Playing on the highest level, “deity”, took a while to figure out, but is more doable then the same level of Civilization IV. Some of the things that helped me:
- A FAQ for the XBox version, which covers basic strategy, tips on each civilizations and tips for each type of victory. Its advice on winning deity is not very helpful, however.
- A detailed tech tree diagram. There is more to this than meets the eye (see “Know your tech tree”, below).
The strategies used by the AI on deity are about the same as on the “easier” level of the game (“emperor”). The main difference is that the computer civilizations are given flat bonuses (for example, their rate of learning new tech seems accelerated) to the point that they are effectively cheating. To compete, you need to take every advantage you can and tolerate no mistakes, especially early on.
The butterfly effect
Avoiding mistakes means that you need to make a lot of use of saving the game. At this level, it is not uncommon to realize that you made a single mistake thirty turns ago that totally screwed you, particularly at the beginning. For example, which direction do you move the very first warrior you build? This might actually turn out to have a huge impact on the game. So, don’t be afraid of going back hundreds of years when you need to. Similarly, save every round. On easier levels, you can tolerate instances when you, say, move a unit a space in the wrong direction or answered a dialog box incorrectly, because your finger slipped or missed the button you wanted. You can’t afford it on deity. Be prepared to curse a lot and restart a turn. Even more if you forget to save a given turn.
An even less “pure” technique for making use of the save feature is to “cheat back” by learning about the map of the game, then starting over on the same map, knowing where key points are. Especially important are the relics. Finding these before your opponents do provides you with a great benefit and denies it to your opponents.
One of the keys to victory on deity is to defeat as many barbarians as you can as quickly as possible. It’s even better if you can reach and kill barbarians around your opponents before they can. While these victories will gain special abilities for your warriors, these are mostly useless. The real reason you do this is that each barbarian village you defeat provides some kind of bonus, and these bonuses make much more of a difference in the early turns in a game. These include:
- Gold, usually 25 or so. At the very start of the game, may be one of the few ways you will be able to make money. The more you grab yourself, the less for your opponents. The main advantage of early gold is that you get your free settler earlier, and that can make a huge difference.
- A caravan. These are useful for quick gold, but they have other uses to the early game, particularly mobility. They are the fastest ancient units (three squares per turn), and they can move through enemy territory. Use them to explore the map, make contact with other civs and find other barbarians. Once you have nowhere else to go, head to the nearest city for 50 gold. The risk here is that the caravan can get nabbed by civs at war with you or the occasional barbarian.
- A galley. The benefit of a free early galley depends on the map. Usually, it will let you find at least one relic before your opponents do. It’s also good for exploration or, if you wind up starting on a small island, claiming some choice continental land with your free settler.
- A horseman. Mostly useful only for faster exploration.
- Technology, usually Horseback Riding. Rare, but happens sometimes.
As you reach certain high-water marks in total gold, you get free things. Know what these are and how they factor into your strategy, particularly if you are shooting for economic victory. The first of these (free settler at 100 gold) is the most important. Hit this as soon as you can. The other levels are listed on the tech tree PDF linked to above. Speaking of that…
Know your tech tree
The basics of the tech tree are clear when playing on any level, but a few things about the tree are not entirely obvious. First, some of the bonuses awarded if you are the first to develop a technology are really nice. Producing Communism first, for example, reduces the cost of any factory you build by quite a bit. Factoring that into your plans can make a big difference. The PDF mentioned above lists what you get for being first with each tech.
Next, the era you are in (e.g. ancient, medieval, etc.) isn’t controlled by the year count, but rather the pace you discover tech. Research five techs, no matter what they are, and you are in the medieval era. Each era changes things like road cost, cost to rush units, initial city population and so on (all shown in a table in the PDF). You can tweak some advantage out of paying attention to this.
The third thing is tech overflow. What happens to the excess science when you discover a new tech? Turns out that it gets converted to gold. So, you can obtain huge sums of cash by watching how your science is accumulating, if you can tune it correctly.
Lastly is tech jumping. The way the techs you can research “unlock” is confusing, but not random. There are set rules that define when you can discover tech without learning a prerequisite first and when you automatically gain the benefit of tech you’ve never actually researched. You can make these work for you.
All victories are domination victories
Peaceful victories are possible on lower levels of difficulties, but barring very unusual circumstances, you’ll need to be fighting on deity. The main reason for this is that your opponents will be advancing much more quickly than you can ever hope to so, unless you can pull off a really early win, you need to take some (most) of their cities to remain competitive.
Another reason is that your opponents will declare war on you at the drop of hat, even when it is stupid for them to do so. (Interestingly, this counters the “peace-nik disadvantage” of Democracy slightly, as you can pretty much guarantee that someone who forced peace on you will declare war again in a few years.)
Peaceful victories might be possible, but I was never able to manage non-domination victory without taking down at least a couple opposing capitals first.
Naval domination is extremely important on deity. Usually cruiser fleets are sufficient. (If you are not basically winning by the time Steel comes around, you are probably screwed. And if you are winning, battleships are probably not necessary.)
This is one area where you have an advantage, because computer opponents rarely collect ships into fleets, which is just stupid. A fleet of ships is much greater than the sum of its parts. I’ve only run into an enemy fleet once.
One drawback is that civs on deity are often smart enough to build cities away from the coast, but this has the effect of limiting the number of ports from which they can build ships of their own. Plus, unless you are going for domination, you can usually be picky in what cities you attack. And, if you are going for domination, all capitals will be on the water.
Plain and simple: naval support wins combats.
Spies are another great equalizer in combat, able to rip down fortifications. This turns a +100% for your opponent into a +50% bonus (or +50% into nothing), so can totally change combats around. On the flip side, other civs on deity will use spies against you like this, so use some spies (or, better, spy rings) for defense of important (or targeted) cities as well.
Don’t be afraid to use the spy’s other abilities as well. Kidnapping great people gains you advantages and removes them from your opponent. Destroying buildings or production also have their uses. Enemy civs will not react to your using spies against them, even if you have a peace treaty (and, face it, they’ll be at war with you soon enough).
Caravans can be used as cover for spies used against friendly civs. Lone spies will get captured, but opposing troops won’t mess with caravans unless you are at war.
You are pretty much guaranteed to run into a situation where the culture of a holdout civ near you is threatening to turn your cities. A spy rush usually works against these cities fairly well. You get a whole bunch of spies within striking distance of a city. The city will usually be defended by its own spy, so build a spy ring out of three of your spies. Hopefully, it takes down the defending spy, leaving no defense against spies. Then send your spies individually into the city. First, keep kidnapping great persons until no more remain. Then start destroying buildings. Usually the cultural buildings get destroyed last, so you need a lot of spies for this.
The only time I’ve seen a computer civ use spy rings is after I did this to a particular city.
Some non-obvious tips about combat. For one thing, the random number generator is seeded. When you save a game, the seed is saved, and restored when you load the saved game. What this means is that if you save a game, then run a combat that doesn’t go your way, if you reload the saved game, the combat will play out exactly the same way the second time.
Second, the big green “advantage!” that shows up in the combat screen is mostly useless, essentially just meaning “this number is bigger than the other number!”. If these numbers are close, you have a chance. If you want to go the save/fight/restore on failure route, it’s worth attacking even if your attack value is a bit less than the defense. This, in fact, is the only way to get great generals, and just one of these can win you a close game.
Other random note: the Oracle of Delphi can really save you some time. In one game, I got the Oracle from discovering the Angkor Wat relic. It radically sped up my game, because you can just always attack, and the Oracle will stop cases where you would lose (not by the odds, but based on the numbers that would actually be rolled). You could do the same thing with the save/fight/restore system, but it is tedious.
You know those strategies that have worked for you that involve getting certain wonders? Like, say, using the Magna Carta for cultural victory? Well, forget them on deity. You are extremely unlikely to be the first to build any important wonder. Pay attention to where they do get made, though, as you may be able to capture them.