The AI is too hard on Deity! What can I do?

Jamesds

Great Scientist
Joined
Mar 14, 2002
Messages
1,158
Location
Highlands, Scotland Uni: St Andrews
The AI is so hard.

Once I played on Deity and made 4 cities, 2 veteran phalanxes in each. An AI elephant came along and took over one, then two, then three of my cities! I was left crying!:cry: :cry:

The only amusing thing was that on one phalanx, while on half heealth, it suddenly disappeared when attacking one of my phalanxes! Cool! :lol: :lol:

So, how do you beat the higher levels? Can someone please help me!:egypt:
 
Your game almost sounds like a proximity problem; in the opening game, it's often a good idea to locate and destroy any nearby civilization before they come at you. In the case of the game you stated, instead of building barracks and phalanxes for the defensive, try to get some huts with military units and find out if you have any nasty neighbors; if you do, try and eliminate them. (Also, if you're learning deity, you might try playing on the large map to increase the opportunity of giving yourself room to grow before encountering an opposing civ.)
 
Winning at deity is very hard. While some players reckon they can always do it; I have always found I need some luck myself.


At Deity the AIs Build, Grow and Research much faster than you.

There are two main approaches according to world size:

For small single continent worlds the best strategy is to build horsemen and chariots and attack first.

For large worlds the best strategy is to outgrow the AIs.
To do this; you must reckon on having about twice as many cities as your nearest AI rival. I play mostly large worlds; advice below.

I recommend:


1. Strategy

(0) Play the Purple civilisation with 6 other AIs.

(1) Build Warriors (to explore and maintain order) and settlers only until you have about 8 or 9 cities while researching Monarchy and Trade. Place your cities close together. Some overlap is OK.

(2) Switch to Monarchy at Oedo year; and build caravans to assemble Hanging Gardens.

(3) Build temples; more settlers and research republic and mysticism.

(4) Switch to Republic and set luxuries to 80% to grow by WLTK for a few turns only.

(5) Build caravans and other wonders (Pyramids is good; if AI has not already built it). Build Marco Polo's Embassy and trade technology (but do not give Monarchy or Republic) to an AI with many cities (or an AI which has met an AI with many cities) rivalling your own civilisation. Let them languish in corrupt despotism. Give or trade away construction to AIs so they will waste time building fortresses. Considergiving them Mysticism so they can build Oracle just before you research Thelogy and obsioleting the Oracle. If you do not ask AIs for cash; you won't get it. Do not agree to go to war; unless bribe very good and enemy very far away.

(6) You need either Michelangelo's or J S Bach's Cathedral.

2. Wonders

You will have to prioritize betwen Wonders as at deity you won't get them all. If on very large continent; you may omit LightHouse and Magellan's; but should consider Great Wall and War Academy; but if on island the reverse applies. In general; you should look for civilization wide rather than single city wonders
(Collossus, Copernicus Observatory, King Richard's Crusade, Isaac Newton's College, Shakespeare's Theatre). However the Collossus can be worthwhile if built in trade rich city; with library and marketplace to follow and if you set up trading caravans to it.
If you get Marco Polo's embassy; do not bother with Great Library and vice versa.


3. Improvements

Temples are essential as first improvement.
Do not build many barracks. One city barracks
can equip between 6 to 10 cuities with veterans.
Do not build granaries.
Do not build city walls unless city under attack. Push your warriors out to form frontiers keeping AIs well away.

Libraries and Marketplaces as second round of improvemrent building.

The third round may be Harbours for Coastal Cities with inhospitable land squares around. Aqueducts for rapidly growing cities; but courthouses if far away and suffering from corruption.

4. Use of Settlers

Firstly build New Cities.
Secondly build Roads
Thirdly Irrigate or mine coal squares.
Fourthly mine hills; never mine mountains.
Fifthly covert jungle or swamp.

5. Tribes (huts)

Hit as many as possible.

6. Wars

Remember that while you are wiping out one tribe;
the other five are growing. Avoid unnecessary wars
until you are very strong and run out of room for cities.


If you are still well behind by 1,000 BC; then start again.
 
Originally posted by EdwardTking
2. Wonders

You will have to prioritize betwen Wonders as at deity you won't get them all. If on very large continent; you may omit LightHouse and Magellan's; but should consider Great Wall and War Academy; but if on island the reverse applies. In general; you should look for civilization wide rather than single city wonders
(Collossus, Copernicus Observatory, King Richard's Crusade, Isaac Newton's College, Shakespeare's Theatre). However the Collossus can be worthwhile if built in trade rich city; with library and marketplace to follow and if you set up trading caravans to it.
If you get Marco Polo's embassy; do not bother with Great Library and vice versa.

I have to disagree with this section a bit. Granted the very best wonders are Civ-wide (e.g., Michaelangelo, Hoover), but the one-city wonders should not be dismissed - in fact many people build strategy around them. If you can set up a super science city (several trade specials, maybe some rivers, Colossus, Copernicus, Newton in same city with science improvments) your one city will likely out-research the other entire civs. This one city can be the engine for dominating the world, even at deity.

Also, I would avoid the Great Library altogether whether you get MP or not. Every tech you acquire makes your research cost higher. While you occasionally might get a tech you want from it, the GL often gives you techs that you are not interested in at the time, thus only serving to slow down your research toward the things you deem important. You'll be going for a tech to get a key wonder and next thing you know, you look at the science adviser and it says something like "advances every 40 turns". If you wait until 2 other civs get the key techs, it's probably too late. Plus, if you get the SSC up and running, you will be getting just about every tech first anyway.
 
It's not true that you need at least double number of cities than your AI's opponents. With half you can win on deity. How? Here are some advises:

1. For each city you found, do the following: first one defensive unit (defends the city and allows you to grow to size two without civil disorder), then one settler to found a new city. After that you can go for another defensive unit or a temple. The objective is to grow without civil disorder. All that you need in your cities in the early stages of the game is two defensive units, a good attacking unit, a temple and very important: city walls. With city walls it's impossible for the AI to take a city.

2. With three units and a temple you can grow till size 6 without problems. Go for monarchy soon so you can get food by irrigating. As soon as you get monarchy build a settler to do so for every city. Also build roads and mines, but only in the squares that you are already using or you envisage to use soon.

3. Build your cities in a place where you have some hills (to produce faster with mines) and grasslands to get food.

4. For the wonders, Michelagel Chapel is very important, make sure you get it. Build some caravans to help the construction. Colossus is fine to increase research. You must build it in a city that have trade. Also Copernicus and Newton if you can in the same city.

If you do that you will have a good start that is very important. If you go to war with any other civil, don't rush. He will attack you and loose a lot of units in your cities as you will have city walls. Build some barracks and the best units you can. Don't risk any unit, the key in war is not loose units. When you have an army good enough attack in their weakest point.

I think this is enough for now. The rest is improving your cities with aqueducts when they are size 7, some marketplaces in cities with good comerce etc ...
 
Sorge is right, matching the AI in growth is unnecessary and sometimes self-defeating. It is possible to conquer the entire world or build a spaceship with only one city, so certainly with 10 to 15. The key is to have each of these cities fulfulling their potential.

I would suggest building only 10-15 cities during the early game, and then use your settlers to modify their land to it's greatest extent (assuming pre-railroad: mining on hills and mountains, road & irrigation on fields, and remember to plant forests if a city is low on production).

While improving this land, build improvements in these cities and trade routes. Once each city has every square modernized, you are in a position to expand. This way you have a home country with improvements and a good land infrastructure which will be packed with trade and scientific advancement, instead of a vast array of small outposts.

Pool all of your engineers into one area on the frontier (preferably have your homeland cities supporting one settler each). Speedily use their combined work to fully improve the 20 squares around a square where you wish to build a new city (colony). Then build a settler in the nearest modernized city, use it to build a city in the chosen square, and move on to colonize further, one by one.

Of course, you can create variations like doing two colonizing pools or having two settlers supported per city, or whatever else. The point is, you must have some core of cities which will be industrial and scientific powerhouses and a periphery of small, developing outposts, rather than having everything be a sad third world. The benefits from the city improvements as well as the higher trade from roads & high population as will get you ahead in science and give you more cash and luxury revenues, which will in turn allow even more investment in science.

Also, having a good land and improvement infrastructure will allow you to go into republic earlier, which is important. With this excellent homeland and the republic's advantages, you can get ahead in science, further improve your cities, colonize if you like, and in the end conquer however you like. That's the civilization building phase, and to find out some good conquest ideas I recommend looking at the ideas on this thread:

http://forums.civfanatics.com/showthread.php?s=&threadid=18057

My strategy there is relevant to this because it can be done even if you are far behind the AI in growth and/or science.

Finally, the issue about whole-civilization wonders. Jorge is also right that one-city wonders can be very important. The way I build wonders starts with picking my most productive city, improving all the land around it, and filling it with each improvement. Be sure to have at one point or another Shakespeare's Theatre and King Richard's Crusade in it, although you should of course build other ones there during the time before you've discovered Medicine and Engineering. This will make it the most productive city in the world until Industrialization, and the Theatre will make it dependable. Therefore, it should be able to build almost every single useful wonder before anyone else. Atop that, it also makes better use of the wonders. For instance, it's better to have the Colossus and Issac's College in one city than in two separate ones, because the extra trade of the former gets more science from the latter. Ideally, you would have The Colossus, Shakespeare's Theatre, King Richard's Crusade, Copernicus' Observatory, and Issac's College all in the same city. If you do it right, you can get every pre-industrial wonder under your belt, or at least every one worthwhile.
 
Here is a one-city-challenge .sav with such a city:
 
Back
Top Bottom