Yeah, Heavenfall's been nice enough to let me use some of the models from his Expanded Factions mod. This should help fill out the Ether group and should work for the illusory units you mentioned.

I got your PM, but I'm not quite to the point to where I'm able to mess around with the custom unit stats. Soon, hopefully. 

*Updated: the mod now has 30+ brand new spells and 20ish rescaled vanilla spells*

Done: Testing. It's been a little painful to discover the ways in which the engine is currently limited, and I've had to adapt as best I could. While there are a few more spells I plan on adding, I think this is about as much as I could come up with, given the current xml functions.

Still to do: Finishing custom particle effects. Assigning new books to canon sovereigns.

Also, I'm still on the fence as to whether non-selected spellbooks should be available through research.

In my opinion non-selected spellbooks should be available via research but there should be something extra to do. Like some quest it unlocks with the end result being that you gain access to the spell book. Or perhaps a choice between the spell book and some other reward. A quest to re-discover old knowledge after researching where it may be found feels some how more right to me. Rather than your non magic researchers miraculously figuring out how to cast a new bunch of spells.

Yeah, if I did include them, I had planned on rearranging the magic tree a little bit to have some intermediate techs before each book tech, which would all be seperate. With a little bump to the tech exponent, I think that would keep the factions from ending up with them all. But, I like your way better, it's just going to take a little investigation on my part to figure out how all the tech/quest xml works.

The main problem that I see is this.

Now, most of the schools have a way to do all the major things... damage, slowing down enemy armies, etc. I've tried hard to make them all operate differently, and where they're absent, it was a deliberate choice to counterbalance a strength of the school. There are really not that many gamemodifier tags that are available for spells, so there is only so much variety that can be coded up. So, with an increased degree of overlap between each school, how many different damaging spells does anyone need? How many movement debuffs? Would the end-game just involve each sovereign using the best spells from each school? What does this do to immersion?

I'm not trying to argue against your point, just curious as to what you and others think.

*Also, don't pay too much attention to the damage calculations attached to the spells, I just started spreadsheeting them, and the progressions are kind of f'ed up, I know. Any thoughts on ideal late-game damage would especially be appreciated. Most of my playing is just mod-testing. I have never won a game of Elemental. Sad, I know. Right now, I have it worked out that Int/1 as a modifier should be the ceiling (except for Tsunami!) for the best direct-damage spells.*

Yeah I get what you mean, it's a bit of a pity there aren't different damage types in the game then at least you could choose the correct damage spell for the target. Unless you actually modded in a damage multiplier to the unit stats for each magic school. Then used the multiplier to reduce damage to the unit (and other effects) from within the spell defs. But it's allot of extra work when it only adds a little to the game and the AI probably won't take advantage of it.

In terms of how many different spells of each type. Generally I'd want a long range (lower powered), Short Range (High powered), Short and Long range AOE as well as high and low mana cost versions of the single target spells. So 6 at least  . Plus some just to have a different effect to see. I'd also want some low power versions for new channelers who haven't got the INT to do any decent casting yet, or if I wanted warrior casters. So you could have quite allot of overlap, especially if you include spell progression. Then there is always direct damage and defended damage, and with defended damage you can (I think) have differing amounts of accuracy.

You should also be able to put cooldowns on spells which would another reason to have multiple versions of them. Obviously not for the weaker spells but for more powerfull spells maybe you can only cast it once every 2 or 3 rounds. Want to cast it twice then eithe have two casters or two simmilar versions of the spell.

hmmn, another idea just came to mind for spells in general. Having a unit stat for the strength of a spell school for the unit that reduces with each cast in tactical combat. So if you over use one school in a long combat then you will be having less and less effect on the enemy units. Would be a reason for wanting multiple books, though it may just be more of an annoyance than anything. Or it could be used as a limiter to how many times a particularly powerful spell can be cast by one unit. You could then have spells that directly target these stats, or only have the stats so that spells can be used to target them to alter spell strength on a book by book basis.

The problem is that, no matter what system is used, eventually it boils down to doing damage to the target to kill it before it can kill you. The trick is to make it interesting somehow.

And finally a random spell idea: A spell that transfers the casting ability of one mage to another mage resulting in more powerfull spells. Sort of like ritual casting. Maybe it even unlocks spells that can't be used otherwise.

Yeah, I think that you're right about the ranges and areas of effect. As it is, the Fire school is the only one that does this adequately. There are quite a few continuous damage spells, but I could probably add an AoE to a couple more books as well as a couple short range, high powered spells.

I finished the rebalancing and all the particle effects today for all the books but Light and Shadow, so hopefully I can get the last few spells implemented and release v0.9 tomorrow.

Alright, version 0.9 has been posted. This is the "beta" version. I think that I got all of the bugs, but there may be some balance considerations to take care of in the late-game. I also plan on adding about 10ish more spells, but I've been having some problems getting them to work.

Feedback would be greatly appreciated!

*Update* Fixed a couple damage calculations. Sirocco and Heat Armor calculations were a bit off.

To do for v1.0:

-Balance spell power for the late-game.

-Split the Ether spellbook into Light and Shadow once I get my head wrapped around some of the aspects of the summoning system

-Add the Transcendence spellbook

-Try to figure out an interesting way in which non-selected spellbooks can be learned.

-Give some of the new particle effects a little bit more love.

-Implement an Affinity system with unit stats.

Updated for ease of installation and to remove some weirdness with updating. Now no core files need to be moved.

Thank god for that.  I'll give it a try now.  I always hate to move about core files.

Sweet. Let me know what you think.

*Updated*

Included optional file to slow arcane research down quite a bit. Now, getting level 5 spells will take some time. With the optional file, sovereigns must meet stat requirements for all spells, which have been lowered slightly from vanilla.