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

[GenericName1998]

I'm playing on Immortal:
- Random leader (Brennus (spiritual, charismatic))
- Normal, standard Pangea
- No tech trading/brokering
- No random events
- Agressive AI
- No barbarians

I was doing very well in the early game and my score, to my surprise, was actually on par with that of the other civs. Around 1300 AD, however, Bismarck declared war on me. I was able to fend of his first three waves, but he steamrolled me with an endless supply of cuirassiers and grenadiers, while I was holding a hill-city filled mostly with longbowmen.

In total, I managed to kill 29 cuirassiers, 9 trebuchets & catapults, 4 grenadiers etc etc., and I lost about 27 longbowmen and 10 pikemen. And most of Bismarck's army was still coming.

How in the world am I supposed to match these kinds of numbers of units, without crippling my economy (upkeep alone would be devastating)? The only solution that allows me to get emperor+ victories is to just hold out and pray that no AI attacks me, until I make it to assembly line and start pumping out infantry with the help of factories.

How do you guys deal with this?

 

[Qactus]

Well, basically you have to be the one with the Cuirassiers and Bismarck the one with the longbows.

If you post a starting save with your preferred settings and intended goal (e.g. Pangaea conquest) people can guide you through the game and you'll see you won't need much defensive play at all.

 

[lymond]

Yep, it really boils down to you having a lot to learn about a complex game. You’ve bitten off more than you can chew though I admire the tenacity to play IMM.

edit: yep..no tech trading is not helping you on high levels. Learn the value of trading techs and trading for gold.

I suspect you don't use or generate great people effectively or have a command of bulb strategies.

Playing with no barbs is weak

Playing with no huts/events is strong

 

[scheines]

A few thoughts:

• Improve your teching speed so you have equal or superior units, instead of defending with losing odds. On immortal you’ll need to be well into the rifling / infantry techs by 1300 AD. This various a LOT based on diplomacy and the map, but if your best defender in 1300 AD is longbows you’re gonna be fighting a losing battle. This date is also very affected by playing with no tech trades on, so your mileage may vary there.
• Are you leveraging the fastest means of production available (slavery, nationhood) or “slow building” units? The latter isn’t usually viable long term; you’ll need to rush some units to match the AIs handicaps.
• You’re inevitably going to cripple your economy in war. If you’re making a wonder in your capital, this takes away from your military presence. Your empire should be working toward a single, unified goal. In this case, every city is making units or supporting war efforts as best possible.
• There’s also various diplomacy / general gameplay skills involved. Does Bismarck have more cities than you? Does he have vassals? Did you expand enough in the early game? Could you have bribed in an ally to deflect some of his units? Could you have avoided a war altogether by giving into demands or begging for 1 gold? Each of these are potential solutions.

I’d suggest taking a look at those parts of your game, and you should start to see the other issues disappear.

Also, until you can handle the tech pace on a difficulty, adding in no tech trades and aggressive AI are going to really complicate things.

 

[GenericName1998]

City Garrison I

A few thoughts:

• Improve your teching speed so you have equal or superior units, instead of defending with losing odds. On immortal you’ll need to be well into the rifling / infantry techs by 1300 AD. This various a LOT based on diplomacy and the map, but if your best defender in 1300 AD is longbows you’re gonna be fighting a losing battle. This date is also very affected by playing with no tech trades on, so your mileage may vary there.
• Are you leveraging the fastest means of production available (slavery, nationhood) or “slow building” units? The latter isn’t viable long term; you’ll need to rush some units to match the AIs handicaps.
• You’re going to cripple your economy in war. If you’re making a wonder in your capital, this takes away from your military presence. Your empire should be working toward a unified, single goal. In this case, every city is making units or supporting war efforts as best possible.
• There’s also various diplomacy / general gameplay skills involved. Does Bismarck have more cities than you? Does he have vassals? Did you expand enough in the early game?

I’d suggest taking a look at those parts of your game, and you should start to see the other issues disappear.

Thanks for the reply,

- I might have bogged down my income early in the game, due to some aggressive settling on my part. That cost me a lot of maintenance (70+% of income went to gold), but luckily I managed to get Civil service not too long after the settling race. I feel like generally, my teching was a bit behind, but still ahead of, say, Montezuma (lol).

- I was using some slavery here and there during military production, but more importantly, I got the Heroic Epic in my capital right before the war started, allowing me to make one unit (longbow, knight occasionally) per turn. I was also able to upgrade my longbows to Guerillia I, II and City Garrison I (Dun gives free Guerillia I upgrade).

- Because I had spiritual, I was able to switch in and out of Theocracy whenever I was going to make units, at which point I had every city produce units. Theocracy was on during the war the entire time, obviously).

- I hadn't made any world wonders in this game so far (almost had the Oracle ).

- I actually think Bismarck had fewer cities than me, and I made sure I expanded rapidly in this game at the start.

- I luckily also managed to get duns (walls) and a castle in the city that was under siege before Bismarck attacked (although that isn't helpful against those cuirassiers).

- Furthermore I had some war elephants, muskets and knights during the end, but it ended up being too late.


The thing that surprises me is that, in order to raise an army that Bismarck raised (lets say around 50 units), he'd need to spend net 5000 hammers on that invasion force alone! Thats 10 Pyramids. Is it even possible for a human player to produce that many hammers and still have the resources to reach the necessary techs?

 

[GenericName1998]

Well, basically you have to be the one with the Cuirassiers and Bismarck the one with the longbows.

If you post a starting save with your preferred settings and intended goal (e.g. Pangaea conquest) people can guide you through the game and you'll see you won't need much defensive play at all.

(I figured out how to attach game files haha)

A few turns after this, Bismarck attacks. I assume you mean post it on the forum, or just here? (Yeah, it was conquest victory only).

 

[scheines]

On Immortal, AIs get hammer discounts on units, among other things. Check out this page, which details all the differences in difficulties. AI related values are yellow: https://forums.civfanatics.com/threads/difficulty-levels-spreadsheet.158130/

Key highlights (someone correct me if I am wrong):

• iAITrainPercent: an additional multiplier for unit costs. On immortal, units cost 80% of the hammers compared to your units.
• iAIUnitCostPercent and iAIUnitSupplyPercent: AIs get discounts on unit maintenance cost, meaning they can amass units for cheaper.

It's possible to out-produce AIs, but you need to heavily leverage whipping and drafting, and strategize elsewhere to assist you where needed (i.e. use a war ally, tech advantage, good military strategery, etc.)

 

[lymond]

City Garrison I


Thanks for the reply,

- I might have bogged down my income early in the game, due to some aggressive settling on my part. That cost me a lot of maintenance (70+% of income went to gold), but luckily I managed to get Civil service not too long after the settling race. I feel like generally, my teching was a bit behind, but still ahead of, say, Montezuma (lol).

- I was using some slavery here and there during military production, but more importantly, I got the Heroic Epic in my capital right before the war started, allowing me to make one unit (longbow, knight occasionally) per turn. I was also able to upgrade my longbows to Guerillia I, II and City Garrison I (Dun gives free Guerillia I upgrade).

- Because I had spiritual, I was able to switch in and out of Theocracy whenever I was going to make units, at which point I had every city produce units. Theocracy was on during the war the entire time, obviously).

- I hadn't made any world wonders in this game so far (almost had the Oracle ).

- I actually think Bismarck had fewer cities than me, and I made sure I expanded rapidly in this game at the start.

- I luckily also managed to get duns (walls) and a castle in the city that was under siege before Bismarck attacked (although that isn't helpful against those cuirassiers).

- Furthermore I had some war elephants, muskets and knights during the end, but it ended up being too late.


The thing that surprises me is that, in order to raise an army that Bismarck raised (lets say around 50 units), he'd need to spend net 5000 hammers on that invasion force alone! Thats 10 Pyramids. Is it even possible for a human player to produce that many hammers and still have the resources to reach the necessary techs?

Most of what you describe here is defensive play. It is not designed to play to win. Longbows are actually not at all a part of my game. I rarely if ever build even one longbow. Or knights for that matter.

HE in your cap is generally very suboptimal.

You likely need to totally rethink your play style and learn quite a few basic gameplay mechanics and concepts. Hard to do that at your stage of that game. The early game is so important and your education should begin there.

General note on AI unit builds. Most AIs are not going to have a ton of units unless they are in war mode. Exception are the psychos with high unitprob, but they are usually in war mode often anyway. The key is hitting the AI before they go into war mode and using diplomacy to keep the peace until you rage on them with your mighty force.

 

[scheines]

Most of what you describe here is defensive play. It is not designed to play to win. Longbows are actually not at all a part of my game. I rarely if ever build even one longbow. Or knights for that matter.

Yep, oftentimes if I end up on the defensive end of a war and am making longbows, I'll play it out but I will eventually come to terms with the loss, and re-load to see where I went wrong to get declared on. Sadly, defensive oriented play is a losing proposition long term.

 

[Fippy]

Players can easily outproduce AIs for units, with slavery (or later SP + workshops).
Usually before / when i attack every city whips an unit each 2 turns, afterwards as soon as they can again.
What matters: AIs will randomly pick which units they get, we can focus on that 1 best unit available.

For example with Cuirs you would build nothing else, other than an occasional defender who stays in captured cities or guards borders.
(i disagree slightly with Lymond on Longbows, i think they are excellent for that task and can be done with overflow)

 

[lymond]

(i disagree slightly with Lymond on Longbows, i think they are excellent for that task and can be done with overflow)

Not sure what the disagreement is here. I never said LBs were bad.. in fact, on paper the are extremely strong for their intended purpose. The point for OP or anyone is sitting there mass producing LBs while playing defensively.

Saying that I never build them though was a bit strong. LBs have there uses, but they don't make an army and certainly not a strategy.

 

[dankok8]

If you're facing a ton of attacking units instead of building all Longbows build some Longbows and some siege. If you had a stack of 20 Cannons you could have smacked them with those and cleaned up with almost any units. Collateral damage is extremely effective against large groups of units.

 

[ArchGhost]

Brush up on your diplomatic game too., the DoW could have likely been prevented or pre-empted with a bribe

Bismarck declares at Pleased or lower. Aggressive AI ensures he won't check relative power to you before rolling a war check on you, though on IMM it's likely a militant like him is going to have more power anyway unless you are heavily cranking units yourself in prep for an attack, or you already crippled him earlier.

No tech trading also doesn't do you any favors on the diplomatic front. It can be easier or harder to force everyone to tech their own stuff depending on the AI's tech preferences (Bismarck's is military, btw), but if you are behind in military techs and get attacked...well you are behind, and it's harder to do things like get him in a war to deflect his aggression since you can largely only use gold to bribe others. As was already said upthread, the solution is you need to be the one doing the attacking, in order to prevent this situation.

The thing that surprises me is that, in order to raise an army that Bismarck raised (lets say around 50 units), he'd need to spend net 5000 hammers on that invasion force alone! Thats 10 Pyramids. Is it even possible for a human player to produce that many hammers and still have the resources to reach the necessary techs?

On the right (huge) map I've produced well over 200 Cuirassiers (20K hammers) along with supporting infrastructure (roughly 350h per city, anywhere from 15-30 cities not counting captures during the Cuir war) by around this point and was ready to swap to Cavalry. You can be much more productive than the AI with similar resources because the AI doesn't whip heavily and it builds a ton of unnecessary crap.

Yes it's easily possible if you are able to snowball and learn to cut the fat out out of your tech path/bulb through and use Liberalism etc. to get there quickly. The most important part is getting there quickly enough to have the advantage over the AI and not the other way around, the problems of production and sequence of who to attack can be ironed out later.

Tech trading on helps you a lot in reaching certain plateaus of tech earlier, especially as the difficulty goes up (and thus the AIs you can trade with tech faster) which you can leverage into a winning position. Don't play a lot of No Tech Trading games but from what I understand beelining military tech is somewhat more important.

 

[GenericName1998]

Brush up on your diplomatic game too., the DoW could have likely been prevented or pre-empted with a bribe

Bismarck declares at Pleased or lower. Aggressive AI ensures he won't check relative power to you before rolling a war check on you, though on IMM it's likely a militant like him is going to have more power anyway unless you are heavily cranking units yourself in prep for an attack, or you already crippled him earlier.

No tech trading also doesn't do you any favors on the diplomatic front. It can be easier or harder to force everyone to tech their own stuff depending on the AI's tech preferences (Bismarck's is military, btw), but if you are behind in military techs and get attacked...well you are behind, and it's harder to do things like get him in a war to deflect his aggression since you can largely only use gold to bribe others. As was already said upthread, the solution is you need to be the one doing the attacking, in order to prevent this situation.


On the right (huge) map I've produced well over 200 Cuirassiers (20K hammers) along with supporting infrastructure (roughly 350h per city, anywhere from 15-30 cities not counting captures during the Cuir war) by around this point and was ready to swap to Cavalry. You can be much more productive than the AI with similar resources because the AI doesn't whip heavily and it builds a ton of unnecessary crap.

Yes it's easily possible if you are able to snowball and learn to cut the fat out out of your tech path/bulb through and use Liberalism etc. to get there quickly. The most important part is getting there quickly enough to have the advantage over the AI and not the other way around, the problems of production and sequence of who to attack can be ironed out later.

Tech trading on helps you a lot in reaching certain plateaus of tech earlier, especially as the difficulty goes up (and thus the AIs you can trade with tech faster) which you can leverage into a winning position. Don't play a lot of No Tech Trading games but from what I understand beelining military tech is somewhat more important.

My god, 200 cuirs xD, that game makes must have taken ages.

The reason I opted out of tech trading/brokering mechanic is because of three reasons:

- It's very exploitable for the human player, by researching techs that the AI often ignores.
- It makes every civ follow a similair research "route", because everyone constantly ends up with the same techs, taking away from a civs "personality" imo.
- It's very boring. If want to take full advantage of it (as I used to lol), you have to try to get the optimum amount of gold out each deal and keep track of what everyone is researhing etc..

I'm also not entirely familiar with how DoW works in Civ 4. I've heard of bribes before, but never used them, because I have no idea how much woud be enough to prevent a war. It seems like theres a certain threshold ratio of your power/their power, where, once you go below it, a civ will declare war. Hard to make sense of though. Imo, civs are (even with agressive AI on) waaay too passive with their forces. They make huuuuge stacks for some reason, but usually won't attack my often nearly empty cities (until they do haha). I'd play on a lower difficulty, but then I can just focus on my economy and stay out of war and steamroll them all in the industrial era one by one ez pz.

You're right about the snowballing, this game is all about figuring out the steepest exponential curve. I think I've got that down decently. War and keeping a big enough army is just where I suck.

 

[GenericName1998]

If you're facing a ton of attacking units instead of building all Longbows build some Longbows and some siege. If you had a stack of 20 Cannons you could have smacked them with those and cleaned up with almost any units. Collateral damage is extremely effective against large groups of units.

Was thinking of doing this (had some cities produce catapults), but you have to either:
-know if the enemy will attack you next turn, or;
-attack the same turn yourself

I would never have succeeded with an attack of my own, because my units were mostly longbows, with their nice defensive bonuses. I'd rather let Bismarck throw his troops against my city walls.

 

[GenericName1998]

Yep, it really boils down to you having a lot to learn about a complex game. You’ve bitten off more than you can chew though I admire the tenacity to play IMM.

edit: yep..no tech trading is not helping you on high levels. Learn the value of trading techs and trading for gold.

I suspect you don't use or generate great people effectively or have a command of bulb strategies.

Playing with no barbs is weak

Playing with no huts/events is strong

I think disabling tech trading actually makes the game more interesting, so I'll just have to take that into account and adjust my difficulty accordingly.

You're right, I never use bulbs. I don't ever see the +~1000bpt being worth a +50%bpt academy, but yeah, perhaps I should look into those strategies. I do sometimes try to go for an early oracle --> metal casting --> forge --> great engineer --> hurry Pyramids. But this obviously works best with an industrious/philosophical leader.

I think I'll enable barbs next time. The reason I disabled it is becasue the AI just gets a really unfair advantage early on by already starting out with 2 archers and some warriors(?) on IMM, while the human struggles with endless waves of barbs.

 

[sampsa]

You're right, I never use bulbs. I don't ever see the +~1000bpt being worth a +50%bpt academy, but yeah, perhaps I should look into those strategies.

This is interesting. I pretty much never see an academy being able to compete with say bulbing philo (which also unlocks a very powerful civic), though bulbing gains value with tech trades on.

I'm pretty sure all your problems stem from mistakes in the early game. People here can help you with that.

 

[GenericName1998]

This is interesting. I pretty much never see an academy being able to compete with say bulbing philo (which also unlocks a very powerful civic), though bulbing gains value with tech trades on.

I'm pretty sure all your problems stem from mistakes in the early game. People here can help you with that.

My reasoning is:
- A city, currently without academy, will on average, over the next 150 turns (games last about 300+ turns before it's mostly decides who's gonna win) generate 50+ commerce/turn.
- Assume 70%/30% research/wealth times 50% = 50*70%*50% = 17.5 extra research.
- 150*17.5 = 2625 research in the long run.

However, bulbs do provide an immediate bonus, whereas an academy takes a long time to pay off (about 60 turns in this example).

Gotta say, I love that civ 4 is still so full of learning points and new strategies. You don't see that very often in games.

Edit: whoops, missed that fact that its only 50% in my example haha

 

[pi-r8]

My reasoning is:
- A city, currently without academy, will on average, over the next 150 turns (games last about 300+ turns before it's mostly decides who's gonna win) generate 50+ commerce/turn.
- Assume 70%/30% research/wealth times 50% = 35*50% = 18.5 extra research.
- 150*18.5 = 2775 research in the long run.

However, bulbs do provide an immediate bonus, whereas an academy takes a long time to pay off (about 60 turns in this example).

Gotta say, I love that civ 4 is still so full of learning points and new strategies. You don't see that very often in games.

Edit: whoops, missed that fact that its only 50% in my example haha

It is a fascinating game. It's full of these apples-to-oranges tradeoffs that are hard to compare, but still make a big difference in the end. The specialists and great people are especially interesting that way.

I think the main thing missing in your analysis is that the bulb doesn't just provide a single, static benefit. Since it warps you forward down the tech tree, it also gives you the benefit of all later techs earlier. Bulbing philosophy gives you philosophy some turns earlier, which gives you pacifism earlier, which gives you more GPP. (you can also trade it, but you said you're playing with trading disabled, so I'll ignore that). It's hard to directly compare GPP vs bulbs, but clearly it's worth *something*. It then leads to paper (trade your map for money!), education (universities), and liberalism (free tech!) earlier. Just getting those things 5 turns earlier can be worth a lot.

But then the other big thing, is that what you really want is a big military advantage. So instead of slowly increasing beakers to out-tech in the late game, just pile on short-term advantages so you can reach the renaissance ASAP. once you get to the Renaissance, you can either go for Cuirassiers, or Cannons, and you'll have a unit that eats medieval units for breakfast. The AI can't cope with them at all, so you can easily conquer a neighbor and double or triple your size. That pretty much wins you the game, and is worth way more than any number of academies.

 

[GenericName1998]

I think the main thing missing in your analysis is that the bulb doesn't just provide a single, static benefit. Since it warps you forward down the tech tree, it also gives you the benefit of all later techs earlier. Bulbing philosophy gives you philosophy some turns earlier, which gives you pacifism earlier, which gives you more GPP.

Yep, and not only that, but you'll also get whatever tech you want to research after that faster, giving you the advantages of that tech earlier as well. But that research boost will fade and be overtaken by the academy in the long run, reversing the effect. Indeed, hard to decide.

It's full of these apples-to-oranges tradeoffs that are hard to compare, but still make a big difference in the end.

I think this is particularly the case with civics as well. Sometimes I just spend some time pondering about what the consequences of switching from representation to universal suffrage will be and if/when it's a good time to do so, because even though they are massively different in the bonuses they provide, they can still, somehow, be compared. It forces you to plan and think of what you want your future economy to look like.