When to build improvements?

[dhokarena56]

So, I'm returning to Civ after a hiatus of a couple months. Right now, I usually play Monarch. I'm quite good at fielding an army, improving terrain, etc. What I would like to ask is: especially for the earlier improvements, what "triggers" do people use as a guide for building them? What conditions does a city need to meet for building a library or a marketplace to make sense?

 

[Cyc]

Potential size of the town/city is the main tigger for me, plus the speed with which it will get there. Location is also important and usually determines the speed I just mentioned. Granaries can really help with Settler and Worker factories, or towns with slow growth. There are many variables. But as a standard, I think all towns need a Temple and a Barracks. But begining towns should be building Settlers and Workers. So when you are done with expansion and Worker numbers are good, then you can start to determine what buildings go where.

I like to biuld a Marketplace before a town gets close to going over size 6. That means a Market before an Aquaduct, but if the town is on a river or lake, then the Market has to be built in a timely manner. You want a Market in a city before it is allowed to grow freely to size 12. Especially if you are planning a war in say the mid/late Middle Ages.

Libraries are important, and you should fit them in when you can. I've even built a Library as a second improvement after a Temple when I needed the extra culture when a neighbor is building close to me. Situations vary a lot, so try to use your best judgement when to start them (core cities first).

I'm a builder of large civilizations and like as many cities as I can floating around the 30 population level. I build Wonders that will enable a large civilization (I try to anyway). For me, growth and happiness are key. If you have those things, you can conquer the world.

You may want other things, so do what you feel is best.

 

[tR1cKy]

There are many variables. But as a standard, I think all towns need a Temple and a Barracks

I have to disagree with this. While it may be true for specific goals (such as a 100k culture victory), in the general case it's just plain wrong, and it will cause far more harm than good.

Temples are next to useless. Spending 60 shields and 1 gpt for one content face is just not worth it. If you need happiness, grab luxuries, use the luxury slider, build markets in productive cities and just put a clown or two in the corrupt ones.

Libraries are only marginally more useful. A few of them in productive cities may help, if you need to boost research, but in corrupt cities they're just a waste of resources.

There are cases in which building a temple or a library in a specific city makes sense, such as when you want to grab a useful tile, a resource, or you want to be safe from cultural pressure from an AS you're not ready to destroy yet, but generally those shields are better invested in barracks, markets or military.

Markets can be quite useful in your productive cities, for they boost outcome and multiply the effect of luxuries. Aqueduct may be useful to bring a productive city without access to fresh water past size 6, but in a corrupt one the 100 shields needed (plus manteinance) are just not worth it.

Let's not forget barracks: building veteran units instead of regular ones do make a difference.

 

[Aabraxan]

My cities need nothing. The question is what the empire needs for a city to have.

Not every city needs a temple. For most of my games, very few do. Not every city needs a rax. Build those things that move the empire towards your victory condition. If you're playing for culture, that quite likely includes temples. If you're playing for conquest, it likely does not.

Several factors should be taken into account in deciding what improvements to build and when: size, corruption, function. A few examples:

A settler factory needs a granary to speed settler production. It probably does not need a market, as it will remain small and crowding is a self-regulating problem there.

If you want your capital to be a super-science city, you'll want all of the science improvements (& the wonders), and a market. Whether you need a temple may depend on your access to luxes and your playstyle. Being a Republic is almost a given for a super-science city, for the commerce bonus.

Aques take on particular value in a Republic, due to the unit support improvement.

Out in the horribly corrupt farms, where artillery is being built, you don't need a rax. It just costs you gold, and arty do not benefit from raxes.

 

[Theov]

A barracks makes produced units vets instead of regulars. And it heals your units and some other things you sometimes need.
So.
You build a barracks and then produce a warrior in 2 turns. That extra health point has now cost you 40 shields and 2 gold for the 2 gold per turn you just paid for the upkeep of the barracks.
If you build 10 cities and you think all cities need a barracks, then you're going to pay 10gpt just to have vet units. Now say you actually needed 20 units. Then why not build 2 barracks, build 10 units each. (or 3 and 7) and
You pay upkeep every turn for your barracks. You might as well buy one from scratch by the time you need one and still it would be cheaper to keep 10 around in case you need them.

 

[tR1cKy]

Theov, who said that every city needs barracks? Of course i build them in core cities only. A 1sp city would take forever to complete the building, and "another" forever to build a unit. Barracks would be useless.

Secondly, it's not 40 shields for an extra health point. It's 40 shields for an extra health point for each unit you produce in the city.

I can't give exact figures, but in my games a single barracks city typically produces no less than 20 units during the game. That would be 20 extra hp for 40 shields. If we pick the regular swordsman as the standard unit, then you would end up with 200 shields worth of health points. And it is a conservative estimate, since some of those units are actually maces or knights.

The downside is of course the upkeep (one gpt, not two). So you have to decide where to build barracks and where not. It's not always worth the cost, but in productive cities it usually is. But here's the good news: as what happens with the shield cost, the upkeep for barracks usually repays itself!

Let's suppose you decide to save on barracks and stick to regular units instead. To have the same amount of hit points, you would have to produce 1/3 more units. That would result in a bigger upkeep for units, generally more than enough to offset the upkeep savings on barracks.

But there's even more than that. A veteran unit is more likely to win than a regular, and if it wins it can heal and regain its lost health in full. A losing unit is lost forever, you have to invest more shields to replace it. Therefore, the profit is more than what the mere numbers tell.

 

[Theov]

Theov, who said that every city needs barracks?

Cyc said that.

Of course i build them in core cities only. A 1sp city would take forever to complete the building, and "another" forever to build a unit. Barracks would be useless.

Secondly, it's not 40 shields for an extra health point. It's 40 shields for an extra health point for each unit you produce in the city.

I can't give exact figures, but in my games a single barracks city typically produces no less than 20 units during the game. That would be 20 extra hp for 40 shields. If we pick the regular swordsman as the standard unit, then you would end up with 200 shields worth of health points. And it is a conservative estimate, since some of those units are actually maces or knights.

The downside is of course the upkeep (one gpt, not two). So you have to decide where to build barracks and where not. It's not always worth the cost, but in productive cities it usually is. But here's the good news: as what happens with the shield cost, the upkeep for barracks usually repays itself!

Let's suppose you decide to save on barracks and stick to regular units instead. To have the same amount of hit points, you would have to produce 1/3 more units. That would result in a bigger upkeep for units, generally more than enough to offset the upkeep savings on barracks.

But there's even more than that. A veteran unit is more likely to win than a regular, and if it wins it can heal and regain its lost health in full. A losing unit is lost forever, you have to invest more shields to replace it. Therefore, the profit is more than what the mere numbers tell.

You're right, I do build them (maybe that wasn't clear from my post), but I try to be efficient with them like you described here.
I did write 2 gold - for the 2 turns you have the barracks. I tried to describe that you have to pay one gold for every turn you have the barracks.
So building a barracks, making a scout warrior and then leaving the barracks in the city that's now building workers and improvements is a big waste.
To go further on this. I usually build about them in all major cities, build a load of units that will go to war soon and then I'll keep one barracks (near the border - or if I have rails in a productive city) where I sent the wounded.

 

[tR1cKy]

Cyc said that.

LOL! Gotcha. For some reason i read libraries
Anyway, understood and agreed.

 

[MysteryX]

I don't know that it is a good idea to think of it in terms of whether there is a situation which calls for building a specific improvement.

The question is whether building that improvement is the best thing that the particular settlement can be doing at that particular time.

One of the main benefits of a library is that it multiplies a city's research output. The settlement needs good research output to begin with, though, to make the most use of that multiplier. So if I were mainly interested in that research multiplier, a library goes down on the priority list until the settlement had grown substantially. If I were interested in the culture, the library goes back up on the priority list; however, a quick temple might serve my cultural needs better than waiting a while for a library.

Marketplaces are almost always great additions to any settlement, because you always need money early in the game. But of the ancient age improvements, they take the longest to build. In the time it takes to build a marketplace, one could build 2 or 3 other improvements, or 5 units. So again I would wait until the settlement was a substantial size to get the multiplier. On the other hand, if I had 3 or more luxuries, either under my control or through trade, the happiness value of those marketplaces make them much more valuable and make them a little bit more worth the wait, moving them up on the priority list.

If you have a trait that cuts the time of building a particular type of improvement in half, that makes a big difference in moving up that building on the priority list. A temple as first improvement can be a good move for a Religious civilization- particularly when it comes to culture, those extra 50 or 100 turns or so of culture build-up early in the game produce big dividends.

Even if your settlement can benefit from a particular improvement, it may have better options available. A particularly fertile settlement might be better off continuing to pump out workers until your land is entirely improved. If your neighbor is controlling a valuable resource just over the border, you might be better off pumping out a few military units to take and hold that resource.

Always start by asking yourself what you need, instead of asking yourself what you want.

 

[vorlon_mi]

Agree with the overall sentiment -- improve each city based on what your empire needs.
In your core cities, where corruption is low, libraries and markets will be productive.
Harbors for the coastal cities will increase the commerce (and food!) from water tiles, making it easier to grow. Put barracks in shield-rich cities, for production of veteran units.

Some observations in the middle game, since you are going to keep planting towns to fill in the gaps, and you need to decide what to do with cities you have conquered. They may be partially corrupt, since they could be relatively far from your capital or forbidden palace.

Growth: If a new town has a river, you could decide to let it grow above size 6. Strongly consider a market, for the happiness multiplier. Be mindful about aqueducts -- do you really want this town to grow above 6? Since beakers from scientists are not affected by corruption, you may be better off to *not* work those irrigated tiles, and hire scientists (or taxmen) instead.

Barracks: These towns may not have enough uncorrupted shields to actually produce units efficiently. However, you may want a few barracks out here to heal units who are injured on the front lines. Before railroads, it could take 2-3 turns to travel back to your core cities.

Courthouses: You will have a few towns that are slightly corrupt, where building a courthouse will pay off. I liked CommandoBob's analogy of a doughnut in another thread.

Temples: If your civ is religious, then temples are cheap. Cash-rushing a temple can be useful in a few key locations to grab the tiles. Later, when you've conquered more surrounding territory, you can consider selling the temple (after, say, 20 or 30 turns) to save the upkeep costs.

 

[Ozymandius]

I believe that the type of victory condition you wish to achieve will determine the improvements required.
My favourite objective is Domination. I build a granary in my capital and one or two barrack cities that specialize in making my military units. Temples allow the expansion of a city so that fewer cities have to be built. I normally give up on any game that does not give me a great leader that I can use to build the Temple of Artemis for city expansion across my empire.
The size of map also affects my building decisions. On a huge map I either build the pyramids (or steal a city that beats me to it) or Zeus for the extra units.
However if I were to go after a cultural victory... or histographic (very tedious) my objectives would be different.
The cost of buildings becomes a major drain on finances so beware of building things that are not needed for your objective.

 

[Theov]

I believe that the type of victory condition you wish to achieve will determine the improvements required.
My favourite objective is Domination. I build a granary in my capital and one or two barrack cities that specialize in making my military units. Temples allow the expansion of a city so that fewer cities have to be built. I normally give up on any game that does not give me a great leader that I can use to build the Temple of Artemis for city expansion across my empire.
The size of map also affects my building decisions. On a huge map I either build the pyramids (or steal a city that beats me to it) or Zeus for the extra units.
However if I were to go after a cultural victory... or histographic (very tedious) my objectives would be different.
The cost of buildings becomes a major drain on finances so beware of building things that are not needed for your objective.

I normally give up on any game that does not give me a great leader that I can use to build the Temple of Artemis for city expansion across my empire.

What.

Really?
Wonder addiction man. Wonder addiction.

 

[Ozymandius]

What.

Really?
Wonder addiction man. Wonder addiction.

It is not a wonder addiction to get the ToA. When trying to get as much territory for the least amount of input a great leader will have the affect of doubling the amount of territory under the control of your empire without the need of building and marching as many settlers to fill the 66% domination limit.

And the reason I abandon the game is that I can't beat my own score in the Hall of Fame.

 

[dhokarena56]

I myself need to break my addiction to ToA. I find it very useful to expand my borders, but...

It is, however, worth it usually to get Art of War, right?

 

[Quintillus]

I generally consider Sun Tzu's worth it. It's costly, but fantastic for either offensive war or plain defence. Although it also depends on the map size. On a Tiny map, I'd probably take Knights Templar in preference to it. On a Huge one, Sun Tzu's.

It does, of course, depend when to build buildings. I'm a builder at heart, so I like to build quite a few. But, in general:

Temple: When a city is unhappy and I want it to grow. Particularly if I want it to be a Wonder city. Or when I need cultural expansion and either don't have Literature, or am Religious.
Colosseum: When I need happiness in a big city and am far from Cathedrals. More often in Wonder cities. Sometimes when I just want happier people for score. But generally, I prefer cathedrals.
Barracks: Basically whenever I plan to build a few military units per city. I guess there's a cutoff at which point I won't be able to produce enough units to make it worth it - maybe around 8 turns per unit for the standard unit of the time? But generally, I build these a lot during wars. Unless it's a designated catapult city.
Library: I like libararies. I'll start building them for science at as little as 4 beakers/turn. I also like to build them even with less science than that as a culture building, figuring the higher culture per maintenance and even a small science benefit likely outweighs the Temple.
Marketplace: If a city has 8 commerce/turn, these are looking fairly attractive. But generally later than libraries, since I try to run > 50% science. Unless my economy's a wreck, then earlier.
Courthouse: I find it's rare (except in the capital) that a courthouse really isn't worth it, since even if it saves 1 commerce it's paying for itself. But when to build it is a another question. If a city is 80%+ productive, a courthouse isn't going to be an early priority - once I have a bunch of other buildings and the city has grown enough that the lower corruption is noticeable, I'll build one. If it's 50% productive, I'll do so earlier. If it's 10% productive, anything I build is taking awhile, but courthouses do tend to be fairly early if I decide to develop the city.
Walls/Aqueduct: Obviously, as needed. Sometimes a bit pre-emptive with walls, of course.
Cathedral: When happiness is an issue. Mostly in larger towns, but smaller towns tend to become larger ones. I have a weakness for building the Sistine Chapel, so this is one of my favorite buildings.
University/Bank: A little tougher to decide on than marketplace/library. Generally if a city has 15 commerce/beakers after a library/marketplace, I'll build one.
Factory: My cutoff is 10 shields. Anything else and it takes forever to build and is of questionable worth. Plus, it's a nice round number as a cutoff.
Granary: It depends. If I have pottery early on, obviously it's nice. But if my cities are already size 4 or 5, and won't be above 6 for awhile, it's not really worth it. Similarly if they are size 9 or 10. I probably procrastinate on Pottery more than is ideal.
Harbor: Whether it affects if the city will grow or not. Might lay off for awhile if it's mostly coastal and I need the city to work hills to build up infrastructure. If it can grow to size 6 without one, it will unless I need a trade port or am building a naval invasion fleet and want better ships.
Offshore Platform/Commercial Dock: No hard and fast rule, but it I don't have a decent number of water tiles, it's not worth it. If I've got a ton (one-tile desert island) and not much corruption, it might be worth rushing (though it probably won't really matter that late in the game).
SAM Missile Battery: Not sure I've ever seen these be worthwhile. I'd rather have fighters or, if need be, FLAK. Similarly, City Defence (the late-game defence building) is almost never worth it, and I only build Coastal Fortresses for fun.
Nuclear Plant: Assuming I already have a Coal, Hydro, or (more rarely) solar plant, only if I want a production champion of a city, and more rarely if it's a free Hydro plant. Similar for Manufacturing plants. But one-turn-modern armor cities are pretty awesome.
Mass Transit/Recycling Center: Probably about 3 pollution saved makes it worth it. Maybe 2. Cleaning up pollution is no fun and I usually have adequate gold by then anyways.

But it really depends as to what order. What traits you have, what shape your economy is in, whether you've got an aggressive leader like Gandhi or Montezuma next door or all friendly neighbors.

 

[bigFRANK]

Temple, if im a religious civ I tend to build these early on (half price quick build).
Coleseum, only tends to be in big cities that i cant make contnet otherwise.
Barracks, everywhere in my core and where i can make units in reasonable time, although i preffer to build/cap sun tuz if possible.
Libraries, round the fringe of my empire in cities that have enough food to put specialists to work.
Markets, most cities that will grow past size 6
Courthouses, tend to be in the 2nd ring.
Walls, dont tend to bother
Aqueducts, i preffer to use rivers if possible but build aqueducts in productive cities if needed.
Cathederals, size 8+ towns where i want the population to make shields for me.
Granary, early cities with bonus food tiles.
Harbour, if 3 or more tiles are costal.
Commercial Docks, rarely make them
Offshore Platform, again rarely make them
Costal Fortress, not worth making in unmodded game, needs a mod to make the CF produce a unit every X turns.
Civil Defence, tends to only get made late in the game just because I can
Sam Battery again only because I can, they are better air defence than flak or fighters though.
Factories, I will build too often but any productive city late in the game will have a factory even if its only putting out 7 or 8 shields.
Power Stations, Hoover dam is one of the wonders i try really hard to build, I therefore often wont buidl power plants indivually, if I miss Hoover its often a case of coal stations all round and lets not worry about the polution.

 

[arc anjil]

Wow! Lots of good info here (for me....).

I usually go for Cultural wins, hardly ever do the military stuff and only go to war when a civ declares war on me. I usually hand out extra luxes to other civs as gifts if I can't trade for something worthwhile.

So I generally put everything I can into all cities to keep the 'cultural" value up (temples, markets, libraries, cathedrals, universities, granaries, barracks, courthouses (do police stations help? Are they worth the research?)). And, I expand like crazy, I crank out Settlers like they're rabbits.

I have a lot of reading to do!

 

[MysteryX]

Coastal Fortress is bugged and does not work, and should never be built. (Unless I suppose in a modded game that completely changes what Coastal Fortress does.)

I have read there is something bugged about the ways fighters engage in air defense, though not the specific nature of the bug. Just by observation, I seem to lose a lot of fighters in air defense, but SAM Missile Battery works much more reliably. Though I would only build SAM Missile Battery in key cities subject to air strikes where I know I cannot move the front. (Such as a city vulnerable to an air strike across the ocean.) Otherwise, I have started using FLAKs, which give decent results (better than fighters, I think), and can move as I push the borders.

 

[MysteryX]

Wow! Lots of good info here (for me....).

I usually go for Cultural wins, hardly ever do the military stuff and only go to war when a civ declares war on me. I usually hand out extra luxes to other civs as gifts if I can't trade for something worthwhile.

So I generally put everything I can into all cities to keep the 'cultural" value up (temples, markets, libraries, cathedrals, universities, granaries, barracks, courthouses (do police stations help? Are they worth the research?)). And, I expand like crazy, I crank out Settlers like they're rabbits.

Police stations reduce corruption and they reduce war weariness. The reduction in war weariness is something I find valuable. It depends in part on your strategy- if you can keep your wars short, maybe you don't need them. Even though I find them useful, I rarely research them- I'll trade for the tech when someone else researches it.

You don't need large populations to crank out culture points. Once you move past your core where corruption is crippling, I wouldn't bother with granaries or courthouses or marketplaces that much. I would use rush-building strategies to set up the culture buildings (disbanding obsolete units, maybe even disbanding current units, pay to complete when either 50% or 75% built). Barracks are something you do not need in all cities, just your core productive ones (or perhaps an occasional one on your border fringes as a place to heal units).

 

[Theov]

Coastal Fortress is bugged and does not work, and should never be built. (Unless I suppose in a modded game that completely changes what Coastal Fortress does.)

I have read there is something bugged about the ways fighters engage in air defense, though not the specific nature of the bug. Just by observation, I seem to lose a lot of fighters in air defense, but SAM Missile Battery works much more reliably. Though I would only build SAM Missile Battery in key cities subject to air strikes where I know I cannot move the front. (Such as a city vulnerable to an air strike across the ocean.) Otherwise, I have started using FLAKs, which give decent results (better than fighters, I think), and can move as I push the borders.

Good point.
Have a suggestion for what I can change a coastal fortress to? Maybe move it towards navigation? You'd think the AI will use it too?