Optimum City spacing

[rookie]

I had this figured out for Civ3.

What is the optimum city spacing for Civ4.

Also, is it better to let your population go to 2 before making the first worker? I just played a game where I built a worker in the first 15 turns and then chop rushed a settler. What is the best build order for the first 45 turns?

Thanks in advance.

 

[Fosse]

Civ IV's resource system and city maintenance costs, in addition to killing the old Infinite City Sleaze style of play has also effectivly killed the idea of optimal city placement (with some additional help from the mandated 2-tile minimum spacing rule). You have a limited number of cities with which to build your nation, and therefore you have more important things to worry about that fitting your cities into an "optimal" grid as you might in previous games.

Those things include:
- Access to fresh water and/or coast.
- Ability to work and claim resources.
- Food, production and commerce potential.
- Distance from your capital, which impacts distance maintenance.
- The potential site blocking other potential sites from being legal.

Each map will therefore have its own "optimal" placement, and any two players will have a different opinion on what that is.

 

[GoodGame]

Its still about 2x City Radius +1, or count off 4 empty squares between cities. Overlap isn't evil anymore, because it takes a very long time (1/3 or 1/4 of the game?) to get a city up to full size, unlike civ3. And full size really depends upon many factors, including what techs you have and what buildings you have built. I'd figure about pop6 is what a city will want to reserve in tiles, with minimal support, and that the natural pop cap might be as low as 7 to 10. So overlapping of the fat cross might be ok unless you want a lot of Specialists/Great people from that city.
Like Fosse said, if you want the city to have fresh water and a huge farm potential, the number of cities you can optimally place will go down. If you want military unit factories with minimal buildings, you'll have trouble getting it to work all the tiles you reserve for it if you reserve full fat crosses.

When to build your first settler/worker kind of depends upon the cities natural terrain. Flood plains make a difference. I generally do some early tech research, so I like to wait until pop 3 before building a worker.

Usually I'm warrior until pop3, then worker, chop worker, chop settler. But sometimes a work boat is better before a worker. And sometimes with 3+ flood plains, it's better to wait to pop4, since it comes way before that first worker could.

Pretty much you have to balance it every game with the city growth rate vs. your early tech race plans (religions) vs. your early city growth plans. Thats why I like CIV4 better than CIV3----recipes are less fool-proof.

 

[razzor7]

I'm a noob to civ 4, so I always place my cities in the blue circles that the comp puts down... is that wrong?

 

[Elandal]

The blue circles aren't probably the worst places to settle. There are a number of problems in how the blue circles are generated, but for as long as there's a high amount of explored (and thus known) space around the circle, it should be OK.

I've turned the unit suggestions off, so I don't see the blue circles myself. But when I still had suggestions on, I made my own dotmaps, then looked at the blue circles and spent a few minutes thinking about how does that circle compare to my dotmap, what possible benefits there would be in different choices, and so on. Except of course when the circles were on my dotmap and I could happily settle where I wanted while settling on a suggested spot

Lately I've looked at my dotmapping practices, and have found a number of inefficiencies there.. I've valued resources a bit too much. First things first: all cities need food. For as long as a city has food, it can become useful something whatever the something is. It may grab resources in it's fat cross (to be worked) or just grab resources (for happiness, health, strategic resources, or for trading) without expectation to work them (resources in the third ring, to be within cultural borders but not in the fat cross, or resources I don't expect to have food to work (eg. ice furs)). It may be just for food, with goal of running specialists (eg. GP farm). Or it may grab cottable land. Or high hammer tiles with food to work them. Or it could be strategic placement (blocking areas, providing transoceanic canal, or cultural warfare - probably lots of possible strategic reasons). Actually, sometimes a strategic city doesn't even need food ("Ice Flipper" in one of Sullla's games being a nice example), it just needs to exist - but that's a rare case. So back to the basics: food.

 

[Grimus]

Not sure how well the AI plans city placement with the blue circles, though I guess you can just look at the AI civs city placement to judge (which isn't that bad).

I disabled blue circles and things right when I started playing. I rather make my own decisions, even if they're wrong, lol. I just wish there was a way to disable the two building recommendations, or that it would take you right to the tech tree instead of asking you what tech you want to research next and giving two recommendations (I need to see the big picture!).

 

[Elandal]

I'm quite certain the AI (and barbs) settle on blue circles. The difference being of course that their blue circles are placed based on their knowledge of the map, not yours The reasoning is fairly simple: if the game calculates a Found Value for a tile, and within a certain range from the settler there are maybe 2 local maxima for the value, those would to human be presented as blue circles, but to the AI they would be the two tiles it'd somehow choose which to settle.

 

[Grimus]

I used to be a city placement nazi, and try not to overlap as much as possible. I would cry over and be disgusted with deserts, tundra, and mountains because I couldn't work them. If AI civs settled to close to me or messed up my future city planning grid, I would be so mad and go to war with them! lol

Now though, I realize most cities won't reach pop 20 anyway, so why bother? Though, you could argue that it gives each city a better selection of tiles to work and choose from.

I prefer to overlap my cities now, even building them right outside of the 2 tile radius between them, the only exception is maybe the cities around my capital where I space them to give my capital plenty of room to grow.

I try to make sure I'm utilizing every bit of land that I can work within my territory. Then I go into each city and assign which tile belongs to which city, some cities I'll favor more than others... and my capital I always give it the entire 20 tiles to work.

 

[Amuze]

I'm quite certain the AI (and barbs) settle on blue circles. The difference being of course that their blue circles are placed based on their knowledge of the map, not yours The reasoning is fairly simple: if the game calculates a Found Value for a tile, and within a certain range from the settler there are maybe 2 local maxima for the value, those would to human be presented as blue circles, but to the AI they would be the two tiles it'd somehow choose which to settle.

Actually, they don't. I was playing a game last night about saw Cyrus' Archer/Settler coming towards the area I was about to lay my new city down at. The two blue circles it offered me were OK, but upon using 1 more turn to settled at my preferred spot, the AI had settled between the 2 blue circles. So much for that theory.

 

[notagoodname]

I try and get 1 bonus food resource per city, simple as that.
It's an easy rule of thumb to follow and with 1 bonus food i'm all set to work cottages/specialists and mines as nessesary.

 

[weimingshi]

At start of every game, I always assess the surronding, so I can found the most optimal specialized cities.

One city for great people generation, This is the place where I can have the most food output, preferably alot grassland with river close by and at least 4 food resources.

one city for commerce to build wall street and oxford univerty. lots and lots of flat land required to build cottages. at least 1 food resource and some luxury resource is a plus.

2 cities for army production. one build heroic epic and west point, the other build iron works and red cross. Those cities needs max hammer output. best location would be lots of hills and enough food resources to grow the city to 20+. Those cities will only build mines and workshops, if along river, build watermill. Workshops would only be build after state property civic.

Those specialized cities cannot overlap each other, because I need them to work the maximum tiles possible. Other minor cities can overlap them though, they just won't be working those overlapped tiles thats all.

 

[paperchase417]

I plot my cities typically 7 spaces away from the nearest city in a spot so that it grabs as many resources as possible within its workable 12, but also not to conflict with other resources that I might be able to fit into the workable 12 of another new city. I hate over lapping, and only do it when its absolutely necessary.

 

[KMadCandy]

Each map will therefore have its own "optimal" placement, and any two players will have a different opinion on what that is.

ain't that the truth. i've been playing team games with hubby lately and we are so opposite. he can't stand to overlap cities, it's like the biggest sin ever in his mind. and he does place the actual cities well so that they get good resources etc. but for example the other day he left a gap with 4 flood plains and 2 forests just to get the resources he was aiming for without overlapping. i said "honey that's fine, but i'm gonna put a city right here to use those FP later on so be sure you're working the tiles in the overlap part when my borders there are gonna expand."

in a perfect world i'd not overlap cities. but to me, wasting filler space that could be productive is a far greater sin than overlap, in a game that will go on long enough that you really want nice cities and it's not just 'kill kill kill'. it's easy enough to manage the "main city" to be sure it gets the right tiles and that the filler city only uses the leftovers. in that particular game he was amazed how well a couple of my "filler cities" were doing by the end.

i mean come on, WASTING four flood plains!!!!!!!!!!!!!!!!!!!!!!!!!

 

[HiroHito]

I'm a noob to civ 4, so I always place my cities in the blue circles that the comp puts down... is that wrong?

No its not wrong, most of the time they choose the best location save strategic considerations.

Try to understand why its the best location before settling, and keep the cohesion of your empire in mind, you can always conquer that juicy city the AI will build later on after all

 

[DarkFyre99]

Actually, they don't. I was playing a game last night about saw Cyrus' Archer/Settler coming towards the area I was about to lay my new city down at. The two blue circles it offered me were OK, but upon using 1 more turn to settled at my preferred spot, the AI had settled between the 2 blue circles. So much for that theory.

By placing a city, you've altered the landscape. Certain tiles are unavailable to found a city on because they are within two tiles of your own, and certain tiles are likely to be unavailable to be worked, due to cultural pressure. The "optimal placement" algorithm seems to take into account current cities when figuring out where to place a new city.

 

[Elandal]

You can actually see a "Found Value" for a tile if you enable chipotle and mouse over tiles with shift pressed. I have no clue how that number pair is calculated or what it means, but seems to me that the blue circles appear where the numbers are higher than in the surroundings.

Anyway, if the AI does NOT use that calculated Found Value for city placement choice, then there's Yet Another (tm) algorithm that is not presented to the human nor via debug interface that is used by AI. I find that quite unlikely, as that'd mean duplicate code (two algorithms for one task, one used by AI, another presented to human).

Now why does that AI not settle on the blue circle you see? Assuming that you both have same knowledge about the map in the area, differences may arise from numerous sources still (I don't know what all is taken into account), but founding a city in the area definitelly changes the plots - a number of tiles become illegal to settle on, while several legal tiles now would have heavy overlap with the newly founded city, and so on. If anyone cares enough, he can test this out: enable chipotle cheatcode, look at the blue circles and the Found Values presented in debug interface, then settle a city, give yourself another settler with WB, look how the Found Values changed and look for new blue circles the new settler would see.

 

[Welnic]

Actually, they don't. I was playing a game last night about saw Cyrus' Archer/Settler coming towards the area I was about to lay my new city down at. The two blue circles it offered me were OK, but upon using 1 more turn to settled at my preferred spot, the AI had settled between the 2 blue circles. So much for that theory.

The AI now calculates those circles just using the resources that it can see at the time. If it hasn't discovered bronze working then it won't be able to calculate any copper in the equation. So if Cyrus' techs were different than yours his circles could have been different.

 

[UniverseZero]

The AI now calculates those circles just using the resources that it can see at the time. If it hasn't discovered bronze working then it won't be able to calculate any copper in the equation. So if Cyrus' techs were different than yours his circles could have been different.

No because maybe the circles are calculated taking the unrevealed resources into the equation too, I don't know if you have done this before but if you open the world editor in a single player game you can see all the resources that you don't have techs to see in the real game.

 

[Elandal]

You can see the resources in worldbuilder (of course - you can edit the map in worldbuilder), but if it's not seen by a civ (no tech to see it), it's not counted in the civs found value calculation.

To confirm this I guess one could open up a game, look up the values, add required tech to see a resource, go back to the game, and look up the values again. You need to enable cheatcode for this to get the debug data.

 

[notagoodname]

Overlapped cities are not nessesarily bad at all. In fact they can be very useful.

It seems most people don't know this but overlapping cities can use the other cities overlapped tiles when the other city isn't using them.

Just go to the city management screen and click the greyed out overlapped tile that is reserved for the other city. It will now be reserved for this current city.

So overlapped cities can actually be used to grow each other cottages while 1 of the cities focuses on production instead.