The AI is missing something

Nope, its terrible at building up infrastructure. Commonly builds troops out of villages and rarely builds the right buildings right away in villages (Cleric, Lumbermill, bell tower, town hall).

As above, AI likes to build troops in villages and other places it shouldn't. Making the troops weaker and stunting the growth of villages.

Frogboy(AI GUY) has said he has had a hard time with this issue. I wouldn't mind the AI getting some under the table help with this on harder difficulties, better than units with a bazillion traits.

It's almost like the AI needs two pools of mana. The first pool is dedicated for 'common' spells. The second pool of mana is for 'larger' spells. Both pools grow, and both pools grow at different rates, depending on how far along you are in a game; a 'weighting' if you will. Throw in some exception handlers in so that the AI can choose to occasionally 'borrow' from one pool to supply the other if it felt it was needed, and then presto...AI can cast small and big spells.

As to the coding of such an idea....I have no idea...I'm not a coder.

@Frogboy:

How I manage my mana -

1. I prioritize mana accumulation (ie. meditation), and aside from filling essence slots, cast as few spells as possible (prime directive).

  A. I focus on retaining enough mana to cast my most powerful spells.

    i. I cast class applicable unit enchantments (ie. evade on low defense melee units), only when necessary; and unless a string of battles is planned, ensure that they are dispelled as soon as possible.

  B. I like summon and strategic spells.

  C. The above rules allow me to use tactical spells whenever they are most effective (per prime directive), without really having to consider mana expenditure.

2. I cast my most powerful spells when they are most effective, ensuring mana reserves are retained for foreseeable battles.

  A. I prioritize different spells as different factions.

Hopefully this helps direct the AI.

I am totally not against this idea, but it moves in an important way from the fair AI approach that Stardock uses, in the sense it starts to simulate how a player plays instead of playing the same game, even with bonuses. I know many strategy games do this. I am pretty sure MoM did it, it didn't actually run an economy or have a build queue based on production capacity, it just added units at intervals and then sent them out. I remember casting some of the negative city spells against the AI and thinking, hmm, that doesn't do anything does it. (This was way back in the release version, it may not be the case in the game today)

I don't think it would be hard to do at all, but it would be a big step as far as AI design for them.

Erm this is describing what many of us have asked for which is a tactical mana pool as well as a strategic mana pool.  One can not be used for the other so the AI always has mana for combat, maybe not much if they had a big battle but they do have mana.  

I'm not sure how Stardock have coded it but having an internal concept of (at least) two pools of mana is probably how I would do it anyway, without making the AI break any of the game rules. One pool to be for used whenever the AI wanted too, the other pool would only be used for critical purposes such as close fights involving the sovereign, close fights defending a city, close fights attacking a city or strategic spells against the AI's biggest enemy of the time.

I think you could also cheat a bit by giving the AI extra mana at higher difficulty levels (on a sliding scale based on difficulty, not just a bonus 50,000 or whatever) without breaking the player's sense of immersion. Certainly less immersion breaking than running into enemy troops which are blatantly breaking the rules with bonus traits, hp, etc.

I think I may have misunderstood with the mana pools. If you just mean mana pools as an internal mana budgeting scheme, that's probably a good idea.

There is a big difference to me between an AI with bonuses and an AI that doesn't play the same game. Here is one economical (in development terms) way to make an AI. First, forget about managing the economy. All you are interested in is the appearance to the player, so just spawn units at the rate and quality that seems to match the difficulty level. You put buildings in the cities along a timeline consistent with playerlike development, so when the player takes the cities it looks like a city that was doing stuff. When the AI declares war, which it does at a pace that the developer thinks is consistent with fun and difficulty, the AI sends stacks directly toward your cities. No scouting or anything, because it is actually a tremendous amount of work to keep and work with a belief state in the AI consistent with the knowledge a player would have in the same situation. The AI expands at predetermined rates for the appearance of human-like play, but there is no resource consideration going on.

I think you get the idea. That kind of AI would save an incredible amount of devolopment, and give a decent illusion of a human-like opponent if done for that effect. And you can make it as challenging as you want it, not hard at all. But that is also not what makes writing game AI for a game this complex one of the most intellectually challenging activities that one could engage in. The AI I described isn't playing the game at all. Giving bonuses alone doesn't produce the situation above, the AI must still utilize them through playing the game.

Playing the game well for a human is extremely hard, making an algorithm of good play that uses the same rules as the human and has belief states consistent with human knowledge in the same position is a task at the edge of humanity's knowledge. I mean trying to make something like a Deep LH. I wrote a thread about that before, but if people took up LH as seriously as say chess, I don't think an AI could be developed that could come close to competing with humans, based on the relative complexities of chess or go versus a game like LH. I don't know quite how the numbers work out, but there may be as many starting board positions (with monster placements, wildlands, map varieties, different starting locations for the factions, ect) with the default sovs in all the combinations of starting variables than there are possible legal board positions in chess. That is the tip of the iceberg in complexity analysis, which I did in more detail here and in the replies https://forums.elementalgame.com/438446/page/1/#3300617 .

"In go, for instance, you can only do one thing in a move, play a stone on the board against a single opponent. But in FE there are way more complexities: multiple opponents, it is possible to make dozens of moves within a turn, far more complex rules, the stochastic elements (the gameboard, placement, and events are random), enormous variety of game "pieces" (spells, units, champions, monsters, resources ect), and the fact the number of things that can happen and the number of things you can do in a turn increases as the game goes on." 

In other words, you can never use minimax on LH.

Yes, part of the reason why i wrote this post is that on Harder and above i think the AI needs Mana producing buffs and maybe a reduction in casting costs on spells. The game can be very magic intensive at points, so the more Mana the AI has at higher difficulty levels the better in terms of making the game a challenge.

I agree entirely. As soon as I see the AI doing things that I just flat out could never do then my interest in a game is dented. Unless of course it is a game or scenario where that is part of the point.

Last time I played FE/LH there was still a bit of that happening, such as settling squares that I can't. The one that really used to annoy me the most was when the AI would settle beside a dragon (or similar) and get away with it... particularly if the settle spot was one I was planning on settling myself but I had to wait until I could kill the dragon while the AI had no such difficulties! Fortunately it sounds like that happens less now.