As a teenager, my favorite game was Ages of Empire II.

You are the general. You start with a town center and a small group of villagers, ready for orders. Click a tree to harvest the wood, click an empty space to construct a building, and use buildings to upgrade your tech and build an army. Send that army at your nearest foe.

It was a real-time strategy game of the highest order. Unfortunately, it doesn’t map well to a mobile experience. You may be the general, but you have to issue “all” of the orders. This doesn’t scale when your portal to the battlefield is a 3.5" iPhone screen. Too bad I play games on an iPhone or iPad most of the time now.

Beyond scale, the tactical aspects of the game bugged me the most. It was frustrating to get in the middle of a three-pronged assault only to realize that half of your villagers were just standing around because you hadn’t told them which forest to attack next.

Microsoft attempted to solve this in the expansion. In Age of Empires II: The Conquerors, the game makes intelligent tactical decisions for you. For example, if a villager is building a lumber camp, they’ll automatically start chopping wood once that’s done, rather than just standing around.

Unfortunately, these improvements only affected the most tactical parts of the game. As the general, you still had to issue every command that was even remotely strategic.

My second favorite game as a teenager was The General (version 4.5e).

This game, like Age of Empires II, is a strategy game. Unlike AOE II, it is not real time and has no animations of any kind. Also, it strays further from the tactical side of things. Instead of soldiers, you command generals. Instead of villagers, you control land. Instead of buildings and technology, you manage a science budget.

I loved the high-level aspects of the game. If AOE II used a similar approach, the commands necessary to attack an enemy would be something like:

1. Build me an army of 1,000 recruits
2. Assign 500 of those recruits to a general
3. General, attack the province to the north of us!

Unfortunately, the attack itself was a pure computer simulation. The odds of winning were a factor of the army’s size and the general’s abilities. Nothing more. If you had a stupid general (or a small army), there was no way for you to impact the course of the battle.

As mentioned, there are two types of commands: tactical and strategic.

Tactical commands are less interesting. They feel like busywork. They are the how.

Strategic commands are more interesting. They require high-order thinking, and–to be effective–require others to interpret and carry out the implementation of the command. They are the intent.

My theory is that strategic commands are the key to a good strategy game. I want to issue high-level commands and let the computer figure out the tactical details. Of course, if the computer isn’t handling the tactics well, I want to be able to tweak the gameplay on a granular level too.

The result, I think, would look much like AOE II + The General.

You start out with a middling AI. Given a strategic command, like “build a city”, the AI will do a reasonable (if suboptimal job) of constructing fortifications while you attend to reconnaissance and army building. At any time, you can take control from the AI and build the city to your exact specifications.

Later in the game, you can train or purchase additional AIs, and assign them portions of your army. Starting with high-level directives, such as “attack orange team”, these generals will begin to set up and conduct an assault. As you observe events unfolding, you can direct certain details of the attack. For example, you might want to use certain attack formations, prescribe conditions for retreat, or construct siege fortifications around the enemy city.

In the same way that most strategy games allow you to choose a certain difficulty level, your hired AIs have their competency tied to their training cost, or experiences in the field. The thing they buy you is time. Time to focus on strategic problems. You are the five-star general. The AIs fill the lower ranks in the command hierarchy.

The net benefit is that you issue fewer commands. However, these commands require more thought and deliberation. Imagine a game where you could watch events unfolding and spend most of your time in deep thought, only intervening to shape the game to your long-term strategy.

By analogy, this is the way that chess grandmasters approach the game. The vast majority of their time is spent in deep thought, looking dozens of moves into the future. Only a fraction of their time is spent making actual moves.

To my way of thinking, this is a much more attractive way to play realtime strategy games (not to mention a more mobile-friendly approach). Personally, I despise strategy games where the top players are those with the fastest reaction times and maximum clicks per minute.

As far as I’m concerned, StarCraft is the speed chess of strategy games.

That was not a compliment.