Overlapping Cities?

[Roblord]

Ok..I've seen a lot of debate on this topic. For those who DO overlap cities...how is it useful for you and how much space do you place between each city...generally? I don't overlap cities at all myself, but I've found that I tend to have less cities than the AI before space runs out. Just moved up to Noble here after being on Warlord for a long time and looking to improve my game.

 

[TheMeInTeam]

Settle to work your resources. Overlap is fine.

There's no hard and fast rule on how much overlap is too much. If the city can contribute more than its costs towards your current goal, settle it. Sometimes even a city with 0 workable tiles in the dead center of your empire is OK in extreme cases (like a really strong mining inc). Usually a couple bio farms or SP workshops is enough of a sell too.

Smaller cities whip and draft more efficiently, but of course more cities costs more. It's a judgment that comes out differently each game.

 

[nanomage]

The thing is your cities are not going to work all of theit 20-tile BFC for very long-long time. place you cities to make the best use of your terrain and resources during the entire game, not for that distant and maybe imaginary time when they will be 20 pop. for the first half of 2/3 of the game, not overlapping cities are just waste of land.

 

[DigitalBoy]

Overlap's not really a big deal. I try to minimize it when I can help it, but if I find a great location, I'm going to plop a city down there even if it has 6-tile overlap with its neighbor. No sense letting good tiles go to waste.

 

[rabidveggie]

I tend to actively look for overlap to fit as many cities as closely as I can since my cities rarely hit 20 pop anyways and hit very late in the off chance that they do.

 

[lymond]

Two great uses of overlap besides what has been mentioned, is overlapping a few tiles to either commerce cities or your Great Person Farm. Commerce cities - for example, a city with a lot of flood plains - cannot work all those tiles to grow your cottages initially. Overlapping a city a few tiles will allow you to work and grow more cottages. Later as the initial commerce city grows you can switch out the tiles each city works and specialize the overlap city on something else. GP farms simply and usually don't work many of the tiles in its BFC, especially if it's focusing an a lot of seafood or big food resources.

 

[z0wb13]

i'd say, don't worry about the outer ring (12 tiles) too much, but be careful about giving up any of the 9 tiles of the inner ring. unless you are trying to culture flip an opponent. so like maybe on average 4 tiles apart, CxxxCxxxC, would be fine on good land.

i think that spreading out cities, without any overlap, is just a bad idea because of all the wasted land. as others have mentioned, your cities won't be much larger than size 12 for about half the game anyway (until biology and food corps). so if that's the case, you just wasted half of your land, in a sense.

don't take this too far, though. if you're talking about tundra/ice (or ocean desert peaks) spread it way out. but any riverside tiles should be worked just about every turn possible.

 

[PartAnon]

In practice there are very few perfect city placements. When it comes to land that a city controls it's about quality, not quantity. A well-placed city can have overlap, or worse than that - some desert and peaks - but if there are good tiles and the right buildings are in place it can be quite productive.

Example of a small-footprint production city:

Spoiler :

2 overlap tiles, 4 deserts, and 1 peak. 13 tiles worked. Not perfectly specialized/optimized but in the late game it has a base production of 55 due to the watermill/lumbermills/levee/railroads for the riverside plains forests. Before that it worked the mines, farms, pigs, using the forests with the Heroic Epic bonus for units, whipping citizens off the forests to build infrastructure. Later when biology kicked in on the farms the pigs were swapped to a cottage city to the north. (There was an Iron Works city also, working 17 tiles, which was plenty for its purposes - I didn't even notice the overlap until much later, and by then I didn't really care).

Overlap also helps you pre-grow cottages in one city to be swapped over another later, as well doubling up on cultural output onto tiles to avoid getting squeezed by a peaceful neighbor.

 

[GreatProfit]

I used to be really anal about overlap, but now it's completely routine. As long as there's enough quality land to support it, you can overlap as much as you'd like. Just remember you can assign which overlapped tile goes to which city using the city screen. Also the BUG mod alt-x dotmapping feature comes in real handy.

 

[NonPrayinMantis]

I try to get every useable tile of land that I can in use by a city. This means some cities are small and are settled in between the larger cities, maybe using 3-6 tiles only. In these small cities I concentrate on building granary, forge, courthouse, then other hammer improvements, then I switch to building wealth. Their tiles are usually workshopped and farmed.

NPM

 

[Baldyr]

I feel that the notion of no overlap - ever, is somewhat a trap that beginners tend to fall into. It does kinda make sense not to overlap, and the player might feel some pride in having figured this out by himself/herself. Personally I gave this whole idea up some 15 years ago (playing Civ2), I believe.

It can help to have some kind of rule of thumb for placing cities though, even for a more seasoned player. How about we formulate such a rule for anyone who feels unsure about overlapping?

Like that every city should have one food resource of their own, or how to easily calculate how many production tiles can be supported by Farms (on average). I know from my own experience that it can be tricky to calculate these things in beforehand, and end up with a city with too little to actually work those intended tiles.

Someone also suggested that every city should keep the adjacent tiles for itself. As a rule of thumb (and not anything carved in stone) I feel this can be reasonable advice; a general rule about not placing cities closer than three tiles apart. (This prevents overlapping the original 9 tiles effectively.)

Lastly, I find it useful to swap high end tiles between cities. The length of a game is enough for a city to pass several phases of specialization. As an example I've had a city with good food resources focus on Scientist specialists initially to get three Academies going in my future research cities. Once this is done, that food gets converted into buildings (Slavery). Once the buildings are in place and the city has regrown to its happy cap, other cities some 3 tiles away can use those same bonus tiles to go through the same motions. When they are done the original city might take the food back in order to produce Workers and Settlers, while the neighboring cities (with better output) can focus on military units. In the later phases of the game these food resources can support specialist, allowing for heavy specialization of the city (for or ).

As an example.

 

[naterator]

As has been mentioned, making a dotmap really lets you see how to max your empire using overlapping to work as many good tiles as possible. Especially to beat the AI to a city location but still place it so you can squeeze in another decent city behind it later. Those second cities will usually be heavily encroached, but will usually end up being well worth it, especially if they have a rescource.

i'd say, don't worry about the outer ring (12 tiles) too much, but be careful about giving up any of the 9 tiles of the inner ring. unless you are trying to culture flip an opponent.

Don't settlers have to be at least 2 squares from a city to found a new city anyway (unless they are on 2 different landmasses i believe)?

 

[PartAnon]

Don't settlers have to be at least 2 squares from a city to found a new city anyway (unless they are on 2 different landmasses i believe)?

Yes, it's a new rule implemented in Civ4 to limit the city sprawling "exploit" used in earlier games.

Here is a BTS game where overlapping is taken to the extreme: Infinite City Sprawl 01 - Manhattan Empire .

 

[Baldyr]

Furthermore, key to understanding why a city doesn't need all the 20 potentially workable tiles of the BFC, is to figure out that not all tiles actually are workable or that not all of them might be worth the effort. Deserts and Tundra are givens (as are Peaks and Snow) but even Plains without a river or Forest are pretty useless. With just a base yield of 1 and 1 its probably better to run a specialist instead. They're really no good for farming either, and if you build a Cottage the tile isn't supporting the consumption of the citizen working it. (If the tile is riverside there's at least an extra .) But also a huge chunk of Hills tiles (especially on a base terrain of Plains) might never be fully worked if there aren't food bonuses available, no matter if there is overlapping with other cities or not.

It could, however, still make sense to let a city have all of its 20 accessible tiles for itself in optimal terrain (lots of Grassland, maybe some rivers, and lots of resources). A size 20 city will require special considerations regarding happiness and health, though. And since not all cities will be able to reach this size, it might be wasteful to invest in these things for just that one city to reach its potential size. And if the city isn't gonna use all those good tiles eventually, they're just wasted.

 

[z0wb13]

@ naterator: in this WB example, yaroslavl has given up all of its tiles.

Spoiler :

 

[naterator]

@ naterator: in this WB example, yaroslavl has given up all of its tiles.

Ya, but only because the tiles have been assigned to other cities. Every city could still work it's entire inner ring with a few clicks.
Out of curiosity, did you assign all the tiles to other cities manually; or did you build the other cities first (and pop borders), and have the computer automatically assign no tiles to yaroslavl because they were all claimed (though unworked) already?

 

[slaze]

Don't think of it as overlapping cities. Think of it as overlapping food.

 

[jaycuhbbb]

The thing is your cities are not going to work all of theit 20-tile BFC for very long-long time. place you cities to make the best use of your terrain and resources during the entire game, not for that distant and maybe imaginary time when they will be 20 pop. for the first half of 2/3 of the game, not overlapping cities are just waste of land.

This is true but I disagree. You can win most games with 6-8 good core cities, it's unnecessary to settle excess cities just to work more tiles. For a good chunk of the game these cities are running at a loss anyway, then when you get to the late game your cities are stagnating and you fall behind. You also have to double up on city improvements to work the same amount of tiles. It's a trade off but I prefer to settle the best cities possible.

 

[Baldyr]

This is true but I disagree. You can win most games with 6-8 good core cities, it's unnecessary to settle excess cities just to work more tiles. For a good chunk of the game these cities are running at a loss anyway, then when you get to the late game your cities are stagnating and you fall behind. You also have to double up on city improvements to work the same amount of tiles. It's a trade off but I prefer to settle the best cities possible.

I'm right there with you about only running a few good or core cities. Its a question of strategy and you should do what works for you. But not every "good" city has to be 20+ in size or work every single tile inside its BFC. You might have to make some sacrifices in order to get all cities this size (health and happiness issues) and some tiles might not even be worth the effort.

I'd say that its better to run a few "optimal" cities rather than a large number of suboptimal ones. And by "optimal" I obviously don't mean cities with a certain size or cities with no overlapping. You really have to take into consideration the makeup of the land you're controlling. And a production city doesn't need all the 20 tiles anyway, at least if the terrain isn't entirely divided into good production tiles and good food tiles. (That sounds more like a good spot for a capitol to run Bureaucracy.)

 

[PartAnon]

This is true but I disagree. You can win most games with 6-8 good core cities, it's unnecessary to settle excess cities just to work more tiles. For a good chunk of the game these cities are running at a loss anyway, then when you get to the late game your cities are stagnating and you fall behind. You also have to double up on city improvements to work the same amount of tiles. It's a trade off but I prefer to settle the best cities possible.

Asking for clarification here:

By "most games," are you talking about 70%, 90%, or 99%? I have not been able to pull off winning most games with 6-8 cities. Not when there is AI out there with 12+ cities and picking up vassals... Except for a couple of rare cultural victories, all of my winning games have had at least 12 cities.

By "core cities" you're implying that there are other, "non-core" cities (e.g., border cities, worker/missionary pumps, resource grabbers, cottage builders, colonies, etc), correct?

And what level are you playing?

If you can indeed consistently win with only 6-8 cities on Emperor+, I would love to see it in live play (this is not meant to challenge you... I really would love to see this).