When to build a 5th city?

[yasinbin]

my playing style is domination. I usually concentrate on annexing the 4th and 5th city and so on. So for me, as soon as I have enough more 'pointy thing' than my opponent/'s.

 

[civ54lyfe]

Coal.

 

[kb27787]

Interesting, I was looking for something like that. But is it from experience or did you do the math (cost of increased tech and policy cost vs what the city would need to return in the same set of turns)?

Considering the hapiness loss (point from KB) , I would guess the 5th city would need to be at least 1 lux. Together with hapiness buildings, wouldn't that be enought to offset the hapiness loss?

Let's see: one unique lux (+4) coliseum (+2) zoo (+2) and stadium (+4) and (-3) from the city itself... at the point where having ALL these buildings... your city would break even when it grew to the point of a whopping..... nine population ignoring happiness policies/buildings (there are not that many anyway liberty will give +1 for a connection and religion maybe can give +2 more from a pagoda) until ideologies come into play really

At early game you don't get stadiums I have games without getting a single zoo, which costs a lot of hammers. This can be compensated for if the 5th city has 3 unique luxes (extremely unlikely)

Personally I like my cities size 30+ or at least late 20's endgame, and depending on merc CS, chances are you'll see your happiness halved, or even quartered, by that 5th city (normally people who play tall are hovering around 30-60 happiness endgame).

 

[Uncle_Joe]

IMO, if you're playing the AIs there is rarely a good reason to build a 5th (or later) city. Sure, there MIGHT be a reason for a critical late game resources but I'm usually so flush with gold that most of the time I can secure anything I need from a CS.

I think the game is too geared towards a 4-city Tradition game. The AIs tend to leave you alone if you do and there really aren't any disadvantages of having a small but tall Civ anymore. The ridiculously high income from trade means that you'll almost always be able to influence CSs to stack up bonuses, resources, and luxuries. And staying small means less aggression from the AI meaning less need to guard your trade and less reason to invest (and waste) money on a military.

Building the 5th city just doesn't pay off IMO. Can it work? Sure, of course it can. Many things that are suboptimal can work. But until/unless the AI is adjusted to react to small/tall Civs, I just don't see wanting to potentially provoke hostility/reaction when 4 cities is MORE than adequate to win a tech victory (or previously, an easy Diplo victory as well).

 

[Matthew.]

Well the game seems to point at 6-8 cities being optimal. But Monarchy essentially makes capital count as two, and academies in combination with NC and/or observatories count as another city or two. So looking at it that way, I do agree that it is easier to just build your 5th+ city within your 4 cities... if that makes any sense.

I guess my point is without Monarchy and forcing a wider game, you still aren't going to go much beyond 6-8 core cities anyway before the "anti-ICS" systems kick in. 4 or 5 is all the same to me, really depends on the map. If there are additional good spots to settle, I will do so without worrying about trying to stick to 4.

@kb2: There are also circuses, stone works, additional luxuries through trades and city-states. Also, the size of a city is irrelevant if half the tiles are junk. I can push population growth in a city to work 10 ocean tiles, but that isn't adding anything besides science from population, in which case it doesn't matter if it is in that city or another. If it is another, that 10 population could be working tiles that actually give bonuses beyond 2 food.

Give me 5 lower pop cities with solid tiles over 4 higher pop cities with a bunch of weak tiles.

 

[Xiao Xiong]

only when you see something called a "fountain of youth"
otherwise, NEVER. A civ with 20 happiness gets golden ages twice as fast as a civ with 10 (and quite often 5 cities will be hovering in single-digit happiness the whole early to midgame), and golden ages are +20% to everything except science, and more than +20% to gold which is pretty much close to having a 5th anyway.

I lost a game recently doing that. I found the fountain and immediately queued a settler, got it as my fifth. I actually had five great locations and no unhappiness thanks to the fountain.

Problem is that it was coveted by two nearby civs. I knew war was coming and immediately queued military but even though I was shaka I just didn't get enough military out before I got dow'd by all the civs on my continent, the two that coveted the fountain city and I guess the third dogpiled on just for fun.

I didn't have a hope, 20+ enemy units closed in on one archer, one spear and one warrior. I got a CB out just in time get annihilated. Oh, and one of them was Attila with a load of rams. Dead.

 

[voek]

Let's see: one unique lux (+4) coliseum (+2) zoo (+2) and stadium (+4) and (-3) from the city itself... at the point where having ALL these buildings... your city would break even when it grew to the point of a whopping..... nine population ignoring happiness policies/buildings (there are not that many anyway liberty will give +1 for a connection and religion maybe can give +2 more from a pagoda) until ideologies come into play really

Good reasoning, so I would say 2 new lux would be needed at least if you are low on hapiness. I don't think the extra hapiness for golden ages will hamper the game. Most game I don't more then 2 golden ages from hapiness surpluss before it ends. And indeed, the tennets can give a city much more hapiness. So if you are good on hapiness I dont think that would be enough to stay with 4.

I think I will check how many bubls I would lose with an extra city (+5% penalty) and would need to generate with the 5th city.

In general I agree that most of the time a 5th city isn't possible or wishfull. But sometimes you have a large backyard where a nice coastal city is possible. In these cases, I was wondering, when would I settle a 5th?

Tnx all for the discussion.

 

[Cicerosaurus]

only when you see something called a "fountain of youth"
otherwise, NEVER. A civ with 20 happiness gets golden ages twice as fast as a civ with 10 (and quite often 5 cities will be hovering in single-digit happiness the whole early to midgame), and golden ages are +20% to everything except science, and more than +20% to gold which is pretty much close to having a 5th anyway.

Is this correct?

I understood each Golden Age took longer to achieve, so eventually (at some stage) they should collide.

 

[HiroProtagonist]

Current game - going for late, post GWB, domination and needed oil. I had 0 oil in my core 4, but had a location to my east with access to 13 oil across 3 fields

It also made for a convenient GWB airbase for taking out most of Persia

 

[bison21]

If I don't go domination, my fifth city will either grab newly found coal/oil/uranium, or a new luxury.

I tend to build a new city late in the game (after research labs) often for strategic reasons and basically as an airpoirt. I once invaded the US, who had dominated the entire continent that I wasn't on. I didn't find a good spot to attack, so I just settled a city at the coast and used it as an airbase to go straight for his capital.

 

[Jesterozan]

IMO, if you're playing the AIs there is rarely a good reason to build a 5th (or later) city. Sure, there MIGHT be a reason for a critical late game resources but I'm usually so flush with gold that most of the time I can secure anything I need from a CS.

I think the game is too geared towards a 4-city Tradition game. The AIs tend to leave you alone if you do and there really aren't any disadvantages of having a small but tall Civ anymore. The ridiculously high income from trade means that you'll almost always be able to influence CSs to stack up bonuses, resources, and luxuries. And staying small means less aggression from the AI meaning less need to guard your trade and less reason to invest (and waste) money on a military.

Building the 5th city just doesn't pay off IMO. Can it work? Sure, of course it can. Many things that are suboptimal can work. But until/unless the AI is adjusted to react to small/tall Civs, I just don't see wanting to potentially provoke hostility/reaction when 4 cities is MORE than adequate to win a tech victory (or previously, an easy Diplo victory as well).

Just wondering on what turn you get the tech victory with 4 cities?

 

[bcaiko]

I still can't believe cities founded in later eras don't just get basic buildings (monument, granary, lighthouse) built as a matter of founding, like they do on an advanced start.

Though folks are right that at least late-game cities don't languish like they did pre-BNW.

...sacrificing your trade routes to pump up a late-game city, though, is still an opportunity cost.

 

[Syntax Error]

I still can't believe cities founded in later eras don't just get basic buildings (monument, granary, lighthouse) built as a matter of founding, like they do on an advanced start.

Though folks are right that at least late-game cities don't languish like they did pre-BNW.

...sacrificing your trade routes to pump up a late-game city, though, is still an opportunity cost.

If you're in the lead, why not? One or two is great enough. If you really want to push it, then put three of your naval trade routes to it (if possible): 2 food and one production. If you wanna go overkill,buy an aqueduct. Even better if you settled that land just for grabbing a rare resource that you needed. You can also probably put Landmarks into consideration.

 

[kb27787]

Is this correct?

I understood each Golden Age took longer to achieve, so eventually (at some stage) they should collide.

They do eventually, but the game should be long over by then (if happiness is clearly 2-3+ folds different)

 

[kb27787]

Well the game seems to point at 6-8 cities being optimal. But Monarchy essentially makes capital count as two, and academies in combination with NC and/or observatories count as another city or two. So looking at it that way, I do agree that it is easier to just build your 5th+ city within your 4 cities... if that makes any sense.

I guess my point is without Monarchy and forcing a wider game, you still aren't going to go much beyond 6-8 core cities anyway before the "anti-ICS" systems kick in. 4 or 5 is all the same to me, really depends on the map. If there are additional good spots to settle, I will do so without worrying about trying to stick to 4.

@kb2: There are also circuses, stone works, additional luxuries through trades and city-states. Also, the size of a city is irrelevant if half the tiles are junk. I can push population growth in a city to work 10 ocean tiles, but that isn't adding anything besides science from population, in which case it doesn't matter if it is in that city or another. If it is another, that 10 population could be working tiles that actually give bonuses beyond 2 food.

Give me 5 lower pop cities with solid tiles over 4 higher pop cities with a bunch of weak tiles.

circuses and stoneworks are situational (not every city can build them) and even with both you'd push the number to a measly 12 pop anyway.

Sure you can find happiness elsewhere, from trades and CS and the like, but you could've done that anyway with just your 4... the 5th really didn't make much of a difference unless you connect several copies of tradables.
You do know that typically "the other half of the tiles" that are not junk is still 20+ tiles? And one more thing: specialist slots, even in the absence of good tiles to work (you've got them all saturated) you can still nab the +2 science per specialist bonus.

Unless you found your city in a coastal desert wasteland with no fish and used food caravans solely to pump it up, bigger is virtually always much better.

 

[fyar]

only when you see something called a "fountain of youth"
otherwise, NEVER. A civ with 20 happiness gets golden ages twice as fast as a civ with 10 (and quite often 5 cities will be hovering in single-digit happiness the whole early to midgame), and golden ages are +20% to everything except science, and more than +20% to gold which is pretty much close to having a 5th anyway.

Sorry, I couldn't let this math slide. Yes, the +20% is "pretty much close to having a 5th city anyway"... during the Golden Age. But that Golden Age is maybe 10 out of 60 turns. The other 50 turns that you're NOT in a Golden Age, that other civ with 5 cities will certainly be outproducing you.

There are many advantages to sticking to 4 cities. Production is not one of them.

 

[Uncle_Joe]

Just wondering on what turn you get the tech victory with 4 cities?

I honestly don't know the turns I win on. It's usually fairly late though but the important thing is that it's before anyone else wins. Sometimes I'm a bit earlier (1970s or 80s) and sometimes if it's been a slower game I've pushed even into the 2000s. But still, as long as I'm able to stay ahead of the AIs, IMO it's still a relatively easy winning strategy.

I'll start recording the turn numbers though because I'm actually curious now too.

 

[Liegence]

I honestly don't know the turns I win on. It's usually fairly late though but the important thing is that it's before anyone else wins. Sometimes I'm a bit earlier (1970s or 80s) and sometimes if it's been a slower game I've pushed even into the 2000s. But still, as long as I'm able to stay ahead of the AIs, IMO it's still a relatively easy winning strategy.

I'll start recording the turn numbers though because I'm actually curious now too.

It is listed in the Hall if Fame

 

[Liegence]

Assuming Tradition is your core SP, then it is generally not necessary IMHO to build any more than 4 cities in standard. A pop in one city could be 2 in your cap. You can spend a lot of hammers and gold getting that new city up to speed, and any trade routes you send could be trade routes elsewhere.

I prefer puppeting.

My rule of thumb on city count, in general, is if I need zoos or stadiums to keep up my happiness I have too many cities.

 

[kb27787]

Sorry, I couldn't let this math slide. Yes, the +20% is "pretty much close to having a 5th city anyway"... during the Golden Age. But that Golden Age is maybe 10 out of 60 turns. The other 50 turns that you're NOT in a Golden Age, that other civ with 5 cities will certainly be outproducing you.

There are many advantages to sticking to 4 cities. Production is not one of them.

Really now... say if (in my normal games) my capitol is size 50 and my satellites are size 20+ each, +20% production in all of them wouldn't get more done than having a 5th may be size 10-15? GAs are not just production you know... CULTURE increase from GA for example is no contest a 5th one slows your policies down (always, regardless of GA or not) and the most it can boost your culture is merely 1 each for each culture building.
Also GOLD of course, is increased by a bit (esp important in GnK where gold is essentially doubled) and if you go freedom GAs will last 12 turns and with a wonder or Persia even longer... those GAs can be worth a lot.


Speaking about production... sometimes you can't just add up the hammers from each city... for example, when building wonders or spaceship parts... only one city may work on one wonder/part at a time... here the 5th doesn't matter at all, and god forbid you miss, say, hubble, when you could've had it if that golden age had happened.

That said, the only reason I would take a 5th is for science.