Advanced ICS questions

[MobileOak]

I'm slowly moving up the difficulty chain, I'm currently winning Emperor games, looking to move up to Immortal, then Deity. I have a few questions about ICS, but before I get to them wanted to give an overview of my understanding of ICS basics so that if I have some misconceptions they can be caught there first.

ICS (Infinite City Sprawl) is a method for settling large amounts of land successfully. The advantages are that it generates more science than many other strategies, while also denying the competition the use of the land. The disadvantages are less frequent policies, and less frequent golden ages.

Effective ICS means you need to cancel out the from both the number of cities and the population of the cities. Each new city generates 4 and each population generates 1 .

Initially a good target population for a city seems to be 4. This is because it rounds well with the Library's science boost, and because the trade route will generate enough money to offset the cost of the 3+ road segments to connect the city. Using the citizen management tool, manually adjust the tiles worked so that food production is STAGNATION when a population of 4 is reached.

A city of size 4 generates 7 . To offset this unhappiness, we use the following tools:
Meritocracy social policy: +1 for connecting the city to the capitol
Organized Religion social policy: +1 for each Monument and Temple (net effect +2)
Coliseum: +2

This gives +5 happiness. Here's where I get to Question 1: Up until now, this -2 I've offset by luxuries, circuses, monasteries, etc. But these aren't infinite, so should I be doing something else? Build a theater? Or is it just understood that at this point, because I haven't researched tech/policies, my empire will be small enough that these should suffice?

So assuming I've reached population 4, and connected the city, I will also have the following buildings (or be working on them): monument, temple, coliseum, library, and then to help offset the maintenance costs, market. So now, we look at the net gold outcome of the city - maintenance costs of the first 4 buildings, 1 + 2 + 1 + 1 = 5. Market generates +2, so as long as the population is generating 3 more, the city will be a net neutral. Question 2 is: Should I be considering any other building as well, as a "always build"?

Once I build all of these buildings, I've been setting the city to generate Science. If I'm running low on gold, sometimes I'll set a few to generate Gold instead. Question 3: Is Science the best thing to set it to at this point?

I tend to play for warmonger/domination games. Question 4: Assuming the city I'm currently capturing doesn't have any significant tiles in its extended range, but fits in with my ICS every four tiles pattern, is it better to keep the city, or resettle a new one? After I fill out the Piety branch, I've been going for Police State. By the time I have it, I usually have a significant happiness boost from all my courthouses (+80 my most recent game), however, this is 3 policies I could apply elsewhere if I never built courthouses. Assuming I keep the city, I've been puppeting it until the unrest wears off, and then when I have enough gold to purchase a courthouse, I'll annex the city and purchase a courthouse straight-up. Question 4b: Is this a good strategy? Question 4c Should I ever starve a city down to pop 4, and if so, when?

Having read a few other posts about ICS, I've seen that as I research some more techs/policies, I could go past 4 citizens as my target population. Question 5: What buildings/policies are standard for this approach, and can it be universally applied, or only to those cities which will be able to generate enough gold to offset Theaters and Stadiums and such?


Question 6: When you're capturing a city, what evaluations do you make to decide whether to limit the city as an ICS city, or to allow it to grow and be a major producer for your civilization?

Question 7: Anything else I'm missing here or misunderstanding?

 

[DaveMcW]

The big thing you are missing is Honor. +1 for garrisoned units, +1 for every defensive building.

ICS as a strategy is dead. You can make it work with a lot of effort, but it won't make you win faster.

 

[vexing]

it still works on settler difficulty =/

ics was great back when libraries had scientist slots and maritimes provided +2 food to all cities. now you generally want to use your available happiness bucket to grow a non infinite number of cities to the point where they can build universities and support scientists.

 

[MobileOak]

The big thing you are missing is Honor. +1 for garrisoned units, +1 for every defensive building.

ICS as a strategy is dead. You can make it work with a lot of effort, but it won't make you win faster.

Can you elaborate or provide a link?

I know it's been responsible for some significant nerfs since Sulla's articles, but I'm not aware of any changes making it a bad strategy, and having watched a decent chunk of Wainy's Songhai playthrough, he seems to discuss it quite extensively, but I figured asking these specific questions here would get me to the answers faster than watching all of his vids.

it still works on settler difficulty =/

ics was great back when libraries had scientist slots and maritimes provided +2 food to all cities. now you generally want to use your available happiness bucket to grow a non infinite number of cities to the point where they can build universities and support scientists.

My understanding was that it was game-breakingly overpowered back then, and now is just a decent strategy.

 

[DaveMcW]

Wainy's Songhai playthrough, he seems to discuss it quite extensively

Can you provide a link?

 

[MobileOak]

Can you provide a link?

From his Songhai playthrough discussion:

Each City has to be at least 4 tiles away from each other. Since after he plans to attack and puppet an empire or 2 He plans on building other cities on those sites to make use of all the available land for as many cities as he can make.

I also want to minimize the number of road tiles between cities. So exactly 4 away is ideal, but 5 is okay too, and 6 and above is mediocre.

Sorry I'm not trying to be combative here. I just haven't seen a definitive thread/discussion here on ICS wherein the undisputed conclusion is that it's a bad strategy now, thus the request for a link, thinking I'd overlooked it somewhere. Is it a totally bankrupt strategy? If so, what do you do instead, given a vast unsettled continent?

 

[DaveMcW]

chgrogers is mistaken. wainy is advocating 1-dimensional ICS along road networks, but not across an entire continent.

Personally I disagree with wainy, I would rather pay for 3 extra roads than deal with 3 extra .

 

[MobileOak]

Additional links to recent posts:

http://forums.civfanatics.com/showthread.php?t=438066&highlight=ICS

Both Martin Alvito and Madjinn discuss ICS.

http://forums.civfanatics.com/showthread.php?t=435044&highlight=ICS

A decent discussion of ICS happiness. It seems that a bank is also a good recommendation (I wasn't sure if using the hammers to build it was good or not).

 

[Martin Alvito]

ics was great back when libraries had scientist slots and maritimes provided +2 food to all cities. now you generally want to use your available happiness bucket to grow a non infinite number of cities to the point where they can build universities and support scientists.

Actually, that was always true. The popularity of ICS resulted because it provided nearly comparable results with a much simpler decision matrix and a larger / base. But you still got out of the game faster by limiting your expansion and focusing on Great Scientist production.

Once we figured out the Research Agreement mechanics, it quickly became obvious that the efficient number of cities necessary was pretty small. The introduction of the National College further reinforced that.

 

[MadDjinn]

chgrogers is mistaken. wainy is advocating 1-dimensional ICS along road networks, but not across an entire continent.

Personally I disagree with wainy, I would rather pay for 3 extra roads than deal with 3 extra .

ICS is non-optimal for a fast win, but it's 'good' for a win, so it depends on your focus.

9 cities (including your capital) seems to be the magic trade route number and aligns perfectly to a GS cascade set for the mid-late game.

 

[vexing]

9 cities (including your capital) seems to be the magic trade route number and aligns perfectly to a GS cascade set for the mid-late game.

9 is a lot of happiness to support. you only need 3 cities to produce the scientists needed for a standard science or diplo victory. gmajor 7, an emperor level diplo victory at epic speed was won with 4 fairly weak cities + 3 puppets. the extra nonpuppet cities granted no unique luxuries.

beyond the bare minimum of 3ish needed cities, i'll only found more if they're granting multiple luxuries, or at least a new unique one so they can start happiness neutral. the increase in culture costs isn't really worthwhile otherwise.

 

[jega001]

First off ICS isn't dead. If that's not your favorite way to play then that's fine but the OP hit it right on the head when he said that you generate more science [and gold] while giving up more frequent policies and golden ages.

1. Yes their will be extra unhappiness not mitigated by Meritocracy and Piety. This is also why Notre Dame is so powerful. To combat this not including wonders you look at the big picture of Stone works, circus, and in the beginning trying to settle a new luxury resource with each city or additional copies you can sell. Its worth mentioning here that in my opinion going to unhappiness for periods of time especially in a domination game is not catastrophic and quite predictable and problems truly crop up once your in double digits and get the combat penalty.

2. This is kinda tough so i'll answer it like this. And per the direct question you asked, i'll say no, though I could totally make an argument for workshops. In cites you decide to grow , you have to always build universities. Period. Filling the slots is in many cases another always do but anyway, I also make a rule for building anything that gives me happiness because the maintenance cost of the building usually is always outweighed by having another citizen.

3. There will come a point when setting cites to gold focus after you truly have built everything necessary will happen (not science.) This however isn't usually until late game. If your cranking out buildings in a city that your setting it to gold (not science) too early then you may want to reconsider what your plans are for that city and develop it into a production city. Another good rule of thumb especially concerning units after any immediately necessary buildings are completed is if you can build something in 10 turns or less it is worth considering doing that. And I'm sorry but one more time, gold not science

Prefix to 4: you said in your ics pattern every 4 tiles, optimally especially with low true ICS and low pop cites, the optimum distance is 3 tiles between the city. Now if you said 4 as in three between and on the forth tile you plop the city down that this was my error in reading what you said. I just wanted to make sure that was clear that's all.

4a. and 4b. With the courthouse "bug" the way it currently is there is something to be said for taking even a 1 pop city with no buildings and annexing it, building a courthouse, and having no happiness costs for the city. However I think that's kinda cheep and don't do that. Small pop cites with no useful buildings get raised in my games. Its rare that they fit in the ICS path anyway. If the city has a good makeup of buildings, a wonder, or luxuries you need right now, I usually keep it. Everything else gets raised.

4c. I usually find pop 5 usually optimal but back to the general should you starve a city. In some circumstances its a good idea. To do this you have to anex the city which moves your policy cost up. Side note; deciding to raise a city immediately and never puppeting it will Not raise your policy costs. This usually isn't a problem until you start getting cites with significant double digit pop. You don't usually find yourself in situations where you do that in a fast moving domination game because deciding to raise to reduce pop usually up's your unhappiness well past the combat penalty and halts your advance.

5. Personally I'd go past four to begin with and stop at 5. The advantages usually offset the disadvantages in most cases. Don't worry about the city generating the gold required to offset each cites buildings individually. In other words if your building maintenance costs per average city is 7, don't be concerned with making sure every single city generates 7 gold. You will find yourself developing Production cites and Gold cites (science comes with the sheer mass of cites you have and the cites running scientist slots with universities and public schools.) I've had high production cites generating no gold and that's fine if you have gold cites just cranking out gold and totally disregarding say production. The offsets will work out to your favor. Also in a domination game im assuming you trading post spam your puppets (get rid of those farms!!) Most of your gold as the game goes on will be coming from those puppets and all the cites in your ICS that aren't production. TP spam around the cites you found that aren't there to produce your army (or in rare occasions to grow and somewhat produce.)

6. Like I said somewhere if there is a good amount of useful buildings, a wonder, or a resource of some kind you want i keep it otherwise it gets raised. Although if it has good potential for production like a good amount of hills it may also be worth keeping.

7. Maybe but I'm not thinking of it at the moment so go ahead and question away!

 

[MobileOak]

ICS is non-optimal for a fast win, but it's 'good' for a win, so it depends on your focus.

I guess this is all I'm looking for initially, a win, not necessarily the most glorious win possible.


Thanks for the detailed response Jega. And yes, I mis-communicated, I am spacing cities with 3 tiles between them, but I always count to four when I do the calculation and put the city on the 4th tile.

I'll try going to 5 pop in my cities and see how that works. I am also going to try razing more cities. If most of my cities are non-captured ones then, what policies would you take instead of my beeline for Police State? (I find I don't usually get more than 3-4 policies by the time the game is over after filling out Piety). Would going towards Professional Army be good, or should I go for Communism?

I will also go for the gold route over science. Roughly how many cities do you grow tall vs just 5 pop? Is it a 'do they have good hills/resources' that determine the decision?

Thanks for all the replies.

 

[jega001]

The professional army or communism debate goes on to this day in a low pop true ICS and I don't believe I can settle it on this post lol. If policy costs are truly where you cant get to professional army then yeah go Communism and Communism gets better the more you truly ICS. It really works out well because you don't have to have even your puppets build something. Its an extra citizen you can have in every city or just a nice late game boost to push you up just enough to keep you out of the combat penalty. The latter the game goes in an ICS domination game the more your truly working against the combat penalty unfortunately.

If social policy costs are manageable and really they should be, Professional army can provide so much more happiness in most cases. Your puppets love to build walls and castles even in cites that arnt near the front. They cost no maintenance. Oh yeah and your upgrade costs are cut by a third. Thats cool too

Yes it is usually a can this city produce units to help me push the front. The exception that comes to mind would be a city that has ok production (it usually has to have decent for this to work without building time problems late game) and a lot of rivers for good early growth and gold. This city usually in my games is the perfect city to have usually filled with the specialists I need usually great scientists or enginers. really its usually scientists.