I agree, before a nation goes to improve its military or magic, first it must improve its basic needs (a.k.a civilization tree), and i think the tech tree is already good in term of its logic.
But as said above, there are some techs in each of those tree which is useless. As said above, right now you just need first few techs of civilization tree to get the basic need to improve your nation, then you can focus on combat.
It is the same with magic tree, you just need first few techs of magic tree, mainly for unlocking spellbooks or shards, then you can leave the remaining unresearched techs in your magic tree. The magic equipment looks appealing, but it is no match if compared with combat tree equipments and army increasing and etc etc. Arcane point building looks useful, but you can substitute it with having many arcane labs.
The point is, most if not all techs in combat tree are useful, while only half or maybe less techs in civilization and magic tree that are useful, you just need civilization and magic tech tree for unlocking few things then you can leave them.
Right now, i don't have any solution to the civilization tree, but i have a suggestion to the magic tree. First, i don't like the idea of increasing the power of spells or lowering their mana cost, because it just make a combat tree focused nation and magic tree focused nation equal or similar in term of magic prowess.
I think magic tree should have techs that can do the following:
1) Unlock a method that increase mana/turn or decrease spells mana cost -> with this, the magic focused nation can cast more early game spells than combat focused nation, hopefully this will equalize their battle capabilities in early game, reliance on spells vs reliance on advanced war equipment. Do note that those that rely on spells still need at least a small army. The more you advance on this tech line, the greater the bonus. But there must be limit to the bonus to make it balanced.
2) Unlock more advanced version of the spellbook tree that grants specific bonuses to the spells and the shards -> some if not all higher level spells need a tech to be researched before you are able to use it at normal efficiency (100% eff.), combat focused nation still can use the same spells, but at lower efficiency (maybe starts at 70%), starting efficiency can be different for every spells (depends on spell tier) . The more you advance in this tech line it can grant bonus to spells efficiency, and maybe grant more abilities to spells (so it's not just improve efficiency), and even grant shards special effect (i haven't thought about it). If possible this bonus must not stack with shards bonus, the more shards you have the efficiency bonus is the same, and higher level spells and area of effect spells have lower bonus than lower tier and single target spell. Again, there must be limit to the bonus to make it balanced.
3) Unlock a way of shard converting -> this present a dillema, as it can destroy the randomness of the map, but maybe the conversion is not 100%. Example is: there are several factor in shard conversion. Water to fire will cost more than air to fire. The cost must be in mana, because if it is not in mana, once you get the desired shard you can immediately unleash what you want with that specific element, so if it cost mana, it can give your opponent enough time to prevent you from casting your desired spells. And shard conversion is not 100%, air shard if converted to fire shard just count as 1/4 fire shard, maybe there must be a way to reverse the conversion but i think it should cost double or more mana and time to do it, so a player must think very carefully before he/she attempt to do shard conversion.
These are just my wild idea though, point 2 and 3 may not be possible to be included in the released version, especially point 3 due complextity involved in implementing it.
And i realize that high tier spell are a bit overpowered if those ideas are implemented, that's why the high tier spell must be toned down. Which made me come up with another wild ideas.
I think few late game spells must be much weaker, and to compensate for it, the mana requirement is also toned down. Those particular spells that i want to talk about are the super spells like volcano, earthquake and their kind, imho they are too powerful, honestly i hate that kind of spells, to me it destroy the fun of the game.
The way i see it, those super spells are the equivalent of nukes in Civilization games, but nukes are limited by resources, and it takes very long time to make one, yet you can spam those super spells as many as you like in just 1 turn, as long as you have sufficient mana for it and of course the shards.
So here is idea for the late game spells:
I think those spells should be weakened, and limited in its use/turn by the number of shards you have. Example: I use volcano spell as example, if you have only 1 fire shard, you can only use volcano once every 8 turn. And perhaps the volcano shouldn't be able to instantly destroy city, instead of that, it just summon a volcano for 5 turns that gradually do the following: decrease population by x value, destroy 10 random improvement, negate training queue, and that city can't produce anything (be it train units or build something), after 5 turns, after that the city can normally produce anything, but 10 of its improvement are destroyed (randomly), and its population decreased by x value, city level can downgrade if the population is sufficiently decreased. This is just an example, but i think it is fair enough and logical enough if you ever know how a real volcano or earthquake or tsunami cause damage to a city.
