Always War Theory

Tomoyo

Fate
Joined
Feb 21, 2004
Messages
9,698
Location
Boston, Mass
After playing a few AW games myself, I would like to share my thoughts, though this comes from playing vanilla civ.

AW stands for Always War, meaing you have to stay at war with everyone all the time. A "Relaxed" AW is when an AI has to come to the diplomacy screen before you have to declare war, but you have to declare before leaving the diplo screen.

A. Kill Zones

A kill zone an area where AI units come at you and you, well, kill them.

There are three types of kill zones, in my experience. The first, the "conventional" kill zone, is where you have one heavily fortified city or fortress, behind flat land. If there is land that is not flat in front of the fort/city, put a unit on it. When AI units step onto the flat land, attack the stacks with slow units, and the single units with fast units. Simple.

The second is one that I rarely see used, that I dubbed the "cheese grater" kill zone. This kill zone requires that the zone not be close to your core, because of its depth. Also, it requires a "bait city", a city left empty for the purpose of luring the AI, although it is somewhat exploitive. (It's an AW tradition though) The best time for a "cheese grater" kill zone to be used is if a choke point is five tiles wide, so the city placement is wcxxxcw, where w is water. A fortress is placed on the middle tile, so the maximum usage of the Zone of Control is used. Then a normal kill zone is operated slightly behind the front line.

The last is the "maze" kill zone. When a bunch of AI units come into your territory, through a "hall", like a funnel of mountains or hills, you shut one end of the funnel with an army, making the AI units go out the other side, only to find that next turn the amry is gone and another has gone to the other side! All while you bombard them to bits. It is very rare to see the terrain necessary for this.

B. The Opening

You can either be very cautious and build a barracks and troops immediately and not explore, but that usually ends in the crippling of your growth and research at the beginning, so I recommend the farmer's gambit.

How do you do a farmer's gambit in AW? Well, you can afford to leave cities empty until the AI's troops arrive, so when you meet an AI, build your walls and troops. Until then, you can build a granary, settle the food rich cities, and so on. Works best on higher levels, relative to the difficulty. My philosophy: if you get found early, you're dead anyway.

C. Other
a. On the higher levels, defense and artillery are more important than offense, but on lower levels, offense and artillery are the most important.
b. Don't skimp on infrastructure
c. Don't skimp on workers
d. cxxc city placement works well because you can shuffle one unit from one city to another in one turn.
e. If you are playing "Relaxed", then you should keep research high so that you will have nothing to trade until Map Making is around, so you can avoid war for some time.
f. I'll add more stuff when I think of it.

Some of this stuff was taught to me by other people...
 
We had an AW game a long time ago on Poly and it was hard. The map was made by alexman IIRC. It had us next to Rome and the Iroq on the other side of the landmass.

Iroq start on horses and Rome started on top of Iron, I think only three of us finished the game, and we all dropped down a level. Declaring on them upon contact was a harsh road to travel.

I was able to capture Rome after they build only one Legion, but had to deal with Mounted Warriors very soon. We had no ohter resources on this large land mass. No saltpetter and no oil.

I remember (it was more than a year ago, maybe even two) England had saltpetter. Anyway it was brutal.
 
I think it also depends on the landmass. I played in an AW succession game (LK61), and we had the bottom half of the continent to ourselves, and a nice little landbridge to pick off Iroquois Mounted Warriors in the jungle. Once we started packing galleons with our cav armies, the game was over quick.
 
Back
Top Bottom