City spacing

[aoav160]

I'm a n00b to Civ and was curious as to what is the best distance to space your cities?

[fanes7]

More space is more good

[aoav160]

Thanks, but do you have an actual number? I noticed the AI has them pretty close.

[vmxa1]

There are many things to consider when using spacing for cities. In general the 19 size works well at levels of Monarch or lower. Tighter spacing use often a better plan above that level.
If you want to war or build has an impact. If you want to war you may go with closer spacing to get more cities sooner and defend them easier. It also lets you crank out units sooner. If you want to make temp cities to create troops from and later disband. I mean it is a complex subject.
If in doubt, go with CxxxC. This can handle most maps and most levels.

[badams52]

As you are new let me give you 3 excellent links in decreasing order of importatnce. THe first link goes over to CFC which produced ring placement.

Ring Placement

City Replacement Redux (for the third time)

Appropriate City Placement

[badams52]

Originally posted by aoav160
Thanks, but do you have an actual number? I noticed the AI has them pretty close.

Actually, compared to many advanced players who win at high levels, the AI spaces it's cities father away.

[aoav160]

Thanks!

[Theseus]

I would recommend pretty tight spacing until you have a good handle on things.

3-tile, or CxxC, adjusted for terrain.

I know that feels too close, but think about it: for much of the early game (which is where you set up to win or lose), this will get you the most optimum usage of your best tiles.

[Gufnork]

I tend to build cities exactly three spaces apart. Makes your defense more flexible in the early game, reinforcements are never more than one turn away.

[fanes7]

It takes me about half an hour, :-)

You can look, the black is your capital,

And red is your city nearby.

And green is "both-use" block.

You can easiely expand by this.

[Theseus]

fanes7' example is of what we call 4-tile and 5-tile placement.

It is fine, but I again recommend 3-tile placement (stretching to 4-tile according to terrain) for a newer player.

Remember, the key is to maximize turn-by-turn use of your best tiles.

[Arrian]

Ok, I'm just reading the "Ring" placement thing for the first time. I kinda wondered what you guys were talking about...

It almost feels like an exploit to me. But, when faced with geographical constraints, plus my tendency to build a FP close to home & rush a palace move later, it actually doesn't sound like a good idea for me.

-Arrian

[alexman]

Arrian, I usually do the same as you with the FP, but this city placement is worth it, even if it's just for the early-game boost.

And the actual pattern is not as restricting as you may think, especially since distances are rounded down to the lowest integer number.

[Arrian]

So 2 and 2.5 are considered the same?

-Arrian

[Arrian]

It would also conflict with placing cities to take advantage of rivers, coast, defensive considerations (building on a hill, for instance, or bonus tiles... wouldn't it?

Hmm... I guess I should have a good look at the map next time I load up a game.

It still kinda smacks of exploit, in that you are essentially deliberately "tricking" the game into thinking all cities founded at a certain distance are one city (for corruption purposes).

-Arrian

[Dominae]

Originally posted by alexman
...especially since distances are rounded down to the lowest integer number.

I've found no talk about this fact on the RCP thread over at CFC. Are you sure?

As you pointed out, if the distance 3 and distance 3.5 rings are equivalent, RCP suddenly becomes quite flexible.


Dominae

[alexman]

Yes, 2 and 2.5 are the same. You can usually find an RCP pattern that's a good fit to the terrain, although some maps are obviously better suited to this placement than others.

And I agree that it's an exploit, BTW, since the AI doesn't do it. Although balancing the RCP placement benefits with taking full advantage of the terrain can provide for some interesting strategic choices.

[Harovan]

Here's an example pattern of using the Ring City Placement approach. Capital letters are the cities, lower case letters the tiles they use. A, C, E and G are size 6 cities (no Aqueducts, only tier-1 improvements like markets and libraries), B, D, F and H are size 12 cities (get Aqueducts and tier-2 improvements like Banks and Universities). X is the capital and has 16 tiles to work. The 4 tiles above size 12 can be used as spare tiles for the other cities in case the terrain makes it necessary (like there's not enough food or similar restraints). The 8 cities in the ring are the same distance from the capital and suffer all the same (fairly small) distance corruption.

. . . a a b b b . . .
. . a a a b b b b . .
. h h a A b B b b c .
h h h h x x b b c c c
h h H h x x x x C c c
h h h x x X x x d d d
g g G x x x x d D d d
g g g f f x x d d d d
. g f f F f E e d d .
. . f f f f e e e . .
. . . f f f e e . . .

This works of course only, if the terrain allows it. A single mountain on one of the ring sqare can disrupt it. But if it is possible to build, this city core will production-wise be equivalent to a much bigger AI empire, while it's very easy to defend. Outside of the ring will be built/captured only a few cities to secure resources and luxuries. Later can with the help of a leader be build a FP city with a similar ring around it. Two of these cores should be able to dominate every game up to a large map.

[Dominae]

Originally posted by alexman
Yes, 2 and 2.5 are the same.

I know it's a waste of time to not believe you alexman, but I need evidence!

Do you have any test results I could look over?

I'm only asking because in a couple of PBEM games I started before I knew about RCP I remember kicking myself for not having placed a couple of cities just 0.5 distance points closer, because they were suffering from different Corruption. I could, as usualy, very well be wrong...


Dominae

[Arrian]

Let me see if I really understand this (having read all 6 pages of the CFC thread now):

1) All cities built the same distance from the palace or fp are assigned the same corruption number, and are treated, for corruption purposes, as the same city.

EDIT: nope, didn't understand it. The first ring of cities is all "#1" but the next ring is assinged n+1, where "n" is the number of cities in the prior ring. See alexman's post below.

2) Moving 1, 3, 7, or 9 on the keypad is a move of 1. Moving 2, 4, 6, or 8 is 1.5.

3) Distance is rounded down.

Therefore, if you manage to get this to work, you would see huge drops in corruption to your mid-range and long-range cities. In Democracy, with courthouses, police stations and WLTKD, this could result in an amazingly efficient empire. And I will admit, I have a thing with trying to build efficient empires.

Any thoughts on how the Commercial trait interacts with this? At first glance, it would seem that this devalues Commercial, since if you are building rings, the OCN boost you get will not be approached (standard map OCN is 16, but 20 for commercial civs... so you would need 17 rings to see a benifit from Comm's corruption reduction). You would still get the extra money, of course.

-Arrian

[Arrian]

It also strikes me as more of a large/huge map strategy. How many rings can you get on a standard map (especially if you want to get a second set of rings around a FP) without ICS spacing? I'm not willing to drop to ICS... 3-tile I can do (even though I'm not especially fond of it).

-Arrian

[alexman]

Originally posted by Arrian
At first glance, it would seem that this devalues Commercial, since if you are building rings, the OCN boost you get will not be approached (standard map OCN is 16, but 20 for commercial civs... so you would need 17 rings to see a benifit from Comm's corruption reduction).

RCP indeed devalues the Commercial trait, but not as much as you indicate above. Cities at the same distance have the same rank against OCN (say 1 for the first ring), but the next ring has a rank of N+1, where N is the number of cities in the first ring.

Dominae, I'll see if I can find some evidence for you, but this has been tested by others as well. Qitai first discovered it over at CFC.

[Arrian]

Oh. That's a BIG difference. So if you build 4 cities in the first ring and 9 in the second, you would get this:

First ring cities are assigned "1"
Second ring cities are assigned "5"
Third ring cities are assigned "14"

Whereas under any other build pattern, you would most likely have:

1
2
two tied at 3
5
three tied at 6
9
...

Something like that.

-Arrian

[alexman]

Arrian: You got the idea!

Dominae: I found the save from my RCP test.

Ravenna is at 3.5 but shares the same corruption as all the cities in the 3.0 ring.

Caesarea is at 5.5 and has lower corruption than the 6.0 ring.

Jerusalem is at 6.5 and has the same corruption as the 6.0 ring.

[Dominae]

Originally posted by alexman
Dominae, I'll see if I can find some evidence for you, but this has been tested by others as well. Qitai first discovered it over at CFC.

Just point me to where it was posted (if you recall)...no need to do any extra work!


Dominae

[Arrian]

Now I want to load up a few games of mine and work out what corruption numbers my cities are assigned, and ponder if RCP would have really improved my empire.

-Arrian

[Lawrence of Arabia]

THis is a cool strat. Whoever came up with it is a genius. I havn't modified my game in like a year, but I think this is a great way of moving up to the next level.

[Arrian]

I'm not so convinced.

I'm not sure that having rings of cities, with the corresponding big jumps in # corruption is always better than having the more standard incremental corruption increase of other build patterns.

Hmm...

Let's take a standard map, with its OCN of 16. Assume the game is on Monarch level. That, if memory serves, results in a 10% penalty to the OCN for the human, which means the OCN drops to 14.4... rounding down to 14.

If that's the case, would the best use of RCP be to do have your first 2 rings have a total of 13 cities in them? Then the third ring, with lots of cities, gets assigned #14 - and keeps your entire empire under the OCN.

I'm still struggling with this. Whether or not it's really worth it ("it" being a combo of the time & effort required to plan this out and the specific terrain issues which may conflict with it).

-Arrian

[alexman]

Arrian, trust me, on most maps it's worth it big time! The reason is that all cities in a ring have less (or equal, for one of them) OCN corruption than they would have if they weren't in a ring.

At the very least, just build your cities as you used to build them, but shift them one or two tiles to keep them in rings. Don't worry about how many cities in each ring.

Take a look at a real example from GOTM 21, which at first glance doesn't look like a good map for RCP (given where Athens stands). This is back when the rounding down of the distances was still not discovered, so I was placing my two first rings at strict distances of 4 and 6.5. (the pattern would look more natural now that we know about the rounding). Players not using RCP didn't do as well in that game.

Of course, I usually use 3-tile spacing, so for your looser spacing the RCP would look different on this map. 3-tile spacing is more flexible than OCP, but that's true even without RCP.

[Arrian]

*shudders in horror*

I will try it out, though.

-Arrian