View Full Version : Getting the Most From Your Cities - A guide to Industry and Economy


Gato Loco
Dec 24, 2005, 11:53 AM
I’m posting this because it doesn’t seem to be redundant with Stuporstar’s city placement article. Read his article first if you’re confused about the numerical values of the various terrains and improvements. Also, this article was written based on version 1.09. I won’t experience 1.52 until I get back from Christmas vacation, but I haven’t seen any references to anything that would drastically change the advice given here.


1. Basic City Placement – Resources

Food – An Army Marches on its Stomach
Believe it or not, food is the basic resource in civ. Even though food is itself quite limited as to its uses, it is the basis on which all other goods are produced. The uses of food are as follows:

-Spend on population by letting it accumulate
-Spend on settlers or workers (along with shields)
-Spend on shields or commerce by assigning specialists or working brown tiles
(For the purposes of this article, brown tiles are those that produce less than 2 food and therefore cannot fully support the population point that is assigned to work them. Green tiles are those that produce 2 or more food.)

The sources of surplus food are:

-2 units of free food in every city
-Bonus food resources and floodplains
-farms/pastures/nets

Usually, gaining extra food requires forgoing some other good, as exceptional food tiles seldom produce more than one coin or hammer in addition to the food (the common exceptions being cows (3 hammers) and nets (2-3 coins). This is especially painful for a financial leader as a high-food tile will probably be a tile you cannot get the commerce bonus for. For this reason, seafood is a financial leader’s best friend. Plenty of seafood bonuses will give your cities the needed food surplus while still getting 3+ commerce per tile, something that cannot be done with farms.


Shields – Should I call them Hammers?

Hammers (a.k.a. Shields) are the most directly useful resource. They are converted directly into buildings and units. They can also be somewhat elusive. Like food, shields are highly dependant on terrain. However, unlike food, it is often difficult to get as many shields as you need for a particular city. The sources of shields are:

-Bonus resources
-mines (usually on hills)
-engineers and priests (and citizens in times or desperation)
-forests and plains
-Watermills, windmills, and workshops later in the game
-chopping forests
-rushing

Of these methods, hills will probably contribute the bulk of your shields. Forests and plains give you only one shield apiece, and generally don’t give much else. A grassland forest is usually not worth using if it could be replaced with a grassland cottage, unless you can build a lumbermill on it. Similarly, a plains tile (or a workshop/grassland) will, in the early game, produce exactly one shield, and if not farmed, will consume food. Plains/forests are somewhat better sources of shields if you have plenty of surplus food. However, in the end, there are some city locations which will be chronically shield-deprived until the modern tile improvements start yielding extra shields (guilds for workshops and replaceable parts for mills, to be exact). If you settle such locations in the classical age, you will be forced to budget a limited number of one-time shield infusions from chopping and rushing. Such cities should never even try to produce units or wonders until the discovery of replaceable parts and chemistry allows them to get a respectable production from watermills and workshops. They can, however, be good sources of commerce if surrounded with cottages and given a few commerce-multiplying buildings. They can also run specialists, but make poor dedicated GP farms since they will not have the extra GPPs from wonders. Also, it is generally agreed that all other things being equal, a large expanse of flatlands should be cottaged rather than farmed to get the best commerce return unless it is dedicated specifically to producing great people.


Commerce

Commerce is the most portable resource. It can be produced anywhere, and generally benefits your empire as a whole as opposed to the city in which it is produced (unless the culture slider is turned up very high). A city that does not produce much commerce is nothing to worry about provided you're getting enough commerce elsewhere. The sources of commerce are:

-Bonus resources (again)
-Cottages
-Specialists
-Water tiles
-trade routes
-river tiles
-mills

In addition, gold (a related good which is generally equivalent to commerce in most cases) is gotten from a few other sources:

-Shrines (and certain wonders)
-Pillaging/capturing cities
-Diplomatic deals

The most productive of these sources are cottages and specialists. Every civ will need one or the other of these commerce sources. Shrines, trade routes, and water tiles are also very useful supplements to one’s economy and should be sought out wherever possible. The commerce from mills comes later in the game and is less than that gained from a cottage built in the ancient era, so these improvements should not be regarded as a primary source of commerce but rather as a source of late-game shields that also produces a bit of commerce.

The main uses of commerce are to pay the city/civic/military upkeep bills (as gold) and to do research (as beakers). Secondary uses are:

-Deals with other civs (tax slider)
-Culture and happiness (Culture slider)
-Rushing construction (tax slider and Sufferage)


2. Spotting resource-rich sites

Bonus Resources
The first indicator of a rich city site is the presence of bonus resources. Bonus resources are essential in the early game for making a small city productive. Once your cities get large, the importance of resources diminishes, but your initial cities should always have some usable resources in the border (note that many luxuries are not useable before calendar).

Good terrain - Cottage Green
Good terrain is terrain that can easily be made productive. These terrains include grassland, hills, floodplains, and rivers. With the exception of hills, these tiles all produce at least two food and they can all be improved to provide multiple shields or commerce. For shields, look for hills balanced by river grasslands/floodplains or bonus food resources. Hills without a balancing food surplus will be useless as the city cannot afford to work them and still grow. For commerce, look for high-food river tiles if you plan to make the city a specialist-GP farm, otherwise look for grasslands or floodplains on which to build cottages. Forests and jungles can be cleared without too much effort, so their presence should not deter you from founding in an otherwise promising location. In fact, forests are a life-saver for a commerce-only city as they can be chopped to provide you with a courthouse, library, lighthouse, or whatever other improvements you might need but are unable to build with the city's paltry shield production.

Mediocre terrain – DON’T farm brown
Plains, lakes, and coasts are not particularly valuable under most circumstances and should not be the only accessible tiles in your city radius. A farmed plain will return 2 food before biology (enough to feed the citizen working the tile), one shield, and one commerce if on a river. This population point is not being particularly productive, considering that he could be producing 2 shields or 3 commerce, plus GPPs, as a specialist. True the specialist will slow down growth, but if the best terrain you have to work is a plain, further population won't be very useful. Building a cottage will eventually give a greater return, especially if you’re financial, but only if you have a sufficient food surplus to still work more “brown” tiles after working enough hills to fulfill your hammer needs. Lakes and coasts are somewhat similar for nonfinancial leaders, better than nothing but nothing to get excited about, giving two commerce which will never improve from tech or civics. Financial leaders will get three commerce out of these tiles and should give them higher priority, though still less than grasslands. Desert hills can also fall into this category, and are a last resort for cities with no other good shield source. It is also important to remember that as time passes, you gain better trade routes, civics, wonders, city improvements, and other ways to increase the value of seemingly marginal city locations, so a seemingly bad city location could eventually look appealing later on.

Bad terrain – Worse than Specialists
Desert, tundra, and peaks. Ugh. Try to avoid them unless they contain bonus resources. Tundra forests will ultimately be marginally useful once you get lumbermills, and tundra cottages could be viable but you need a river to be able to build them. That’s all there is to say about these terrains.


Phases of Expansion – How to REX responsibely
The first phase of expansion starts almost immediately (unless you have a special trick up your sleeve such as religion hogging or one-city warrior rushing) and is the land grab phase. At this point, your goal is to cherry-pick the best city sites and establish a defensible border. Every city you found should have immediately useable resources and/or strategic importance. Expansion should halt before you go over the OCN for your map size (9-10 cities on a standard map for version 1.09). You will deliberately leave many tiles unworked because the tradeoff of colonizing them is too great, given limits on city size and number.
The backfill phase begins gradually as you acquire certain technologies that improve the productivity of less important tiles. These are:

-Civil service (farms on non-river spaces)
-Guilds (workshops give 2 shields)
-Calendar (use of advanced luxury resources)
-Machinery (Mills) and Replaceable parts (Better mills)
-Railroad (Extra shields)
-Communism (Less upkeep, better workshop/watermill)

In addition, the building of courthouses in every city paves the way for responsible expansion above the OCN, and the discovery of galleons will open up previously uncolonized islands.
The lands settled in this phase will be those already claimed in your borders but neglected due to unproductivity, as well as recently discovered islands and the now-vacant sites of razed cities.


Overlap – Yes it’s a good idea in this game too
One of the vital lessons every civ3 player had to learn was to love overlap. With Civ4, overlap is seen by some as a bogeyman to be avoided if possible. This is generally a mistake in my opinion. The days of CxC city placement are over by decree from the Firaxis programmers, but some overlap is quite viable. The thing to remember is that it takes a long time for a city to reach size 20 (the size needed to work every tile in the radius). And I mean a long while, because happiness and health limits will take their toll if you even try to grow a size 20 city in the medieval era. In addition, certain tiles aren’t valuable anyway. Don’t bother shifting your city location just to encompass an extra plains tile. Instead, cities should be placed to maximize the number of good tiles that can be worked within a reasonable period of time, a goal which is not incompatible with a CxxxC placement scheme, or even a CxxC if it gets you an extra wheat tile.


Production management – Watching cottages grow
When deciding how to manage your cities, the most important question to ask is what is limiting their growth. Cities can, generally speaking, be limited by food, happiness, or space. At different times in the game, different cities will hit different limits.


Food limits and Population Budgeting – Think before you go into food debt
Food is a very common limiting factor for your cities, and the most complex. A city is food-limited if it is prevented from adopting an otherwise appealing tile/specialist-working scheme because it would lead to starvation, and in which the city size is still beneath the unhappiness cap. This is most likely in cities that work large numbers of hills and plains, or that have large numbers of specialists. Under these circumstances, production assignments that produce less than 2 food (specialist or hill, for example) must be paired with assignments that produce surplus food (farm-grassland) to find the true productivity-per-worker. For example, a grassland-hamlet produces 2 commerce per pop point, while a mined grassland-hill produces 1.5 shields, per pop, as it must be paired with another worker assigned to a grassland-farm who would otherwise be doing some other productive job. A mine-plain-hill would produce 1.33 shields, as it requires two grassland-farms to support it. A farmed plains still produces only one shield per worker, underlining how worthless plains are under normal circumstances. These calculations would be different if the best unused food producer available were a farmed floodplain. The exact number can be found by taking:

[yield]/(1+(2-[food_a])/(3-[food_b]))
[yield] = base yield of hammers or commerce
[food_a] = base food yield
[food_b] = food value of the best unused food tile available, usually 3 or 4

[food_b] must be at least 3 or the equation will result in a division by zero. In game terms this indicates a city that is unable to produce any surplus food and is very limited in its ability to utilize tiles of less than 2 food production..

While making a comprehensive graph of various tile types using this formula leads to a large, unwieldy spreadsheet, a few conclusions can be drawn. First, the ability to work food-poor tiles or hire specialists depends greatly on the presence of floodplains. A city with floodplains and a city with only grasslands should not even be considered as being in the same class, as the floodplain city will be able to work tiles that would be unprofitable for the grassland city. Second, there is a real difference between a tile producing 1 food and a tile producing zero food. If supported by grassland-farms, shields from a 1-food producing tile are 150% as valuable as those from a 0-food tile, making a grassland hill more valuable than a plains hill. If supported by floodplain farms, the difference drops to 133%, and if supported by floodplain-farms after biology, it is a meager 116%.

Population Limits and Tile Budgeting – Specialists are finally useful
A second circumstance is when a city has plenty of food but a low population limit due to happiness, and you are unwilling or unable to increase happiness. This tends to happen early in the game before calendar resources, the culture slider, and multiple religious buildings are available. In this case, maximizing output is a simple task of choosing the best possible tiles without utterly breaking the bank on food. It is not necessary to use up all available food under these circumstances, although it is usually advisable if specialists or plains-hills are available. In fact, this is the circumstance in which specialists hold their own even without GPPs, since food is not a consideration. If a city is managed in this way, care must be taken to rearrange it to allow for growth once the happiness limit is raised.

Tile Limits and Improvement Budgeting – Tear down those farms
Finally, cities can be limited by running out of space. If there are no more worthwhile tiles to work, you have no choice but to keep assigning specialists until the food runs out and the city stops growing. At this point, the most important question is which improvements should be built to make the best use of available food. While farms certainly allow the population to keep growing, they are of limited use if the only role for the extra population is to become specialists. A mature town improvement (making 4+ commerce depending on various factors) is probably more valuable than a farm which will allow half a specialist (for 1.5-3 commerce, depending on civics). Additionally, a tile making less than 2 food should only be worked if doing so gives a greater return per food consumed than assigning the worker to be a specialist. This is probably true of mined hills. It is also probably false for pre-guilds workshops or arctic-furs camps, for example, as a 1 to 1 food-shield or 1 to 1.5 food to commerce tradeoff is equivalent to the value you get from a specialist without representation, but without great people points thrown in.


Cottage or Specialist? – Cottage unless you know what you’re doing
This question should be in your mind as you plan every commerce-producing city. While every civ will use some cottages and some specialists, there is a substantial grey area in which decisions must be made. In terms of raw commerce produced, the totals are:

Cottage – 1-4 unimproved, 1-6 (+1 shield) with appropriate tech and civics
Specialist – 3 unimporved, 6 with Representation

So pound for pound, specialists win under a strictly happiness-limited caste-system economy due to the fact that they start out at full capacity. However, this is a fairly specialized situation (Lots of food, population strictly capped by unhappiness). Under a strictly food-limited situation (High population, lots of happiness, limited food), cottages will produce much more commerce as much of the population in a specialist-based city would be devoted to farming to feed the specialists. When deciding how to use one’s population, it is important to consider the strengths and weaknesses of both specialists and cottages.

Cottages:
Take time to reach full capacity
Food-neutral if placed on grassland
Can produce shields under Sufferage
Require the presence of flatlands to work
Need 2 civics and one tech to get full bonuses
Can be pillaged
No special buildings required

Specialists:
Instantly reach full productivity
Only one civic/tech required to get +3 bonus
Don’t make their own food
Require specialist buildings or an extra civic change (Caste System)
Cannot be pillaged
Independent of terrain
Produce Great People

The great people are the real wild card in this decision. Pacifism, the Parthenon, the National Epic, and the Philosophical trait can all multiply the usefulness of specialists. On the other hand, the cost of GPs will increase as time goes by, making the specialists less useful later in the game. In addition, splitting up specialists into multiple cities will slow down the rate of GP production.

Relevant questions to ask yourself include:

-Which GPs would be useful to me right now? (esp. for prophets and scientists)
-What are my leader traits?
-Will the turning point of this game come in the Medieval or Modern era?
-Is there space to build cottages around the city?
-How many wonders are there in the city? How many more will I build?
-Is there another specialist-farm city already hogging all the GPs?
-What civics do I plan to run in the future?
-How safe is this city from pillaging?
-Will the city’s productivity be limited by food, space, or happiness?

Galumphus
Dec 28, 2005, 10:40 AM
Wow, a lot of good ideas in this post. I'm starting a game today, and will try out your city placement and green/brown ideas.

Cort Haus
Dec 28, 2005, 10:55 AM
Good post, Gato Loco. :b:

I don't know if working irrigated plains is so bad, especially if on a river. A hammer is a hammer, and for a low-prod coastal commerce city moving from 2hpt to 3hpt while still growing is not to be sniffed at.

Gato Loco
Dec 28, 2005, 08:55 PM
Good post, Gato Loco. :b:

I don't know if working irrigated plains is so bad, especially if on a river. A hammer is a hammer, and for a low-prod coastal commerce city moving from 2hpt to 3hpt while still growing is not to be sniffed at.

A hammer's a hammer, but the farm isn't giving you a hammer, it's giving you a slice of bread. Your coastal city won't be building many improvements, period. A lighthouse would be the only essential, followed by a courthouse and library when you can get them. I'd try if at all possible to hurry up the build build by chopping/slavery/cash-rushing depending on the stage of the game. For this reason I usually stay in slavery for quite a long time unless given a specific reason to switch. Serfdom seems to be inferior to simply building more workers and emancipation comes too late for the cottage growth bonus to be really useful. I don't really know about caste system. The typical super-specialist city can generally build enough buildings to support its specialists, but other circumstances could force you to adopt it. Ideally, if you're embarking on a deliberate program of founding hammer-poor coastal cities without access to seafood you should try to run sufferage and pump some gold into those cities. If you're at the point in the game where you've taken all the good spots and are reduced to squeezing in a few more low-yielding towns you probably have the reuired tech.

However you are right that it's better than nothing. I'll irrigate some river plains too if I run out of food before happiness and really don't have anything better to do with the extra happiness. The capiton "DON'T farm brown" refers to a common civ3 rule about mine/farm placement (mine green, farm brown) which no longer applies to civ4 and shouldn't be taken as a blanket dismissal. It's all a matter of what the next best thing is. If you've got grasslands left they'll usually give you more for your population than plains, especially once you get bonuses to your workshops.

In general, I'm not a really big fan of farms (except on grain resources of course). They give you a measly +1 food for most of the game while mines and cottages give multiple hammers or coins. Your best cities will be built next to food resources that will give them a respectable supply of food to "spend" via food budgeting. These cities probably should steer clear of farms unless they're going heavily into specialists or mining or if you're milking and want to artificially run up your population. The place you get into farms is when you're food-limited and want to budget population towards a net food-losing task. This'll generally happen in extremely specialized cities or in marginal resource-free ones later in the game (when you may also be able to get biology), or in cities founded in food-poor areas mainly for the purpose of collecting non-food resources. Just ask yourself what that extra food costs you (by not building a different improvement or working a different tile) and what you'll be able to get from it (by moving another population to a food-poor tile or by growing the city). The real point I was making is that farms are a second-best option and that cities that need them (in the absence of super-specialization) should be classed as second-rate at best, and city spots that can make do without them should be considered more valuable and settled/conquered first.

Quantum7
Dec 28, 2005, 09:38 PM
Nice article. Mostly because it helps forcing me to think explicitly about what's limiting my city right now. Also, the remark about grassland hills actually being better than plains hills is very counter-intuitive (for me at least). Interestingly enough, until 1.000 AD or so it almost always seems to be population instead of food that's limiting most of my cities. Which means that for most of the early ages, plains hills are better than grassland hills.

Some remarks:
- While cities are smaller, you can have two sets of tiles they can work; i.e. a set of tiles with farms & mines (for production) and a set of tiles for commerce. That way you can switch when needed. Variations ofcourse include pure farms (for very fast growth). When it gets bigger, those tiles can be remade into cottages, watermills & windmills. As long as you have enough workers this can be pretty effective.
- I'd *stress* the (initial) importance of rivers even more. Good production / GP city sites with a number of food resources can become mediocre due to the inability to build farms (until Biology or in the best case Civil Service). Making it impossible for it to use some extra hills or build a few more essential farms.
- Flood plains. For a good commerce site they are probably even more important than food resources. They tend to get undervalued a bit as they don't have a special resource.

emancipation comes too late for the cottage growth bonus to be really useful.

I strongly disagree with this remark. In most of my recent (tech-oriented) games I barely have 5-6 towns before Emancipation becomes available. It pretty much became the most important civic possible to increase tech rate ;).

The real point I was making is that farms are a second-best option and that cities that need them (in the absence of super-specialization) should be classed as second-rate at best, and city spots that can make do without them should be considered more valuable and settled/conquered first.

This is i.m.o. indeed one of the most important things to remember. I now judge future city sites based on the number of cottages they can support; unless ofcourse the goal of the city is production / GP oriented.

Gato Loco
Dec 29, 2005, 01:14 AM
- While cities are smaller, you can have two sets of tiles they can work; i.e. a set of tiles with farms & mines (for production) and a set of tiles for commerce. That way you can switch when needed. Variations ofcourse include pure farms (for very fast growth). When it gets bigger, those tiles can be remade into cottages, watermills & windmills. As long as you have enough workers this can be pretty effective.

This is an agonizing tradeoff if the commerce tiles are cottages that can't grow when they're not worked, but if you've got gold/gems/dyes/etc in the commerce position it's all good, especially if there's an overlapping city that can pick up the mines when the main city isn't using them. Yet another reason to love overlap.


- I'd *stress* the (initial) importance of rivers even more. Good production / GP city sites with a number of food resources can become mediocre due to the inability to build farms (until Biology or in the best case Civil Service). Making it impossible for it to use some extra hills or build a few more essential farms.

Right on the mark. You don't miss rivers until you need them.

- Flood plains. For a good commerce site they are probably even more important than food resources. They tend to get undervalued a bit as they don't have a special resource.

Of course. 1 floodplain = 1 grassland rice for raw food output. Someone should really mod the game so those little resource baloons pop up over floodplains too. With a flood plain/cottage you get to have your cake and eat it too. Plus the cottage/floodplain combo leads to a lower population so you won't go over the health limit like you could farming all those floodplains for a GP factory. Actually a corollary of this could be that after obtaining one rice for health purposes you should cottage all the other rices since they're more or less equivalent to the floodplains which you've probably already cottaged. Any good reason they made rice give less food than the other grains?

I strongly disagree with this remark. In most of my recent (tech-oriented) games I barely have 5-6 towns before Emancipation becomes available. It pretty much became the most important civic possible to increase tech rate ;).

Looks like I spoke too soon. I suspect this depends on your expansion plans. Old, mature cities don't benefit much but new colonies would. The question is whether you're better off building cottages or wind/watermills for the younger cities. Thogh when you note is that with Emancitation plus the financial trait newly minted cottages get up to three commerce after five turns, and 5 commerce after 15, I can see where you're coming from. I'd also wonder whether emancipation is really the most important tech-rate civic when compared to free speech, representation, or state property. It all depends on what kind of game you're playing, as well as what era you're in. I suspect what you really mean is that it lets you increase tech rate with little tradeoff as long as you're not wedded to slavery or caste system, unlike the other science/economy civics which require you to forego other useful civics to run them. You could be running police state/vassalage/mercantilism/theocracy and still be growing your cottages with emancipation. (An emancipated police state? :confused: In theory I suppose...)

Thanks for commenting. Any new opinions are welcome.

Quantum7
Dec 29, 2005, 03:32 AM
I'd also wonder whether emancipation is really the most important tech-rate civic when compared to free speech, representation, or state property. It all depends on what kind of game you're playing, as well as what era you're in.

I'd put the question more like this: When wouldn't you want to switch to Emancipation the minute it becomes available? (otherwise you're considering era's in which the Civic isn't even available)

The answer probably is that if you're following your guidelines, you'll have a significant number of cottages and it will be very useful. Slavery / Caste System / Serfdom tend to get less useful the further along you are (with the possible exception of Slavery).

I suspect what you really mean is that it lets you increase tech rate with little tradeoff as long as you're not wedded to slavery or caste system, unlike the other science/economy civics which require you to forego other useful civics to run them. You could be running police state/vassalage/mercantilism/theocracy and still be growing your cottages with emancipation. (An emancipated police state? :confused: In theory I suppose...)

I meant to say that it's the most important civic to keep on maximizing your tech rate. Whether or not it's the most important tech-civic all together is hard to say (or to calculate). It may very well be, considering people winning (should) have 100+ cottages in most standard games.

What might be very interesting is looking at the Civ IV HOF pages. You can (graphically) look at the GPT rate's for all of the top-submitted Gauntlet I rates. Most of the top 5 tried to get to Democracy as quickly as possible to maximize tech rate. Their tech rate pretty much seems to double in a few hundred years.

Note: I was actually running the max science Civics (Free Speech, Free Religion, Emancipation, Representation, Mercantilism) in my HOF attempt. I had roughly 100 towns by 1300 AD, with roughly 40-50 specialists (mercantilism + statue of liberty). Free Speech was probably giving the most science, followed by Representation in combination with Mercantilism, with Free Religion doing alot less (due to no multipliers).

Finally, don't forget that every cottage that becomes a town earlier due to Emancipation gets another 2 coins from Free Speech.

Chillaxation
Jan 04, 2006, 03:36 PM
Any good reason they made rice give less food than the other grains?

Not other, I'd think, than that rice give less yield of kilocalories per hectare than maize or wheat.

Don't mind me: I was FFA president in high school. This is a great article.

Personally, I've found ideal production cities to be floodplain or grass-river cities with 3 or more hills in the city radius. Floodplain is immensely valuable regardless of the way the city will specialize, because even cottages allow an extra food somewhere.

Cort Haus
Jan 04, 2006, 05:34 PM
Another thing to consider in this is discussion is military support costs. Approx 4 pop points = 1 free unit. I suppose it may seem marginal, but larger pop and growing pop does have a commercial return, and food does effectively have very small amounts of gold attached.

It's also something I've not seen in pop-rushing discussions. Marginal again, but every gold can count in the early game.

Bezhukov
Jan 04, 2006, 07:36 PM
The optimal strategy I've found is to minimize farms (they produce no hammers or commerce) and to focus all strategies on that goal.

That doesn't mean I never build them, they are of great use to get cities up to size in the midgame, for instance, but those same cities produce worker to hold pop once they get to max size (enough citizens to work all available tiles, no specialists), and switch the farms to workshops, then to cottages once infra is built. Mix in various mills to taste.

In the early mid-game, hammer-poor cities may keep the farms to support slavery/drafting, but long-term farms are sub-optimal.

fed1943
Jan 05, 2006, 04:25 AM
Great work,Gato Loco.You have post a technical study,the rigorous data,the foundation;over that,and just over that ,strategies can be thought.
So,food to be able to play,hammers to make things,commerce not to lose and win.BTW,your study shows me overlap is bad:few cities,few precious tiles...).
Best regards,

Rimfrost
Jan 09, 2006, 01:38 AM
Hi ,
I converted the article to Word format to be able to print it, here is the result. Changed nothing but the format.

Gato Loco
Jan 10, 2006, 04:12 PM
@fed1943
Thank you. It's good to see that what I'm doing is useful. As for overlap, I generally to end up with overlap later in the game, but I make sure that my original cities get to work everything inside their radius because they already have multiplier buildings in place.

@Bezhukov
The optimal strategy I've found is to minimize farms (they produce no hammers or commerce) and to focus all strategies on that goal.

That doesn't mean I never build them, they are of great use to get cities up to size in the midgame, for instance, but those same cities produce worker to hold pop once they get to max size (enough citizens to work all available tiles, no specialists), and switch the farms to workshops, then to cottages once infra is built. Mix in various mills to taste.

In the early mid-game, hammer-poor cities may keep the farms to support slavery/drafting, but long-term farms are sub-optimal.

More or less, assuming that you're talking about mature cities. Farms become a much better deal in the late game after biology, especially in newly-captured cities that the AI didn't build towns around. At this point there's less return from building new cottages, so it's a bit more of an open question unless you're running state property (in which case workshops and watermills become the default improvement for empty flatlands).

Cort Haus
Jan 10, 2006, 06:11 PM
I've been using spies to watch the AI economy during the late game, and have been struck by its heavy use of specialists and farms - especially PHI civs, and especially in Oxford / Wall St cities.

Some players have been alarmed by their auto-workers farming over their precious towns, but now I know why they are doing this. Using scientists, the AI was more-or-less matching my cottage + research lab level cities. Their production was low, but who needs that when you've got a zillion units and cheap upgrades. OK, so their space programs can suffer, but I was wondering whether anyone had tried turning towns to farms in their Oxford / Wall St cities and going specialist-tastic. In my last game Peter had made 20 Gt People, which is a vast number of GPP over the game.

After that game I then looked at some early saves in WB and could see Peter running specialists wherever & whenever he could. As the game progressed it was almost like any time he'd built something that could run more specialists, he'd irrigate another cottage->town to run them.

baboon
Mar 15, 2007, 06:40 PM
Great article, i played some civ4, then quit, and bought the expansion. I liked this article more then other strategy articles for some reason, it contained more stuff I didn't know. I guess I like the emphasis on food ;)

edit: I read it again, and for me as a Prince level player it's really great info, gave me some great ideas beyong the usual build a gpp farm, a commerce farm and a hammer farm stuff.

blitzkrieg1980
Mar 09, 2009, 09:58 AM
This is a great article. However, you seem to rip on the plains tiles often. This is a huge fallacy ESPECIALLY when running a specialist economy. Plains tiles with workshops can allow landmasses with very few hills to have high production capacities.

A plains tile with a workshop, Guilds, Chemistry, and running Caste System (Specialist economy) will net you +5:hammers: per turn. That's the equivalent of a mined plains hill with railroad (which comes much later in the game). Even without chemistry, this tile produces 4:hammers: which is equivalent to a mined plains hill. Generally speaking, if you're running a Specialist Economy, it's because you have a food heavy map. With 2 food resources and a couple farms, a "brown fat cross" can turn into a military powerhouse of production. In fact, in most of my specialist economy games, the Ironworks/Heroic epic city has more plains flatlands workshop tiles than hills.

Zerodoc
Jun 25, 2012, 06:51 AM
[yield]/(1+(2-[food_a])/(3-[food_b]))
[yield] = base yield of hammers or commerce
[food_a] = base food yield
[food_b] = food value of the best unused food tile available, usually 3 or 4

[food_b] must be at least 3 or the equation will result in a division by zero. In game terms this indicates a city that is unable to produce any surplus food and is very limited in its ability to utilize tiles of less than 2 food production..

If [food_b] is 3, this still results in division by zero. What's wrong with this equation? Shouldn't a [food_b] value of 3 return a result other than undefined?

Believe it or not, I'm just starting to find time to learn Civ4, though I purchased it when it came out. Great guide!