At Emperor, it's Basically Just a War Game, no ?

[SorvinoBackhand]

I play at the Emperor level and standard size maps at the standard speed with randomized leaders. Here are my general observations:

1. If you don't focus on rapid expansion early, you get crushed. Stonehenge and the Pyramids are nice, but forget it at the hard levels. There is no way you have the time and can risk the production. (If you are lucky, you can take the wonders in the soon-coming war.) The AI is all over you quick, and you need to prep for your first invasion. Wonders, exploration, fighting the barbs for city-state influence, etc. are all just luxuries you can't afford anymore. So forget Honor or Tradition, you gotta go for Liberty to get the expansion bonuses. In short, at the hard levels, my opening moves increasingly don't vary much. As a result, your first invasion goes like this: out of nowhere, some civ you've barely interacted with declares war and throws 10 units at you simultaneously. Because you have been building frantically and bought Liberty, you have almost no military and you lose two cities early. But then the AI's terrible battlefield choices (point 4 below) and your building of military units in all your cities turns the tide. At this point, your domestic production all but stops. You become the Soviet Union, building a massive war machine with little domestic infrastructure (libraries, granaries, etc). This allows you to nearly defeat the first invader, when suddenly another AI nearby attacks you also. Their city-state allies get in on the act (you don't have the cash to buy them, because you blew it buying emergency archers to survive the first wave earlier), as do the other AIs on the continent eventually. And pretty soon your fighting to take over the whole continent in a massive land war that occupies your from turn 50/75 to 300.

2. If you don't find research iron and settle around it, you're in real trouble. Increasingly I find myself bee-lining for iron out of necessity. Otherwise, I bee-line for crossbowmen as the only way to ward off swordsmen and long swordsmen I can't build.

3. You can't automate workers (or scouts) anymore. At the higher levels, the AI's quantitative advantages are so enormous, that you must take advantage of your superior human reasoning ability and micro-manage. I find this somewhat tiresome, as it slows down the pace, but it's the only way to prevent workers running all over the map, wasting three moves in pursuit of some tile far away.

4. Your only hope is to tactically outhinking the AI in combat. One recurrent problem in the Civ franchise is astonishingly bad AI military behavior. Units wander around in front of you cites to get hammered by your ranged units. AI units will embark for no particular reason, even though ranged units are brutally effective against embarked units. Wounded AI units don't retreat to get healed; they hang around the battle space for you to pick them off. The game makes up for this by giving the AI ridiculous quantitative advantages at the higher levels of difficilty. This is why about half-way through the game you get that astonshing pop-up that tells you that you have 400 gold, while the leading AI state has 50,000. The only possible way to overcome such insane stratetgic-quantitative advantages, is for the human player to tactically outfight the AI on the battlefield. This is why you must fight - its the only way to run down the AI's insane resource advantages. And this is also why the AI attacks you all the time at the hard levels: the difference engine of the AI reads a massive AI advantage over the human player and then it opts for war. Hence you get attacked so much, the game just turns into centuries of war (which is wildly unrealistic of course). The game would be far more enjoyable if the AI played better at the harder level and didn't just get more and more of everything in order to simply swamp the human player.

5. Diplomacy flies out the window except for resources swaps. Allegiance and cooperative play is all but impossible, because your AI 'allies' will turn on you on the drop of hat. This annoys me to no end, because it really shows how the AI is just a mathematical difference engine, and not a 'person.' Real world alliances don't just evaporate overnight like this; states don't march all the way across the world to fight someone they barely know. But the AI ignores the obvious geopolitical imperative of fighting neighbors and building sticky alliances. That way, it can attack you for no particular reason. Don't anthropomize the AI, which unfortunately turns the game into an exercise in game theory. Too bad MP doesn't work well too alleviate this, because at this point, I say no to just about every diplomatic solicitation, because I expect the AI to defect with no warning.

6. War of course is hugely expensive, so even after you take control of your continent and close in on the number 3 or 2 position, your happines and gold problems are so severe, that it becomes almost impossible to catch up. At this point, the leading AIs, who have dispatched their rvals with vastly greater ease than you ever could, build the UN or the spaceship and you lose. You try to build the huge fleet you need to invade the other island, but by that time your domestic unhappiness and bankruptcy have paralyzed you. Your techs and culture are way behind, because you were building units and war-buildings instead of domestic production (you're the USSR). You struggle to catch up, but there's no way you can buy the city-states you need to survive the UN vote (because you're broke and couldn't spend the culture buying the patronage tree). And you are loathe to switch you rationalism, because you had to buy piety just to control your spiraling unhappiness from all that imperialism. So sometime around turn 400-425, you lose. On the other hand, if you didn't fight all those wars, you never would have been big enough to take on the AI leaders anyway, so you're stuck.

At least, this is more or less how my games turn out. Thoughts would be appreciated.

[Seyhan31]

The real sh*t fight is when you get to Immortal difficuility. You shouldn't automate your workers on ANY difficuilty. Rapid expansion is really a situational strategy. You should reason weather you should expand outwords or inwards. One massive city can be just as effective as a commune.

AI rely on numbers, not brains. For every 1 unit you have, at emprorer they have 3. At deity it's pretty much impossible unless you've got the envirement to your advantage(The only time I've pulled off a deity conquest was when I waited for Montezmu to assume control of the world, then sent a small army to capture his capital).

At the same time, your war-time tactics should shift from mainstream battles to ambushes and Total war tactics. Attack EVERYONE. Even if it's a worker you don't need, you can destroy and disband him and still get 30 gold for deleting the unit. You should NEVER resort to guirella, unless your in the rare situation where the Iraquious national ability is actualy useful. You can only really fight guierlla defensively, and by the time you've defeated one wave as a guirilla nation, the enemy have already prepared 10 more.

Don't try and capture cities unless they've already been weakened. Expanding outwards too quickly will force the AI into direct combat, and they will completely disregard diplomocy, friend or foe. At the same time, the AI at immortal difficuilty will have Rifles before you have swords.

[Dactyl]

SorvinoBackhand, thank you. Your description of your experience matches mine. I read all these strategy articles and forum entries, especially on CivFanatics, where people talk about some particular favorite approach. I try to put these strategies in effect, and they work after a fashion, but rarely with the effectiveness that their proponents claim for them.

[chemist]

I have won games on Emperor without combat. Not so on immortal, but everything else written about the immortal level above is wrong. You have to play really badly for the AI to have "Rifles before you have swords".

When you first move up to a new level, try cooking the settings in your favor to start with.

[hackerjack]

This seems actually quite ridiculous to me, I'm guessing you wrote this after only just moving up to the level.


I play Emperor mostly and yes my early games descended into desperately trying to survive being attacked and never catching up. But seeing that the strategy I had wasn't working I looked at others and found that actually in 90% of cases a "build" strategy works much better.

Generally for me this involves several early steps.

1. Build a worker - Provided you are not in a god-awful jungle/marsh starting location then getting a worker to pump your population up with farms and connect any resources is paramount. Without it you fall behind every turn on gold, production and science.

2. Do NOT build more cities / Grow your capital - I know it's tempting to grab land before others do but it's a crap strategy at this level. Basically the first 1500-2000 years or so should be as a city state unless there's a magical location nearby (Barrier reef for example). Building a settler stunts growth of your capital, more cities mean more territory to defend, higher cost of policies, higher cost of national wonders and hitting the happiness cap earlier. You can get a granary, library or 1.5 warriors for the time it will take to build a settler.

3. Tech for research, not war - What is the point of surviving early with chariots and swordsmen if you are sacrificing the long-game? You can defend a single city quite well with a small number of early units. I tend to beeline for the National College as early as I can, it allows me to keep pace early on until I can afford research agreements. In an ideal game I might even go for the Great Library rather than a standard one as well if I have the right combination of surroundings to make Civil Service work for me (any other medieval level tech - the min you want to use the GL for - would require too many other early techs to be worth delaying the College for), but that is a hit and miss strategy.

4. Small armies used properly - A single archer allied with two warriors can repel attacks of 10+ units especially if you have a decent defensive position and let the enemy attack you (a warrior on a forested hill in your own territory is ideal and can easily take on several opponents).

5. Right policies - Tradition is essential. At least the header policy which gives you +3 culture is anyway. Not only does it allow you more policies but grows the territory around your capital (You can achieve the same with Liverty + a monument but that wastes good money/production time that could be used better elsewhere). Ideally you want your warriors nestled next to your city with a 1 tile buffer of your territory around them. After that it depends on the situation.

5a. Policy for defence - Ogliarchy gives you a nice extra bonus on defence thanks to the extra bombard power of the city from the archer you are going to have in it, those two combined can take 2-4HP each from invading Warriors before they even hit your own melee units. Also gives you one extra goal per turn not paying for the archer.

5b. Policy for immediate growth - Go down the middle Tradition path to get monuments and the food bonuses.

5c. Policy for delayed growth and culture - Open up liberty and head for the policy that reduces policy cost per city (I forget it's name), on the way you'll pick up +1 culture, faster workers and a free worker.

6. Cash in=win - Sell everything, absolutely everything. Luxuries are pointless with a 1-city early strategy, you can get around 8-10 gold p/t up front for them (always go up front where possible). Open Borders bring in 50 gold and you don't have enough of a territory to stop people just going around you anyway, it also reduces the liklihood of war, or at least delays it. Strategic resources are more difficult and viariable, try to keep a couple irons back.


Using this strategy I nearly always have to fight off two early wars, but those are over and done with by 1500 BC and usually don't overlap too much (try for peace each turn, take any reasonable offer), the AI won't come back as soon as the deal expires and declare again unless you give it reason to. By this time I have a capital with a Library, Granary/Lighthouse, National College and maybe a Monument, 2 Workers, 1-2 Archers, 2-3 Warriors (only replace as/when you are losing them) and should only marginally be behind in the science race. This is the time to get your second/third cities underway (usually I hang on until I have the policy mentioned in 5c) and your workers should be free to be able to help build up the land around them.

Provided you get Iron you have a decent chance from then on. If you don't you're screwed either way.

[chemist]

I only play Immortal but I've read a fair bit about the strategies deity players use. My thoughts relevant to your strategy are below.

I play Emperor mostly and yes my early games descended into desperately trying to survive being attacked and never catching up. But seeing that the strategy I had wasn't working I looked at others and found that actually in 90% of cases a "build" strategy works much better.

Generally for me this involves several early steps.

1. Build a worker - Provided you are not in a god-awful jungle/marsh starting location then getting a worker to pump your population up with farms and connect any resources is paramount. Without it you fall behind every turn on gold, production and science.

Most high level players just go Liberty, so they don't need to build a worker. Alternatively (or as well), use your scout to steal a CS worker. Doing this more than once has risky diplomatic consequences. Building a scout is a higher priority than a worker.

2. Do NOT build more cities / Grow your capital - I know it's tempting to grab land before others do but it's a crap strategy at this level. Basically the first 1500-2000 years or so should be as a city state unless there's a magical location nearby (Barrier reef for example). Building a settler stunts growth of your capital, more cities mean more territory to defend, higher cost of policies, higher cost of national wonders and hitting the happiness cap earlier. You can get a granary, library or 1.5 warriors for the time it will take to build a settler.

Take the free liberty settler. Two cities wont hurt you much and will increase the lux you have to sell. I never build a settler. The third city is built late if at all and by then I have enough gold to buy a settler.

3. Tech for research, not war - What is the point of surviving early with chariots and swordsmen if you are sacrificing the long-game? You can defend a single city quite well with a small number of early units. I tend to beeline for the National College as early as I can, it allows me to keep pace early on until I can afford research agreements. In an ideal game I might even go for the Great Library rather than a standard one as well if I have the right combination of surroundings to make Civil Service work for me (any other medieval level tech - the min you want to use the GL for - would require too many other early techs to be worth delaying the College for), but that is a hit and miss strategy.

I agree, tech for tech. Often you can skip the whole bottom of the tree after mining until you are in renaissance. You may want/need masonry for marble or if your start has lots of stone. You will have to keep a low profile and work the diplo to stay out of trouble because you will be weak for some time. Oh, and have plenty of (chariot) archers.

Playing a "chivalry trick" civilization (Mongol/Arab/Songhai) fits well with this strategy. If all goes well you will have researched all the prereqs for Chivalry by the time you finish the Liberty tree. Choose a GS and bulb chiv. Most good players endorse rushing the GL and using the free tech to bulb philosophy, then starting on the NC. Turning off ancient ruins considerably increases your chance of winning the GL race on high levels.

4. Small armies used properly - A single archer allied with two warriors can repel attacks of 10+ units especially if you have a decent defensive position and let the enemy attack you (a warrior on a forested hill in your own territory is ideal and can easily take on several opponents).

You muist be much better at combat than I am. I like to have a minimum 3 achers per city as defence. I research pottery then archery. I never build a warrior unless playing a warrior trick civ.

5. Right policies - Tradition is essential.

You finish your first tree faster if you go straight Liberty. The free stuff generally beats what Tradition has to offer.

6. Cash in=win - Sell everything, absolutely everything. Luxuries are pointless with a 1-city early strategy, you can get around 8-10 gold p/t up front for them (always go up front where possible). Open Borders bring in 50 gold and you don't have enough of a territory to stop people just going around you anyway, it also reduces the liklihood of war, or at least delays it. Strategic resources are more difficult and viariable, try to keep a couple irons back.

Selling stuff is correct. If you have two cities you will probably need to hold back at least one lux. Settling near horses, phants and stone helps with happiness.

Provided you get Iron you have a decent chance from then on. If you don't you're screwed either way.

Depends on the civ. Iroquois and the chiv trick civs don't need iron.

[hackerjack]

Hi Chemist, some good points there.

I forgot to mention the goody huts. I usually play with them turned off (because I really don't like that a players entire game can be made/wrecked by a couple of random occurrences), which greatly reduces the need and usefulness of a scout.


I have tried Liberty only and it works well if you plan on completing the tree and/or don't want to build a worker. I just tend to prefer the earlier border growth/bonus culture + not waiting 50 turns to get a worker going. Tradition also opens up Ogliarchy (if under attack) or Aristocracy (GL/NC). I also often move onto the Commerce tree ASAP, often by the 6th policy.

I dare say that leads to a slower initial buildup than other players in some cases.


Can't say I'm brilliant at combat, I tend to just do it very defensively early on. Park a few melee units on defensive spots, cover with the Archer/City, let incoming units attack the city itself occasionally and pick off what's left. It doesn't work past the Warrior/Chariot/Archer age though unless you build more.


Never really tried much of the Chivalry trick civs because none of them have UAs that really appeal to me, also I prefer to invest an early G Scientist in an academy if I can get away with it. Again probably not a tactic that will work on higher levels because you wont last to see it pay of long term.

[chemist]

Hi Chemist, some good points there.

I forgot to mention the goody huts. I usually play with them turned off (because I really don't like that a players entire game can be made/wrecked by a couple of random occurrences), which greatly reduces the need and usefulness of a scout.

Agree the goody huts just add a stupid random element to the game (if an AI hutlucks Writing, they will beat you to the GL). However, the scout is still good for finding CS, AIs to trade with, and a good spot for c2. I could imagine starting with monument, though.

I have tried Liberty only and it works well if you plan on completing the tree and/or don't want to build a worker. I just tend to prefer the earlier border growth/bonus culture + not waiting 50 turns to get a worker going.

In my current game, 37 turns with monument as the second build.

Tradition also opens up Ogliarchy (if under attack) or Aristocracy (GL/NC). I also often move onto the Commerce tree ASAP, often by the 6th policy.

I often open with Tradition for the border growth then switch to completing the Liberty tree. Sorry to say, but all the gurus say Commerce sucks. Racing to renaissance so you can open Freedom or Rationalism seems to be standard. Tradition is handy for absorbing any surplus culture picks (I don't tick save culture). You can try to exploit Legalism by holding it back until you have built monument/temples/opera houses in your first four cities (captured cities count), but I usually have better things to spend a social policy pick on.

I dare say that leads to a slower initial buildup than other players in some cases.


Can't say I'm brilliant at combat, I tend to just do it very defensively early on. Park a few melee units on defensive spots, cover with the Archer/City, let incoming units attack the city itself occasionally and pick off what's left. It doesn't work past the Warrior/Chariot/Archer age though unless you build more.

On immortal it can be harder. If your city is in open terrain you can find your 2 warriors dead and most of your 1-ring occupied by enemy units rather quickly. The AI is still stupid though and will do things like attacking with 6 archers and one warrior. Kill the warrior and they're not taking your capital.

Never really tried much of the Chivalry trick civs because none of them have UAs that really appeal to me,

Any kiting (shoot and retreat) unit kills the AI in combat. I stopped playing Arab because I was winning too much. Their UA isn't much, but UB provides double quantity for all luxuries, which is great and basically gives you all the money you need via trading. The catch is that it requires Currency. I would generally try to work things so I could get both that and chivalry as free picks.

Mongol Keshik is ridiculously OP.

also I prefer to invest an early G Scientist in an academy if I can get away with it. Again probably not a tactic that will work on higher levels because you wont last to see it pay of long term.

It doesn't add up unless you are playing as Babylon. The fixed science trickle becomes insignificant as the game progresses because the cost of techs escalates so much. A better strategy is to stockpile GSs. I tend to save them until I can grab a technology I think will either give me a significant military edge or boost my growth to the next plane. Chemistry, rifling, dynamite, education, economics, scientific theory, electronics, atomic theory, fission...

One popular deity strategy revolves around stockpiling GSs until you can rush to electronics. Get the prereqs to Biology and Railroad, then burn 5 GS in one turn! I think it's boring, but it's one reason deity players like Freedom.

[hackerjack]

Yeah I had a feeling quite a bit of what I was doing wouldn't scale up to higher levels, hence the issues you mentioned.

I really should try moving up ne of these days but I find I play solidly for a few weeks then drift onto something else.

[awightman]

I have not played for that long, but today I just played my 4th game on Emperor and I won BIG. I won with a cultural victory in 1954 AD, at one point it said I had 29 policies and 2nd place was 20. I was also #1 in land, #1 in literacy, was "Allies" with every City-state in the game, was 3rd in military strength, and like 2nd-3rd'ish in GNP, crops, and population.

I have found it extremely effective to stick with just 1 city until you've unlocked Tradition and Liberty, got the free worker policy from Liberty, then the one that starts a golden age and reduces culture cost increase from founding cities by 33%, then the settler one which will create your 2nd town around turn 50-60. You want to make sure you are growing and happy first and foremost, and then priority on culture buildings, science buildings, and then go for wonders as soon as your capital has the production to produce them in under 18-20 turns. Go for the Great Library, The Oracle, Stonehenge if you feel like you are ahead of the curve early, Oxford University, the Porcelain Tower, the Great Wall, Brandenburg Gate, etc. Any wonder that gives free policies or research, go for it ferociously. Be pissed when someone else gets it first. Even use workers to chop forests for production to make sure you get those early wonders.

You should keep only 3 total towns until at least around 200+, after which only settle a new town if its a great location. Get ranged units if you get attacked. Trade any luxury resources and GPT you have to other players to get all their gold in the treasury, and use it to immediately purchase walls and other units. Get the Tradition policy where garrisoned units increase city strength by 100%. Fire away, stay just out of range for them to attack you each turn, and just within range that if they move close you have a city and 2-3 ranged units killing them in one turn. In these levels if you don't do complete kills they will often level up and use it to fully heal and take out your weakened unit.

Now your main goal is to make sure your Capital builds wonders, your other 2 towns have all the research and culture and growth buildings available, and then move onto barracks and walls and happiness (but always do happiness right away if your empire is unhappy, you never ever want an unhappy empire unless you are totally desperate).

Since you've built wonders, you will occasionally get great people. Policies and wonders will give you free great people as well. Use every Scientist to learn the most expensive technology available. If you are a few turns from learning one that will open up an even more expensive one than is available, save the scientist until you've learned the one you're on and then learn the new one. You can enter new Eras very quickly this way which will allow you to bust into the Piety and Patronage trees. Completing Piety tree is very important, as it gives you lots of culture, a free policy, and 10% reduces cost on all future policies. Patronage is also very important as you just have to make sure you have lots of gold and you will get tons of units, great people, culture, science, food, and resources from the City-States. Also if they are allies they will back you up if you get attacked.

I got attacked real early but I made 3 archers and picked off 6 warriors without losing a unit and made peace for 200 gold, then later had Arabia attack me a couple of times, but I already had the Great Wall and Himeji Castle, plus lots of gold to purchase Knights vs his archers and warriors and 1 pikeman. By the time the battles was over I had 4 cannons and took over 2 of his towns. I made puppets - he attacked me a little later and I took 2 more towns of his and made puppets again.

I think I was in the lead in score the entire game, every single turn, and part of it was a great starting location with lots of resources, the other part was focusing on all the free science and culture I could get, using Piety and Patronage and very few towns building wonders and spending great scientists on technologies and engineers hurrying wonders and merchants doing trade missions, etc.

All the "freebies" from the wonders/great people/policies are what really puts you ahead of the AI. And if you are superior technologically and have gold output, then you can fend off any attack from this AI, ever. The easiest way is build on the cost and get a strong navy....

[Chemical Ollie]

Perhaps I was lucky, but on the few Emperor games I've played this weekend, I had no problems fending off the enemy Stack-of-Doom with just a warrior or two. The city itself is good at bombardment, so the AI needs several tough units to crack it. It also helps if you have allied a nearby City-State, which the enemy has to pass on his way to your land.

On the other hand, I have failed the same thing on lower levels, but those were my first Civ games in years, so I blame it on the rust.

[sgk2004]

Has any of this changed since G+K?

[ambalamps]

Has any of this changed since G+K?

As a regular G&K player, I still find a lot of these tips useful as I try to move up to Emporer level games. I would think some advice would change. For example, Stonehenge now provides faith instead of culture, so it might not be as advantageous to build it early - unless you had a strategy that called for using religion.

Spies become a lot more useful at the upper levels of the game, as if you are lagging behind in science, they can be used to great effect to narrow the gap to other AIs. Also somewhat useful in retaining a particularly useful city-state (although I've never found spies to be particularly good at that).