possible: Pangaea, Normal size, Normal difficulty, Normal speed, etc.

Faction: Ardent Mages (no tile bonuses or penalties, no FISDI bonuses, no "normal" Era I techs unlocked)

Starting city: 9 , 8 , 7 , 4 , 1

* The city is a good test case because food is almost always higher than industry, but they are close enough to each other for a relevant comparison. The high science makes up for Ardent Mages not starting with two “normal” Era I techs like other factions. The dust production is average.



- Played until the build order and tech order are complete and it's Summer, then compared

- No governor; Hero and units kept together to minimize maintenance and used to explore home region, then parked on capital, but not in garrison (will simulate maintenance)

- No luxury resources consumed

- No bonus approval anomaly

- No quest rewards

- No production purchased

- No settlers produced

- No free minor faction population, because pacifying speed varies widely from one faction to the next

- No influence spent on Empire Plan since different play styles will prioritize different plans



"Food first" Test



- All population assigned to food

- Population growth is the top priority in this test case



Build order

- Founder's Memorial

- Seed Storage

- Mill Foundry

- Public Library

- Empire Mint

- 2x Spice Extractors

- 1x Glassteel Extractor

- Sewer System

- Telsem Warlocks as production placeholders



Tech order

- Seed Storage

- Mill Foundry

- Public Library

- Empire Mint

- Open-pit Mine

- Alchemist's Furnace

- Sewer System



I happen to have a "turn log" where I recorded what turns population grew, technologies finished and improvements were built. I was planning on putting it in a spoiler tag so it could be looked at if anyone was interested without making this thread longer than it already is. Sadly this forum can't do spoiler tags, so I just left them out, but if anyone is curious and wants to see how the tests unfolded turn-by-turn just let me know.



“Food first” Summary



Build order completed: Turn 55

Tech order completed: Turn 49

Population: 8 with new population in 9 turns

Population growth at turns: 4, 9, 17, 23, 32, 41, 52

Output: 52 , 20 , 23 , 28 , 3 , +2 Spice/turn, +1 Glassteel/turn

Dust: 326(+8), Influence: 103(+3), 33 Spice, 11 Glassteel, 1 Telsem Warlock finished

If all population switched to production: 11 , 62 , new population in 41 turns



“Industry first” Test



- All population assigned to industry

- Industry output is the top priority in this test case



Build order

- Founder's Memorial

- Mill Foundry

- Seed Storage

- Public Library

- Empire Mint

- 2x Spice Extractors

- 1x Glassteel Extractor

- Sewer System

- Telsem Warlocks as production placeholders



Tech order

- Mill Foundry

- Seed Storage

- Public Library

- Empire Mint

- Open-pit Mine

- Alchemist's Furnace

- Sewer System



“Industry first” Summary



Build order completed: Turn 46

Tech order completed: Turn 44

Population: 5 with new population in 8 turns

Population growth at turns: 5, 14, 21, 36

Output: 17 , 46 , 23 , 28 , 3 , +2 Spice/turn, +1 Glassteel/turn

Dust: 275(+0), Influence: 85(+3), 31 Spice, 9 Glassteel, 9 Telsem Warlocks finished

If all population switched to food: 43 , 20 , new population in 3 turns



Analyzing the “Food-first” vs “Industry-first” tests



* "Industry-first" hit 5 population (almost 6 considering it finished earlier) without a single population assigned to food, and finished faster!

* “Industry-first” finished the tech order 5 turns faster and the build order 9 turns faster

* “Industry-first” managed to produce 9 Warlocks (!) versus “Food-first” which only made 1 Warlock

* By turn 56, “Industry-first” would have had 41 Spice and 19 Glassteel due to how quickly the extractors were built

* The extra industry spent on Warlocks (84 industry each) between technologies could have been spent for a Minor Faction village (120), a Settler (134 + 1 population), or a Borough Street (150)

* “Industry-first” went into negative net dust at times due to all the Warlocks being produced, but these could be sold once the Mercenary Market tech is unlocked. Just be aware of how a large force early on drains your extra dust per turn

* Rather than build settlers, “Industry-first” could have used its Hero + 11 Warlocks to rush a nearby AI depending on difficulty. Rather than build cities, let the AI do it and then just take them. On higher difficulties (Serious and higher) this is practically required for at least one or two AI's if you hope to stand a chance of keeping up

* The biggest advantage “Food-first” had was ending with the option to increase its industry up to 62 due to its 8 population compared to “Industry's First” 46 industry, a difference of 16 industry per turn, which is nothing to scoff at. Era II could have been a very productive era for “Food-first” if it chose to “flip the industry switch”

* “Industry-first” generally seems to have the advantage over “Food-first” considering how many ways you can spend industry compared to food only giving you more population for the sake of getting...more population!

* Each population consumes 2 food regardless of what the population is used for, so food only has a 50% net return (base value, before modifiers), whereas population on industry gets the full 4 base value. In this way, industry, science and influence are “purer” resources when compared to food and dust, which both get deducted for various reasons



Conclusion



If the timing of some of this seems slow to you, you would be correct. Remember that many things were left out (approval anomalies, free population from minor factions, luxury resources, quest rewards, etc.) to create a “clean” baseline free of these more random elements.



In a real game, it is wisest of course to mix the two together, and adjust your population based on what you will get the most out of. However, this test demonstrates the power of industry, and how population growth slows down tremendously around 5-7 population due to the population growth curve discussed above.



In my own games, I usually have 1-3 population on food and the rest on industry for most of the game. My cities seem to hit the population “soft cap” around 10 population, and I finish most games with 10-12 population on my cities.



If you made it this far, I hope this was helpful or educational in some way, thanks for reading! Criticisms and discussion are welcome of course. The next thing I'd like to do is pick a faction (probably Vaulters since I haven't gotten a victory with them yet), and min/max a few different specific build orders based on goals (wide, tall, militaristic) and post it as a discussion here.



TL,DR



All-in industry is usually better than all-in food, especially after 5-7 population. You can neglect food completely if you'd like, but 1-3 population on food is usually plenty.

Good work, this is pretty much what I thought already. The only thing I wonder is if food is better until like 2 or 3 population or if industry from start is always better.



Also the food first seems decent if your goal was to do a dust based economy (using a custom race ofc with businessmen/dust effecency etc that makes dust build viable). Industry is probably still better till you have 8+ dust per person though.

@Isaac: Thanks, glad to confirm for you. As far as 2-3 population on food or to go industry right away, I'd say it depends on how much food vs industry you have after founding your city before improvements. In most regions, food will outnumber production, sometimes by a lot.



Quick number crunch should help:



14 food/turn = Pop 2 in 2 turns, Pop 3 in 5 turns, 7 turns total

10 food/turn = Pop 2 in 3 turns, Pop 3 in 7 turns, 10 turns total

7 food/turn = Pop 2 in 4 turns, Pop 3 in 10 turns, 14 turns total



Era I improvement = 75 industry



11 industry/turn = building in 7 turns, 2 buildings in 14 turns

13 industry/turn = building in 6 turns, 2 buildings in 12 turns

15 industry/turn = building in 5 turns, 2 buildings in 10 turns



Based on this, even if your food per turn starts out as low as 7, you should still assign your population to industry and just prioritize a Seed Storage as the first building. Another nice thing about industry and buildings is their costs don't increase with each one you make. All Era I buildings will cost 75 industry and you get the full 4 industry per pop as opposed to only getting 2 food per pop prior to a Public Granary.



As far as dust vs industry, I haven't looked around to see if anyone's done a thorough analysis, but the quick and dirty is that 1 industry costs roughly 2.5 dust each. I've seen it vary from as low as 2.1 to as high as 2.7 though. In any case, the rule is usually that's it not more efficient to buy industry until you can make 10 dust per population (4 industry x 2.5 dust per industry.) The problem with this is that dust is used to pay for maintenance of units and cities across the empire, while industry directly benefits the city it's created in. The other thing to consider is that industry is only used for industry, whereas dust has multiple uses across the empire (heroes, luxuries and mercenaries to name a few.) Once again, industry is usually the better choice for population. I think to make dust work as an efficient alternative to industry, you need to combo it with bonuses to building and unit cost reduction, as well as having a high amount of dust profit, not just a high amount of dust/turn.

- Rokdog your work is a brilliant demonstration of a great gameplay problem. The curves are clears. As an intuitive player (as most players) I knew that focusing on food is inefficient, but any objective demonstration is useful to show it to sceptic people.



- As a player, there are only two situations where I focus on food : when I am pop 1 and have enough food to go pop 2 in one turn by focusing (without I would stay at pop 1). And when I play Necrophages and I got at least 2 villages pacified + 1 slaver governor (so each pop in food give 2 (natural)+4 (cull the herd) + 2 (slaver) which give me a very strong growth. After that, when I got pop 8-10, I stop and I go for industry (or science if needed).



- In my opinion the consequences of your demonstration should be important on the game design reworking. It proves that the gameplay about food is entirely corrupted by this kind of logarithmic progression :



- Why investing on food citizens, if food became prohibitive ?

Why taking advanced food tech and building, if it's very slightly increase your faction output, compared to any production or science tech and building ?



- If Endless Legend does not change the food and growth mechanism, It will largely reduce the diversity of gameplay, between strategies of the factions. Everybody will tend to only aim the pop 10-12 and after that, will abandon the food gestion, which off course destroy any difference between decisions, tech choices etc.



1. Short comparison between growth mechanism and growth limit mechanism of Endless Legend and another 4X, Civilization 4 (or 5)



- In Endless Legend, the progression of food cost for new population point is near from the exponential curve. But, to compensate it, there is no happiness or health cost malus for these new population.



- In Civilization 4, the progression of food cost for new population point is purely constant. Each new population point cost the same cost than any other. But, to limit the growth, the food only come from the land, never from specialists citizen. So the maximum population sustainable per city is hardly limited by the size and the fertility of the 21 cases near your city. And, there is two others counter mechanisms in the growth : each new population point in the city increase the unhappiness in the city, which cause unworking citizen if you reach the happiness limit of the city (which can increase with building, luxury etc.).



- Conclusion : As Endless Legend cannot and should not change to adept the mechanisms of growth control of Civilization, I think we (we, as persons interested by Endless Legend development and balance) should propose new mechanism to reward the growth and food focus by keeping a close adaptation to any Endless Legend game design. For example, it is not possible in Endless Legend to limit the food output, because the Devs made the choice to create an infinite amount of potential worked hexagone (a city can occupy an entire region, by districts not just only 21 hexagones).



Here is few ideas to give a contribution for a change of the system about growth and population, in the objective to give more advantage to a food focus/large city strategy :



2. Create new stades of city evolution city size quality level



- Theses stades would be : Town(from pop 1 to 5)->Big Town(from pop 5 to 10)->City (from pop 10 to 20)->Big City (from pop 20 to 30)->Metropolis (from pop 30).



- When a city reach the limit of the new phase of evolution, it gains access to new buildings and a further output specialization (food, production, science or dust).



- These new building are, first, the traditionals buildings we can obtain by the obtention of new technology, but are limited to certain development of city (for example, you can build an Arsenal military building only when you got a Big City (from pop 20). Or you can have a Dust Refinery if you are City (pop from 10).



- The others building are, two, special buildings for each factions. For example, lets imagine that : Vaulters : Teleportation Anchor : Need a City (from pop 10). Give the ability to a teleporting army of Vaulters to select this city (this building could too resolve a bit the overpowering teleporting feature of Vaulters). By this, Vaulters could have a big interest to focus on food before production, or to focus on food technologies.



Other example : Drakkens : Throne of the Great Ancestor : Need a Big City (from population 20) : Give to Drakkens access to a new late unit, the Dragon. A strong hero unit, which cannot be used with others units, and what you can only have one at a time. A super unit which can only be used to defend the drakkens land (cannot leave any drakken territory).



Last example : Necrophages : Lair of the Great Worm (don't forget these names are bad, and my English is bad too ^^) : Need a Metropolis (from population 30) : Give to Necrophages access to a new kind of teleport, offensive teleport. A titanesque worm inspired from Dune which can be activated to dig in three turns a tunnel to any point of the map, with a limit of 20 hexagone from the use. Necrophages units can then use the tunnel for 3 turns only. Can only be used one time, kill the worm after, and need a long time passing to recover one worm.



- The output specialization : When a city become a Big Town, then a City etc. it is given to the player the choice to pick an historic specialization output percentage for the city. For example, a town become a Big Town, and gain access to one of these specialization : Food, Production, Science, Dust or Military Units. Food specialization give +20 % food output. As you can have 4 specializations at max, you can have +80 % in food output at max.



The idea behind that, is the specialization of the cities. It is more interesting to manage specialized cities than to have uniformed cities where you can put and unput the needed governor where you want. Off course, it is not enough to specialize the cites or solve the food/growth problem, but it can help.



3. Limiting the number of food specialists according to the fertility of the worked square of the city



- It is very unlogical that any region even arctic region can have any citizen focused on food production. We can admit an unlimited number of industry worker, of scientific or of dust merchant, but not of food I think. Only natural fertile hexagon should be able to sustain larges population, and as the current food system is uniform, any glacial region now could be a Metropolis. So in this idea I think the number of food specialist could be limited by the natural fertility of each hexagon.



Only hexagon with at minimum +1 natural food output would give +1 food specialist slot. It would change that in arctic or desert region, it would be hard to grow by specialist. You'll need specials buildings (as granary or new others) to reach a bigger city, but you ll probably not reach more than the size of a City (ten population point). In others regions, in forest, grassland and others decent fertile land, you ll can reach easily the biggest sizes.



4. Limiting the food bonus accorded by building when the hexagon is not fertile



- Another way to limit the growth in sterile region would be to don't give any food bonus per building when the hexagon is bad in food (so a natural food output of zero). It would not be possible to cultivate potatoes in these regions.



5. Create a more rational food evolution



- If we think that each new population point is just a simple add (of 10 000 residents for example) to the total of population point, there is no reason that this add cost more food than when passing from 1 pop to 2. But if we consider that it is not a simple add of 10 000 but a geometric population expansion, it become more understandable, and it feat with the idea that a city is not more a town, but a big town, then a city etc.



- So, what I want to say it's this exponential progression should be logically explained, and this idea of new quality stade of city expansion look me acceptable. Anyway it should be increase more slowly than an exponential progression. A more rational evolution of the food needed to growth.







I would like to know yours ideas about these propositions, and possibly, a feedback of the devs, to know if our efforts as thinking people can be useful, and used in a certain way, or not.

Bump ! I like your ideas Jojo !