City specialization in GEM 1.12?

[Tomice]

Hi! I never had the feeling there was as much need for city specialization in Civ5 as there was in Civ4, but I'm pretty sure you can still profit from doing so.

In the new version of GEM, how do you focus your cities?

The only regular habit I have so far is to create dedicated military cities, since all those XP buildings can't be built everywhere. They are of course production-rich and ideally at least one of them is coastal.

My capital usually ends up being the production powerhouse for wonders and also the largest city for science. While production and science are sort of a tradeoff, there are several nice scientific wonders making this an adequate combination IMO. Due to large population and wonders, GPs are also produced here.

The rest tends to be some sort of general purpose city, maybe focused on gold more than other things. But I feel like I don't really have a clear strategy for them, so I'd be glad to hear a few thoughts o this topic.

Which city types do you have?

 

[Naeven]

Capital production/wonder/military. Also ends up as Science city about 50% of the time.
2nd or 3rd city become Science if there is lots of jungle or other terrain that favors village spam + lots of food. I also had a few good Science cities next to lots of mountains due to the sci on mountain event, again only if there is enough food to support it.
I always create a holy site spam city between city nr 2-4, often coastal for the food. This becomes my culture city, I often only use faith to generate more holy sites until renaissance-ish then I save it for Great people buying. The holy city is often placed in lesser terrain as well if there is land I want to grab for resources/strategy.

 

[mitsho]

Yes, it's a bit a problem that production is linked so tightly to wonders. I usually try to build science wonders with Engineers, but that only works so long. I agree that there's less specialization going on in civ5/GEM. This is often due to not being able to focus these cities so quickly since I do need "a unit there" NOW! or a cultural building there to push some borders or the market everywhere to pay my bills before being able to specialize the city...

But generally, I try to have: a military city, the capital as a wonder/production one, Science and Gold cities go the rest and often overlap (villages). If I have a good coastal spot, I try to build many gold boosting buildings and wonders there with the help of the inland tiles before refocusing the citizens out to the sea to grow and fish for gold and reassigning those land tiles to other cities that can reach them. Sea cities can be very powerful commerce creaters, but need a sizeable navy as well...

I've never thought of a city so far, either I try to get them everywhere or these buildings are only a last priority... It seems to me that you can ignore religion quite easily without much repercussions...

Oh, and there are also "get Ressource/Natural Wonder" cities, though fewer and "make AI angry/provoke them into a war and blast their army away while they attack this" city. That last tactic works surprisinlgy well on foreign continents besides Tall Civs that were allowed to grow and thus have a science bonus. It's amazing how you can decimate an enemy army if you can lure them into your territory over swamps and forests...

 

[Thalassicus]

I place cities in big clusters of common terrain, where I can get massive bonuses out of just a few buildings or wonders.

In my current game my capital's surrounded by tons of grassland, so I focused the capital on villages, academies, and science. My second city had 3 wheat, 4 hills, and many forests. I turned this into an early wonder powerhouse. Wheat feeds hills for production, and forests chop for big +50 boosts to wonders.

My third city has 9 hills, so I got Banaue and pump out my land army here. I didn't do early wonders there, because it was a low-food area until Banaue got built. The fourth city is a port with lots of coastline and a nearby production/gold natural wonder. Water gives tons of gold, so I focused this city on gold and shipbuilding.

I get my cities around size 8-10 and halt their growth for a while to control unhappiness during classical wars, so my capital doesn't have any particular advantages. There are very few direct capital bonuses in GEM. In my current game, my second city is much better for wonders than the capital, since the second city has the great wheat/hills/forest combo.

 

[Tomice]

Interesting thoughts, thx! It's especially good to hear the designers own strategy.

I now remember how you introduced the system that every bonus ressource should have a building boosting it, to make the terrain more varied/interesting. Vanilla Civ5 had the design principle that every spot should be just as good as another for every purpose, or so it seemed.

So, let's do a bit of brainstorming. We have these roles a city can fulfill:

SCIENCE:
A lot of population/food, science-boosting buildings

GOLD:
Luxuries, yield-increasing buildings, gold-increasing buildings. And rivers! Every improvement benefits from them, but no matter what we do, those tiles always get extra gold.

CULTURE:
Dunno? Apart from culture buildings, is there any prerequisite (e.g. regarding terrain)?

FAITH:
Same as culture. Sure, there are beliefs like "faith from tundra", but do I have to focus cities on faith? Are there many +XX% faith buildings?

MILITARY:
Lots of production, XP buildings, production booster buildings

WONDERS/GENERAL PRODUCTION:
While the military factory cities are probably busy producing units and XP buildings, we still need somewhere to produce workers, settlers and locaation-independant wonders.

WONDERS:
Most wonders should be built depending on the city roles mentioned above, for the rest see "general production"





So, is there something I forgot? Do you have any strategies or must-have buildings for certain city types?
What about good combinations? In a science-focused city, is it better to grow the population as high as possible with farms or is it better to build villages for a strong science/gold combo?

 

[mystikx21]

Culture has effects from pantheon beliefs and from some events (mountains, pastures, fisheries) that can slightly increase culture, but most will be from buildings and landmarks/holy sites.

There are two +% faith wonders (Grand Temple national wonder, usually only built in your capital, and Lalibela, which also adds faith on nearby hills).

 

[Tomice]

Isn't mosque or monastery percentage-based? No matter what, neither culture nor faith seem to depend much on location as far as I see. And you probably only need to specialize cities on them if you're pursuing the corresponding victory I guess.

 

[Thalassicus]

The Amphitheater gets extra culture from quality costumes. I push for that if I start near cotton/silk/fur/dye. Luxury resources often clump together in a small area, so if that happens, you have an ideal cultural city site.

I consider ports the best gold cities, even better than rivers. Coast with a harbor is 3 1, same as farmed river grassland. With a Market we can use that huge food supply to work a Merchant specialist, or some village tiles on land. The difference between grass-rivers and ports is grass-rivers tend to not have big food resources like Fish. Seafood helps cities get a quick start.

 

[mystikx21]

Port+river if i can get it is always gold. Otherwise its ports all the way (extra merchant slot for one).

Monastery is +% faith. Mosque is per pop.

 

[Tomice]

Very interesting! Let's draw a few conclusions:

- It might be a good strategy to focus one city on economic (national) wonders, another one on scientific stuff. When combining all percentage-based boosters in a large city, the output can skyrocket and dwarf all other cities. We'll probably work the corresponding specialists there and generate quite a few great people, which might be best used for their tile improvement, since its output may easily be doubled later on.

- Those two cities should probably get the two best spots available and be our largest cities. Especially the commercial city should be on a river mouth with lots of (sea)food and luxuries, while the science city mainly needs food from whichever source available. It seems like our capital might be the best spot for the commercial center in many games, since few other spots have such a density of bonus ressources.

- If needed for a certain victory type/strategy, we can follow the principle above (concentration of %-based boosters) for a religious or cultural city too.

- All cities not specialized this way are probabably hybrids for science and gold generation, working as many farms as we need to get them to decent size and villages for the rest. We'll probably not want those cities to get too large since we need the limited happiness for our two main hubs.

- Of course there need to production cities, which can be subdivided in military and civilian/wonder production (see my earlier post). I guess one civilian production center should be enough, since most wonders we can get will be needed to improve our main hubs.



What do you think? Am I right with my "Two main hubs" assumption?

 

[Naeven]

A small note on Culture city terrain. Just like I sometimes use Mountains to farm the +5 event, one could take the +5 option.
All the events might actually need to be rebalanced more according to the new specialist yields distribution.

Culture is kinda anti-specializing tho, since most wonders give culture. Wonders will be built in production cities or other specialized cities.
So for best culture effect, you just need 1 super city stacking wonders. (This effect amplifies if you go Aristocrazy, which is fine since the tall style is super city).

 

[mitsho]

On Opportunites:

Spoiler :

I agree that all the opportunities would need rebalancing and also new opportunities added to the pool so that they are era-appropriate and diverse. However, there's nothing wrong with one or two of the events being strong so that you chose to change your playstyle (buy a useless mountain tile in the hope for such an event). They are kinda like beliefs in that there can be difference in strength since those are interesting and makes us play differently. (and none of those are a I-win button like discovering the Great Barrier Reef as Spain on turn 5 in the vanilla game is)

The two main hubs seems reasonable so far...

What is the timeline for specialization in your games? Normally, it takes me a lot of time til my empire is set up and I can focus the cities truly. For example, even if I'm peaceful, I need an army of some sort to deter my neighbours from invading. I need that one usually as soon as possible, so that when I'm built up, I will not need a military city at all (since I will not have a army anyways and I can get other units by military city states).

So it takes time until I am set up anyways, how is it with you?

 

[Tomice]

While I'm a relatively good player (winning easily on king and often on emperor) I didn't follow much of a concept in terms of city specialization until now. It just happened.

But I guess it's perfectly normal to not think as much about this in the very beginning. For me, there are two major stages in a civ game:

- The opening, where the initial land- and ressource-grab happens. Every tech, every unit, every building counts. I play very aggresively towards a certain goal (usually to catch as many ressources as possible to sell them), and I'm very vulnerable if something unexpected happens.

- The consilidation phase, when I claimed my core lands and I start choosing techs or buildings I don't urgently need for survival, and when my army is in decent defensive positions and able to hold of mid-sized AI armies well.


City specialization surely is more of a topic for stage 2, also because the corresponding buildings don't come that early in the tech tree. But you sure help yourself if you think about roles before you position your cities.

Remember that CIVUP contains a "mark tiles" feature accesible by pressing "Ctrl-X". This helps me a lot when placing my cities.

 

[Naeven]

I'd say I specialize fairly late, in Medieval you can maybe start to see it somewhat.
I don't place a city in mind to make it specialized (Jungle might be an exception), I place a city to get as much resources/yield out of the terrain as possible. The specialization is just something that comes naturally as more events/resources/GP's comes along.

This could describe my generall way of playing tho. I don't really decide beforehand if I'm going peacefull/conquest/tall/wide/other. It's just something that naturally develops as the game continues.

 

[Thalassicus]

It depends on what we mean by "specialize." I usually choose a specialization for cities before founding them, so I can decide which exact tile to settle (this one or the one next to it?), and what to build first. I get a Granary first for a science or production city, and a monument for most other cities.

 

[Tomice]

Interesting, I haven't really considered granaries that much of a priority building, but I used to build wide in my last games, so happiness was more of a limiting factor than food.

Of course, not wasting possible food yield and building more mines, villages etc. could be beneficial. Also the kickstart allowed through early granaries. But I guess it wouldn't work without strict usage of the "limit growth" button.

 

[Tomice]

I realized a few things since last posting here, so here are some additional thoughts:

• There really isn't much of a way to specialize before the middle ages. Sure, you can prioritize markets and such things, but there are only one or two early buildings that could be seen as "special" for a certain city type. And often they are just one of many buildings/things an early city urgently needs. A granary, a worker or a defense unit might be way more important early on even in your #1 gold city.
  
• There are 2 very important national wonders: The NatTreasury and the NatCollege. They are both rather easy to get since libraries and markets are among the buildings almost any city can need (might change with the new rules for building upkeep planned for GEM 1.13, though!). They are also very useful for the two main hubs mentioned above (gold/science).
  
• Other national wonders have prerequisites that are less likely to be built in every city (barracks for heroic epic!), so in many games we might be stuck with those two national wonders.

 

[mystikx21]

I end up with Heidelberg most games (as a late game free tech, it is quite nice in saving some time on space race or getting to yet another wonder). Hero Epic almost every game, Foundry almost every game and National Epic every game, and the Grand Temple any game that I have a holy city. The trick is getting any of these up before they start getting really expensive.

I miss the 75% requirement, I think that would help some of the specialisation.

 

[Tomice]

In which era do you tend to get those? Would you agree that NatTreasury and NatCollege are usually easier to get / come earlier than the others?

I'm surprised by Heroic epic, I didn't think so many build more or less pointless barracks just to get the bonus. Have to ask in the other thread when/how you pull it off best.

 

[mystikx21]

I would agree NatTres/College come earlier, mostly because markets and libraries are a high priority building and because those are the best two in my opinion.

I tend to get Hero Epic and Nat Epic around the same time or not much after, and they're probably the next two in power for me. Barracks are cheap and give an engineer slot early, which works well with liberty finisher to boost production. And Nat Epic requires amphitheaters, which I usually can get most of them with the tradition policy that gives cultural buildings (I've taken to using all three early trees at the same time, usually finish commerce/knowledge before honor though). Circus Max I get sometime in the early-mid Renaissance, as I tend to need the happiness around then anyway. Grand Temple and Foundry I usually can get before I start expanding in Industrial era. I could get the Temple sooner, if I put a higher priority on temples. I find I don't need them that much to rack up a lot of faith points and I don't need a lot of faith points until the late game when I can pop great people.

I could also get Hermitage sooner, but it takes more tiering than the others to work through on any new cities/conquests. If I play less wide/aggressively, I get it a lot more often and easily.

I get Heidelberg late, around Atomic/Information era, to time the free tech as something expensive. That's true often even if I can build it sooner.