[BTS] Huge/Rainforest/Marathon/Settler Rapid Expansion+Research

[snowern]

Assuming a rare cherry picked triple river gems start, what would be required to Lib Steam Power in the BCs with ~24 cities at that time?

I'm trying to imitate Tristan_C's playthrough at https://forums.civfanatics.com/threads/most-super-specialists-by-turn-500.562931/#post-14809407 which means that not too many non-merchant specialists and GPP polluting wonders can be used, as most GPPs are reserved for Great Merchants. Also, not too many forges either, to avoid global warming. Usually, the only special stuff I build is an Academy, Shrine, and the Oracle, before I get to Liberalism.

Here's the start:

There's also pigs to the left.

I got Steam Power in 510 AD, which is chronically bad. The usual tech path is probably similar to Tristin_C's playthrough:

• Iron Working,
• Currency,
• Code of Laws (Religion+Caste System),
• Civil Service (Bureaucracy),
• Priesthood (Oracle),
• Monotheism (Org Religion),
• Calendar,
  
• Oracle Chemistry (workshops),
• Banking (Mercantalism),
• Philosophy (Pacifism),
  
• Replaceable Parts,
  
• Lib Steam Power, and
  
• Economics (Corporation Merchant).

I can sometimes get Currency on time, but everything else gets slightly further away from schedule. All the while, I beg and trade for gold from the 17 AIs to keep research going.

Save files attached, just to be extra sure it's not a map problem.

 

[sampsa]

How good is your city management? What difficulty level are you able to beat consistently?

 

[snowern]

On playthroughs of this kind of setup, I either try spamming cottages in lots of cities, or, cottages in just the first 2 or 3 cities and farms/workshops for the rest.

Because I'm spending most of the time in Caste System, I'm not doing much whipping. On my next attempt (same map, right now), I'll stay in slavery most of the time until somewhere around Chemistry, and whip Warrior+Settler+3xWorker from each city once it gets to 6+ population, and hope the whip anger doesn't cause too much of a problem. Maybe I'll skip the warrior. This should make expansion go faster, but I'll still have to decide whether to switch in and out of Caste System for workshops or depend on cottages. I could also produce Research or Wealth, but that would slow down expansion. Maybe I could produce Research after whipping those 5 units?

For specialists, I only run enough for a Great Scientist + Great Engineer + Great Prophet. The rest go into merchants, for all cities, once in Mercantilism. In a previous game, I got 120 or so Great Merchants and a 128 pop city, which was my end goal, but that was with Next War (which gets an extra +1 food for late game farms), and I would like to improve the early/mid game. Using scientists to bulb technologies might be a good idea, but that would take away from the future Great Merchant pool. Same for the Great Library.

For city growth, I only mess with the tiles if a city isn't growing, working only a few of its cottages, or getting no production.

For buildings, I currently build a Granary in every city. The capital also gets a Library and, typically, a Forge for the engineer. The second or third city builds the Oracle. And then, some monasteries are sprinkled about before Scientific Method. The 20 floodplain site might get the Globe Theatre for happiness before Future Techs, to grow faster. I might also put the National Park in there too, because I think the no unhealthiness is better than using it in another city, where 20 farms might be better than the 20 free specialists.

The only military units I build in the early/mid game are warriors/archers and a few stronger units to capture the 20 floodplain site. I typically get Steam Power before the AI starts getting in the way.

For city placement, I'm currently packing all cities as close together as possible, with the occasional adjustment to get riverside access or a resource. Maximising the city count should maximise GPPs because of Cereal Mills. One of the earliest cities is settled for production, to get the Oracle, which is not built in the same city as the shrine, to reduce GPP polution.

For the expansion rate, I currently only build settlers and expand if existing cities have at least some improvements. But maybe having a few unimproved cities around isn't so bad as the trade routes might make up for the maintenance on this difficulty. City maintenance isn't a problem, but there are times when the AIs don't have enough gold to fuel my research. The shrine+Mercantilism merchants makes this much easier.

For the worker ratio, at 510 AD in the previous game, I had 67 workers for 23 cities. Seems like more workers than necessary, but there was still plenty of unimproved tiles at the borders. Probably the jungle.

Before this, I made many attempts at Huge/Pangea/Always War on Noble with much more reliable wins if I choose a more water heavy map like low water/snaky continents so that the AI's unit spamming doesn't come from all directions. I think I had the same problem then as I do now in that I just don't research fast enough. For Always War/Noble, with tech trading enabled, there were times when I would be beaten to Liberalism. For a more ordinary Normal/Pangea setup, I don't know what I would be like.

 

[sampsa]

I wouldn't advice whipping a warrior. I do not understand why you build monasteries but not libraries. I don't understand why you don't build forges.

 

[snowern]

The intention is to keep going until 2050 AD without global warming, which means very few forges and other unhealthy buildings, assuming Tristan_C's global warming details are correct. Although, Huge/Rainforest is actually an 84 tile wide Normal sized map, but I should still be able to build 200 forges because the numbers round to zero rather than to nearest, after looking at CvGame.cpp. The +25% production bonus should make up for itself after 2 archers + 2 settlers + 6 workers. So, yes, it seems I've avoided forges more than I needed to. I'll see if this gets me anywhere next time.

Monasteries are just for the ~300 required missionaries when in Pacifism. Currently, most cities end up with workshops, which doesn't give much for the libraries to multiply.

And, I don't whip warriors. I typically whip a settler for 2 pop with 80-100 hammers overflow into a worker. Anything that's not a settler or worker is typically built without whipping to allow the city to grow.

Anyway, I've got the Steam Power date to 390 AD at 20 cities with 292 beakers/turn. About 50-70 turns slower than I would like.

 

[sampsa]

OK. I admit that I don't fully understand your goal thus can't give great advice. Monasteries just for missionaries seems kinda wasteful to me though. There must be some other way (OR?).

 

[snowern]

Only at least 3 monasteries are required some time before Scientific Method, because the missionaries are a national unit. Pacifism is required for its extra GPPs, and I'll be running it as soon as possible all the way to 2050 AD. As can be seen in that post I linked, the end game involves settling as many cities as possible for more GPPs, and pacifism needs the religion in each of those ~300 cities. This is all largely irrelevant though, before this Steam Power milestone, as I could simply build the 3+ monasteries just before Scientific Method.

The problem is that I don't know why there's such a big difference in my research rate and settling rate, compared to that post. Right now, I'm using the Steam Power date for comparison. I'm not as fused about the number of cities, because there will come a point where I can rush buy them by the truck load.

 

[sampsa]

The problem is that I don't know why there's such a big difference in my research rate and settling rate, compared to that post.

Probably due to sub-optimal city management? A save from ~1000BC might give clues.

 

[snowern]

Here's the nearest one, for the 460 BC Oracle+Chemistry and 390 AD Steam Power.

 

[elitetroops]

Assuming a rare cherry picked triple river gems start, what would be required to Lib Steam Power in the BCs with ~24 cities at that time?

Skip the ~24 cities requirement. With these settings, and the goal you have in mind, you'd want to have 24 cities much much earlier and probably closer to 100 by 510 AD. More cities means faster research, it's really as simple as that. On settler difficulty there is no maintenance to speak off, so every new city is always a net gain.

Hopefully you have the "no city razing" option enabled and maximum opponents on the map, then your warrior can walk around and pick up free capitals in the early game. You should start by building a second warrior for even more free capitals, which makes this start quite bad with no hammers. Even if you start with worker it's bad, as there are no 3 yield tiles in the first ring.

With a goal like yours, it's quite common to use all the cheesy options the game offers. On settler that means huts! Huts will provide you with settlers and workers. You really want to pop a fast worker very early and use it a while for scouting and popping more huts. Another good trick is to choose AI that don't start with hunting. No AI scouts means more huts for you. AI that (edit: don't) start with hunting are also slower to archery.

 

[rah]

I assume you meant AI that start without hunting. Or I'm confused.

 

[snowern]

I've had another, final, go with No City Razing and no Hunting civs. Walking into enemy capitals is only possible for the first 1 or 2 AIs because they get a quick warrior from their initial production bonus. Chariots upto Musketmen were used to clear up the rest with the final AI falling in 1015 AD. I have a feeling this is rather slow, but it all worked out in the end.

3910 BC - Bronze Working from hut
380 BC - Chemistry
220 AD - Steam Power
705 AD - 20-floodplain site captured
760 AD - Refrigeration, Cereal Mills
845 AD - Railroad, Mining Inc
1240 AD - Future Tech
1610 AD - All improvements done for 369 cities
1886 AD - All cities have both corps and the religion

Globe Theatre + National Park boosted population growth and the number of specialists before Future Tech. The final food output of Cereal Mills was +51/turn.

And here it is: