Next-level City Specialization

Everybody knows about a GP farm and a Hammer central, but what about the rest? In a peaceful game, maybe all you need are cottages and commece centers. Or, maybe not so in a heated game where military units are constantly consumed / spread.

Since I mostly play Pangea and Terra maps, I usually find a shortage of units, navel, missionary, catapult, bowman... you name it. Therefore, I am just thinking perhaps I would be benefited to plan ahead to bring specialization to the next level. I am testing it out in my current Deity game and it works quite well.

Type 1 -- Production Cities

1A. Unit

This one is the traditional hammer central. Obviously one needs to put Heroic Epic here. If you are also an expansion-type player, you probably can't see many turns of Ironwork or Weset point in use before you win the game. I would prefer the West Point since Iron work also applies to non-military productions.

Requirement: High number of mines, workshops/watermill, and enough farm to support all hammer tiles to be worked plus maybe a few Mechanics/Priest specialists (hammer bonus). No cottages at all.

1B. Wonder

Early wonders can be rushed by Great Engineers, but soon this won't be enough. It would be a huge waste to make the unit production center building a wonder, which takes many turns. Therefore it should be beneficial to dedicate a second production center just for Wonders. In the rare case when you run out of wonders to build, you can build military units. I would put the Ironwork wonder here when it becomes available.

Requirement: Same as above.

1C. Navy

A strong navy is always important. It would be nice to dedicate one city to produce a lot of navies. Obviously you need to put a Dry Dock here when it becomes available, and forget about dry docks in any other cities. However, usually it is more practical to build a certain number of Galleys/Caravels early, and upgrade them when available.

Requirement: Still, all hills are mined. However, I would probably spawn cottages on the plains instead of workshops/water mills, because you don't really need that many ships unless you are playing a water world. Therefore, this city is in essense a coast commerce city with a reasonable number of hills.

1D. Missionary

This one is only needed when you found a religion and build the speicific Shrine. Each city that has that religion gives you 1 gold per turn to the holy city, even if it is not your state religion. You are allowed to have 2 missionaries of the same type at any given time, so you can actually make more than 1 cities dedicated to producing missionaries for the maximal spreading speed. If you are running Organized Religion, you don't even need a monestary to build missionaries.

Requirement: Any small city. You certainly don't need adquaducts, markets, grocers and harbors in very small cities, since they won't reach your max. health and happiness limit in the next 500 years. I find it best to use them to build units that aren't related to barrack. If you don't have a shrine, then you can use them to build ships and other stuff, such as a scout, a settler, etc. Collectively they will give you some good output.


Type 2 -- Commerce Cities

2A. Research

This is usually your capital, since you have that +8 gold bonus from the palace, and the longest time in the game to build something useful in it. You are going to cottage spawn most squares. You pop an Academy here with a Great Scientist. Finally, you build an Oxford University in it and enjoy 400+ beakers every turn under bureaucracy.

Requirement: Cottages, Windmills. Anywhere with abundant food and a good number plains (for cottages). Cities with enough food will reach your max. population fast, so you need to keep a reasonable production on the hills for a Market, Grocer, Bank, University, Observatory, and maybe the Oxford U. there. For smaller cities, you can adjust the mine/windmill ratio to give you the fastest pop. growth possible (so they go work on cottages).

2B. Wealth

This one is basically the same as the Research City. The only difference is that you will build a Wall Street here to collect more money. A Holy City with a popular Shrine is a perfect candidate. If you don't have one, then it is no different than a Research City. I guess everybody knows that all other cities that you don't find a particular use of it is a research/wealth city.

Requirement: Same as above.


Type 3 -- GP Farm

3A. The GP Farm

Nothing special - the good old GP farm you know. The key to locate a GP farm is not how much food surplus you have, but how much food surplus relative to how LITTLE grassland you get. A city with 5 food sources and 15 tiles of land is a perfect research/commerce city, while a city with 4 food sources and 3 tiles of land is a perfect GP farm. You get the catch here - you have a lot of population sure, but they can't work on any cottages - so they must become specialists - so they contribute to great people points.

Also, you don't need to build the Globe Theater here. Chances are you have so little hammers, that you have to WHIP (enslave) to get things done all the time. You only need to keep a reasonable population, so all your food is worked on, and you are hiring as many specialists as possible. The highest population city in your nation deserves the Globe Theater, not here.

Requirement: A lot of Food + a serious lack of workable lands.


That's about it... Welcome comments and criticisms.

I think this is a good summary. Glad you call 2B a wealth city instead of a commerce city since it seems like the game refers to commerce as the thing cottages produce before it gets converted (by the research and culture sliders) into research/gold/culture.

I have two questions these days about city specialization:

1) If it's a research city, do you use slavery or an existing forest or mined hill to build your (minimal) infrastructure (e.g. library, later university, etc.)

2) More importantly, what's the ratio of such cities you're looking to achieve? If you have half commerce cities and half production, you'll go broke. If you have only one or two production cities, how do you gear up for war?

Nice overview, maltz.

1D.
You are allowed to have 3 missionaries of the same type at the same time, not 2.
I don´t get the idea of this. Why not using 1A, 1B and 1C? You can only have 3 missionaries in production + on the map. They need some time, to arrive at the targeted city, so you are able to use these cities also in their dedicated way.
No need for small cities...

2A.
A compareable way is to take a high-food-city and go the way with scientist-specialists.
http://forums.civfanatics.com/showthread.php?t=158437

3A.
There are some other ideas on these GPFs around in this forum. Some of them are quite interesting. For example: Combination of Science and GPF?
http://forums.civfanatics.com/showthread.php?t=161673

Opensilo is correct about the difference between commerce and wealth/gold

1D I disagree with completely. You don't (ideally) build missionaries in production cities. You build them in Science cities.

The point to specialization is leveraging the synergies available. Missionaries come from monasteries, which carry a science bonus and a culture bonus - neither of which is particularly valuble in a production city. But an extra 10% kick in a city with a lot of beaker sources is valuble.

1B I think I disagree with. The only real difference between 1B and 1A is the presence of a barracks. And since you won't always have wonders to build, you'll likely end up building military here anyway. I'm not sure it's critical, though.

1C I think the main point to make here is that coastal cities are natural commerce cities (sea tiles bring in commerce, the trade routes are more valuble (especially with a harbor). So creating naval shipyards is, as a general rule, fighting against the natural alignment of these sites - and you don't want to do that very often.

Beyond that, though, a naval city is really a unit city that happens to have an ocean view.

Commerce/Science/Wealth cities are just a mess, because the sliders are national, rather than local. An ideal commerce city has all of the science buildings and all of the gold buildings, and you just maximize the commerce.

When you have to start making choices, though... OK, a city with a shrine should be going toward gold buildings. A city with the great library should be geared toward science buildings. That covers two or three cities, what about the rest?

As far as I can tell, the answer is "that's complicated". The number of shrines available is going to be a big factor. The Financial and Philosophical traits are also relevent, because of the building bonuses. Also important is how you intend to dedicate your GP Farm - which is going to depend on the previous two points. Note that GPFarm/Science is a lot easier than GPFarm/Gold (see http://forums.civfanatics.com/showthread.php?t=162195).

That determined, you should be able to figure out where the science slider is going to run, and everything else comes from there. If you are going to be running high science, then "everywhere else" wants science buildings (and your great scientists start spam academies), and your gold cities perhaps reconfigure themselves to run merchant specialists (great merchants may get attached here, so that you can run more specialists).

On the other hand, if you are going to run high gold, then everywhere else wants banks, your science cities start looking into specialists, and your great scientists get attached to cities instead (great merchants go on trade missions).

In practice, I've yet to find a situtation where specializing a commerce city for research or wealth seemed to be necessary. So the only remaining benefit seems to be moving the culture slider about depending on which cities need an upgrade.

example: I need to build theaters in all my commerce cities. So first I turn off the science slider, put the science cities into production mode and build theaters there, and the wealth build up. Once that's done, the science cities go back to commerce mode, the wealth cities go to production mode, the slider goes way up. When that's done, everything goes back to equilibrium.

Ok thanks. Changed 2B to "Wealth".

opensilo said:

1) If it's a research city, do you use slavery or an existing forest or mined hill to build your (minimal) infrastructure (e.g. library, later university, etc.)

Click to expand...

I guess it depends on how much food surplus you have. If you have A LOT (like 4 food sources), then slavery + windmill (extra $ for research, extra food for whip) may give you a better result. If you don't have a lot of food, then probably you have to resort to just mined hill. Forest is best for chopping...

Also, I find I usually end up buying universities and observatories (sacrificing a few turns of research). That is when I have the Pyramid.

opensilo said:

2) More importantly, what's the ratio of such cities you're looking to achieve? If you have half commerce cities and half production, you'll go broke. If you have only one or two production cities, how do you gear up for war?

Click to expand...

I think one only needs 2 full-time production cities. The production of ships and other units, as I mentioned in the original post, can be distributed to smaller commerce / hybrid cities.

I don't think it is easy to file bankcrupcy. For example, in my current game (Small Terra, Deity) I had to drop to 0% research after nabbing my 9th and 10th city, and still run a -8/turn deficit. But after I get Code of Laws and other good stuff like Calendar, I can run a 50-60% research now (no more expansion in between, thinking of that, I just got Civil Service... )

Thanks for the inputs again.

Syntherio said:

1D.
You are allowed to have 3 missionaries of the same type at the same time, not 2.
I don´t get the idea of this. Why not using 1A, 1B and 1C? You can only have 3 missionaries in production + on the map. They need some time, to arrive at the targeted city, so you are able to use these cities also in their dedicated way.
No need for small cities...

Click to expand...

Maybe it is a map size issue? I was playing on Small when I hit the limit of 2. Or maybe I got it wrong.

In my game, 1A and 1B were busy building something else (catapults, forge, the Hanging Garden, etc.), so I used some small cities to build missionaries. They work very well. But if you don't need missionaries in a hurry, you can generate missionaries faster from those spots for sure. Maybe I shouldn't really call 1D "specialization" because it is a lack of specialization temporarily...

Syntherio said:

2A.
A compareable way is to take a high-food-city and go the way with scientist-specialists.
http://forums.civfanatics.com/showthread.php?t=158437

Click to expand...

I think in the long run, the cottages (towns) will pay off. But if one is in a hurry, specialists is ineed the way to go. In my current game I also switch between scientists and cottage depending on my financial situation. When I just expanded, my research rate is low, so I switch to scientists to compensate for that.

The science GP farm is a good idea. I am pretty much doing the same thing now.

Thanks for your opinion. You have a good knowledge of tuning specialists, which I don't have too much experience.

VoiceOfUnreason said:

Missionaries come from monasteries, which carry a science bonus and a culture bonus - neither of which is particularly valuble in a production city.

Click to expand...

I think 1D is just a minor issue, because you only build missionaries in a short time frame. They can come from anywhere as long as you get your holy city hotels booked full. As I mentioned in the original post, by running Organized religion you don't need a moastery. To tell the truth, I have never built a monestary (always got something else to build).

VoiceOfUnreason said:

1B I think I disagree with. The only real difference between 1B and 1A is the presence of a barracks. And since you won't always have wonders to build, you'll likely end up building military here anyway. I'm not sure it's critical, though.

Click to expand...

I don't know about your games, but I always have an endless list of wonders to build. After chopping the Pyramid, it is the Great Lighthouse. Then it is the Hanging Garden, Notre Dame, Taj Mahal. You can certainly put a cheap barrack in 1B. Later on, maybe a Red Cross here? (I've never had a game long enough to see it).

maltz said:

I don't know about your games, but I always have an endless list of wonders to build.

Click to expand...

Ah - I'm actively trying to break myself of that habit (I'm also, I notice, playing at a lower difficulty level than you are). It finally sank in that most wonders are just as effective if you capture them. The Oracle and the Taj are the only two that absolutely matter (oops, Hanging Gardens - I always forget that one); Notre Dame, Statue of Liberty, and 3GD if you care which continent they are on (not relevent on one-continent maps), the Great Library to put it in the right city.

But who really cares where the Spiral Minaret gets built, as long as it is under your control?

I'm not a war monger myself, but from reading some other discussions here I believe they have a succinct term for cities like 1B: targets.

But on a similar topic, you might want to address rushing wonders at the GP farm (where the Epic multiplies the GP bonus).

Actially I try to double wander city with GP farm. Reasoning - flexibility and wanders gp point use.

Useally if I have a desert city with lots floodplance and hills.
No wander to build - specialists. Wander avalible - convert specialists into shield producers.

I useally finish having here National Epic, that increatse wander GP's to.

As result I am getting very efficient GP producer which double as production city in a pinch.
Shields tend to be mach less important in Civ4 then in previous civs, so there non optimum use does not matter.

Just to toss it out there as an idea, are there any real differences between commerce cities based on towns, and commerce cities based on other sources (sea, lakes, oases, perhaps also resources).

Intuitively, GP farm is a good wonder museum because wonders provide GP points. If I create an "ideal world" using the map editor, I would give my GP farm all the wonders in the world, so they give me like 400 GP points per turn.

In reality, there are many factors to consider:

1) Do you build the wonders in the GP farm, or you use GEng to rush?
-- If you build them, will the AIs beat you? In high difficulties, there is almost no chance that you can finish any wonder before the AIs if you don't chop / rush with Great Engineers (GEng).
-- If you rush them, do you have enough GEng to rush them out? To get GEng you need the Pyramid, the Hanging Garden, or a forge. Can you get a GEng early? Maybe not.

2) Each specialist is 3GP. Each wonder is 1-2 GP. The effect of a wonder is not as good as a specialist. Plus if you run Representation, specialists are really helpful in early research.

3) If you have a lot of food and land tiles to work on, this town may be a better commerce center (cottage spam).

If you only have 4-5 cities in total, then perhaps you can combine the wonder town and the GP farm. However, if you have more cities, say 10, then it is really nice to have a separate GP farm.

I think it's tough to be serious generating GPs in a city where you're trying to build wonders. For a maximally producing GP farm, you want to work tiles that give you excess food so you can run specialists. Any production would go to building buildings that let you assign specialists. If you're running caste, then you just farm and generate GPP.

In contrast, a wonder producing city will, once you run out of forests, need to be a production city. That means all extra food goes to running mines not specialists.

If you are not actively pursuing GPs, though, the wonder city will generate some for you. But I certainly wouldn't call that a GP farm.

I suppose it's possible to convert a former wonder city to a GP farm if you've got a good hybrid site and you've got all the wonders you want. If you don't care about controlling the type of GP you generate, this could be optimized by building national epic there and running pacifism and a state religion. This might be best if you want to run late-game golden ages to accelerate the space race and you basically ignored GP generation up till now (so that both the GPs and the golden ages are cheap).

VoiceOfUnreason said:

Just to toss it out there as an idea, are there any real differences between commerce cities based on towns, and commerce cities based on other sources (sea, lakes, oases, perhaps also resources).

Click to expand...

I think commerce cities that are based on grassland/rivers (with towns) are the best to go for. Other commerce resources are sometimes good for an early edge, but are not really necessary for commerce cities. Often an improved commerce resource will give you less commerce than a fully developed town on that tile anyway.

VoiceOfUnreason said:

1B I think I disagree with. The only real difference between 1B and 1A is the presence of a barracks. And since you won't always have wonders to build, you'll likely end up building military here anyway. I'm not sure it's critical, though.[/i]

Click to expand...

The main difference between the two is the national wonders. The military one gets Heroic Epic and later West Point, while production gets Iron Works. You are right though in that your military city isn,'t the only place you create units from, but if I have nothing pressing to build I'll usually build all my units there and make my production city produce wealth or research.

I do agree with most of these cities other than missionary. There's only a relatively small time frame early in the game when they're really important. Cities aren't nearly as specialized that early in the game. I usually don't have just one city that builds them, I spread it around depending on which cities don't have other pressing needs.

Another advantage of a GP farm is as a settler/worker factory. Early in the game you can't make many specialists anyway, but a city that has several improved food resources can really churn out settlers quickly early on, it's a nice change from pure chop rush strategies.

PieceOfMind said:

I think commerce cities that are based on grassland/rivers (with towns) are the best to go for. Other commerce resources are sometimes good for an early edge, but are not really necessary for commerce cities. Often an improved commerce resource will give you less commerce than a fully developed town on that tile anyway.

Click to expand...

Unless you trade the resource to another civ for gold-per-turn! I frequently give them cows for 12 gpt.... Keeps me researching or stomping my way to their front door, whereupon I will expect them to show me the beef.

But I digress. The only benefits to coastal/oasis squares are their early returns and the fact that they can't be pillaged. And resources can be quickly (less quickly for sea resources) returned to full operation. It's painful to lose those towns.

opensilo said:

Unless you trade the resource to another civ for gold-per-turn! I frequently give them cows for 12 gpt.... Keeps me researching or stomping my way to their front door, whereupon I will expect them to show me the beef.

Click to expand...

Of coures, but then the resource doesn't need to be in a commerce city particularly. The gpt is raw gold put into your treasury and isn't affected by the city's buildings. The improved resource would be just as good in any type of city.
What I really mean to say is that just because you see a commerce resource, it doesn't mean you need to make a commerce city specially for that resource.

I agree that surplus resources are good for extra gpt from other civs.

Its really hard to mess up a commerce city..

Cottages, Food for cottages, Lighthouse if near water.

Of course, it depends why you need the commerce as to what other improvements you build.

A specialist-based economy doesnt need a commerce city for direct slider research purposes. He needs the commerce city for indirect research purposes. He wants to throw a chunk of the commerce into the happy slider:

More Happy (leads to) More Specialist (leads to) More Beakers (and other powerful Specialist Voodoo)

Its very.. how shall I say.. 'dramatic' how different a commerce cities needs will be based on what the rest of the empire is doing.

PieceOfMind said:

What I really mean to say is that just because you see a commerce resource, it doesn't mean you need to make a commerce city specially for that resource.

Click to expand...

Agreed. Conversely, just because you have one or two non-commerce resources inside an otherwise commerce-prone city doesn't mean you shouldn't improve them. You just don't have to work them.

This thead has helped.

I'm getting better at matching the city to terrain. I get the Great Person Factory ideas but am missing the finer points of specialists versus cottages.

Now, I've only played the GOTM games on Civ IV. The last one was Emperor - which felt about my level. I've won the games to date, always shooting for as fast a Domination victory as possible. In the last couple of games I've tried to focus on specializing my cities.

Looking to improve, I downloaded a number of the winning Domination submissions in GOTM 3. Looking at Grey Cardinal's submission who won the month with the highest socre I noticed an almost compete lack of cottages (I think I counted 2 or 3) - almost entirely farms with a myriad of specialists.

Do the better players 'time' their win - ie. use cottage income to fuel research / gain the techs needed to win then farm at the very end to max population. Or is it that when shooting for a fast win the cottage / village / town progression is too slow and so the better player turn to specialists.

I would accept that GOTM is a special case - i.e many players might tune their play to max their GOTM score.

Gunner Jones.

Gunner Jones said:

... but am missing the finer points of specialists versus cottages.

Click to expand...

Cottages provide long-term advantages, but if you are aiming for an early win (for a very high score), it is likely that the cottages are not as good as specialists.

All specialists provide another 3 beakers under Representation (enabled early by Pyramid). The best research specialist is the scientist, under Representation they provide +6 beaker each.

A library, which is available very early, makes 2 scientists possible. Therefore, any city with >= 4 food surplus can feed 2 scientists, which is 12 beakers of research, or 6 if there is only 2 food surplus. If you run Caste system (enabled by Code of Laws), all cities have unlimited number of Scientists, but it is not useful unless you have a lot of food surplus (more common on a crowded map).

For example, we have 4 riverside grassland plains:

4 Farms (3F + 1G each): 12F + 4G, feeds 2 scientists.
Outcome: 12 beakers + 4G

4 Cottages (2F + 2G each, 3G if commercial): 8F + 8/12G
Outcome: 8/12 G

4 Hamlets (2F + 3G each, 4G if commerical): 8F + 12/16G
Outcome: 12/16G

4 Villages: 8F + 16/20G <-- a quick domination game usually ends around here
Outcome: 16/20G

You can see that, Cottage starts to break even on the Village level, or Hamlet level for a commercial leader. Therefore, if you are using a financial leader, I would say go for cottages. Otherwise, farm + scientists are the way to go. If you are philosophical, those 2 specialists will help a lot (giving you Great Scientists for academy & tech discovery); However, if you know you will play till the late game, cottages should be better since you will get various techs/civics to boost the production of cottages.