How many cities do you need to keep up in the late game?

[Fippy]

Usually it's more like ~1k beakers from an occ super city (700 can already be achieved in a regular game imo),
but yup ofc it's very important how trading with the AI is managed.
At friendly they will trade all techs, and you should have a big tech lead early (unless playing deity) cos in occ there are no costs and you go wonder hunting / GP generating.

Easy? Nope i guess, but in theory you really need just 1 city for a succesful space race.

 

[pi-r8]

Usually it's more like ~1k beakers from an occ super city (700 can already be achieved in a regular game imo),
but yup ofc it's very important how trading with the AI is managed.
At friendly they will trade all techs, and you should have a big tech lead early (unless playing deity) cos in occ there are no costs and you go wonder hunting / GP generating.

Easy? Nope i guess, but in theory you really need just 1 city for a successful space race.

Even a friendly AI won't trade techs that they have a monopoly on, or space race techs when they're aiming for a space win ("we'd rather win the game").
There's also the small matter that giving away techs to someone competing with you in the space race might not be a good idea. It's very unlikely you'll be able to set up multiway deals in the late game like you can earlier.

You can get a good tech lead in the midgame, but even a really big lead of like 10000 beakers is like 1 late game tech on deity. It shrinks disturbingly fast when the tech-hungry AIs start to rack up those era discounts.

 

[Fippy]

Even a friendly AI won't trade techs that they have a monopoly on

Incorrect i think, monopoly should be unimportant.
Mansa can be used as example, he will trade everything and friendly AIs are like him.
If they build a wonder from that tech, they will not trade thou.

Also you can skip several techs for Space, while AIs will research anything.
+ they are much less focused (can be nicely seen on Sulla's Survivor games).

 

[krikav]

"Even a friendly AI won't trade techs that they have a monopoly on", yeah they will.
I think they refuse still if they are building a wonder though. "We have our reasons..."

 

[Anysense]

80-100 is a lot of raw roduction:! 100 would be uh... 25 caste system grass workshops. Better hope for a lot of production bonus resources! Not to mention that the research falls away whenever you build a building or spaceship part.

Using corporations would be a significant improvement, provided you have sufficient amount of resources. As far as I know there is no better way to play Space. Should you find something more efficient that would be quite a revolution.

 

[pi-r8]

OK, I stand corrected on the monopoly thing.
Of course, it's not a given that they'll be friendly (are you going to convert to their religion and stop trading with everyone they dislike?).
And I'm pretty sure even Mansa won't trade Space techs if he's going for the space win.

 

[Fippy]

Giving in to "stop trading with our worst enemies" is one of the biggest traps in this game
I almost never do that, diplo penalty of -1 can be ignored in this case (and can also fade again).
If there's an AI hated by everybody else, i might open no borders and have no resi trades with them which results in others never asking.

Open Borders (+2), we appreciate your resources (+2) and fair trades like a tech gift (+4) are already +8.
Those can easily be achieved over time, then you can decide who you want friendly with other options like Reli or favorite Civic.
SPI shines here.

 

[krikav]

Spoiler :

Finished internet 4 turns ago. (competitors doesn't have plastics), got flight+rocketry from that and started apollo. (Got a whole lot of other techs too, mostly backfill, but also fission.
HC built UN and Charly got the seat (stupid me, should have voted for Capac).
So now they have voted to make both universial suffrage and enmancipation mandatory. Didn't dare defy because I'm severely limited in resource trades, especially since ivory is obsolete. So research is more sluggish than it would be.

Spearheading to semiconductors now, without waiting to get the discount from refrigeration. Feel like it's important to get laboratories up in time.
After that, I'm thinking that maybe fusion is the way to go?
Not often that I am at this stage of the game, feel kind of lost. But strategy is to just be patient and let the techs drop in for free when they do, and not boost the AIs with trading if I can help it.

 

[krikav]

Attached start and finish saves too in case someone want them.

Spoiler :

Went for ecology last, speeding up statis chamber by a turn with a chop.

Everyones best friend

So, with confidence I can aswer the topic with "Less than 11."
But as always, "It depends...."

 

[georgjorge]

Great example, krikav! Thanks!

Was any AI close to winning in 1836? I've seen Culture and Space go earlier than that a fair number of times on Deity.

Currently closing in on finishing my game, in which I have only nine cities (and one of them doesn't produce anything) - we'll see if it gets me to Space in a reasonable amount of time.

 

[krikav]

Great example, krikav! Thanks!

Was any AI close to winning in 1836? I've seen Culture and Space go earlier than that a fair number of times on Deity.

Currently closing in on finishing my game, in which I have only nine cities (and one of them doesn't produce anything) - we'll see if it gets me to Space in a reasonable amount of time.

Not really. Capac went for culture half-heartely it seems, his third city was 60 turns away some time when I was busy building spaceparts, bribed him to go into burocracy a pair of times, but then he got squashed by charly.
No AI was running away.

And I think we where all pretty set back by the fact that UN banned representation.

The point I see being made, is that you really don't need 2000+ BPT, I was never over 1500, even in GA.
Also, having enough hammers is important.

 

[pi-r8]

Nice one @krikav
I have to admit I'm surprised how fast you were able to tech with relatively low BPT. Were you trading/interneting a lot of techs? In my current game I'm up to 2100 beakers now at turn 278 (see spoiler) but still a ways to go. I guess the main difference is I'm playing this on immortal so there's not much late game tech I can usefully get from the AI. And it was just kind of a slow teching game in general.

Spoiler :

 

[krikav]

I traded alot prior to internet, brokering alot back and forth, swapping around religion and/or civics during GAs to get different AIs up to friendly.
Got the entire steel->RR->Combustion line through trades, also radio.. Self-teched industrialism, plastics and computers.

After internet I did utilize that alot by instead choosing to self-tech such tech that where deep into the tree, such as toward fusion, while waiting to get other cheaper techs for free.
I think I mentioned in a earlier post here, that I think this is one (almost crucial?) tactic for getting through with a space victory for a small empire.

 

[Fish Man]

I traded alot prior to internet, brokering alot back and forth, swapping around religion and/or civics during GAs to get different AIs up to friendly.
Got the entire steel->RR->Combustion line through trades, also radio.. Self-teched industrialism, plastics and computers.

After internet I did utilize that alot by instead choosing to self-tech such tech that where deep into the tree, such as toward fusion, while waiting to get other cheaper techs for free.
I think I mentioned in a earlier post here, that I think this is one (almost crucial?) tactic for getting through with a space victory for a small empire.

Congrats; I didn't expect nearly that speedy of a win! In fact you might've won faster here by space than even trying to take over land, since whipping your cities to the ground for 50 turns to get enough cannons to take down 2-3 AIs in iso seriously sets back your tech pace by a lot. Seems like you finally revealed the niche situation for replicating the tall Civ V meta in this game .

 

[Anysense]

Nice one @krikav
I have to admit I'm surprised how fast you were able to tech with relatively low BPT. Were you trading/interneting a lot of techs? In my current game I'm up to 2100 beakers now at turn 278 (see spoiler) but still a ways to go. I guess the main difference is I'm playing this on immortal so there's not much late game tech I can usefully get from the AI. And it was just kind of a slow teching game in general.

Funny how teching on Immortal can be more difficult than on Deity because AI is useless. The compensation is that conquering is a lot easier on Immortal compared to Dety.

Edit: Apparently you tend to fall into "just one more building trap". Also, your cities seem somewhat overgrown. It's not worth it growing to size 20+ just to work another coast tile - too much trouble with and , not to mention the amount of food required for growth.

 

[Fish Man]

Nice one @krikav
I have to admit I'm surprised how fast you were able to tech with relatively low BPT. Were you trading/interneting a lot of techs? In my current game I'm up to 2100 beakers now at turn 278 (see spoiler) but still a ways to go. I guess the main difference is I'm playing this on immortal so there's not much late game tech I can usefully get from the AI. And it was just kind of a slow teching game in general.

Spoiler :

View attachment 539977

Four immediate mistakes I see from that image:

1. Researching refrigeration. It's much preferred to get to superconductors from techs that are actually required to win the space victory, since you won't be throwing away beakers for nothing. Plus health can be overcome late game by bio farms, or hospitals if you must.

2. Not settling the fish. Wasting such an easily obtainable 6f resource slows down your development by a lot...especially when you're stuck on an island with rather limited food choices (except for the capital which has a decent amount, I guess).

3. Growing your cities too big. If not conquering, health and happy will be far bigger constraints since you can't get all resources in the game or trade for them by virtue of controlling 60% of all the land. So smaller cities are more beneficial since you won't be wasting food sustaining citizens that do nothing while consuming 50% more!

4. Corps. Corps have little to no place on a regular standard speed game...you need state property for the production, and at any rate, again - the smaller your empire, the less able you are to trade for the corp resources, defeating the point of founding them in the first place. Plus the opportunity cost of the great people...the GE could've been used to rush something like Broadway for happiness + trades (or start golden age), the GM for a 1700+ gold trade mission.

Besides that, I don't know the details but I think your tech path and/or build orders could use some adjustment. I can't really tell though just with that pic.

 

[Fish Man]

@Fish Man
Lots of wonderful insights delivered!
However... 75 turns to space from communism sounds more like a HOF situation or perhaps a really lucky ordinary game.
Although most of what you say is direcly translatable to the situation investigated here (tiny empire, no land advantage), I do believe that there might be nuances that change things.

But for example... burocracy is all but guaranteed to be even more mandatory in a small empire than it is in a large, for obvious reasons.

@Fish Man
Yeah, that's a good analysis of the advantages of state property/caste system workshops. But it seems like you're talking about the scenario where you've conquered a ton of cities. That's probably the best strategy, but I'm going for a variant here, where I conquer few, if any, cities.

Still, a space win 75 turns after communism is very fast. Did you delay communism or something? How many cities did you have?

When I have just 6-10 cities that I settled early, developing towns is no problem. I don't even need emancipation for it really, so I can still use caste until the emancipation unhappiness gets too bad.

Getting hammers is definitely a problem with cottage cities. What I've been doing is running universal suffrage until I have sid's sushi/mining inc everywhere, then switching over to representation. And then gradually replacing them with workshops to make the spaceship parts. It's not ideal, but it's something. It's enough to get libraries/universities/observatories/laboratories everywhere, and I'm usually running a very high research rate. So it's pretty close to your "ideal" scenario, with 14 research from the towns vs 8 from workshops. Except it's 16 since I like financial leaders, and the towns have the flexibility to also build other stuff, while the workshops have to build research. The bigger problem is keeping up with health, especially if you build coal plants and industrial parks for extra production.

So yeah, hammers are a problem.. but I feel like the biggest pain point is just raw beakers, to research all those expensive late game techs that the AI won't trade me. There might be benefit to switching civics during golden ages, running caste/rep to max specialists, and otherwise using emancipation/US to grow and build. I think you're overrating bureaucracy a bit though. In the best case, If you have a fin/riverside town, it'll produce 7*1.5 = 10.5 under bureau, vs 9 under free speech. Not much better... except free speech also boosts all your other cities, with lower maintenance, and the cultural bonus is often useful too.

The 75 turns from communism to win is actually not an exaggeration, in my experience. To efficiently go space, you must conquer as much land as possible, as fast as possible. The sooner you get those 25-35 cities online, the faster you can start snowballing with the hammer economy. Plus the longer you delay conquest the more you risk some AI running away in a fashion you can't control. So the way you play is actually similar to domination, on deity. Start with a compact but research-focused empire, 8-10 cities usually. Pawn aesthetics to catch up in the classical era, then GS spam to win lib race and lib a key military tech. Afterwards, whip your cities to the ground to pump out an army to take over the world (well, not all of it...be mindful of the dom limit especially with regards to vassals' lands). Your research rate will suffer - this is inevitable, and it will hurt, but it is inevitable if you want to reach the 3000-5000 bpt range by the end of the game needed to take full advantage of the SP hammer economy. After that key military tech, then go to communism (or beat it out of someone you vassal) - don't delay it for anything else! By the time you reach communism you should already be done or nearly done with the killing phase, just in time for a switch to SP-caste-FR (hopefully aided by a golden age) to take advantage of your newly acquired lands. Workshop everything and let the hammers flow in! The research you get from this method is truly unparalleled, and quite enough to get you to a space win within 75 turns from communism a lot of the time (though a lot of the finesse afterwards is learning to direct vassal teching and choose the right tech path to go down yourself).

As a case study: a Shaka deity map (an old NC, IIRC). I impi-rushed Darius, but then stayed put until cuirs/cavs - after which I blitzed Saladin, Surry, Kublai, and then Rammy before any of them got rifles. The time from rifles to SM I was barely limping along research-wise, and I probably would've reached communism faster on 12 cities. But after adopting SP the tech pace was unparalleled. I did in fact win soon after, exactly 73 turns after researching communism - aided by my vassals getting steel and astro for me. So yes, delay communism after you get the steamrolling military tech you need (usually cuirs for speed - MT is a dead-end tech but cuirs pay back for themselves tenfold in beakers and hammers because of their speed and dominance against medieval units). After that though, straight to SP (before electricity or AL), and from then on with 30 or so cities you'll find the win surprisingly within reach.

The communist revolution:

Spoiler :

A win very soon after (I attacked Justin w/ modern armor the turn of, just for funsies):

Spoiler :

Ah, sorry, I got carried away explaining communism there.

As to your other points - that's an interesting strategy for sure. I'll have to test it along with other methods to see which yields the best results. But industrial parks are a mostly useless building (the cost hammer and health wise is too large for what is essentially merely 1 extra engineer the rest of the game). If health is a problem try to settle your cities either coastal or river so you can get either harbor or hydro plant to mitigate the unhealth (or both). Of course, in large empires the ideal wonder for this is the Three Gorges Dam...but it's a far worse hammer investment in an empire of, say, 12 cities compared to 36.

 

[Fish Man]

Nice one @krikav
I have to admit I'm surprised how fast you were able to tech with relatively low BPT. Were you trading/interneting a lot of techs? In my current game I'm up to 2100 beakers now at turn 278 (see spoiler) but still a ways to go. I guess the main difference is I'm playing this on immortal so there's not much late game tech I can usefully get from the AI. And it was just kind of a slow teching game in general.

Spoiler :

View attachment 539977

Got the t0 save or WB save so I can give this a spin, with the explicit rule of no aggressive wars or conquests? Apologies for the triple-post, btw.

 

[pi-r8]

Four immediate mistakes I see from that image:

1. Researching refrigeration. It's much preferred to get to superconductors from techs that are actually required to win the space victory, since you won't be throwing away beakers for nothing. Plus health can be overcome late game by bio farms, or hospitals if you must.

2. Not settling the fish. Wasting such an easily obtainable 6f resource slows down your development by a lot...especially when you're stuck on an island with rather limited food choices (except for the capital which has a decent amount, I guess).

3. Growing your cities too big. If not conquering, health and happy will be far bigger constraints since you can't get all resources in the game or trade for them by virtue of controlling 60% of all the land. So smaller cities are more beneficial since you won't be wasting food sustaining citizens that do nothing while consuming 50% more!

4. Corps. Corps have little to no place on a regular standard speed game...you need state property for the production, and at any rate, again - the smaller your empire, the less able you are to trade for the corp resources, defeating the point of founding them in the first place. Plus the opportunity cost of the great people...the GE could've been used to rush something like Broadway for happiness + trades (or start golden age), the GM for a 1700+ gold trade mission.

Besides that, I don't know the details but I think your tech path and/or build orders could use some adjustment. I can't really tell though just with that pic.

Bear in mind, I was intentionally doing this as a variant/challenge to try and play with as few cities as possible. I'm sure it would have been better to settle more cities here, but I'm trying to do something different. Actually my original goal was to play with just 6 cities, I chickened out and settled 8. There's more land available on this map than I expected.

Given that I'm restricted in number of cities, I might as well grow them as big as possible. I guess I could have skipped refrigeration, but I wanted the health. Corps pay for themselves pretty quickly... sushi was an extra 5 specialists in each city, with representation. Roughly 480 wealth/beakers per turn, with my 8 cities.

I've attached the initial save (and my save from that screenshot) if you want to try. I know you can win much faster by making a ton of cities, either peacefully or attacking someone. But just for a change, try *not* doing that. Beelining cuirassiers, especially, is just unfair to the AIs IMO.

 

[Fish Man]

Bear in mind, I was intentionally doing this as a variant/challenge to try and play with as few cities as possible. I'm sure it would have been better to settle more cities here, but I'm trying to do something different. Actually my original goal was to play with just 6 cities, I chickened out and settled 8. There's more land available on this map than I expected.

Given that I'm restricted in number of cities, I might as well grow them as big as possible. I guess I could have skipped refrigeration, but I wanted the health. Corps pay for themselves pretty quickly... sushi was an extra 5 specialists in each city, with representation. Roughly 480 wealth/beakers per turn, with my 8 cities.

I've attached the initial save (and my save from that screenshot) if you want to try. I know you can win much faster by making a ton of cities, either peacefully or attacking someone. But just for a change, try *not* doing that. Beelining cuirassiers, especially, is just unfair to the AIs IMO.

I don't understand, then. Is the goal to see how fast you can win with under an X amount of cities, or how few cities it takes to win? If it's the former, then fair - I'll give it a shot limiting myself to 8, same as you. If it's the latter, then you can pretty easily make it on immortal with just one, even without OCC on, I think.