AI not using its full alotted movement

AI seems to be saying " :help: :help: :help: me move!!!"

Hehe, Oaty posted saying that when the AI attacks weak rebel stacks, they retreat, and the AI doesn't follow up, thereby wasting armies on dumb chases. Well this is a very general problem for the AI. Unless engaged in a siege type situation, the AI only uses enough movement to attack once, after that, its army stack doesn't move, no matter how much alotted movement it has left :dizzy2: :embarassed: Well this generally really sucks, because this means that when faction A is twice as powerful militarily as its neighbour faction B, it will take it forever to conquer faction B because all that happens for years on end is armies poking each other out of the way, as the weak stacks of faction B retreat away from the strong stacks of faction A. Meanwhile if faction A were controlled using the full movement of stacks, and multiple attacks when possible, faction B would be conquered in a matter of a dozen turns or so. Bah, this really makes the AI very very very slow to use its power base to expand, and thereby makes it that much weaker an opponent, which in IMHO is a major bad thing.

In conclusion, patch please ~D

It seems the A.I. army has 3 possible goals from my experience. There is a get your army out of my land mode and do'nt pursue that army across a border. Annialate an army mode where it will pursuay that army across the map until it no longer exists. Numidians have it the worst here as I noticed they had chased some rebels from near the city of Lepsis Magna all the way to Siwa. In the other parts of the map rebels will always get put in a mess for someone else to deal with. And there is capture a city mode and if it runs into an anemy army it will attack it if the odds seem in there favour, unfortanately the army does not continue on to the city until next turn.

By and large the AI seems to do each factions turn in a "batch" mode when it really needs a sequential mode. It issues orders to all units, and they move. They do not perform follow ups with remaining movement points for the most part. They do not do staged moves: have an army board a boat, move the army across the pond, land them, combine with anybody nearby, by mercs, and perhaps engage or besiege all in the same turn (possible at times.) They do not send out an agent as a recon, then send an army out behind them to act on the recon all on the same turn. After a landing/invasion, the AI does not combine forces then beseige all on the same turn. The human gets to "steal a march."

The weakness here is that as humans we use are using our forces in series. We also more appropriately pick forces to perform multiple tasks/battles and STILL end up where we want them at the end of the turn. For instance, I will leave a settlement with a full force to attack a big stack, win, then send my foot units back, engage some scrap with my cav units/general, then move that force back to the settlement (the infantry could not complete the move in unison.) Or I'll buy mercs near the end of the movement points, and send the merc sub-army off to finish chasing down rebels or the like.

The AI think time is extremely short on the strategic map. There is a lot that could be added to the strategic AI without appreciably slowing it down.