I haven't been very (well, at all) active as of yet, but this topic interested me enough that I felt I'd weigh in. Mechanics and realism are subjects of interest to me. I may go on a bit long, so skip to the end for the conclusion if you just want my proposal.
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
The way I see it, there's 3 main objectives any successful armor system must achieve:
Functional diversity
Realism (I will not say verisimilitude, because it is a word that I cannot stand)
An acceptable level of balance between types
These are all tied in with each other, and unfortunately are somewhat opposed to each other.
Functional Diversity: This means having real differences between armor types, not merely cosmetic (although cosmetic differences are important, too!). In many systems (including D&D in large part) there is a single statistic that determines everything about a piece of armor, often known as Armor Class or something similar. Chain mail and Plate Mail aren't any different - chain mail is just worse.
Realism: This isn't the same thing as historical accuracy. Realism in this context can be thought of more as reasonableness or even just adherence to logic. An example of an obvious failure of realism is D&D 3rd edition, where a mace and a katana have the same effect on a suit of chain mail (actually, in D&D the katana would have more effect).
Acceptable Balance: This is the one that doesn't play nicely with the other two. The real world of course isn't balanced, and too much focus on balance can easily destroy diversity. But it doesn't have to...
I believe that the core issue with most armor systems is their basic approach - a single unified armor class statistic. Yes, it's easy to use, but that argument doesn't hold nearly so much weight in a computer game - the player only needs to understand that their armor is better against slashing weapons than against bludgeoning weapons without having to look it up on a table (the computer can do that, and it doesn't mind).
For example, let's examine a popular type of armor: the chainmail huabark. In most systems, chainmail occupies a spot in the mid-range of armor. But chainmail is notoriously unequal in its protection; nearly sovereign against slashing attacks, unhelpful against bludgeoning attacks and anywhere from unhelpful to actually harmful (depending on who you believe) against piercing attacks. You can't accurately reflect this relationship with a single number, however it might be structured.
As a side note, a system such as this also helps weapon type viability and diversity - worried that the mace isn't very good compared to the katana? Try them both on a properly armored target, and it's the katana that will only produce a few nicked links of metal. Such an interaction between weapons and armor introduces a myriad of tactical options, makes armor choice more than a matter of picking a single number and at the same time enhances weapon diversity.
Another important element of armor is fatigue and mobility. Although as is oft mentioned the medieval knight could indeed do cartwheels in his full plate, knights also suffocated to death on the field from fatigue and their own heat exhaustion (I'm not saying that should happen to players, although it would be hilarious on hardcore mode ).
One easy solution to this would be some sort of fatigue stat, which as it adds up makes the character less and less effective - so the knight in full plate armor might begin the battle untouchable and just as spry as the swashbuckler, but after five minutes of solid battle the knight could barely swing his sword while the swashbuckler would be just as quick as ever. Heavier armors would accelerate the accumulation of fatigue.
The final factor I think is important to armor balance is cost. This was the real limiting factor in the years of plate mail - nobody except for a nobleman could afford the stuff, it had to be custom-fitted and it took forever to make. The cost I think should be sharply additive for higher armor complexities; platemail shouldn't cost twice a suit of chain, it should cost ten times and take a month to make. Realistically, it should also be custom-fitted to a certain character and require an armorsmith to refit, although with a crafting system that would be a lot less of a restriction.
Another thing that seems sort-of on topic but I wasn't sure where to put is shields: Shields aren't useless for crying out loud - they're the single best piece of protection you can carry! The rules should reflect that. In D&D, carrying a shield earns you a well-deserved mocking.
Conclusion (Armor Balance): The most important component to making armor diverse, realistic and maintaining the viability of lighter armor types is to make armor react realistically to different attack forms. Factors such as fatigue and mobility are also important. In essence, if a real-world reason for preferring a different armor type isn't supported by the rules, they're probably doing something wrong. Realistic and reactive armor also increases weapon viability and diversity, while keeping certain weapon types from outshining others when they shouldn't.
As to the question of armor type advancement (Purified Steel Cuirass or Breastplate +1?) I would be inclined to say use both, but sparingly. In a world where magic is rare I think it's important to have the ability to differentiate between armor without needing it to be magical, which a D&D based system isn't really designed to do.
There are, however, two big problems I see with a material-based system:
Complete obsolescence of lower tiered materials when better materials become available
Possible disconnect between rules and reality
I'll start with the second, as it's quicker to explain:
Rules and Reality: What I mean by this is situations like are found in Dragon Age - the best weapons are Volcanic Aurum, and so your weapon (expressly crafted from the bones of a dragon) is somehow made of Volcanic Aurum to fit the tier system. It doesn't make any sense, and it's jarring.
Keeping All Materials Relevant: My proposal for keeping materials relevant is related to my main armor proposal; instead of a linear bonus, make the material have an effect on the properties of the armor. To use two common fantasy examples, Mithril armor could be lighter (Less fatiguing) and more responsive to speed enchantments while Adamantine could be harder (Breaks less easily? Needs less maintenance? More resistant to piercing attacks) and more difficult to enchant. That part of the idea is still in its infancy, but I think it has some real promise.
In addition to materials and magic, I think the skill of the craftsman who made the armor should be important - but not too important. The masterwork system of D&D doesn't really get the job done (too insignificant), but something along that idea, or maybe the fine weapons of Icewind Dale I.
Conclusion - Part II (Armor Materials & Style): I think materials should be the main differentiating factor between armors of the same type, followed by craftsman skill. Materials should not give linear bonuses, but instead offer particular advantages and affinity or resistance to certain types of enchantments. I do think that armor appearance and armor ability should be closely tied - I'm always annoyed when I find max'd heavy armor only to discover that it actually only covers half of my character's torso, or when the armor that looks like it offers the best protection actually sucks.
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
I hope that came out reasonably coherently . If I get time I might post a better summary or some follow-up ideas.
[EDIT] Ooh, I almost forgot!
What does 'Armor' mean in a game?: I strongly support a system where Armor and Defense (or whatever you want to call them) are not the same thing. Armor absorbs hits and prevents weapons from damaging a person (much like Damage Reduction) - Defense prevents hits from connecting in the first place (much as Armor Class operates in D&D). I think having a divide between these two also helps to make different fighting styles viable; yes, the warrior in a breastplate might not be as well armored as the one in full plate, but he relies on mobility and chose his armor to provide a strong protection against missiles.
As with many of my earlier points (and the whole concept, in fact), this also applies to weapons. Some weapons are more precise and easily controlled or threaten a larger area and therefore hit more easily, but that doesn't necessarily make them any good at armor penetration. The reverse is also true.
I don't usually comment on update, but that was very impressive. The beautiful terrain I was expecting, but the sunset was excellent and the water effects are gorgeous. And I really liked getting a look at some characters. Keep up the good work! (My head didn't explode, but I took precautions )
