Way back when we started planning our buildings system, so very long ago, we had this grand idea that each faction would have a certain number of complexes dedicated to showing how that faction expanded and conquered areas. This would mean that each faction would have a custom government/governing system, which was tiered. So, for instance, the Romans could have built an "allied" government type which would be relatively quick to implement, and upgrade, granting various bonuses and penalties, and another which would be "romanization," essentially turning the province Roman, granting citizenship, with different bonuses, penalties, taking a lot longer to build, and so on. It would create vastly different recruitment options, first and foremost, and affect everything in the empire.
The goal, of course, was to grant huge diversity across our factions as well as to make gameplay more fun and interesting, challenging for the player, and more realistic in its depth, in a sense.
Unfortunately, several obstacles caused us to scrap this system. First and foremost was the building complex limit. Not only could we not have multiple complexes per faction, but it wasn't even realistic to limit the governments to one level only and have one complex per faction. Compounding this was our second fateful issue; the AI does not destroy buildings, and we have absolutely no way to destroy a building via scripting or any other game mechanic.
So then, the challenge was to provide some level of depth and choice for the player when conquering a province, but not to overwhelm the AI, when we had to share government complexes due to the complex limit. Without being able to destroy buildings, this was quite problematic for us. It took us many, many months to come up with a working system which, we feel, gives at least a little of the functionality we envisioned, but is also workable for the AI and, of course, doesn't put us over the complex limit.
There are really two different styles of government, at the moment; the nomadic/settled type, and the type that applies to all of the rest of the factions. Because of the way steppe factions conquered land, we wanted to show the choice these people faced - in steppe and other areas, the faction remains nomadic, moving the herds and staying a fluid people. In "mixed" areas, the leader has the choice of staying nomadic, or becoming more settled. And, in areas which are neither, the leader is not given this choice. This affects, primarily, unit recruitment, though hopefully in the future will have other implications.
(Side note - we have discovered a VERY annoying problem, where some buildings which handle unit recruitment cannot have other capabilities, such as bonuses to law. We have no idea why this happens, why some buildings can have these while most can't, and it only manifests itself as a random CTD during AI turns, and sometimes also when the player right-clicks on the building level which contains the problem. This is yet another problem which limits what we can do.)
So, in order to show that some factions can have certain government types in only some areas, through scripting we place "precursor" buildings, which enable the building of those more home-oriented government types. To enable the AI to utilize the governments, and without the ability to destroy buildings, we have had to, through scripting, place governments on behalf of the AI when it takes a province, and to 99% damage existing government buildings when the province is taken. It is considered an exploit (which we cannot prevent) for the player to utilize these existing government buildings. This, so far, is the only viable workaround to the problem which we have found, and we welcome suggestions on how to improve it.
Regarding unit recruitment, for multiple reasons we have decided to create what we lovingly call the "Military-Industrial Complex," or MIC for short. The primary purpose of this building is to represent the military infrastructure of a province. This infrastructure is independent of city size, and is very important a faction such as the Germanic tribes, who did not live in large settlements but were spread out in large numbers of small ones, to represent the ability to raise an army from the countryside. In addition, we feel it better represents, combined with other building requirements in unit recruitment strings, the way in which units were recruited. We must still use some RTS conventions, such as tiered recruiting based on building new buildings, but this is a game device we use which we can't really get around. We can't be more flexible in allowing recruitment based on other factors, so the time/money for MIC tiers is the best we can do, along with other buildings, of course. It did not make sense at all to keep the barracks/stables/archery range/etc convention; A state would not raise skirmishers independent of spearmen, and so on. Plus, we save on building complex space by doing it.
So, our recruitment system is based on an individual unit determination of how that unit would be raised, and by whom. This would be demographics (For instance, many better Dacian units are drawn from the noble class, and the noble class would be exported to conquered areas.), availability of requirements (you can't find elephants everywhere, though we have tried to reflect how a few states were able to import elephants), government type (Romans can build Auxilia in various areas through one government type, legions through another, and native units a third), and various other factors. Recruitment areas vary per unit by faction, and even by faction by government type. Hopefully soon we'll have a spreadsheet up and running which will show the choices we have made, and enable discussion of what those choices were and how we can better reflect things.
We've also chosen to do several things. First, we're severely limiting weapon and armor bonuses. Through testing we've found that they affect things to a -huge- degree, completely tossing out the "balance" of battles we have tried to create, to reflect the historical way in which various troop types were used, especially in regards to missile units. Second, we will probably be reducing the experience scale, for the same reason. These things are creating completely artificial results, and aren't in line with the philosophy of the mod. The weapons/armor bonuses will be the most limited, probably not occuring at all for missile units, and at most +2, though more likely +1, for other troop types.
So, having said all that, there are some obvious inefficiencies in what we've been forced to do, and are always open to suggestions on how to improve it. In addition, some things cause some real concerns for us, including:
- Possible 10-or-so limit to hidden resources in a single province (anyone have any insight here?), which will really put a damper on our way to reduce wait times through making EDB more efficient.
- Inability to port over even the simplest building to 1.5 (Until we can do this, all thoughts of moving to 1.5 are completely dead. If you guys want to ever see this mod moved to 1.5, we could really use assistance on porting the most basic of our features, much less our more complex stuff)
I welcome discussion about this.