Time to bump it up
Strategic Thinking (how did I discover them), part III
In this part I'll talk about how I wrote those strategies for scenarios. A scenario usually depicts a war, so I needed to think differently -- a wartime economy is different from a peacetime one.
Objective: win the scenario, i.e., capture certain cities in the shortest time.
Breakdown of the objective:
a) for a target city, move enough troops to the vicinity of the city within the shortest time
b) win necessary battles to capture (or destroy) the city in the shortest time after arriving
c) protect my own cities, including the newly captured ones
d) I don't need to get every city in the shortest time; only need to minimize the time for the last target city I am going to capture.
In certain scenarios I actually planned capturing the farthest city, and moved down the list to the second farthest one, and so on, until the closest one. But this is a job for a mathematician, not a player (well, it is so complicated that sometimes things will go out of control and even Einstein will have a problem). So I had to find another way to attack the problem.
Instead of how to reach the objective, let's talk about how to enhance my ABILITIES to reach the objective. Of course the higher the abilities are, the better the chance to reach (or at least close to) the objective. A breakdown of the abilities:
a) Logistic: ability of moving to the front quickly
b) Assault: ability of scoring more winning battles
c) Defense: ability of thwarting attacks
Note that I used "assault" instead of "attack" for the second ability. Assault includes attack (the ability to knock out a component) and momentum (the ability to attack continuously). A higher momentum will let me score more victories in each turn.
Logistic consideration:
If possible, build some transports and engineers for moving troops quickly. In rare cases also build carriers and subs for transporting. If there are more than one type of transports, pick the ones with the most transport ability, which is defined as the product of "carry" and "move" (move needs to be adjusted by wonders and techs). The number of transports and engineers need to be just enough to keep up with military's advance, i.e., at the end or each turn, I neither have any military troops unnecessarily using their own moving points for traveling, nor have any spare engineers/transports who cannot help the military to advance faster. In an ideal scenario, all military units' moving points are used for fighting, not moving around.
Only possessing abilities is not enough; cleverly using them is also very important. Here are several tricks:
a) ship chain. This is a well known trick so I won't discuss it here
b) teleport troops. If I have alliance with a civ, then resolving the alliance will automatically move my troops located closer to that civ to the nearest cities of mine
c) move-and-work. Suppose I want to change a city square from grass to forest for better defense, and I have two settlers 6 squares away from the city. The two places are connected by road. I need two turns to move the settlers to the city then 5 more turns for transferring the terrain, that makes 7 turns, right? Not really. I can move two squares then use the last 1/3 move to do mining for 3 turns, then get the job done using another two turns. 5 turns total instead of 7
d) temp city. Instead of using 3 or more engineers for RR a square, put down a temp city. The down side is that an engineer is lost permanently, but sometimes I have enough of them being built and will be available the next turn. Thus, for covering a longer distance in the current turn, building temp cities every other square would give me an edge.
Assault consideration:
What is the most important factor to win a war? It will be nothing other than the number of winning battles I can throw at the enemies. So I need to do all I can to increase this number. Again there are several tricks (both for increasing the ability and for using it):
a) select the right unit to build. The bulk of my military should be composed of a type of units which is effective (can win a battle most of the time) and have the highest movement points; only build a few heavy but slow units for knocking out really tough targets. Assume I have two units which are equally effective in attack but one has move 2 while the other has move 3. The first one will most likely only be able to attack once per turn, while the second one can attack twice, sometimes 3 times per turn. Hence the second unit is in effect equal to two of the first ones
b) If I need to research several techs in order to build my effective attacking units, I'll not go all out for offence, but put my effort on scientific research. I'll hire a lot of scientists, and actively deliver commodity caravans.
c) always rush build. Remember I said that it is a war economy? For each attacking unit which is built one turn earlier, I can score one or two more victories. Those victories will convert to occupied cities and I can gain benefit from that, which will be more than compensate to the money spent for rush buying. Hence, I never hesitate to sell city improvements for the cash to rush buy -- kill the hen to get the egg, if that's what I need to do
d) don’t' waste the chance to recover. When using a move 3 unit to attack, always pulse after attacked once and use another unit to continue the attack. Then, when I have emptied a city, use the first unit to occupy it to recover, and that unit will be more likely to survive the next battle, and perhaps have the chance to score the 3rd time in the same turn
e) camels may be helpful. Because the importance of rush buying units earlier, I always consider establishing a profitable trade route. This is a complicated problem so I'm not going to talk about it here. Hopefully in the future I'll have a chance to revisit it
f) diplos may be helpful as well. If most of the enemy cities are of sizes less than 4, then it is better to bribe them. After eliminating its capital, a size 3 city will cost merely 270 gold (assume using diplos and enemy civ has 0 gold in treasury), even if it has a courthouse. The city will probably have 3 city improvements that I can sell, plus some free units. That should be a good deal
Defense consideration:
AI is very stupid so usually I don't need to worry about defense. The best defense is to block enemy advance. If they cannot reach you , they cannot attack you.
Hence, the strategies of winning a scenario will have the following parts:
a) plan before move. The first turn may provide some opportunities, like, an empty enemy city to catch with a hefty sum of money plundered. Grab them and I'll be in a much better shape for the rest of the scenario. Send out empty ships early and form ship chains so later on troops can be moved to target cities easily. If act too late then I may have to spend several turns just for capturing the last remote city
b) identify what military units to build, and if the tech is not available, how to research it ASAP (hire all citizens as scientists will be considered; if cities are small then using settlers to add to cities to size 5 may also be considered)
c) decide what city improvements to sell and in what sequence, to raise money
d) rush build military troops (and barracks, if necessary) in one turn if possible
e) find a possible beneficial trade route and maximize the trade bonus by building up the source city (when caravan rehoming is allowed)
f) weigh the benefit of using diplos against using military for attacking the enemy
g) building engineers and transports (a lot of them) to increase mobility
h) take care of counter-attacks by blocking enemy advances (pillage road or use bombers)
i) last but not least, consider talking to other civs and try to gain advantage.
