ICS (Infinite City Sprall) will, for the purposes of this post, be defined as a strategy of building an empire of small cities packed as tightly together as possible.
For contrast, some alternatives include:
* A smaller number of very large, well developed and efficient cities.
* A large, conquest fueled empire of conquered cities.
* A large, spread out or even sparse empire
A lot of people dislike ICS, as it gives a very bland and repetitive empire. Personally, I think ICS should be a viable strategy with the right social policies. However I also think there should be alternatives which are competitive. At the moment, it seems that well managed ICS is simply streets ahead of the alternatives.
Here I will attempt to list all of the factors that contribute to ICS being better than the alternatives. I tend to use civ4 as a reference point, partly because it's the most recent comparison point, and partly because I've played a lot of civ4. I am by no means saying that any one of these factors should be changed, I'm just listing them. To me it's the sheer bulk of factors that is the biggest problem, not any specific one of them. This list can hopefully help explain the problem to people, and maybe even help provide inspiration for solutions. I'll try and update the list with more suggestions that people might post here (as long as I agree with them).
This is also not another civ5 sucks thread, I quite enjoy civ5. There's just some things that I think are worthwhile improving about the game in potential future patches.
So, the list:
* Colliseums are the best value happiness building. If the happiness buildings got better rather than worse, then you'd be rewarded for in-depth development of less cities rather than lightweight development of more cities. More expensive later buildings is fine, but for all those extra hammers a Theatre gives the same 4 happiness and costs 5 instead of 3 upkeep.
* Unhappiness per city is supposed to be the big ICS killer, however it can be easily removed or in fact reversed by social policies and the Forbidden Place (which the AI also doesn't seem to build as quickly as many other wonders, although maybe that's just my experience). The upshot of all this is that a small city with a colliseum can actually be a happniess profit.
* Too many flat per-city bonuses. Maritime city states and a number of social policies give per-city bonuses, rewarding a player for building more cities, and making 1 population cities very productive. The "city centre" hex is by far the best hex available.
* With land tiles generally a lot weaker, specialists come out better value by comparison than in civ4, even before certain easily accessable social policies help them along. With maritime food, you don't need to develop a large number of farms and food resources and grow a huge city to be working a lot of specialists, you can have a small city quickly grow to 4-7 pop and work as many specialists as you have slots available. So ICS gets a lot of specialists (particularly scientists), and specialists are good (particularly scientists).
* Infrastructure is "underpowered". It's generally less useful than in previous civs, and production is so scarce that it's a lot harder to come by. In previous civs, it was practical for cities to build a full set of specialised infrastructure, and once they'd gone to the trouble of doing so you wanted those cities to work as many of the appropriate tiles as possible. With infrastructure so scarce it makes little difference which cities are working the tiles.
* National wonders are underpowered (no need for "quotes" on this one). A specific case of the above, but national wonders seem to provide bonuses that mostly aren't even better at all compared to standard buildings, at significantly higher hammer price. In civ4, the planning of many national wonders was critical to optimal play, because they were so useful. In civ5 they're often not even worth building. This also means the "X in every city" requirements, which could otherwise hamper ICS, aren't such a big deal because missing out on the national wonder doesn't hurt so much.
* Cultural border expansion is slow. Buildings do not provide incidental culture the way they did in civ4, instead there's specific culture buildings which aren't good for anything else. ICS in a lattice avoids the need for any cultural border expansion entirely, which saves a lot of hammers for culture buildings or gold for buying tiles.
* Growth at high populations is painfully slow. Partly because the new "granary" is so far up the tech tree, but mostly because the food costs of growth rise so rapidly. This means that to get more population it costs much less to just settle a new city than to grow an existing one. It's also another reason that the dream of mega-cities, with well developed infrastructure and working many tiles to get the most out of their % bonuses is very hard to realise.
* City location is no longer all that important, so there's no real penalty to just sticking them down in a lattice. This is a culmination of a whole lot of factors:
- Resources are weaker in yield, so the benefits of claiming bonus resources are low.
- Tiles in general are weak, so settling in desert/tundra is no big deal. The hills in desert/tundra are as good as anywhere else too, and food tiles aren't necessary for ICS with maritime food.
- Cities don't grow quickly past 10 food, and rarely make it as far as 20. Working most of the 36 theoretically available tiles is reserved for OCC, maybe the capital, and games that set out specifically to do it. This means there's no real need to avoid "wasted" tiles in core cities, as they'll always be able to find worthwhile tiles to work.
- Location dependant buildings (observatory, circus, monastery, coastal buildings, some wonders like Machu Pichu etc.) do attempt to encourage planning of city placement. However, those buildings generally aren't any better than their alternatives (circus < colliseum, monastery ~= temple, observatory ~= university, mint ~= market etc.) and with production so limited it takes a long time to build both. ICS doesn't lose out on those buildings anyway, it just doesn't stack them up cleverly.
* Trade routes provide a lot of gold, but are offset by the costs of road maintenance. ICS minimises the road maintenance for the amount of trade route gold generated.
* The trade route equation also favours more cities given the same population (thanks PieceOfMind for this point). The equation is (1 + 1.25*city size), making the '1' effectively another flat per-city bonus. The free '1' is presumably to partly offset the cost of more roads to more cities, but two small cities packed into the same space as the alternative one bigger city actually need less roads in general, as the city centre tiles get a road "for free".
* Social policy culture costs are the one thing that actually genuinely hit back at ICS. One thing that mitigates this a little is cultural city states. A few early cultural city state allies can generate enough culture to buy some of the key ICS social policies (which with the exception of some of the order tree are generally quite accessible) before the number of cities gets out of control.
* Tightly packed cities are an ideal defensive set up.
For contrast, some alternatives include:
* A smaller number of very large, well developed and efficient cities.
* A large, conquest fueled empire of conquered cities.
* A large, spread out or even sparse empire
A lot of people dislike ICS, as it gives a very bland and repetitive empire. Personally, I think ICS should be a viable strategy with the right social policies. However I also think there should be alternatives which are competitive. At the moment, it seems that well managed ICS is simply streets ahead of the alternatives.
Here I will attempt to list all of the factors that contribute to ICS being better than the alternatives. I tend to use civ4 as a reference point, partly because it's the most recent comparison point, and partly because I've played a lot of civ4. I am by no means saying that any one of these factors should be changed, I'm just listing them. To me it's the sheer bulk of factors that is the biggest problem, not any specific one of them. This list can hopefully help explain the problem to people, and maybe even help provide inspiration for solutions. I'll try and update the list with more suggestions that people might post here (as long as I agree with them).
This is also not another civ5 sucks thread, I quite enjoy civ5. There's just some things that I think are worthwhile improving about the game in potential future patches.
So, the list:
* Colliseums are the best value happiness building. If the happiness buildings got better rather than worse, then you'd be rewarded for in-depth development of less cities rather than lightweight development of more cities. More expensive later buildings is fine, but for all those extra hammers a Theatre gives the same 4 happiness and costs 5 instead of 3 upkeep.
* Unhappiness per city is supposed to be the big ICS killer, however it can be easily removed or in fact reversed by social policies and the Forbidden Place (which the AI also doesn't seem to build as quickly as many other wonders, although maybe that's just my experience). The upshot of all this is that a small city with a colliseum can actually be a happniess profit.
* Too many flat per-city bonuses. Maritime city states and a number of social policies give per-city bonuses, rewarding a player for building more cities, and making 1 population cities very productive. The "city centre" hex is by far the best hex available.
* With land tiles generally a lot weaker, specialists come out better value by comparison than in civ4, even before certain easily accessable social policies help them along. With maritime food, you don't need to develop a large number of farms and food resources and grow a huge city to be working a lot of specialists, you can have a small city quickly grow to 4-7 pop and work as many specialists as you have slots available. So ICS gets a lot of specialists (particularly scientists), and specialists are good (particularly scientists).
* Infrastructure is "underpowered". It's generally less useful than in previous civs, and production is so scarce that it's a lot harder to come by. In previous civs, it was practical for cities to build a full set of specialised infrastructure, and once they'd gone to the trouble of doing so you wanted those cities to work as many of the appropriate tiles as possible. With infrastructure so scarce it makes little difference which cities are working the tiles.
* National wonders are underpowered (no need for "quotes" on this one). A specific case of the above, but national wonders seem to provide bonuses that mostly aren't even better at all compared to standard buildings, at significantly higher hammer price. In civ4, the planning of many national wonders was critical to optimal play, because they were so useful. In civ5 they're often not even worth building. This also means the "X in every city" requirements, which could otherwise hamper ICS, aren't such a big deal because missing out on the national wonder doesn't hurt so much.
* Cultural border expansion is slow. Buildings do not provide incidental culture the way they did in civ4, instead there's specific culture buildings which aren't good for anything else. ICS in a lattice avoids the need for any cultural border expansion entirely, which saves a lot of hammers for culture buildings or gold for buying tiles.
* Growth at high populations is painfully slow. Partly because the new "granary" is so far up the tech tree, but mostly because the food costs of growth rise so rapidly. This means that to get more population it costs much less to just settle a new city than to grow an existing one. It's also another reason that the dream of mega-cities, with well developed infrastructure and working many tiles to get the most out of their % bonuses is very hard to realise.
* City location is no longer all that important, so there's no real penalty to just sticking them down in a lattice. This is a culmination of a whole lot of factors:
- Resources are weaker in yield, so the benefits of claiming bonus resources are low.
- Tiles in general are weak, so settling in desert/tundra is no big deal. The hills in desert/tundra are as good as anywhere else too, and food tiles aren't necessary for ICS with maritime food.
- Cities don't grow quickly past 10 food, and rarely make it as far as 20. Working most of the 36 theoretically available tiles is reserved for OCC, maybe the capital, and games that set out specifically to do it. This means there's no real need to avoid "wasted" tiles in core cities, as they'll always be able to find worthwhile tiles to work.
- Location dependant buildings (observatory, circus, monastery, coastal buildings, some wonders like Machu Pichu etc.) do attempt to encourage planning of city placement. However, those buildings generally aren't any better than their alternatives (circus < colliseum, monastery ~= temple, observatory ~= university, mint ~= market etc.) and with production so limited it takes a long time to build both. ICS doesn't lose out on those buildings anyway, it just doesn't stack them up cleverly.
* Trade routes provide a lot of gold, but are offset by the costs of road maintenance. ICS minimises the road maintenance for the amount of trade route gold generated.
* The trade route equation also favours more cities given the same population (thanks PieceOfMind for this point). The equation is (1 + 1.25*city size), making the '1' effectively another flat per-city bonus. The free '1' is presumably to partly offset the cost of more roads to more cities, but two small cities packed into the same space as the alternative one bigger city actually need less roads in general, as the city centre tiles get a road "for free".
* Social policy culture costs are the one thing that actually genuinely hit back at ICS. One thing that mitigates this a little is cultural city states. A few early cultural city state allies can generate enough culture to buy some of the key ICS social policies (which with the exception of some of the order tree are generally quite accessible) before the number of cities gets out of control.
* Tightly packed cities are an ideal defensive set up.
