First off, hello everyone.
I like the fact that you can design your own units. That is a very, very good mini-game for a turn based strategy game to include, in my opinion.
I like the thought of having units cost components rather than some unified, fungible resource like gold or credits. It was certainly cool in "Lords of the Realm 2"; it would be cool here as well.
I like the concept of having local amounts of resources/components rather than a universally accessible stockpile that all of your cities tap into. That is a rarely used mechanic that would help Elemental to distinguish itself from other turn-based strategy games.
However, I am concerned about the micro-management (MM) issues that could be introduced when you combine all three game mechanics.
Most games that include combat have a "build a unit" mini-game. In fact, the build-a-unit mini-game is often one of the central mini-games in strategy games. In Homeworld you needed enough resources, time and the requisite tech to build a unit. Then, you click a button, your mothership consumes the resources, spends time building and finally the unit pops out; neither the time nor the cost element changes in regard to circumstances. In Civilization you needed the requisite building, the requisite tech and a variable amount of time to build the unit. Units didn't cost resources unless you wanted to hurry along their production by sacrificing money or city population (enslaved to rush-build). Now let's talk about Elemental...
There are three questions I need to ask when I am going to decide where to build a unit:
1) How much will that unit cost at location X?
2) How long will it take that unit to be built at location X?
3) How useful will the unit be at location X?
The answer to 1 is, as far as we know, fixed. A "paladin" (as defined in the original post) costs one sword, one set of armor, one shield and one helmet. That cost does not vary from city to city... as far as we know. This could vary but no mechanics for that have been introduced.
The answer to 3 is game-context specific and is not of concern here; I mention it only for the sake of completeness.
This game is being deliberately designed so that the answer to 2 will vary from city to city depending on, among other things, the availability of components. It looks like there are three or more stages to creating a unit: collecting the raw materials (weapons, armor, mounts, etc) into one location, assembling those raw materials into a unit and letting that unit sit around in the queue improving its stats, a.k.a. training. What is required to get all the necessary components (10 swords, sets of armor, shields and helmets) to city X so I can build ten paladins there? We don't know but I want to call attention to this because it's ambitious.
What I am worried about here is the game becoming too MM intensive. If I have to track how long it'll take to get swords, armor, shields and helmets to my city then I am essentially playing four additional mini-games just to complete the "build a unit" mini-game for one particular unit. Do I first have to manage my iron mining and iron distribution to ensure I have adequate sword production to even consider training paladins? How automated will resource management be? How many steps do I really have to consider if I want to build 10 paladins? That's a number that can quickly become too large.
Or not. Turn based games have done all sorts of stuff to streamline; keeping the strategic depth that having addition steps creates while introducing game mechanics that lessen the amount of attention required by the player.
In short, I like the ideas presented by Brad and company, I just want to express my concerns about the challenges surrounding those design decisions.