You said it. Some of these ideas ARE pretty crazy. Disregarding the insane amount of coding involved in some of them (not that I'm a programmer), they would turn the balance of FFH II on its proverbial ear. Here's what I see as the problems:
Amurites: They already have plenty of power and flavor if you approach them correctly. Making every unit a caster sounds like a great idea until you're faced with an army of hunters carrying an equally large army of hawks all capable of summoning big, fat elementals to rain destruction down upon your capital. Sure, they'd have to leave a few to protect the stacks, but this suggestion is immensely overpowered. Let's not even start with the potential horrors of the fireballs or meteors such units could eventually unleash. "Ouch! I'm on fire! Make stop the burnings!"
Belseraphs: The mime ability seems benign enough at first, but such a thing would be quite overpowered and horrific in a game with such potential for very powerful world units and such. One could use one t4 mime unit under sufficient escort to wipe the world clean of heroes and the like! Think of what happens when mimes defeat Acheron or worse yet, an even more powerful late-game unit. It's a whole new avatar of wrath or Hyborem or Basium, sitting inside a city, waiting for your army to come make him very, very weak by throwing themselves at him. Well, perhaps it's not so overpowered after all. Perhaps it's simply sloppy and useless.
Bannor: If you're playing the Bannor and your capital has somehow been attacked to the extent that it is no longer defended, you haven't been playing very well. Having the palace continually eat population to spawn warriors strikes me as a particularly Calabim thing to do. There are already plenty of ways to hurry-build defenders with civics a Bannor player under Crusade would find most appealing. There is no need to implement such an idea, even if it wouldn't be the coding nightmare I expect.
Calabim: Losing xp over time due to hunger doesn't make much sense. Losing attack/defense strength over time to hunger would be more appropriate, but that would require coding things so that a vampiric unit must win a battle every X turns or suffer Y% damage. This is one of the better concepts you've presented, but is still a bit more complexity that the game doesn't actually need to be playable and balanced.
Clan of Embers: This isn't such a bad one, either. It might not even be that hard to code. I wouldn't know about that. The only issue I see is that perhaps with all these higher-tiered orcish units spawning inside one's own borders, what happens if you'd like to employ units with hidden nationality to weaken your rivals? They would be doomed from the start, or at best have an immense pool of xp from slaying countless orcish crossbowmen or whatever. Thus it is unbalanced, one way or the other. Try a Terra map if you like mid- to late-game barbarians. You'll arrive on the New World to hundreds of orcish crossbowmen or their equivalent.
Doviello: Regardless of the fact that Charadon may already receive your proposed barbarian buff, this one seems to be based on good intentions. Tundra tiles and certain civs' affinities for them should be addressed in some way, though I'll wait for Ice before I go too far down that road. One population working two one-food tiles would be about as effective as an undeveloped grasslands city, though it would only ever reach half the population of such a city. The best way to build on tundra is to find some deer near a river on a coast where one can farm the tundras with access to fresh water, camp the deer for extra food, build a lighthouse for the coastal food and gold, and hope for forest spread to lumbermill. The basic problem of civs' affinities for tundra does deserve some attention, which I suspect it may be given in Ice.
Elohim: The Great Wall concept is a bit much for a civ who already has the defensive trait. Placement of one's troops is paramount to surviving attack. With the new fort mechanics, I wonder how effective a mounted rush can truly be against any well-played civ, particularly a defensive one.
Grigori: There is a reason that adventurers are scarce in the late game. No other civilization has access to such overwhelming numbers of heroes, and rightly so. The Grigori have this to make up for the fact that they can neither adopt nor found religions, and it is balanced enough as it stands. With the new world spell, Cassiel's people will have even more of these very powerful units which can be upgraded. Adding this new specialist only serves to further complicate an already challenging-to-play civilization.
Hippus: Having more horses doesn't seem to justify easier withdrawal. Each soldier can still only ride one horse, after all. There is already ample reason to have multiple sources of horse as the Hippus: simple trade. Horses are a strategic resource which can be traded for mana, iron, and other very useful goods. Adding an affinity makes the already high withdrawal rate of Hippus units even more open to abuse. "My city was attacked by 20 horsemen. Only two of them died."
Infernals: Think of using this against the poor Khazad. Suddenly an entire stretch of continent must be sailed 'round to access. No more need be said.
Ljosalfar: I agree with you on this one. Being able to grow forest on existing Elven improvements makes a lot of sense concerning both flavor and balance. Why should the elves have to tear down unforested improvements just to make way for a bloom spell? For that matter, it should apply to Svartalfar as well. Though a forested city still should not provide the city tile itself with any appreciable bonus beyond what it has already as a city tile. Let's not see a city built on deer in an ancient forest granting 4+ food or any such silliness. Might be difficult to code the game around this, though...
Luchuirp: Storing golems seems a phenomenally bad idea. Imagine your dismay on surrounding a Luchuirp city with two or three defenders only to see dozens of golems spring up and trounce your army with the aid of a few repair spells. The game does not need this new complexity, and the "hidden" nature of such an idea flies in the face of the Svartalfar/Sidar theme of Shadow mana. Golems are meant to be used up and thrown away, not hoarded like Acheron's wealth.
Malakim: While I sympathize that desert affinity is quite possibly as crippling as tundra affinity, one should not forget the simple Scorch spell. This one tiny spell makes it possible to plant a stack of units in enemy territory and give them a virtual instant fortress with the ability to summon sand lions from within. How would one "build" an oasis, anyway? that just doesn't make very much sense to me. Besides, anyone can build a water node for those desert squares that simply must be worked. Leave a buffer of desert squares containing forts to discourage incursions. That works well enough.
Mercurian: Not rightly sure what you are hoping to accomplish with this. Basium is plenty powerful enough already. Having him sanctify nearby tiles, while certainly flavorful, serves little real purpose in-game other than to further annoy the player of Hyborem (of which he does a fine job as it stands). The nature of Basium's summoning already provides an advantage over Hyborem if the builder of the Mercurian gate is willing to sacrifice a sizable city to the Mercurian. This fellow needs no additional help, far as I can see.
At any rate, it is not my intention to insult or anger you, it-ogo, but merely my thoughts on your ideas. As noted, some are good ones. Some are bad. Brainstorming like this makes FFH a better mod for all of us. I know I've had my share of bad ideas, so no hard feelings. Happy Shadowing!
Amurites: They already have plenty of power and flavor if you approach them correctly. Making every unit a caster sounds like a great idea until you're faced with an army of hunters carrying an equally large army of hawks all capable of summoning big, fat elementals to rain destruction down upon your capital. Sure, they'd have to leave a few to protect the stacks, but this suggestion is immensely overpowered. Let's not even start with the potential horrors of the fireballs or meteors such units could eventually unleash. "Ouch! I'm on fire! Make stop the burnings!"
Belseraphs: The mime ability seems benign enough at first, but such a thing would be quite overpowered and horrific in a game with such potential for very powerful world units and such. One could use one t4 mime unit under sufficient escort to wipe the world clean of heroes and the like! Think of what happens when mimes defeat Acheron or worse yet, an even more powerful late-game unit. It's a whole new avatar of wrath or Hyborem or Basium, sitting inside a city, waiting for your army to come make him very, very weak by throwing themselves at him. Well, perhaps it's not so overpowered after all. Perhaps it's simply sloppy and useless.
Bannor: If you're playing the Bannor and your capital has somehow been attacked to the extent that it is no longer defended, you haven't been playing very well. Having the palace continually eat population to spawn warriors strikes me as a particularly Calabim thing to do. There are already plenty of ways to hurry-build defenders with civics a Bannor player under Crusade would find most appealing. There is no need to implement such an idea, even if it wouldn't be the coding nightmare I expect.
Calabim: Losing xp over time due to hunger doesn't make much sense. Losing attack/defense strength over time to hunger would be more appropriate, but that would require coding things so that a vampiric unit must win a battle every X turns or suffer Y% damage. This is one of the better concepts you've presented, but is still a bit more complexity that the game doesn't actually need to be playable and balanced.
Clan of Embers: This isn't such a bad one, either. It might not even be that hard to code. I wouldn't know about that. The only issue I see is that perhaps with all these higher-tiered orcish units spawning inside one's own borders, what happens if you'd like to employ units with hidden nationality to weaken your rivals? They would be doomed from the start, or at best have an immense pool of xp from slaying countless orcish crossbowmen or whatever. Thus it is unbalanced, one way or the other. Try a Terra map if you like mid- to late-game barbarians. You'll arrive on the New World to hundreds of orcish crossbowmen or their equivalent.
Doviello: Regardless of the fact that Charadon may already receive your proposed barbarian buff, this one seems to be based on good intentions. Tundra tiles and certain civs' affinities for them should be addressed in some way, though I'll wait for Ice before I go too far down that road. One population working two one-food tiles would be about as effective as an undeveloped grasslands city, though it would only ever reach half the population of such a city. The best way to build on tundra is to find some deer near a river on a coast where one can farm the tundras with access to fresh water, camp the deer for extra food, build a lighthouse for the coastal food and gold, and hope for forest spread to lumbermill. The basic problem of civs' affinities for tundra does deserve some attention, which I suspect it may be given in Ice.
Elohim: The Great Wall concept is a bit much for a civ who already has the defensive trait. Placement of one's troops is paramount to surviving attack. With the new fort mechanics, I wonder how effective a mounted rush can truly be against any well-played civ, particularly a defensive one.
Grigori: There is a reason that adventurers are scarce in the late game. No other civilization has access to such overwhelming numbers of heroes, and rightly so. The Grigori have this to make up for the fact that they can neither adopt nor found religions, and it is balanced enough as it stands. With the new world spell, Cassiel's people will have even more of these very powerful units which can be upgraded. Adding this new specialist only serves to further complicate an already challenging-to-play civilization.
Hippus: Having more horses doesn't seem to justify easier withdrawal. Each soldier can still only ride one horse, after all. There is already ample reason to have multiple sources of horse as the Hippus: simple trade. Horses are a strategic resource which can be traded for mana, iron, and other very useful goods. Adding an affinity makes the already high withdrawal rate of Hippus units even more open to abuse. "My city was attacked by 20 horsemen. Only two of them died."
Infernals: Think of using this against the poor Khazad. Suddenly an entire stretch of continent must be sailed 'round to access. No more need be said.
Ljosalfar: I agree with you on this one. Being able to grow forest on existing Elven improvements makes a lot of sense concerning both flavor and balance. Why should the elves have to tear down unforested improvements just to make way for a bloom spell? For that matter, it should apply to Svartalfar as well. Though a forested city still should not provide the city tile itself with any appreciable bonus beyond what it has already as a city tile. Let's not see a city built on deer in an ancient forest granting 4+ food or any such silliness. Might be difficult to code the game around this, though...
Luchuirp: Storing golems seems a phenomenally bad idea. Imagine your dismay on surrounding a Luchuirp city with two or three defenders only to see dozens of golems spring up and trounce your army with the aid of a few repair spells. The game does not need this new complexity, and the "hidden" nature of such an idea flies in the face of the Svartalfar/Sidar theme of Shadow mana. Golems are meant to be used up and thrown away, not hoarded like Acheron's wealth.
Malakim: While I sympathize that desert affinity is quite possibly as crippling as tundra affinity, one should not forget the simple Scorch spell. This one tiny spell makes it possible to plant a stack of units in enemy territory and give them a virtual instant fortress with the ability to summon sand lions from within. How would one "build" an oasis, anyway? that just doesn't make very much sense to me. Besides, anyone can build a water node for those desert squares that simply must be worked. Leave a buffer of desert squares containing forts to discourage incursions. That works well enough.
Mercurian: Not rightly sure what you are hoping to accomplish with this. Basium is plenty powerful enough already. Having him sanctify nearby tiles, while certainly flavorful, serves little real purpose in-game other than to further annoy the player of Hyborem (of which he does a fine job as it stands). The nature of Basium's summoning already provides an advantage over Hyborem if the builder of the Mercurian gate is willing to sacrifice a sizable city to the Mercurian. This fellow needs no additional help, far as I can see.
At any rate, it is not my intention to insult or anger you, it-ogo, but merely my thoughts on your ideas. As noted, some are good ones. Some are bad. Brainstorming like this makes FFH a better mod for all of us. I know I've had my share of bad ideas, so no hard feelings. Happy Shadowing!
Of course, some of your ideas are insane (Amurites, Balseraphs, Infernals),
undead vampires that they ALWAYS feel a strong bloodthirst even if they drunk much and physically unable to drink more. And that make them crazy.
