A few numbers (maintenance costs for # of cities)

[Velociryx]

Have been fiddling around with expansion paradigms, and while I disagree that ICS is "dead" I can say that it really doesn't pay--at least, not in terms of my playing style.

Ran a brief stint with no Civics to muck up the numbers and got the following on a standard sized map:

Cities 1-2 - No maintenance costs

City 3 - Maintenance Cost = 2gpt

City 4 - Maintenance Cost = 4gpt (research lowered to 90% to compensate)

City 5 - Maintenance Cost = 8gpt (research lowered to 80% to compensate)

City 6 - Maintenance Cost = 12gpt (one of these was for units, and it went away the following turn...not sure what's up with that) (research lowered to 70% to compensate)

City 7 - Maintenance Cost = 14gpt

City 8 - Maintenance Cost = 17gpt

City 9 - Maintenance Cost = 20gpt (research lowered to 60% to compensate)

City 10 - Maintenance Cost = 24gpt

What I'm seeing is a gradual increase in costs, and I'm finding that I can keep pace with those rising costs fairly well with an early emphasis on cottages and other commerce boosting terrain enhancements.

Game score (and thus, the way other AI civs react and respond to you) seems to be driven largely by the number of cities you have/amount of land you control. After this wave of expansion (now, granted, 10 cities isn't exactly hard core ICS, but this was just a quick test to see what the maintenance curve looked like), I was at the top of all the power charts, and was on tech pairity, if not slightly ahead of all my near rivals.

More research needs to be done in this area. I'll be the first to admit that I am NOT a typical ICS-er, but if I can cope pretty well in the early game with a relentless expansion drive, then I'd imagine that a hard core ICS-er could do even better.

In any event, the long and the short of it seems to be that city maintenance seems to plod along on a linear curve, rather than a geometric one, and there's a tickle at the back of my brain that says this might not be enough to truly kill off ICS...when some of the hardcore guys get here and play with it, we'll see what they have to say. In the meantime, it's something to mentally chew on...

-=Vel=-

[Maniac]

Could you please post the numbers seperating between maintenance due to # of cities, and maintenance due to distance from capital? I assume it's especially the increase due to # of cities that would have to make ICS unprofitable.

Though personally I hope ICS is still possible. I'd like it to be not THE but A strategy to play the game well.

[Velociryx]

Will have to run another test for that. On the financial summary screen, I didn't see a breakdown of those costs, differentiating between distance from the capitol and raw numbers of cities, however, I did note that the cities further from the capitol tended to be my more expensive ones (city number 9, for example, added 4gpt to my cost, and was quite far from the capitol. City 7 was much closer (I was back-filling) and only added 2gpt. I suspect then, that they're rather "rolled together" but one test isn't enough to verify that for certain.

-=Vel=-

[Madcaptn]

just for the heck of it i'll go make a 1 vs 1 game on a huge terra map on quick & see how the cost for up to 20 cities goes.. if i dont run bankrupt heh

*edit* did 20 cities on Warlord (1 of them being the capital so I did 19 settlers) & the numbers were kind of weird..

From Capital to 2nd city = 0g
From 2 to 3 = 0g
From 3 to 4 = 1g (went up 1g)
From 4 to 5 = 2g
From 5 to 6 = 3g
From 6 to 7 = 4g
From 7 to 8 = 5g
From 8 to 9 = 7g (went up 2g)
From 9 to 10 = 8g (went up 1g..)
From 10 to 11 = 10g (went up 2g)
From 11 to 12 = 12g
From 12 to 13 = 15g (went up 3g)
From 13 to 14 = 17g (went up 2g)
From 14 to 15 = 20g (went up 3g)
From 15 to 16 = 22g (went up 2g but the next turn I was still at 16 cities but it went from 22g to 24g...and that's only cities maintenance that I'm calculating)
From 16 to 17 = 27g (went up 3g)
From 17 to 18 = 30g
From 18 to 19 = 34g (went up 4g)
From 19 to 20 = 37g (went up 3g)

This was only the cities maintenance ( it doesn't count civics cost & such, just plainly the cities maintenance & this was all without distance maintenance )

The numbers are probbly even higher on higher levels of difficulty too & like I said this was only cities maintenance but I was paying much more than 37g per turn trust me heh

[Brizey]

This is going to be hard to crack empirically.

What we know:

Lots of low integer increases (means rounding error is huge)
Slowly increases something close to (NoC)/5
Some other factor modifying this basic formula

It could take hundreds of examples to crack the formula.

What you really need to do is to do lots of game saves and try founding new cities in differeny locations, etc, starting with the same game save.

It looks like the "other" part of the formula creeps in around 9 or 10 cities. I would guess this is probably where you get beyond a basic ring of cities and have a city that is two "city spacings" away from the capital. But we would need to test it to be sure.

[Fried-Psitalon]

You really think ICS is a viable strategy in any form? It leads a civilization first to tech stagnation, then rapidly towards bankruptcy, which leads to mass disbanding of units.

If you could explain how you still find ICS to be a legitimate strategy, I'd love to hear it - I've had people try mad REXing in MP and watched them go bankrupt and get killed by the barbarians before I even lift a finger.

[The diplomat]

We should not confuse ICS with expansion. ICS is dead because it is no longer possible to just build new cities real close together forever. For one, the game does not allow you to build cities less than 2 tiles apart. Secondly, if you did build lots of low pop cities, the maintenance costs would bankrupt you.

Intelligent expansion is not just a viable strategy but it is a neccessary one. Players need to expand but they need to make sure that the new cities can support their maintenance costs. So, you'll need to develop your cities so that they produce lots of commerce so that they can support further expansion.

I would add that all your cities in civ4 should be high pop cities. In civ4, 1 size 12 city is better than 12 size 1 cities. Having any city that is stuck at less than ~4 pop is a bad idea!

[Fried-Psitalon]

That's my point. Vel knows the difference between the two, and he's saying ICS - the maximum placement of cities at minimum distance regardless of the terrain, including ice and desert - is still a viable, reasonable strategy. I have a lot of respect for Vel, but I have to seriously question that claim.

[jkp1187]

Originally posted by Fried-Psitalon
That's my point. Vel knows the difference between the two, and he's saying ICS - the maximum placement of cities at minimum distance regardless of the terrain, including ice and desert - is still a viable, reasonable strategy. I have a lot of respect for Vel, but I have to seriously question that claim.

But he did say that the strategy doesn't pay.

Glad to see that Vel is back, though. His strategy guide for SMAC was awesome....

[Velociryx]

Hey guys!

Granted, I've not played a sufficient number of games to form any definitive conclusions yet, BUT....based on my observations thus far, an increase in maintenance costs of +2gpt - +4gpt (the max I have seen) has not been sufficient, thus far, to slow my expansion down in the least.

Further, in continuing to expand un-interrupted through the ancient and classical eras, I've encountered the following:

* The AI is positively terrified of me (I'm guessing this is based on the fact that I have more cities than they do)

* I can keep tech pairity (so far, I've been able to maintain at least a 60% research rate)

* I've not had to invest in much infrastructure, because my city pops are low (a classic strength of ICS, as we all know)

I've been able to get around the increasing maintenance costs by doing the following things:

1) Gold mines - so far, in both of the games I've experimented with pushing the number of cities, I've had sources of gold (4 the first time, 6 the second....that influx of gold does a BUNCH to counter the gradually increasing maintenance costs as you get more cities)

2) Cottages

3) put some of my spam cities to building wonders I know I won't complete (and when I don't, I get a gold "refund" for shields invested.

4) Trade LOTS of resources for gpt to the AI (keeps my relations up, and allows me to continue to spam cities).

Now, there may come a point when the costs for adding a new city simply overwhelm the benefits of having yet another center of production. Again, I've not played with it long enough to be certain, but I can tell you this....as someone who DOES NOT use ICS, as a general rule, I'm hardly the best person to evaluate it....a die-hard ICS'er could, no doubt, achieve greater efficiencies at it than me, but I'm not doing bad with that which is supposed to be dead and buried....at least not so far....

-=Vel=-

[Velociryx]

And I should also add that it is WRETCHEDLY un-fun....but...for the sake of experimentation....

-=Vel=-

[Fried-Psitalon]

What difficulty are you playing on? It gets more painful on higher levels. Also, those gold sources are very lucky things.

[DeepO]

Vel, keep one thing in mind please: it is not the goal of CIV to get rid of ICS in itself, the goal is to make ICS unworkable as the 'best' strategy for CIV.

So, while you can probably have some softcore ICS in your games, you should really compare these to 'normal' games, and see in which you are doing best. If you conclude that ICS is the only winning strategy, CIV has failed. Concluding ICS is a workable strategy but not the best might be interesting as an experiment, but doesn't really tell you a lot.

For those comparing maintenance: you have to split out between number of cities, and distance corruption. You also need to standardize city sizes, as maintenance will depend on size. You can't compare a 1 city empire to a 10 city empire if that also means your first city has grown to size 6 in between. The best would be to compare everything at e.g. size 10. Or 6, doesn't really matter, but too low and you're seeing too much effect from rounding issues.

BTW, everyone is always talking about the effect an extra city has on maintenance, and this is of course important. But there is a similar thing at work with the city sizes: if you let a city grow, without caring for the right improvements or buildings, that city will cost you more than it will bring. This mechanism also poses a limit: a 4-city empire, all cities at size 10, without any CHs, markets, cottages, or any other form of money generators is going to go bankrupt just like a 10-city empire, all of size 1. If those same 4 cities of size 10 would have had some building focus on gold, they are generating cash instead.

DeepO

[vmxa1]

It looks ICS is not great idea, at least. I am not too sure at this point even Rexxing is a great plan.

It seems you will suffer from having more small towns with little infrastructure. Getting culture going is much more useful than in III.

Great People are so important that is looks (not saying IS) that is a better way and getting larger pop is one means of increasing it.

I have not played enough games too make an definitive statements at this time.

In my first game (played all the way out and voted) I was looking to switch civics a lot, but now I am not sure that is all that useful.

[Velociryx]

Started a new game, random civs (I got Germany, Frederich). Average world size, noble level.

13 cities so far, some of them founded ON THE FAR SIDE of another civ (my main holdings ---> English ---->batch of new cities to the south of england).

Maintenance costs just passed 40, but I've got courthouses available now, which should rein them back in.

Basically, I haven't stopped expanding so far, and am in *total* control of the continent, and the three rival civs sharing it with me.

ICS might be weakened, but it ain't dead....at least not from what I've seen.

-=Vel=-