The next thing we need to think about, building on the ideas from last post, is how many different “classes,” meaning character roles, we are going to incorporate. There are a few different extant paradigms for this: the warrior/wizard/thief/cleric breakdown from D&D forward through most fantasy games, for instance, has become relatively ubiquitous. I understand D&D4 tries to get out of this mold, but I’m not well-read enough on the topic to explore it; what I do notice, in many successor games, is the tendency to want to broaden character options by selectively combining traits from the different types. The clunky but obvious version is the idea of “multi-classing”, but the proliferation of hybrid classes speaks to the same desire: the paladin, for instance, is the warrior+cleric, while the bard is the thief+wizard.
Hopefully, instead of having static multi-class options, our dice system so far allows a very flexible and diverse range of choices, wherein you simply buy the dice for the type of actions you want to perform under each role, and how large/many you want them to be.
(A digression, or perhaps a backtrack: difficulties go up, so you want larger dice to perform more complex actions, but having more dice means more actions to perform or combine. It’s perfectly reasonable to model a novice mage by spending your starting points buying multiple d4s or d6s from the wizard column, meaning you know lots of minor spells but nothing big yet, or by buying the biggest single die you can get, meaning you have committed one significant spell to your grimoire but haven’t branched out yet. It’s easier to buy several dice from different columns and patch them together into a combined action: the backstab as a warrior die + a thief die, the fiery sword spell as a warrior die + a wizard die, the silence spell as a wizard die + a thief die, etc. ad nauseam.)
It occurs to me, then, that the roles or classes are simply types of expected actions, and you either always behave in ways that adhere to a particular list of actions (the single-class character) or you do things not in keeping with your laundry list and thus pay a balancing price by not being as good at either of them due to splitting your focus (the multi-class character). There’s an implicit value judgment at play in such a construction, particularly in those games that make you pay an “experience point tax” to multi-class, or limit in any other way the amount you can flex your character’s advancement. Our dice system is at least value-neutral, in that you simply have the same finite amount of resource (points to buy dice) and allocate them as you see fit, but it doesn’t cost anything extra on the surface to branch out.
So, now we need to come up with a framework. Let’s try to divorce the discussion from the fantasy tropes we’ve been using so far — instead of warrior/wizard/thief/cleric, we can reskin these as focused on combat, special abilities, espionage/stealth, or support. As most fantasy game players will note, there’s a tendency for one or two of those options to outstrip the others as they progress, particularly the specialist (TVTropes dubs this phenomenon “Linear Warriors, Quadratic Wizards”). We should note that there are often auxiliary factors that attempt to balance this out: the more useful and lethal specialist pays the price by not being able to take as much as they can dish out, for instance. Our system thus far only talks about the actions one can perform, and nothing else. We may or may not factor such other considerations in as we go, but that can come later, I think. (Or everyone may just have the same capacity for damage and whatnot. I like the simple answers best whenever possible.)
I also like appropriating symbols or systems from other incongruous places at times, so my first instinct is to use the ROYGBIV visible light spectrum to map to the dice, assigning a role to each of the colors, with red on the one extreme being fighting and violet on the other being diplomacy. My gut instinct is that that’s too many choices, though, that will result in overlap between the different role purviews in the middle of the spectrum. It’s also too much like I’m ripping off Immortal: The Invisible War. (Not that that was such a bad system, though it had its flaws, but I want to come up with a more distinct result.)
Thinking over different genres of games and fiction, we could say a good spread of diverse choices would be five roles: fighting enemies, making/fixing things, interacting with/fixing people, getting information, and a fifth role that falls under what I would call “meta-game specialist”: a lump together of spellcasters in fantasy games, hackers in modern games, etc, because they all tend to deal with manipulating mechanical phenomena in the game system. They generate and hand out bonuses or penalties to people performing actions (themselves or others) as distinct from actually performing those actions. Since in this project, we’re also looking at anyone being able to incorporate elements from all of the roles, this allows us to model a host of effects or character types where some of them are based on pure training and talent and others get special bennies to enhance a slightly more reserved training curve.