Fishing in the Desert, or Where Not To Build Some

[FriedrichPsitalon]

One of the most frequent topics that people seem to ask me about is “What should I build when? I’m never sure what I need to build in this or that town at any given time- I waste a lot of time deciding.” In all honesty, there have been times that I’ve captured towns and been quite surprised with what I found inside them. This article, then, is devoted to that topic: When to build which city improvements, and why.

The usual caveats:
1) This article is intended for “usual” MP games- between 2 and 3 hours, 2-5 players, elim or “classic” (Points) rules. You might use it as a guide to other situations, but it is intended for that.
2) I’m not claiming to have all the answers- feel free to speak up. Play styles differ, and obviously, I’m going to be somewhat biased towards my own.

I’m not going to use any particular spiffy organizational method this time- just talk about each one, right after another, in no particular order. Bear in mind before I begin that I am talking about the MP environment and the needs of same- meaning, you devote cities to specific purposes, you keep your maintenance costs town by not building relatively unused structures (Sure, barracks in every single town is nice, but it’s also a fast way to increase your maintenance costs for no good reason), and you keep your intentions sharply focused by staying “on task” to your goal while not wasting time with things you don’t need.

Temple Makes citizens content, 2 culture per turn, 60 Shields to create, 1gpt Maintenance

Temples are a moderately high priority item, but they aren’t absolutely essential. Temples should be built in towns that are on the borders of your empire, or in areas that are likely to experience very rapid growth and/or have little/no access to luxuries. The border expansion that temples provide can be handy for spotting incoming attacks, and if you’re planning on outlasting your foe to win on points, the land area gathered by a temple is very important. Bearing in mind the culture-size 1 city borders compared to a culture-size 2 city border, a temple gains you 8 squares- only 1 less than a brand new city! You do NOT need to build a temple if your city is inside your boundaries already (either due to ultra-close city spacing or high culture cities nearby, or possibly coastline) and/or you have a great deal of luxuries around. If you have a sentry net out, aren’t concerned about land area, and have roads to more than one or two luxuries, temples COULD possibly be skipped. Watch out for culture flips if that option is on, though.

Barracks Veteran unit production, fast unit healing, allows for upgrades, 40 Shields to create, 1gpt Maintenance

One of the more common mistakes I see is the “Build Barracks Everywhere” syndrome. You do NOT need a barracks in every town! If your town is producing workers and settlers, you do not need a barracks. If your town is producing catapults and galleys, you do not need a barracks. If your town is nowhere near the front lines and has low production, you do not need a barracks. Barracks are for (surprise) military centers. If the city in question is highly unlikely to come under direct attack, decide what you intend for this city to build. If the answer is workers, settlers (A high food/low shield area), or catapults (low shield/low food area- possibly taken for a luxury, etc) don’t bother with barracks. Naval production centers can also skip out. Being able to instantly upgrade a unit wherever it stands is all fine and good, but that maintenance cost can add up, and it isn’t THAT big a deal to shuffle garrison units around a bit to upgrade them. Save yourself some shields and cash- put barracks only where they’ll do some good.

Granary Halves food storage required for population growth, 60 shields to create, 1gpt maintenance

Granaries, on the other hand, are probably created too rarely. Expansionists with two good starting squares should consider building one of these right off the bat, and using their scout to assure their early safety. It may result in a slightly slower start, but within 30 turns or so, your population and settler production will be well above your non-granary equipped rivals. Granaries are best used in towns that experience rapid population FLUCTUATION – not rapid growth! If a town is going to proceed up to size 6 or 12 and then stop, never going down, a granary is a waste of time. 60 shields, particularly in the early game, is a lot to burn on something that will sit useless when a city reaches maximum size. The Pyramids are wonderful because your city starts with a granary and grows quickly- but having to build one manually means your city will probably be size 3 before it finishes- halfway to maximum non-aquaduct size! Worker and settler factories, on the other hand, benefit greatly from a granary. Depending on how strong the production in a city is, a granary can potentially double the rate at which these units are produced. Bottom line: If your opponent is seriously out-expanding or out-developing you and you can’t figure out why, there is probably a granary (sometimes even two!) at the bottom of it. I’ve even seen players with quickly built granaries out-expand people who get first and second turn free cities from goody huts.

Walls Land bombardment defense of 8, 50 percent bonus to defense, useful only to size 6, 20 shield production cost, 0 gpt maintenance

Walls are another item that is overused. It always confuses me at the end of the game when I see deep interior cities in a player’s empire – or coastal cities he/she knows I cannot reach – with walls around them. Did you need something to do for a few turns? Churn out a worker if your city is at maximum size. Build a spear and send it out on Sentry duty. Don’t waste time on walls you’ll never use! Many people like to build walls early as a safety precaution, but there is a point at which safety becomes paranoia. With a proper sentry net you should always get a turn or two of warning (at the very least) and at only 20 shields to produce, you can often rush walls if nothing else. If you’re working with a city that is certain to remain your furthest border and you have nothing else to do there that would be smart (say the city is too far out with too much corruption for major improvements) then go for the walls- otherwise, wait on these until you need them. It’s not that there is a maintenance cost involved – apparently stonework doesn’t degrade over thousands of years? – it’s just that they’re so quick, that you can throw them up quickly if need be. Don’t waste valuable development time unnecessarily. (You’ll note that I’m not belaboring the obvious point that if your city IS going to come under attack you should build walls. I’m hoping THAT is obvious- for city defense, they’re mandatory to build at some point.)

Aquaduct Allows city growth beyond 6 for cities without fresh water, 100 shields to produce, 1 gpt upkeep

The question on whether or not to build an aquaduct focuses on, oddly enough, whether or not the city is likely to come under attack. Obviously, for cities that are relatively safe, growth is good, and you should try to get the Aquaduct going around the time your city hits size 5. In most cases a city just hitting size 5 will probably finish the aquaduct midway or late into its growth of size 6, allowing it to move smoothly without being “stuck” to size 7. It is probably not a good idea to build an aquaduct in a “front line” town, however. Many people mistakenly believe that a size 7 city is just as good as a size 6 with walls- this is NOT the case. You lose the bombardment defense of 8, making catapults suddenly much more likely to hammer your city back down to size 6 – and the smaller bonus that comes with it. If for some reason you know your opponent does not favor catapults, there is more reason to build an aquaduct. (Personally, if I see cities over size 6, I always pack a few….) If you happen to own The Great Wall, your loss is even greater – the Great Wall’s bonus does not extend to size 7+ cities! In effect, a 6 size walled city that grows to 7 loses defense with the Great Wall in place!

[FriedrichPsitalon]

Courthouse Reduces corruption/defends against propaganda 80 shields to produce, 1gpt upkeep

The key word when talking about Courthouses is “significant.” Corruption can be a major factor in MP games, which rarely if ever see the final two forms of government. Courthouses, obviously, are the best way to fight corruption’s effects. There are many situations in which a courthouse would be useful, but there are also a few where it will not be. Obviously, cities which are so far out in your empire that they have 99 percent corruption are likely candidates, but be careful: too far out into the distance, and that 99 percent corruption is really something like 120 percent – meaning your courthouse will NOT lower the corruption beyond the 99 percent maximum. Check this by looking at your production and commerce levels - if you have only one shield and one gold that isn't lost to production, and you've got 10 or more commerce total, you're probably looking at a "max corruption" city. Unfortunately, you can't tell if that "max corruption" is at 99 percent or 91 percent or 120 percent from that city alone. As a judge, use cities that are between your capital and the candidate city to decide. If the city ‘next inward’ is not at 99 percent corruption normally (meaning, without a courthouse) then you can probably get some use from getting a courthouse in place. If the next-in city IS at 99 percent corruption, you probably shouldn’t bother – your candidate city is probably well over 100 percent "true" corruption, meaning a courthouse won't put a dent in it - and unless you are going for land area points, the city itself may be a dubious build. Similarly, cities that have almost no corruption (capitals, forbidden palace cities, and immediate neighbors come to mind) may also not be worth building a Courthouse in. If you’re only losing 2gpt to corruption, the courthouse will eat half of that in maintenance. Don’t you have ANYTHING better to do with 80 shields? Go build a swordsman to ‘forcibly increase your land’ or a worker to improve commerce with more roads. The best candidates for courthouses are your ‘middle cities’ – the ones with about 50-80 percent corruption. These cities have enough production to get the courthouse actually built, but enough corruption to make it worthwhile. Pop-rushing courthouses is painful, and they aren’t cheap to rush moneywise either.



Marketplaces 50 percent increase in tax revenue, increased effectiveness of multiple luxuries, 100 shields to produce, 1gpt maintenance

Marketplaces are useful to have, depending on your choice of government. If you must remain in despotism, a marketplace is actually pretty weak for the production cost. True, the added luxury benefit is nice, but if you’re having unhappiness problems that 2 garrison units (considerably cheaper in shield cost) and a temple can’t fix, then you’re having more serious problems than a marketplace needs to address. The corruption rampant in Despotism makes building marketplaces in many cities pretty much a waste of time. Monarchy improves this somewhat, but corruption can still be pretty problematic. Add a few more cities to the list of “worth it to build a marketplace”… but we’re talking larger size cities in any case. The reason is that you need a great deal of tax revenue in order for the production/maintenance cost to be worth it – a size 4 city in Monarchy/Despotism is probably only netting 8 gold per turn (upwards maximum of 10 assuming river AND road), and if half that is going to research, and 20 percent to luxuries, you’re getting 50 percent of not quite 2 gold as a bonus- hardly worth the trouble in most cities. In a Republic, however, the bonus to commerce that the government enjoys makes a Marketplace a very lucrative proposition. A size 4 city could be generating as much as 15gpt, 1/3 of which is 5, with a 50 percent bonus of 2 per turn… this adds up nicely in a series of cities, and if you focus on building them in your largest cities, later on (as you should, Marketplaces, however good, are still low priority) you’ll be making much larger gains. In summary- Despotism: Probably not to nearly never, Monarchy: probably not to occasionally, Republic : Occasionally to Often.

Library 50 percent increase to city’s science output, 3 culture per turn, 80 shields to produce, 1 gpt maintenance
I’m guessing that my comments on the library are probably going to draw a bit of fire, but hey… Very simply: Unless you’re a scientific Civ or are absolutely crying for extra land area out to sea (if it’s land area on land just build another settler), don’t bother building a library in the typical 2-3 hour MP game. Literature as a research is VERY low priority for me- so low that I generally don’t even bother with it, along with any other research I don’t need to get out of the Ancient Era, unless I need/desire a government change. If I stumble across it somehow, libraries follow much the same rules as Marketplaces, but with one larger problem: the ‘bonus gold’ they give you in research has much less utility value than marketplace ‘bonus gold.’ Marketplace gold can rush buildings, upgrade units, investigate cities… library gold can only increase your research. By the time you have Literature (in most games) the bonus research you’ll get just isn’t worth it – the game will be over before it has much effect. Also realize that the same rules for effectiveness/corrpution in Marketplaces DO hold true for Libraries- meaning Despots REALLY shouldn’t bother, Monarchs probably shouldn’t bother, and Republicans could bother, but ought to build a marketplace or worker or “Peacekeeping Unit” instead. A scientific culture that ran straight for Literature and then Republic could probably do well with a nice set of libraries – Comm/Sci cultures, with lower corruption, could actually do quite well (Go Korea!) but other than that…. Skip the books, bring the bombardment.

Harbor +1 food from salt water squares, veteran naval units, allows trade over salt water squares (with tech), upgrades sea units, 80 shields to produce, 1gpt maintenance per turn

Harbors are nifty little structures that have all kinds of uses. Having said that, though, the mantra I’ve repeated before comes back into play: you don’t have to build one in every single town that can build one. Look at what a harbor really does and decide if any of those uses suit you: if you’re not expecting/planning any sea engagements, you don’t need veteran naval units. If you don’t have any island townships or need of naval trade routes (I think I’ve needed a naval trade route once ever in a typical MP game), you don’t need to build a harbor for that reason either. If your town is surrounded by grassland, floodplain, or irrigated plains, you probably don’t need the extra food a harbor brings – unless your city shares squares with a neighbor. Now what makes a harbor neat is that it can do all these things, but if you don’t have any use for them- don’t build one! Harbors are nice things to have in towns surrounded by desert or tundra- they may never be production giants, but with a harbor and everyone working the waves, you can have a size 6 town with a respectable commerce income relatively quickly. (I call that technique “Fishy Funding,” because it can be quite effective with only moderate expenditures to set it up- harbor, temple, marketplace, maybe courthouse, you’re done. Line your coast facing away from your foes and enjoy.)

Colosseum Makes more unhappy citizens content, 120 shields to produce, 2 culture per turn, 2 gpt maintenance

In a word: eugh. There are precious few reasons to build a Colosseum in a normal game, and even less in MP games. The earliest structure with a 2gpt upkeep, a Colosseum only gives as much culture as a temple, and only slightly better ‘population control.’ Also carrying the dubious honor of “highest non-wonder construction cost in the Ancient Era,” the reasons to avoid a Colosseum grow larger every time I look at it. If you have a city that’s huge, you have no wonders or luxuries, and you made the mistake of relentlessly beating the people recently, then and only then, maybe, you could consider a Colosseum in a 2-3 hour MP game. (Or you could just build a few workers or a settler to get the population down a bit… or you could just move up the luxury slider one notch. Five or six colosseums often approaches the cost of a slider-bump, but at least this way your smaller towns get the benefit, too.)

Cathedral (Temple Required) Makes more unhappy citizens content, 160 shields to produce, 3 culture per turn, 2gpt maintenance

Honestly, I almost want to say “If you got to the Feudal Era in a MP game and are having difficulties controlling unrest, something weird is going on” but to give the benefit of the doubt…. Cathedrals are almost as bad as Colosseums, except that they’re slightly more effective, the Sistine Chapel can make them very effective if you actually get into a position where you can, need, AND want to build it, and religious Civs might actually find these worth building. If you’re grubbing for every inch of ground in a score race, this is a better build than the Colosseum. Otherwise, MP games probably shouldn’t see many of these built except in very large cities with unusual unrest problems – go find some luxuries or build 5 swordsmen for nearly the same cost and take some.

Enjoy!

[Drang]

I don't have the links at the moment, but I noticed that someone did an experiment (that I haven't yet repeated) that indicates that walls DO probably give a bonus to size 7+ cities. This is probably a bug, but I've been known to build walls in size 6 cities before they expand if I plan on letting them get attacked, or may not have a choice in the matter, in the future. Made me revise the walls/aqueduct equation.

[FriedrichPsitalon]

I saw that article, and you are correct, it did seem to suggest walls continued to have an effect. This article was reproduced without modification, and in its original form I hadn't heard about that yet.

[Kimmitt]

Size 7 cities give an inherent +50% defensive bonus, whether or not they have Walls.

I don't know how a Size 7 city with Walls works with the Great Wall.

[Dr Shot]

Sorry if I missed the Boat here, as I only skimmed the posts(have read them b4). Once walls are built ( I observed this ingame yesterday, after reading this post)in a city under six, of course, and the city grows beyond six, subsequently is reduced(by any means i beleive) to six or less, the walls reappear.
This may not be new data for any of you but it can be useful.If an inavders attack requires a 'pop' for a unit, the walls and the % bonus will be still in effect(after city is six or less).

[Drang]

This is true, Dr. Shot, and I've actually pop rushed building stuff just to get population below 7. The question is that there may actually be a bug that continues giving the wall bonus after it shouldn't due to size. If I ever see the post again, or get the effort up to reproduce the experiment, I'll make a note of it here.

[FriedrichPsitalon]

It actually isn't much of a question; it's fact. The original tester was examining the effect of the Great Wall, but stumbled across this information when setting up his "control group" in his data.

I believe the number of tests was *well* over 1000, and the statistical data he found for walls at 7+ vs no walls at 7+ was beyond question; WELL beyond simple random chance. I wish I could find the post as it is hard for me to explain, but this wasn't the numbers being funky- the sample size for his data was massive, and the results consistent.

Very simply: A bug introduced in the 1.21f patch makes regular walls REMAIN effective in a city from size 7-12, even though they appear to disappear, and you can't build them new at that size. This is unique to 1.21f, and was not testing for a metropolis (size 13+.)

[SirPartyMan]

Dear Friends:

Never underestimate the value of the wall. It keeps on working past size 6.

From Apolyton: "In reality, it's only the bombard defense that expires when the city reaches size 7. The defense bonus remains until the walls are sold or destroyed, no matter what the size of the city".

I read the article on CivFanatics that Fried is referring to. The gentleman who wrote it ran thousands of tests to prove his thesis and I found it very credible. I tried to find a link for it using search but was unsuccessful. Does anyone have it?

Best,
SirPartyMan

[TacitExit]

Heres the link in case anyones still interested. forums.civfanatics.com/showthread.php?s=&threadid=47118&perpage=20&display=&pagenumber=1

This was supposedly fixed in PTW 1.27f; I was going to rerun the tests but now C3C has come out and the Great Wall no longer has the 'Doubles City Defenses' ability, the original aim is redundant.

I may do it just to check if the wall fix 'took' and didn't 'untake' with C3C, however (as some other bugs have, I've heard).

Anyone think its worthwhile?

[whiplash]

I think these paradoxical test results support my long-held belief that the RNG produces a result that "sticks" for the entire game.

[FriedrichPsitalon]

Whiplash, see this thread:

apolyton.net/forums/showthread.php?s=&threadid=103730

Straight from my former boss, Jesse Smith.

Apparently C3C's RNG was a bit streaky - but it'll be fixed in the beta patch next week.

I gotta admit I like this: they want to get the fixes out as fast as they can, so they'll release a beta-patch to get SOMETHING into people's hands while they're working on a more solid version. Now that's customer support.

[CiverDan]

a bit streaky..i was wondering why I lost two straight vet spears to jags....

[FriedrichPsitalon]

That was a year ago, Civer-D. Someone ressurrected all my old strat threads and pinned them. Not that I mind; a lot of questions lately are answered in one of these that were buried, but I should probably go back through and update them a tad... Civ theory has advanced considerably in the last year.

[SirPartyMan]

Dear Friends:

I thought the Game Strategy Tips section needed reorganizing. I moved the most popular threads up to page 1 and pinned them, to make this section more useful.

Best,
SPM