City specialization is not where it needs to be: how to fix this?

[Arms Longfellow]

When Civ 6 was first announced and I heard about the districts system, I was excited, because I was hoping that it would mean the return of specializing your cities like in Civ 4. In that game, it was so fun to stake out the terrain with a settler and think: "Look at all this food, and there's some decent production as well. This is the perfect spot for my Great Person city!" Then you'd settle, grow the city to a huge population, build National Epic and spam wonders. Elsewhere in your empire, you'd have your military city which was high in production, and you built Heroic Epic along with all the military buildings to churn out highly experienced units around the clock, ignoring things like libraries. Then there was your financial capital which had tons of cottages, merchant specialists, and ideally was also a Holy City so you'd also get the income from spreading your religion. And so on with the super science city, and production cities. This was great because all of your cities had an important and UNIQUE role to play in your empire.

Civ 5 was a step back because it introduced the OP strategy of abusing Tradition by making your capital city a massive juggernaut which did everything, and every other city existed just to feed it with trade routes. Yes I'm aware that going Liberty was also viable and changed this somewhat, but still, Civ 5 felt like all of your cities were just boring copies of each other, with some being superior or inferior based on the terrain. It was more about how many luxuries you could secure rather than genuinely caring about the cities themselves.

Civ 6 is a slight improvement but again, your cities don't feel unique like they were in Civ 4. You don't build a campus in your specialized science city like I was hoping would be the case; instead, it's more like you build a campus in EVERY city because you're trying to win a science victory.

One of the causes of this problem is the use of flat + bonuses rather than % bonuses like Civ 4 had. Civ 4's % bonuses encouraged players to run up the multipliers to ridiculous levels, making city specialization a powerful strategy. Adjacency bonuses are nice and one way of encouraging specialization, but it's not enough. Sure, it's nice to get a couple extra science points by building a campus in an ideal location adjacent to mountains, but is this bonus really enough to stop a player from building a random campus in the middle of nowhere when he's going to be getting a bunch of science anyway after he builds all the campus buildings?

Possible ways we could fix this: Bring back National Wonders, which would do much to give each city a unique identity. Make more world wonders that only provide a local bonus, because right now most of them give you some empire-wide bonus (for pete's sake, even Venetial Arsenal applies to all your cities). Buff specialists so we actually have an incentive to run them, and provide more % bonuses rather than only + bonuses. Right now, there are almost no buildings that give % bonuses except for Oxford University, which is a World Wonder, so it's tough to even get.

Right now, Civ 6 is: "I want to win a cultural victory, so I'm going to build a Theatre District in every city." What I actually want is this: "I'm going to work really hard on making this city the world's premier hub for culture, like Florence was during the Renaissance." Right now, there's really nothing you can do to make a city a premier hub for culture other than building a Theatre District (adjacent to some world wonders for a piddling amount of extra culture), all the buildings, and I guess running 3 artist specialists for some reason even though they barely provide any benefit. Do others even share my same sentiments about city specialization, or are you cool with the way city dynamics work in Civ 5 and 6, with more of an empire-wide focus?

 

[king of nowhere]

+1 on this.
more % bonuses would encourage specializing cities.
it is also worth noting that civ4 gave you more freedom to build up your land. you could make cottages on grassland and mills on mines for extra gold, or farms on grassland and mines on hills for extra production. for most spots, you could choose many different improvements, and often there wasn't a clearly better choice. in civ6, you only get some choice after you get civil engineering, and by then you're mostly set. And even after you get conservation, the best choice is clearly to build enough food to grow the city and then sawmills everywhere else. because working tiles get you nothing but production and food, and excess food doesn't help you much.

 

[Arlequin]

I agree with you and proposed an other solution in this thread:

https://forums.civfanatics.com/threads/specialised-cities.614814/

% is a way to go but national wonders as they were in civ5 are not a solution for civ6: I forgot about civ4 ones. I am currently making a mod (still halfway through) using "urban policies" which are a replacement to national wonders. The mod focuses on specializing cities and generating gpp through specialists.

 

[Infixo]

1. Get rid of yields from population or make them very small. Right now you've got 0,7 science / pop! It gives you later >50% of your entire science yield.
2. Make adjacency bonuses not so common but little more powerful. So it would really matter where you place your toys. If you make more of them, the effect will be just the opposite - you'll get one no matter where you place things.
3. Make city-level policies (see above what @Arlequin wrote, most dfficult to implement).
4. Make more exlude-type buildings. There's a thread about it. So, you would have more choices on how to specialize your city.

%-based bonuses were removed from Civ6 to avoid inflation effect. I agree with this approach.

5. National Wonders - they will make a small difference, just for a few cities. And you can't have them too many - again, inflation effect. Probably one per district type, so you can boost a chosen city in a specific area.

I'd say that only no. 3 would introduce some problems for AI (new mechanic), but the rest is using existing game mechanics, so it should be ok.

 

[Rosty K]

For Cultural Victory, there should be some bonus (adjacency or I don't know) for Great Works tourism output too. So it would actually matter where to keep your Great Works. Right now it's 'build museums everywhere', but there should be something to really make you want to build it here before there. Same could work for Campus buildings (give adjacency bonus not only to the district itself, but to buildings inside, and maybe even to the project).

In the current state of the game there are some districts I almost never build, or build one (Aerodrome, Entertainment, Encampment), and I end up having the rest in every city if the population allows. Otherwise you just can't keep up on science, culture, faith and gold.

 

[UWHabs]

To me, the best way to encourage specialization would be to greatly increase the effect of adjacency bonuses. As it stands now, sure, if you can find that +3 or +4 campus early, that's a massive boost when your empire only gets 10 science. But if your empire is generating 50 science, then the difference between a +1 campus and a +3 campus doesn't matter at all. Or if I'm generating +500 gold per turn, I can throw a commerce hub in the middle of the desert and not even care if it missed out on a +5 adjacency spot.

So the possible ways to make adjacency stronger:
-Make the adjacency an exponential value. So a "+3" campus would actually yield 9 science. Pro: really matters adjacency Con: Could be way too much bonus early
-Have the Adjacency apply to all buildings in the district (sort of like the Shipyard). If each building has a small base yield plus some factor of the adjacency, then it matters more. Right now a fully loaded campus gives 11 science plus adjacency. If each building instead gave 1/2/3+adjacency, then a fully loaded campus would instead give 6 science + 4 times the adjacency. So in this case, a campus with no adjacency bonus would give only 6 science total, but one with a +3 adjacency bonus would yield 18 science.
-Have the adjacency bonus apply to the great people points. So if you build a campus with no adjacency, you don't get any great scientist points from them. If you build a campus in a +3 spot, then the district and each building would produce +3 great scientist points. Raw science doesn't change much, but at least the more specialized the city the more it contributes to great people.
-Have more regional buildings. If every district was like the industrial zone, where the final 2 buildings were regional ones, then that could encourage you to make sure you find the right spots. Especially if you combined it with my 2nd suggestion. So, for example, if you're debating IZ location, and one spot is a +4 adjacency and another is only a +1, right now you might still pick the +1 if it means it reaches extra cities for the regional effect. But if placing it in the +4 spot gave a lot more production to the factory to the cities it reaches, then you're going to be a lot more likely to use the +4 spot, even if it means having to find a spot in those other cities to build a factory.
-Have the adjacency also apply as a % bonus to the city. So a +3 campus would not just yield +3 science, but also +30% science to the city. This would work sort of like my second point above, but would just be a slightly different mechanism for how the bonus gets applied.

I don't really like the first point, but any or all of the last points could be combined together. Another option could be to change the policy cards. Currently, because you can get a policy card that doubles all campus adjacency, or doubles all campus buildings, it encourages you to build campuses everywhere, for example. If instead, the card was something like "triple your best campus" or one that was basically "double the best adjacency for each district", then there would be an extra advantage to choosing the absolute best location, and at least a notion of specialization would come into play.

 

[Candido Rondon]

Percentage boosts could be one way to address this. One reason I don't like this though is that I think it maybe gives too much emphasis to city placement, which can potentially hamstring a player early on.

I think the best way would be to make it so that district buildings don't give flat bonuses but instead increases the yields from each citizen working in that district. In this model, a Campus would max out with a lot more Research/Turn than in the current game (since you'll presumably have fewer Campuses), but you'd have to invest a much higher portion of the population in that district to get that benefit.

So a city with an absolutely huge population might be able to get away with running most districts at full strength, but anything below that would have to specialize in a few districts and concentrate their citizens in them.

 

[Rosty K]

Percentage boosts could be one way to address this. One reason I don't like this though is that I think it maybe gives too much emphasis to city placement, which can potentially hamstring a player early on.

Civ6 doesn't (well, almost) limit the expansion, so percentage boosts could potentially lead to 'trouble'...

 

[playshogi]

Just as districts are limited by pop: 1,4,7,10... so to should the district type be limited by number of cities: 1, 4, 7 10....
So if you want 4 campuses, you need 10 cities. Additionally, the 2nd building in each district should provide regional effects like, the factory and zoo.
Universities for campus, etc.
Problem is what to do with districts in captured cities? Player asked which districts to keep and destroy the rest, seems gamey, destroying all seems severe.

 

[Arms Longfellow]

Some good suggestions in here. Another thing I want to add: I don't like how food has always worked since the very first Civ game. It doesn't make sense that the city that's surrounded by nothing but farmland is your empire's largest metropolis. That food should be exported to cities that actually need it because there isn't enough room for farmland. I want trade routes that give a city food to also DEDUCT that same amount of food from the exporting city. Obviously this would not be an appealing use of trader units, that's why it ought to be separated from trade routes entirely. Maybe make it an option in the city screen itself, to export food to X city. The whole trading system needs rehauling anyway because no one finds managing 30 trade routes in the late game fun. But taking food away from one city and giving it to another would create an interesting specialist city: your breadbasket. The one that you want to have a big enough population to work every farm tile, but anything in excess of that is destined for your heavily urbanized cities that need it. Maybe the feature to export food isn't enabled until a certain tech is researched.

 

[king of nowhere]

Some good suggestions in here. Another thing I want to add: I don't like how food has always worked since the very first Civ game. It doesn't make sense that the city that's surrounded by nothing but farmland is your empire's largest metropolis. That food should be exported to cities that actually need it because there isn't enough room for farmland. I want trade routes that give a city food to also DEDUCT that same amount of food from the exporting city. Obviously this would not be an appealing use of trader units, that's why it ought to be separated from trade routes entirely. Maybe make it an option in the city screen itself, to export food to X city. The whole trading system needs rehauling anyway because no one finds managing 30 trade routes in the late game fun. But taking food away from one city and giving it to another would create an interesting specialist city: your breadbasket. The one that you want to have a big enough population to work every farm tile, but anything in excess of that is destined for your heavily urbanized cities that need it. Maybe the feature to export food isn't enabled until a certain tech is researched.

up until the modern era, transporting large amounts of food over large distances was just unfeasible, or too expensive to be done except to feed an army on the march. the only case of a city really being fed from imports before the modern time is ancient rome, and even then the grain that arrived from africa was only part of the food of the city, and the roman empire was huge and it had a lot of resources to afford dooing that kind of stuff. and it eventually fell also in part because of large unproductive expences like keeping their sprawling capital fed and entertained.

 

[steveg700]

Foodstuffs were very much a transported commodity in empires throughout history. Rome's supply from Egypt was not an isolated case, just a notable one for being an intercontinental distance. Having said that, historically population centers did tend to be linked to the immediate availability of a food supply. Angkor Wat was the largest pre-industrial megacity because of the vast farming going on around it.

But all of that's academic. Until the modern era, civilizations also didn't upload a bunch of virtual currency into the cloud either. Military units are cranked out of clusters of mines and sawmills. But in the game it is necessary because of Civ's abstracted method of portraying empires. While I used to advocate similarly putting food into a bucket, I guess the problem is that it would lead to newly-founded cities ballooning too quickly. Trade routes are not a bad way to represent moving supplies around, I just wish a single route could make multiple stops. It makes a great deal of sense to me that a city with a commercial district or harbor owns its trader rather than producing it and the shuffling it off to some other city. As the city develops commercially, it can send its trader out further to make more stops. Of course, it should also be beneficial to the destination cities as well--another idiosyncrasy.

 

[steveg700]

To me, the best way to encourage specialization would be to greatly increase the effect of adjacency bonuses. As it stands now, sure, if you can find that +3 or +4 campus early, that's a massive boost when your empire only gets 10 science. But if your empire is generating 50 science, then the difference between a +1 campus and a +3 campus doesn't matter at all. Or if I'm generating +500 gold per turn, I can throw a commerce hub in the middle of the desert and not even care if it missed out on a +5 adjacency spot.

IMO the best thing to do is to unlock additional adjacency bonuses through techs and civics. There is plenty of room for this, since there are still lots of stops on these trees that are pretty fruitless. So, for instance, campuses get adjacency bonus increases from mountains after getting astronomy, and from rain forests after chemistry.

The price of this is that the UI would have to do a better job of informing players of suitable spots for districting. Either that, or just expect players to do all the advance planning for optimization.

 

[Arms Longfellow]

Yeah I wouldn't mind if food exporting didn't kick into gear until some modern era.

. While I used to advocate similarly putting food into a bucket, I guess the problem is that it would lead to newly-founded cities ballooning too quickly

I think this is okay because it gives players some incentive to bother founding cities in the late game. Right now, with the ridiculous scaling costs of districts, the only way you can make them work is to help them out with trade routes. This wouldn't be unrealistic either since, for example, the Chinese city of Shenzhen was just a small town up until the 1980s, and now it suddenly has a population of 12 million people. Clearly they assigned lots of trade routes from there

 

[steveg700]

I think this is okay because it gives players some incentive to bother founding cities in the late game. Right now, with the ridiculous scaling costs of districts, the only way you can make them work is to help them out with trade routes. This wouldn't be unrealistic either since, for example, the Chinese city of Shenzhen was just a small town up until the 1980s, and now it suddenly has a population of 12 million people. Clearly they assigned lots of trade routes from there

Well, that population increase didn't come from sudden excessive procreation in that city. Certainly it didn't come from excess food, which only boosts population indirectly by way of making excessive procreation feasible, and perhaps attracting people to come in from the sticks. That's a migratory pop increase, which you can't effect in this version of Civ.

That seems like another benefit that could be tied into trade routes as ages progress. But like I said, I think trade routes should work a bit differently. You should help the foundling city by directing trade routes there from established cities.

 

[king of nowhere]

well, as for growing new cities, the best way was if you could make population migration. in civ3 you could join a worker to the population, increasing it by 1. since building a worker costed 1 population to the city, it was a fair deal. I liked that system, no idea why they removed it. maybe because you could make workers with a small, quick-growing city and send them to a bigger city, whose population would require much more food to grow? possibly, but it would be simple to implement something like "pop 1-8: needs 1 settler to get +1 pop. 9-15: needs 2 settlers to get +1 pop. 16 and above: cannot be grown in that way". given how expensive settlers are in this game because of gradually increasing costs, it seems too weak as a mechanic.
as for growing the city withy trade routes, i wish there was an effective way of stopping and reassigning them when needed instead of having to wait for them to expire and then needing to reassign them anyway

 

[Candido Rondon]

well, as for growing new cities, the best way was if you could make population migration. in civ3 you could join a worker to the population, increasing it by 1. since building a worker costed 1 population to the city, it was a fair deal. I liked that system, no idea why they removed it. maybe because you could make workers with a small, quick-growing city and send them to a bigger city, whose population would require much more food to grow? possibly, but it would be simple to implement something like "pop 1-8: needs 1 settler to get +1 pop. 9-15: needs 2 settlers to get +1 pop. 16 and above: cannot be grown in that way". given how expensive settlers are in this game because of gradually increasing costs, it seems too weak as a mechanic.

I would say that joining cities should just be restricted to Settlers. I would also suggest that joining Settlers to a city should do something else too like maybe giving 50% of its total production cost to the target city. Otherwise it would be a pretty steep cost for just adding a population point to the target city and people would rarely make use of it.

----

Whether migrating Settlers to boost city population or spamming food trade routes, you're probably going to need to start importing some food anyway because those population points need to be fed.

Making cities more dependent on food trade routes later in game not only makes sense historically but works well for gameplay IMO. In order to simulate the growth of shipping foodstuffs over longer distances, you could have technology bonuses for buildings like Granaries. For example, Granaries could give +1 Food/Trade Route with Steam Power, then maybe +1 Food/Trade Route with Electricity.

In late game, as cities grow to the point that they use up all of the available land, they'll start bulldozing Farms to build more beneficial structures, and these higher-producing trade routes can fill in the gap.

-----

But going back to the earlier topic of other yields produced through districts, I tend to favor population-based bonuses over adjacency bonuses particularly when those adjacency bonuses are based on the underlying terrain. Terrain bonuses are fine to nudge a city's development toward one specialization or another, but too much and you risk wildly unbalancing the game. If the only way to have a viable science city is to happen to have an ideal location to place a science city, there's something wrong there.

But if you say that a Library gives +1 Research for every Citizen working in the Science Campus, then you're not only requiring that a player invest in the cost of building the district and the buildings, but you're also requiring them to make the investment of people into the district to make it productive. It strongly incentivizes players to focus on a few districts per city rather than spreading their labor power over a bunch of different districts in each city.

It requires no arbitrary caps, and it's driven largely by player choice rather than external circumstances.