[BTS] How to compete with the AI's absolutely enormous armies?

[Fippy]

Academy + Astro double bulb (so 3 GS) is usually optimal in Iso.
With Pyras an Engi can also be great for Machinery bulb.

 

[Donald Duck1]

More on academy vs bulb...on a standard cuirs game you'll probably get 3 great scientists before lib without being PHI. Two should bulb part of edu and lib itself, but the third (or rather your first) could either build an academy or bulb philo.

Shouldn't one try to get more Great Scientiest even without Philo? I'm usually at 4 or 5: 1st for academy, 2nd for philo, 3rd-5th for paper/edu/lib. Paper is only like 600 beaker, but the 3rd GS costs only 300 GPP, so the scientist specialist generating the 3rd GS yields 6 beakers/turn for paper bulb + 3 beakers/turn raw science, for effectively 9 beakers/turn. That's better than working most tiles.

And for cuirs, Music is required, so you can launch a Golden Age, and if the situation allows it run communism to get a 200% bonus on GPP generation on top of that. So the scientist specialist generates 6*3 + 3 = 21 beakers/turn if bulbing paper (and double that if bulbing education).

Is this only realistic with the creative leaders (cheap libraries) that I always play? Or is the Golden Age better used for some other purpose?

 

[Pangaea]

On the contrary, Optics>Astro are important bulb strategies in iso

I didn't talk about isolation games. In those, as repetitive as they are, naturally double-bulbing Astro is strong.

 

[dankok8]

@GenericName1998

I took a look at the save you posted. I noticed some things with your game.

1) You went for Polytheism as your first tech to found a religion. This is a big no-no on higher difficulties. The most important thing is improving your food resources early on in your case the Corn. To make matters worse, the religion did absolutely nothing for you here and you adopted another religion which spread more widely. Don't found an early religion.

2) Researching the Wheel before Bronze Working was another big mistake. BW enables Slavery for whips and chops both of which are extremely strong.

3) You settled your second city super late on turn 47 (2160 BC) and very far from your capital. You settled your fourth city really really far probably with the idea of getting Horses. Settle your first few cities closer to your capital to reduce maintenance costs and enable you to connect the cities for trade routes. I would have probably settled the second city 1 W of the Copper around turn 35. Generally when you have cities around you within 5-6 tiles from your capital that have food, found them ASAP. People often have 5-6 cities by 1000 BC and a very compact and connected empire. The way you settled you may have claimed more land (debatable) but you killed your civ early with higher maintenance and didn't profit from your powerhouse cities early.

4) You are building way too many buildings! I'm looking at your capital that has like 20 buildings. I would have built the Granary, Library, and Forge. That's about it. You aren't going the military route here and thus don't need Barracks, Stable, Dun etc. and a lot of other buildings especially like Markets, Grocers and religious buildings just aren't worth it.

5) You've got a ton of obsolete units costing you gold in maintenance. 1 Gallic, 7 Axes, 4 Archers, 3 Spears, 1 Chariot... These units aren't going to do squat against any army that attacks. Disband them and you save 16 gold per turn. You'd be several turns ahead in research by just doing this many turns ago. Units are generally too expensive to upgrade.

6) You focused a lot of your research on the bottom of the tech tree. If you made a focused Liberalism --> Military Tradition beeline you'd be a lot better off. You're behind pace technologically for 1300 AD because of this. Not tech trading does slow things down a bit though.

 

[civac]

Good points. I would like to object on two issues.

It is indeed a mistake to build too many buildings. However, Markets and Grocers can be worth it in no tech trading games in your strong economic cities. They have a much longer timeframe to pay off, you often want more cities before breakout than in a TT game so you have more maintenance and you cannot sell techs to keep your research slider high.

The Military Tradition beeline is much weaker under these settings. You need a tech advantage for it to work and having to backfill everything yourself slows you down significantly. It is of course still doable but not necessarily superior to other military breakout strategies.

 

[SnipedSoul]

Can you post the starting save?

 

[GenericName1998]

@GenericName1998

I took a look at the save you posted. I noticed some things with your game.

1) You went for Polytheism as your first tech to found a religion. This is a big no-no on higher difficulties. The most important thing is improving your food resources early on in your case the Corn. To make matters worse, the religion did absolutely nothing for you here and you adopted another religion which spread more widely. Don't found an early religion.

2) Researching the Wheel before Bronze Working was another big mistake. BW enables Slavery for whips and chops both of which are extremely strong.

3) You settled your second city super late on turn 47 (2160 BC) and very far from your capital. You settled your fourth city really really far probably with the idea of getting Horses. Settle your first few cities closer to your capital to reduce maintenance costs and enable you to connect the cities for trade routes. I would have probably settled the second city 1 W of the Copper around turn 35. Generally when you have cities around you within 5-6 tiles from your capital that have food, found them ASAP. People often have 5-6 cities by 1000 BC and a very compact and connected empire. The way you settled you may have claimed more land (debatable) but you killed your civ early with higher maintenance and didn't profit from your powerhouse cities early.

4) You are building way too many buildings! I'm looking at your capital that has like 20 buildings. I would have built the Granary, Library, and Forge. That's about it. You aren't going the military route here and thus don't need Barracks, Stable, Dun etc. and a lot of other buildings especially like Markets, Grocers and religious buildings just aren't worth it.

5) You've got a ton of obsolete units costing you gold in maintenance. 1 Gallic, 7 Axes, 4 Archers, 3 Spears, 1 Chariot... These units aren't going to do squat against any army that attacks. Disband them and you save 16 gold per turn. You'd be several turns ahead in research by just doing this many turns ago. Units are generally too expensive to upgrade.

6) You focused a lot of your research on the bottom of the tech tree. If you made a focused Liberalism --> Military Tradition beeline you'd be a lot better off. You're behind pace technologically for 1300 AD because of this. Not tech trading does slow things down a bit though.

I don't remember the game exactly, but here are my explanations/justifications:

1) Yeah, considering I was playing against 5(?) AI's, I reckoned that I only got one chance to found a religion, so I went for it straight away. I later ended up converting to Judaism because literally every civ had it, so I made some missionaries and switched (spiritual, so no anarchy).

2) I think the reason I got roads before BW is because I had marble in my capital, and wanted to go for an early oracle.

3) I settled late because I wanted to grow my capital first. I just had some really good tiles there, and I figured: if 1 extra tile gets me, say 3 extra food+production, then assuming a settler takes 100 food+production, it shortens the amount of time needed to get the settler out (so tile "growth-->settler" might be faster than just "settler"), however, the larger your city, the less this applies. What I did was basically: develop capital--> pump out some 4 settler one after the other.

4) Yeah, I probably go for too many buildings. Keep in mind that my beaker/gold was about 70/30, and even less earlier on. Markets give as much (+25%) as libraries (though they cost about double), so, particularly in your capital they seem like no-brainers to me. The Dun was useful because of the promotions, barracks definitely paid off as well. Monasteries give +10% research, so again they seem like a good investment for the capital to me. Temples basically give the repression +1 without having to put a unit (costing upkeep) in there.

5) Yeah, this is a difficult one. The thing is, If I got rid of them, I'd be even more vulnerable. Ideally I'd want to replace them with better units, but until that time, I keep them, because that 5 damage +50% against melee is still significant. Nevertheless, archers are really pushing it by that time haha. Probably not worth the upkeep.

6) Liberalism isn't all that good imo. Keep in mind that it costs 1400 beakers, which is a lot. It's only justified to get it if you can immediately research something like astronomy, chemistry, nationalism, printing press, because liberalism itself is near worthless that early on (US and free speech don't become useful until late industrial). those expensive techs that I just named aren't all that useful either. I'd much rather have banking or engineering.

You definitely make good points though, and I was probably too focused on getting early religions/wonders/territory. Thanks for the analysis!

PS: I don't get notified when someone replied here, so that's why the response is so late. Whoops!

 

[FlyinJohnnyL]

6) Liberalism isn't all that good imo.

Oh my god, I think you just broke the forum.

 

[nate46]

If you consider his setttings it makes sense though... with aggressive AIs and tech trades not allowed, rushing liberalism is kindof pointless unless you're planning to build oxford relatively soon.

 

[sampsa]

I also think liberalism is slightly overrated. By that I don't mean that it's not a good line (I pursue it in most of my games), just that it's not necessarily always the best play. Without tech trades it does indeed go down in value, though cuirs are still cuirs.

 

[FlyinJohnnyL]

I forgot that he was playing with no TT, that indeed sours Liberalism quite a bit.

 

[Fish Man]

I forgot that he was playing with no TT, that indeed sours Liberalism quite a bit.

Doesn’t this make libbing cuirs even better, though...AIs will even be slower to rifles, and everyone won’t have replaceable parts or rifling like two turns after just one person gets it?

 

[nate46]

But cuirs don't get defensive bonuses, which might be nice to have with aggressive AI... might be hard to win a liberalism race if you're under attack all game.

also if OP wants banking, I guess that's for mercantalism? If you're at war with everyone, the free specialist helps a bit more than foreign trade routes.

 

[Pangaea]

Doesn’t make libbing cuirs even better, though...AIs will even be slower to rifles, and everyone won’t have replaceable parts or rifling like two turns after just one person gets it?

This was my thinking too, although I'm not sure how it would play out as I normally don't use that option (it is sometimes ON in BOTMs, like the #197 I recently linked to in a different thread). Means it's slower to get to Cuirs ofc, as you can't trade for anything. BUT, the human can still bulb there, which the AI (generally) doesn't, so it should be possible to get there in decent time, certainly compared with the AI.

Maybe something like Trebs would be better, though -- depends on the map etc etc.

Anyway, lacking defensive bonuses don't much matter with Cuirs, because you're killing everybody on offense anyway. Sure, a city here and there can get hit by stray units, but with two movements it's pretty easy to slam into a stack trying to assault you. And with two movements, Cuirs can fork cities and take several in one turn, totally overpowering the AI. They don't get stopped until Rifles, and with no AI tech trading, that will probably take quite a long while.

 

[civac]

If you go for Rifling you naturally pick up forges, macemen, knights, crossbowmen (requires Iron Working and Horseback Riding aside from the beeline). With the Cuirassier beeline you don't naturally get any new units. Against aggressive ai this is quite an issue.

 

[Qactus]

But cuirs don't get defensive bonuses, which might be nice to have with aggressive AI... might be hard to win a liberalism race if you're under attack all game.

also if OP wants banking, I guess that's for mercantalism? If you're at war with everyone, the free specialist helps a bit more than foreign trade routes.

If you go for Rifling you naturally pick up forges, macemen, knights, crossbowmen (requires Iron Working and Horseback Riding aside from the beeline). With the Cuirassier beeline you don't naturally get any new units. Against aggressive ai this is quite an issue.

This sounds more like Always War. From my experience with the "Aggressive AI" setting, staying out of war by handling diplomacy well is not really more difficult. What happens though when you do go to war is the AI have a lot more defenders. And that would lead me to making sure I'll have a crucial military advantage such as Cuirs even more and avoid medieval slogs even more.

 

[Fish Man]

If you go for Rifling you naturally pick up forges, macemen, knights, crossbowmen (requires Iron Working and Horseback Riding aside from the beeline). With the Cuirassier beeline you don't naturally get any new units. Against aggressive ai this is quite an issue.

This is pretty irrelevant, as ideally you'll want to build 0 military units before cuirs besides your warrior fogbusters/garrisons. So it doesn't matter if your best defenders are axes or knights. And if you think someone's plotting on you, just grab machinery, in which case you have crossbows and maces now - more than enough to defend.

 

[civac]

You may well be right. Ideally you don't want to build any units. But I don't always get ais to like me in time outside of gifting cities to them.

I like rifles better as they aren't obsoleted in the same way cuirs are. Maybe, I should try the cuirs beeline in a NTT a few times.

 

[Windsor]

I very much agree that AggAI+NTT significantly nerfs Cuirassiers (but they're obviously still relevant). Both the Aesthetics-Literature-Music and Paper-Education-Liberalism line loses a lot of value in NTT, while the bottom half of the tech tree gains relevance. So either just beelining Knights or going for Cavs are options. I don't find AggAI alone makes staying out of wars that more difficult, but NTT does limit your control over diplomacy so the combination of AggAI+NTT gives you more games where having a decent army makes sense.

But rather than ditching Cuirassiers for Cavs, AggAI+NTT tends to send me down the path to steel. That's a strategy that also make leftover axes and maces keep their relevance.

 

[sampsa]

So now that we got that out of the way, let's focus on things that might help OP actually improve.

1) Yeah, considering I was playing against 5(?) AI's, I reckoned that I only got one chance to found a religion, so I went for it straight away. I later ended up converting to Judaism because literally every civ had it, so I made some missionaries and switched (spiritual, so no anarchy).

The thing is that an early religion is slowing you down, because you get all the key techs later. On rare cases you can make a run at being first to mono, but medi/poly I'd never go for.

2) I think the reason I got roads before BW is because I had marble in my capital, and wanted to go for an early oracle.

BW is an absolutely crucial early tech. The ability to chop generates 5 per turn (one turn of movement, 3 turns of chopping generates 20) which is huge this early. Whipping can be good even pre-granary, post-granary it's very powerful. Connecting marble for Oracle is way less urgent. Your for Oracle should often be generated by chopping anyway.

3) I settled late because I wanted to grow my capital first. I just had some really good tiles there, and I figured: if 1 extra tile gets me, say 3 extra food+production, then assuming a settler takes 100 food+production, it shortens the amount of time needed to get the settler out (so tile "growth-->settler" might be faster than just "settler"), however, the larger your city, the less this applies. What I did was basically: develop capital--> pump out some 4 settler one after the other.

2nd city T47 is massively late. Getting a new city working good tiles is much better than growing an existing city to work good tiles, because city center gives at least 211. I'd try to found 2nd city around T35 or earlier, but of course depends on various factors.

Also, I suppose you understand that 1 extra tile gives you 3 extra + if it's a 5 yield tile, because a citizen working the tile eats two . Growth->settler is faster than straight settler only with extremely strong tiles (like several wet corn or pigs).

4) Yeah, I probably go for too many buildings. Keep in mind that my beaker/gold was about 70/30, and even less earlier on. Markets give as much (+25%) as libraries (though they cost about double), so, particularly in your capital they seem like no-brainers to me. The Dun was useful because of the promotions, barracks definitely paid off as well. Monasteries give +10% research, so again they seem like a good investment for the capital to me. Temples basically give the repression +1 without having to put a unit (costing upkeep) in there.

Market is a horrible building, because 150 is an insane amount of . It doesn't give as much as a library, because your slider position should often be at 100% when a market does nothing. Duns, barracks OK in some cities, sure. Monasteries in capital maybe, elsewhere no. SPI temples are OK, but nothing stellar.