How to pick city locations: Discussion

[Yosh]

When it comes to developing my civ, my greatest problem is perfectionism: I want every city that I found to be in *just* the perfect spot. I take loving care to decide on a spot that is exactly outside the "fat cross" of any other city, and within reach of food and production-heavy tiles. I ensure that each tile that city will work is at least average, and will supply the city with generations of usefulness.

Obviously, my strategy has some drawbacks.

My question then, dear friends of Civ4, is how do you decide where to place cities, and when it is "ok" to overlap, because it has become clear to me that it is not very efficient to spread your cities out in this way.

As a side note, I have been able to dominate all the difficulties up to Prince, at which point I'm getting smeared all over the place!

[Dog of Justice]

"Exactly outside the fat cross" -- you're not even willing to accept one or occasionally two squares overlap??!!! This forces you to simply not use quite a few solid tiles in your own territory. Even a perfectionist should be willing to employ (4,2), (4,1), and (3,3) spacing.

Generally I try to space cities as far apart as possible while using 98+% of the decent tiles in my territory (and I even count grassland-jungle here). As the difficulty level increases and the health/happiness limits become more constraining, it might be reasonable to use slightly closer spacing.

[Blake]

It depends on available space. If there is plenty of space I'll expand to good resource spots only (along with rivers). Otherwise slightly denser spacing may be called for.

I should probably backfill a bit more, but it's often more tempting to get more cities by conquest. In my opionion, it's mostly the resources that count.

[Yosh]

ok so I used to be an absolute perfectionist (the Civ 3 ICS really got me ticked) about city spacing, I do space things sometimes where there is small overlap...

still though, I think that the initial 30-60 turns are some of the most important turns in the game; deciding WHEN to expand has also historically (no pun intended!) been a problem with me

[DeepO]

Originally posted by Blake
I should probably backfill a bit more, but it's often more tempting to get more cities by conquest. In my opionion, it's mostly the resources that count.

Globally, you're right. Resources and coastal tiles.

Locally, rivers and fresh water are the other issues, but the global things take priority.

DeepO

[Utwig]

If you're going for conquest or domination, very loose spacing should be desired:

You hit population limit for domination at about half land area, and since more cities cost money, very loose spacing will net you more area with lesser cost.

Otherwise I stick to OCP almost exclusively.

[Glostakarov]

I don't mind a square or two of overlap, but I do my best to minimize the number of worthless tiles in my city radius. Mostly that means desert and peaks, but to a lesser extent arctic, or ocean for a city that isn't on the coast.

More important for the placement of my early cities is the location of my neighbors. I agressively place my early settlers so as to limit rivals' expansion, so the roadblock factor of the city borders is of chief importance there. The balancing act between creating a useful city, blocking enemy expansion, and avoiding overlap with the nearest enemy city makes these the most challenging cities to place. With backfill cities I can take a lot more care to arrange them for best advantage.

[cal_01]

If it's going to be a specialist city, then I typically go for no-overlapping squares. If not, I don't really mind, since I typically use my other cities to secure resources or to force borders back.

[Stageon]

I think there are as many viable city arrangements as there are strategies. Depending on the map/game type/difficulty, your order of priority will change. Priorities include using cities to increase cultural borders and hem in a neighbor, picking prime sites for specialist cities, grabbing resources, holding down strategic locations like a coast or bottleneck, even little things like maybe laying down that sixth city so you can build your sixth theatre and be able to build shakespeare's. Lots of things to think about - I doubt I ever do it the same way twice.

[khumak]

Re: How to pick city locations: Discussion

Originally posted by Yosh
When it comes to developing my civ, my greatest problem is perfectionism: I want every city that I found to be in *just* the perfect spot. I take loving care to decide on a spot that is exactly outside the "fat cross" of any other city, and within reach of food and production-heavy tiles.

I used to do that as well but when you really think about it how many of your cities ever actually wind up capable of using ALL of their tiles and getting more value out of every one of them than they would with a specialist or two? My rule of thumb now is to maximize the amount of good terrain I have within my city radii without missing any special resources. I'm willing to tolerate up to 3 or 4 useless tiles including overlap and peaks/desert.

When I have patches of desert or peaks in my territory I try to leave gaps between my cities just large enough to avoid having most of those tiles in any city radii but small enough that the AI can't squeeze a city inside that hole in my culture. One notable exception is that I'm willing to stick a city or two in a horrible location to grab a key strategic resource.

Having access to copper or iron is especially important and for some reason I seem to have terrible luck getting access to it lately. My last 4 games in a row I've wound up having to research the dead end horse archer path because I had absolutely no access to either copper or iron. Unless you have barbarians turned off you're screwed in that situation unless you grab horse archers. Warriors and Chariots just don't cut it against axemen.

[Yosh]

Thanks for the input. I've been reading Sulla's walkthrough, and also S & G's Cuban Isolationist Variant game, and that's yielded a WEALTH of knowledge!

I suppose what it comes down to, like most aspects of Civ, is a balancing act; in this case it's between how much effort you have to expend to create a new city, between what you can do with the time spent NOT founding a new city (and improving, etc)...

So all things being equal then the wild card that would have the greatest effect on your *civ* (NOT the city!) would be timing then? I know there are other threads on this board discussing timing of when to build cities, so I won't go into it much

[Bobby Chicken]

I put my cities where I want them regardless of overlap. The truth is that very few of your cities are going to be over 20. Don't be afraid to put cities well away from the fat cross either; your culture will grow and they will culturally overlap later; or you can build more cities. Is there something I'm missing? Is there an advantage to exact spacing? I don't see any.

[Addled Platypus]

my strategy is to scout out the land first, fot two goals

1. where are the resources
2. where are the other civs

I mostly place my cities close enough to create a solid border, to divide "them" from my territory

You have to be fast if your are near a coast location

Connecting your cities with roads early as possible helps alot

[Snotty]

What I have been trying is originally loose spacing of citys so I can land grab with culture. Then if Im still at peace, I back fill cities mostly to grab any empty coastal areas and where I can next to my borders so I can border push with culture. Im ending up with plenty of overlap, with a few large cities and more medium cities.

[Drachasor]

Hmm, here's what I do...

Overlap: I avoid this, but I am willing to accept up to 2 squares of overlap with up to two other cities (so up to 4 total). Anything more than this requires an exceptional reason such as hemming in the AI, excellent resources only the new city could use, or some other strategic reason. Usually this means any clumps of unused land are never more than 4 or 5 squares. Often they are much smaller (since I do allow some overlap).

Ice/Mountains/Desert: I try to place my cities to have as few of these squares as possible. I'll occasionally accept up to 4 or 5 of these sorts of squares in a city radius, but usually I avoid this pretty well. This leaves such squares inbetween cities, which means I can still use those desert tiles for oil if it pops up, even if no city can directly use it. It's good to keep in mind that a bonus resource square is quite nice, but 2-3 normal tiles is better. I try to place my city on one of these squares if I am going to have a city in the area--this turns a worthless square into a useful one.

Coast/Ocean. I avoid these squares for non-coastal cities. For coastal cities I like to get as much coast and as little ocean as I can. I tend to play financial civs so those coasts give me 3 commerce instead of the 1 from ocean tiles.

Tundra: I only build on tundra if I have a clean water source so I can irrigate it, or if enough good squares are there to make sure the city can grow to a decent size. I like to have forest on my tundra, as this lets me put a lumber mill on the tundra, turning it into a pretty good square. Hills also work. The key thing is that it needs some non-tundra squares for a decent food surplus.

Resources: I am ok with having a few resources in my territory that are not in any city's radius. Sometimes it is better to give a city more normal squares than to place the city in overlapping or bad terrain.

Naturally, with all of these strategic reasons can trump the above concerns. This doesn't come up that much however. Usually strategic concerns only affect the order I build my cities, not where I place them.

-Drachasor