ICS best of the best!

[Paeanblack]

Also don't forget with Gandhi you suffer double CityNumber unhappiness. So ICS'ing with Gandhi is still out of the question.

India is one of the best civs for ICS. The first two spaceship launches under 200 turns were both with Gandhi doing a full ICS. The only civs in the same league are France and Babylon.

FP doesn't come till Renaissance. Even bee-lining directly to it you'll be nearing turn 200 by the time you build it (unless extremely lucky and get an engineer). I feel it may come too late for effective ICS in this mod, but I'll be interested to hear feedback.

By turn 200, you can be researching Future Tech. The FP can be built much earlier.

What you are describing is not ICS. ICS is the unfettered spamming of cities. Building a city "when I have the opportunity" is managed growth. The goal is to kill the power of the TRUE ICS'er, the person who has 30+ cities by turn 200. ICS power comes into its own mid-game, not late-game as you are talking about. Setting up ICS is the first 200-250 turns. That what defines ICS.

The issue is about games with 60+ cities by turn 180. The first 200 turns aren't 'setting up ICS'; the first 200 turns are the entire game.

The reason people aren't terribly interested in the mod is that your posts indicate that you don't have a good grasp of the strength of ICS and what makes it so strong. Buffing theaters and stadiums won't have an effect.

 

[Ahriman]

What I would do, in fact, is to implement your proposition and to make happiness buildings work kind of like the proposed system: Colosseum +0.15/pop, Theatre +0.15/pop, Stadium +0.2/pop (total of 0.5).

I would also reduce the increase in the food bucket to a linear relationship and reduce culture spread cost to linear. Lastly, I would add an unsellable building to each new city (government representation) that costs 10 upkeep or something.

I think these all together would go waaaaay too far in the other direction, meaning that having a very small number of truly massive cities would dominate every other strategy, and that you'd be better off razing most cities you captured.

 

[alpaca]

I think these all together would go waaaaay too far in the other direction, meaning that having a very small number of truly massive cities would dominate every other strategy, and that you'd be better off razing most cities you captured.

I don't think you really thought that through. Growing still takes a long time with a linear increase. Civ4 has a linear bucket increase and a few massive cities weren't the only way to go. The thing is that as long as each city can be made profitable by building things there, once you have the buildings set up it turns out a profit.

An example: If you set food growth to 10 + 10*n where n is the number of citizens already living there, you still need 2300 food to grow to size 20. 2300 food is 200 turns at 11.5 food surplus, which is more than most of my cities have unless I run a ton of maritimes (which have to be nerfed, by the way)

The two questions that determine whether it then makes sense to build are: Can you afford the down payment you need in order to set it up? And will you be able to make it turn a profit before the game ends? If both are yes, settle away. If either is no, destroy the city.

As it is right now, a new city requires a very small down payment (only a Colosseum) so only the second question is relevant. If you remove the first question, you get quick expansion ICS. By re-introducing it, you have to balance long-term benefits versus overexpansion problems.

To look at this another way: Did few huge cities dominate civ4 strategy? Definitely not. Yet it had similar mechanics to the ones I proposed.

 

[Deep_Blue]

The issue is about games with 60+ cities by turn 180. The first 200 turns aren't 'setting up ICS'; the first 200 turns are the entire game.

I never tried ICS myself so Is there an example game or game guide to look at?

 

[Ahriman]

I don't think you really thought that through.

I don't think you thought through -10 gold upkeep cost per city.

An example: If you set food growth to 10 + 10*n

Ok, I misunderstood what you were proposing. I thought you were proposing linear cumulative food requirements for city sizes.
As in: each new pop requires X food. So with +Y food surplus, it takes you the same time to grow from size 1->2 as from 10->11.
Sorry, my bad. 10*n is not so unreasonable (though by my calculation 2300 food gets you to size 21 - you start at size 1 with no food, 20 food for size 2, 30 more food for size 3, etc). I agree that the food requirements in Civ5 are punitively high for larger cities.

Yet it had similar mechanics to the ones I proposed.

No it didn't. In Civ4, there is a check on number of city growth through a non-linear city growth function.

And the city upkeep costs weren't *massive*, particularly for your second or third city. It didn't take a large city just to pay for the maintenance cost. A flat -10 gold cost per city would limit city growth far too severely, particularly in the early game.
When would you build your second city? Remember that trading posts are multiple techs in. Your gold surplus is very small in the early game. You'd have to delay a second city dozens of turns, or risk bankrupting yourself.

 

[TheHarshax]

Debate aside, who has run this mod? Does it stop ICS?

 

[atteSmythe]

Debate aside, who has run this mod? Does it stop ICS?

The goal should not be to stop ICS, but to come up with a way of preventing it from always being the right option (not always right, but not always wrong). As it is, placing a new city is always worthwhile, regardless of where or how small it is. So long as new cities are always good, placing them as close together as possible will always be optimal, as that lets you place the most cities.

I'm skeptical of a lot of these proposed fixes. At a glance, most would either make puppeting everything you come across the next best option, or would throw out puppeting along with ICS.

 

[kelemvor]

I never tried ICS myself so Is there an example game or game guide to look at?

Find yourself a good half hour of free time to sit and read though this thread.
http://forums.civfanatics.com/showthread.php?t=392335

 

[alpaca]

I don't think you thought through -10 gold upkeep cost per city.


Ok, I misunderstood what you were proposing. I thought you were proposing linear cumulative food requirements for city sizes.
As in: each new pop requires X food. So with +Y food surplus, it takes you the same time to grow from size 1->2 as from 10->11.
Sorry, my bad. 10*n is not so unreasonable (though by my calculation 2300 food gets you to size 21 - you start at size 1 with no food, 20 food for size 2, 30 more food for size 3, etc). I agree that the food requirements in Civ5 are punitively high for larger cities.


No it didn't. In Civ4, there is a check on number of city growth through a non-linear city growth function.

And the city upkeep costs weren't *massive*, particularly for your second or third city. It didn't take a large city just to pay for the maintenance cost. A flat -10 gold cost per city would limit city growth far too severely, particularly in the early game.
When would you build your second city? Remember that trading posts are multiple techs in. Your gold surplus is very small in the early game. You'd have to delay a second city dozens of turns, or risk bankrupting yourself.

I'm not set on the number (was just a quick idea) and I would prefer a non-linear relation, too, but I'm not sure if it's possible without DLL access. The 10 were coming from reasoning that you should have something like size 5 to break even with a city. Size 5 means you get 5 science and the five citizens should easily net you 5 gold, hammers or additional science if you run a specialist from a library.

Early game can be helped by giving the palace some more gold, say 6. Don't forget that deficit gold is taken from science so running some deficit for a while in the early game is similar to expansion stifling research in Civ4.

The third or fourth city need some additional sources of money, however, but another thing I would do is boosting resource yields. Since you settle near luxuries anyways, you shouldn't have much trouble getting some additional money rolling in to fund your expansion. These are the (undeveloped) yield bonuses I currently have in mind

Resource	Base yield	Improved yield
iron	2H	4H
coal	2H	4H
horse	1H2G	2H4G
oil		1H2G	2H4G
aluminum	1H1G	2H1G1S
uranium	1H	2H2S

Wheat	2F	4F (plus CS, etc.)
Cow	1F1H	3F1H
Sheep	1F1G	2F2G1H
Deer	2F	4F
Banana	2F1G	4F1G
Fish	2F	4F

Whale	1F1G	1F2G2H
Pearls	1F2G	1F5G
Gold	2G	3G1C (6G1C with mint)
Silver	2G	5G (8G with mint)
Gems	3G	6G
Silk	3G	6G
Fur	2G	5G
Cotton	2G	5G
Marble	1G1H	2G1H1C
Incense	2G	2G2C
Dyes	2G	4G1C
Ivory	1G1H	3G1H
Spices	1F1G	1F4G
Sugar	1F1G	1F3G
Wine	1F1G	2F2G

Of course, this will have some global upscaling of the economy as a result, which allows me to increase building maintenance so that road maintenance is cheaper in comparison. I'll also have to boost some other stuff but this is currently still at the ideas stage.

Debate aside, who has run this mod? Does it stop ICS?

Except Dale, I guess no-one? I don't think he has released it.

 

[Dale]

Debate aside, who has run this mod? Does it stop ICS?

I haven't uploaded it yet. As I mentioned, we went out last night so didn't get a chance to do that.

I did a run through late last night and am making a couple additions to this. The building change DOES in fact impact ICS, and with a couple other changes will reduce the incentive to ICS. In fact, with this building setup tall cities (pop 20+) are easy to keep happy.

No access to the DLL is cramping what can be changed (eg: SP effects are coded not lua) but I'm able to work around most of the issues.

 

[civcivv]

I personally feel alpaca's changes would be best to limit ICS.

Say 8 upkeep per city, palace + 2 gold at start.

Then FP + Order Policy reducing unhappiness by only half their current value, so when you have both its -50% as opposed to 100%. By the same token, increase base per-city unhappiness by 1 but increase difficulty default happiness by 2.

Nextly you want to reduce per-luxury happiness to 3, so that with policy you get +4 max.

If you finally then lower colosseum+circus happiness to 2 each and buff Theater+Stadium to 5 and 7, you make ICS a lot less viable without totally killing it, by introducing what is essentially start-up costs per city.

 

[Dale]

I personally feel alpaca's changes would be best to limit ICS.

Say 8 upkeep per city, palace + 2 gold at start.

Then FP + Order Policy reducing unhappiness by only half their current value, so when you have both its -50% as opposed to 100%. By the same token, increase base per-city unhappiness by 1 but increase difficulty default happiness by 2.

Nextly you want to reduce per-luxury happiness to 3, so that with policy you get +4 max.

If you finally then lower colosseum+circus happiness to 2 each and buff Theater+Stadium to 5 and 7, you make ICS a lot less viable without totally killing it, by introducing what is essentially start-up costs per city.

The problem with linear costs such as these as you expand, it once you get setup to cover those costs there's nothing to stop infinite expansion. That's a lot of the problem with Civ5 is, all the linear costs / benefits. If some of these were exponential like food per city size is, then wide empires will eventually hit a wall where they cannot cover the cost of the next city. Once that occurs it's a simple case of balance the values and voila, suddenly tall and wide empires become equally viable (my goal).

 

[Ahriman]

The 10 were coming from reasoning that you should have something like size 5 to break even with a city.

Ugh. I think that's a terrible design principle. I don't think cities should ever be an unambiguous burden. And I don't think that building your second city should make you worse off. This is a key problem with linear penalties; if they're high enough to worry about when building a 20th century, then they're crippling when building your second.

A better way to go is to make it slower to get to pop5 (by nerfing Maritime CS) rather than making cities below pop 5 into deadweight.

I think we also need to distinguish between some terms.

There is nothing wrong with expansion being favorable; a larger empire with more population and sufficient infrastructure to support it should always be desirable, except maybe in how it accumulates social policies, so that there is at least some tradeoff.
But the main tradeoff should be between lots of tightly packed cities, or fewer but larger and more spaced out cities.
Ideally either of these will be a decent strategy, the problem is currently that one strategy (lots of small cities) currently dominates.

The problem with linear costs such as these as you expand, it once you get setup to cover those costs there's nothing to stop infinite expansion.

But your model also has this. With your model, with Colosseum + Theater + Stadium you can have not just infinite cities, but unlimited population within each city (bounded only by food).

 

[bobbyboy29]

I know I said I'd get this out tonight, but the weather was fantastic so we had fish 'n chips on the beach.

People on the forums will be happy to note that we're supposed to get thunderstorms tonight, hopefully that will keep you modding indoors!

The problem with linear costs such as these as you expand, it once you get setup to cover those costs there's nothing to stop infinite expansion. That's a lot of the problem with Civ5 is, all the linear costs / benefits. If some of these were exponential like food per city size is, then wide empires will eventually hit a wall where they cannot cover the cost of the next city. Once that occurs it's a simple case of balance the values and voila, suddenly tall and wide empires become equally viable (my goal).

You hit the nail on the head here Dale. This is the exact reason that no other strategy can currently keep up with ICS, my question is, can you mod it to make science and trade routes increase exponentially (or at least quadratically or something if that's not possible) with growth as well? I recall seeing a city growth exponent constant somewhere in the SDK files, you could decrease that a bit at the very least...

 

[Valkrionn]

That's something I would like to address, personally. At the very least the Science.

I hadn't thought it was possible, but I was thinking in terms of civ4... With civ5 it may well be.

ATM I'm working on my own changes to luxuries, meant to be used concurrently with this. Basically, luxury happiness is dropped to 3, from 5, but excess luxuries grant 1 happiness.

 

[Dale]

But your model also has this. With your model, with Colosseum + Theater + Stadium you can have not just infinite cities, but unlimited population within each city (bounded only by food).

Couple with one of the other changes, the incentive to build up not out will increase the more you expand. Eventually outward growth will be un-affordable and upward growth with a theatre and stadium will become affordable.

That's not linear.

 

[Dale]

People on the forums will be happy to note that we're supposed to get thunderstorms tonight, hopefully that will keep you modding indoors!

New NCIS on tonight.

You hit the nail on the head here Dale. This is the exact reason that no other strategy can currently keep up with ICS, my question is, can you mod it to make science and trade routes increase exponentially (or at least quadratically or something if that's not possible) with growth as well? I recall seeing a city growth exponent constant somewhere in the SDK files, you could decrease that a bit at the very least...

Food requirement per pop as cities get bigger is insane IMO. I may look at it in "stage 2".

 

[alpaca]

The problem with linear costs such as these as you expand, it once you get setup to cover those costs there's nothing to stop infinite expansion. That's a lot of the problem with Civ5 is, all the linear costs / benefits. If some of these were exponential like food per city size is, then wide empires will eventually hit a wall where they cannot cover the cost of the next city. Once that occurs it's a simple case of balance the values and voila, suddenly tall and wide empires become equally viable (my goal).

Actually with the city growth thresholds, they overdid things a bit in my opinion. A quadratic amount of total food (linear increase to get to the next pop size) is probably sufficiently penalising to slow down growth significantly later on.

I do agree that trade and science being linear are big problems. I think adding more library-like buildings and removing scientist slots from libraries might make it more worthwhile to go for larger cities because they would not be cost-effective in small cities. So at least for science, you could add a stronger non-linearity in the number of cities.

Trading is a bigger problem. You could add National Wonders of the Chichen Itza type that increase your trade income without you having to add population (I think you can allow multiple copies of NW by setting the MaxPlayerInstances higher than 1, right?) Reducing the flat trade route income to -1 or -2 but increasing the pop multiplier could also be worth investigating because it will make hooking up small cities ICS-style not worthwhile but could still turn in a decent profit for larger cities. I wouldn't go lower than -1 probably because otherwise you get an undesirable situation where you actually want to avoid linking size 1 cities to your trade network.

Ahriman: In Civ4, most cities were a liability for a while after being built. Of course, it would be desirable to have a super-linear relationship and maybe something like that can be implemented.

Dale: For food, I've been testing the above-mentioned 10+10*n, and so far it feels good but it might require dropping the food storage buildings down a bit (25%+25% are fine imo). If you'd like to plug it in, the formula is a bit bugged which generates the weird values you need to use. Here's my code

<!-- Removed the superlinear behaviour and replaced it with a linear increase of 20 + 10 per citizen already in the city. The formula is
			BASE_CITY_GROWTH_THRESHOLD + CITY_GROWTH_MULTIPLIER * (n-1) + (n-1)^CITY_GROWTH_EXPONENT
			I suspect this to be a bug but anyhow, it's now changed to 20 + 10 (n-1).
			To grow to size 20 you need 2300 food now (w/o hospitals and such).
		-->
		<Update>
			<Set Value="19"/><!-- 15 -->
			<Where Name="BASE_CITY_GROWTH_THRESHOLD"/>
		</Update>
		<Update>
			<Set Value="10"/>
			<Where Name="CITY_GROWTH_MULTIPLIER"/>
		</Update>
		<Update>
			<Set Value="0"/><!-- 1.8 -->
			<Where Name="CITY_GROWTH_EXPONENT"/>
		</Update>

 

[Paeanblack]

The problem with linear costs such as these as you expand, it once you get setup to cover those costs there's nothing to stop infinite expansion.

Eventually covering the costs is not why ICS works...ICS is already stronger than vertical growth without any policies, maritime allies, or buildings.

If you have a budget of 3 happiness to expand with, you can use it on growing 3 new pop or building a new city. Those 3 pop are working 2 farms and a mine to be food-neutral. The new city is working the city tile and a mine to be food-neutral.

Except for missing out on 2 science, the new city wins in every other category. It yields more hammers, more gold, more cultural tiles, and more combat capacity. It requires fewer worker-turns to get running. It also requires much less food to grow once you more happiness available. All that extra food the big city needs to gain one pop has an opportunity cost in lost hammers. The food needed to grow a city from 5 to 6 could be a warrior if you worked hammers instead. Growing from 7 to 8 costs you a swordsman. 10 to 11 is a Longswordsman. These hidden costs rapidly outweigh the costs of the settlers themselves.

Note that all of these benefits come before buildings, allies, policies, and golden ages are thrown into the mix. ICS doesn't need Colosseums to be stronger than vertical growth...it already is stronger. Tweaking the happiness buildings won't address your goal.

You need to attack ICS at its heart: strong city tiles and cheap settlers.

Take away 1 hammer from the city tile and add it to the palace. Remove 1 gold from the city tile and add 1 to the palace. Charge maintenance for the road underneath the city and add 1 more gold to the palace. Double the cost of training settlers and give the palace +100% to settler production. Bump up the rush-buy penalty for settlers by 50%

 

[Slowpoke]

New NCIS on tonight.



Food requirement per pop as cities get bigger is insane IMO. I may look at it in "stage 2".

You know, you really oughta try mega mod Dale, if only to check out the ICS system in place there. I mean most of the ideas in this thread are already adopted in that mod and have been for several weeks, like linear growth, no minimum gold per city, scientist nerfing, increased palace yields, etc. And the "All India" happiness solution (from all the way back in MM version 1) also effectively works just like your colosseums. It almost makes me feel sad really that ideas I came up with a month ago are becoming normal only now. Says to me people aren't even looking at the mod