city layout

[MadScot]

Depends on whether you like overlaps, really.

If you would tend towards an OCP-style, where there is little overlap, and where you maximize the late-game potential of each city, then I'd say something like the 4.5 or greater for the first city, and don't fill in all the locations, either.

If you prefer a closer packing, then something like the 3 initial ring, again probably not all filled, and 6 or so at the next ring.

Personally, what I do is try to visualise where the locations might be for the 3, 3.5, 4 and 4.5 rings. Then depending on how the terrain inteeracts with those locations, I try to pick one. So if 4.5 gives me lots of nice river or coast locations then I might pick that one.

Slavishly picking a fixed radius is not a good idea; there'll always be a map where it doesn't at all work, or where you should do some other kind of build entirely. (An isolated island start almost demands priotising coastal locations, for example, and a lot of tundra would tend to force ICS)

 

[cgannon64]

Mix of terrain and OCP, with terrain being the main factor (I don't mind a little corruption).

 

[toothpaste100]

I have a question, on the picture I have attached, would the squares I have placed a blue dot equal 6 as well as 4.5 squares away from the capital city in the center marked with red? If so, would it recieve the same corruption as the squares that are bordered with blue?

 

[MadScot]

All the "blue dot" squares are 6 "corruption distance units" away from the centre - all the green open circles would be the RCP=6 ring.

They would all be more corrupt than any of the open blue circles, which are all at RCP=4.5

edit: you always take the fewest number of diagonal steps in determining the distance in RCP-units

 

[toothpaste100]

What about NE 2 squares and then NW 5 squares. Wouldn't that equal 4.5 and 6 squares away depending on how you look at it?

 

[Arturo]

RCP works like a charm for me. I usually (when I say usually I mean in my previous 2 games!) space my first ring 4.5 tiles away from the capital. So far, due to terrain (mountains or ocean) I have inly managed max of 5 cities in the first ring. My second ring is about 6 tiles from capital, and in between the first-ring cities.

After that it is anywhere I can settle for resources and expansion.

 

[MadScot]

Originally posted by toothpaste100
What about NE 2 squares and then NW 5 squares. Wouldn't that equal 4.5 and 6 squares away depending on how you look at it?

No, because corruption uses the shortest distance to the capital.

So NE 2 and NW 5 is actually 2 tiles N and 3 tiles NW (or 3 tiles NW then 2 N, if you prefer). So it's 2*1.5+3*1=6 RCP units away.

Any other path works out to a longer distance, so is not used. If you go 2NE and 5 NW that's a total of 7, for example. There's no way to get there in as few as 4.5 units.

What you have to do is visualise 4 arms along the NE, NW, SE, SW directions. These are along sides, so cost 1 point per tile. That's the best cost, so you move as far as you can until you can get to the target location along a diagonal. It's better to pay 1.5 for a diagonal than 2*1 to go along-then-up, so once you get to a diagonal you just move up the diagonal. 1.5* the diagonal move plus 1 * the along-the-arm move is the RCP distance.

I had a grid showing all the costs somewhere .....

 

[toothpaste100]

There is a very good grid example in the Ring City Placement thread in the strategy articles, page 2, mid-way down the page I believe.

I am thinking about doing 3 rings, with 4 cities on each.
First ring at 3, Second Ring at 4.5 and the Third ring at 6. Will this inhibit my growth at all?

 

[MadScot]

Well, toothpaste, you'' have 4 pretty much corrupt-free cities, 4 more at a moderate level, and so on.

But you will be constricted when you get to the industrial age, because you only have 9 workable tiles per city, yet you'll be able to build much bigger cities by then. But until then it looks reasonable enough.

OK, can't edit it to attach it, here's the RCP distances in the Eastern segment (including the NE and SE axes)

 

[Lord Beverage]

@toothpast100
Your squares with blue dots would equal six. Do they also equal 4.5 you're asking? Obviously not, for they equal six. Since the blue ones equal 4.5 they would recieve less corruption than the green /w blue dot.

Now having mistakenly read MadScott's post as '2, 4, 6" and somehow missing the 4.5 on the image as well, I made a correct image for 2, 4, 6 also including 1,3,5,7 and OCP for comparison. I now realise I'm an idiot and can't read english. Anyway, to make myself feel better I'm going to post it anyway.

RCPs should be obvious; black and black with green diagonal (RCP-5 overlap) are OCP.

[edit]
And of course in the time it took me to make it toothpastes question was answered five times over.

[edit2]
Gah! And a better image is up too! It's just not my day...

 

[MadScot]

Originally posted by Lord Beverage
Now having mistakenly read MadScott's post as '2, 4, 6" and somehow missing the 4.5 on the image as well, I made a correct image for 2, 4, 6 also including 1,3,5,7 and OCP for comparison. I now realise I'm an idiot and can't read english. Anyway, to make myself feel better I'm going to post it anyway.

Nice graphic

I find it interesting that OCP puts 4 cities at an equal RCP distance, then the other two at the same distance as each other. Are the people who disdain RCP as exploitative going to abandon OCP too, since it is (accidentally, but quite definitely) using the corruption mechanics to improve the civ's production.

And since the AI seems to like a pseudoOCP also, is the Ai using a pseudo RCP also??

 

[Lord Beverage]

I had always been inadvertantly using OCP since Civ2 many many years ago, never for exploiting corruption, but because it creates a nice pattern of cities with a minimum of overlap and no unused terrain. I had always perceived that as the true benefit of OCP until learning how the corruption mechanic truely worked.

 

[Neomega]

I have used OCP since Civ I, but lately I have been trying the 3^2> strategy. Which basically packs in 6 cities around the center city, and shares 10 tiles amongst 7 cities.

 

[Quasar1011]

I voted 'other'. Terrain mostly influences my city placement; however, I often place cities in not-so-optimal spots, in order to culturally capture another civs city. Often they built it one square from where I was going to build. The nerve of them!

 

[JonathanValjean]

My city placement generally approaches OCP, but it is slightly more compacted than that for a little less corruption.

 

[mad-bax]

The problem with RCP is that it's easy in theory but difficult in practice. Once the ring radius is selected you have little choice over city placement. Early in the game you may not have explored enough of the map to choose the correct radius, or you may have rolled a (near) coastal start. This means that some of the tiles that you need to place cities on are not available (sea, lake, mountain) and some of the best tiles will not be available until culture is put in the cities. the effect of this is that instead of having say 6 cities in the inner ring you have only 4 if you're lucky.

I am playing an SG now where we are basing the RCP on a city that will one day be our capital. This method still has its' problems, but in some situations could perhaps be effective.

I think that RCP can be the right way to go in some situations, but otherwise I use a slightly dense build of around 11 tiles per city focussing on sites which allow a mix of city types, settler factories, worker pumps, commercial centres, unit production et al.

The last point I would make about RCP is that it's benefits increase the longer the game goes on. If you want a quick military win I wouldn't use it. If I were going for a 20K then I would if I could.

 

[Stapel]

It is all about terrain to me! I build my cities close, so they can share tiles. This very important when mm-ing a lot.