View Full Version : Reimagining the classes

2019-09-04, 01:35 PM
Warning: this is a bit stream of consciousness.

Recently I've come to the conclusion that I kind of hate pretty much every class in D&D for one reason or another. My grievances are fairly edition-agnostic, as while balance is a concern I feel that before balance can be addressed one needs to understand the design space you are working to balance, and for historical reasons this is a question that never quite got asked until 4th or 5th, and by then there was so much tradition that needed to be maintained that it was impossible to fix.

For simplicity's sake I'm going to basically assume that the 3.5/PF/5e engine remains essentially intact, Vancian warts and all, because this is about fixing the classes in D&D, not creating a whole new game. Instead I guess the idea here would be to create the outline of something Pathfinder-like; basically the same, but legally distinct and hopefully better-designed (insofar as it matters I mostly imagine a return to 3.5 form/mechanics as the height of D&D, but with a PF/5e-like de-emphasis on PrCs and lots of new base classes and instead relying on ACFs/archetypes, and preferably borrowing a few of the better ideas from 5e and the Tome of Prowess fan splatbook).

Similarly, I believe that D&D kind of stumbled into inventing its own genre of [medieval] Heroic Fantasy, which includes some frankly ridiculous levels of [especially magical] power and an ability for basically anyone to potentially achieve it; with levels being the way they are, D&D is fundamentally more a sort of western xianxia than anything resembling sword and sorcery, and this needs to be clear from the beginning to the designers, DMs, and players. Or, to put it another way, for there to be a balanced game, every class needs to at least reach tier 3 if not tier 2 as a conceptual baseline. To put it yet another way, yes, this is largely me complaining about melee not getting nice things, as part of a larger discussion on how genre conventions define design.

To start, among magic users, I find there's a nice kind of trichotomy between the Wizard, who studies to gain power, the Cleric, who gets power externally, and the Sorcerer, for whom power is intrinsic. There are different ways to build this trichotomy, such as a Bard/Warlock/Psion baseline, but we'll get into those kinds of distinctions shortly. This trichotomy, which I'll henceforth call the Skilled, Channeller, and Inherent respectively, addresses probably the most fundamental question about how a player conceives of their character, which is their process for solving problems (not just combat of course) and growing as a person. This should probably be thought of as a heatmap rather than three hard options, as it's easy to imagine hybrids, but to as large a degree as reasonably possible we want to keep things simple and iconic lest one end up with classes that occupy niches too narrow to make viable high level characters.

Only when this fundamental high-level concern of how one conceives of problem-solving is addressed, I feel, can one begin to discuss such lower-level concerns as the specific power source or combat role one wishes to occupy (this is where 4e class design began to fall flat, I think; had this been considered I imagine it would be rather less likely to fall into the trap of having such completely unified combat mechanics), assuming one wants there to be more than half a dozen core classes.

I do kind of like that 4e went as far as it did to codify power sources and somewhat reimagine party roles. For my purposes, I think a successful class should have the inherent ability to fill multiple roles with different builds, and to keep from ossifying a specific metagame at the design stage I think it might be a bad idea to formalize specific roles in the design stage, and instead describe problems that one can expect to have to solve and let the players define their own meta. I'm also somewhat sceptical that thinking in terms of power sources is necessarily the best way to divide things at this level, but I don't have anything better so it's what we're using.

4e's list of power sources, which did evolve out of existing genre conventions and was designed to be general enough to run just about the full gamut of possibility in fantasy, is probably a good place to start.

1. Martial
2. Divine
3. Arcane
4. Primal
5. Psionic
6. Shadow

I'm going to scrap Shadow and fold it into Arcane, partly because traditional cosmology, partly because it's just a limited version of what Arcane can do.

Each of these power sources can occupy at least two of our three conceptual problem-solving categories.

Martial is obviously Skilled, but it can straddle a border with Primal to become Inherent, and there are a couple of ways to push it into a sort of Channeller. It's also worth noting that there are straight up a ton of ways to do Skilled Martial, making it easily the most versatile single category, and badically any Martial class will be at least hybrid Skilled pretty much by definition. .
Divine is obviously a Channeller thing… except for the gods themselves, for whom the Divine is Inherent. I'm not sure if an Inherent Divine class is really in the cards, as usually that's more of a racial power thing, but Draconic Bloodline Sorcerers are a thing so a Demigod class should definitely be workable.
Arcane is classically a Skilled power source thanks to the Wizard, but also appears in the form of Channelling Warlocks and Inherent Sorcerers.
Psionics tend to straddle the line between Skilled and Inherent, and tread into Channeller territory very rarely.
Primal is also capable of being Skilled (Incarnum), Channeler (Druid), and Inherent (Fey blood, etc).

From here, we can make a sort of preliminary list of base classes that should cover pretty much the entire design space, which can then have PrCs/ACFs/Archetypes (or even new base classes if a new core mechanic can be created a la the ToM classes) erected around them to allow more player options (and naturally the ability to publish more splatbooks).

Martial Categories:
The Skilled Martial. An extremely versatile concept, though unfortunately trivial to overspecialize into uselessness. The Fighter and Rogue are the classics here, and I retain a temptation to merge them into a single class.
The Inherent Martial. Both the Fighter and Barbarian can occupy this space easily. That the Fighter is so non-iconic as to fit into both of these categories is perhaps an indication that it should be abolished as a category distinct from other classes.
The Channeller Martial. There are a couple of ways to go here: reliance on equipment (which means making items swiftly and in great volume) and reliance on people/animals. The Trapmaster/Gunslinger/Alchemist, the Beastmaster/Chevalier, the Samurai/Knight (either from their lord or from having retainers).

The perennial problem with Martial characters is that it's hard to create character concepts capable of believably keeping up with even a mid-level Wizard, especially when magic martial artists like Monks are a thing they have to compete with as well. It may well be best to make a sort of omnicompetent base class with lots of room for specialization; this is often what happens in fiction anyway, even for what would be relatively low level characters, but the more that capacity for omnicompetence is relied upon the less iconic the characters become. I think the best way to go here is with very few base classes, each relying primarily on a different facet of the Channeller conceptual space while backing that with some Skill and Inherent:

Martial Classes:
The Knight. Here's our Fighter/Barbarian/Thug archetype space. Social power whether by getting help from the nobility or from retainers is critical for any real versatility, and combat power could be achieved via a mount or by a Rage/Action Surge mechanic, as well as numerous strong manoeuvres for applying BFC or the occasional debuff. Far from the noob-friendly barbarian of yore, this would be one of the most demanding and DM-dependant classes at high levels, but it by its nature also comes with plenty of RP hooks and carries an inherent badassery with it.
The Ranger. Here the Channeller options consist of alchemy, pets, and traps. There is a ton of room for specialization and versatility here alongside the gloriously Skilled power of stealth, though specializing in pets can make urban adventures very tricky. For obvious reasons, the Ranger straddles the line into Primal.
The Factotum. Specializations here range from going full Channeller with guns, traps, and/or alchemy as an engineer, focusing on Skillful stealth as a thief or assassin, straying into Arcane as an Arcane Trickster-type, going the social-fu/mastermind/charlatan route, or blending all of them.

Awesome as ToB is, I think sticking to expansions of the core mechanics of skills and manoeuvres a la ToP is the way to go, but to help streamline the mechanics of the classes and help distance them from the magic even as they get special treatment.

Divine Classes:
The Divine Channeller, the Cleric. Caster par excellence, but mostly lacking the self-buffs that result in CoDzilla. Plenty of room for the Domain specializations that make Clerics awesome; actually, Domain spells should always remain a hefty chunk of the spell list as well, to keep their inherent majesty from getting too out of control.
The Inherent Divine, the Demigod. A class so common in myth and legend and yet so weirdly absent from D&D outside of some half-assed racial stuff. I figure the way to go here is to give something loosly akin to the classic Paladin as class features rather than remaking the Divine Soul, which is basically a reskinned Sorcerer anyway. Actually, I'm imagining a ton of touch attack and ray SLAs and self-buffs, and likely somewhat different spell lists depending on ancestry.

Huh. Divine was easier than I thought. Sure there's probably room for some hybrid classes like the Divine Soul or Archivist in splatbooks, but for a core list this feels remarkably tidy having come from the utter mess that is designing workable Martial classes.

Arcane Categories:
The Skilled Arcane. This is the core of the Arcane. Both the Bard and Wizard have long had a home here, and there remain a ton of specialist niches like the Eldritch Knight/Gish, War Mage/Evoker, Necromancer, Truenamer (not that anyone ever wants to touch the Truenamer), and so forth to rebalance the Wizard away from being the hilariously OP thing it usually is.
The Arcane Channeller. There are two basic ways to do this. Either you have the Artificer archetype or the Warlock/Binder archetype.
The Inherent Arcane. Here be Sorcerers.

Arcane Classes:
Wizard. Like with the Cleric, specialization is a must here to prevent Tier 1 Syndrome. Some schools are definitely better than others though, so as with the Cleric a general pool to pull from is also critically important, whether that be from a secondary specialization or delayed access to everything but the specialized school.
Bard. Like a Wizard specialized in a blend of some of the, alas, weaker schools, but able to largely make up for it with some great buffs. The jack-of-trades angle is a classic, and depending on build could either be Arcane or straddling into Martial.
Binder. Various Arcane entities to bind with and swap between, for a wide range of decent abilities. The Warlock belongs here as a specialist version devoted to a single entity rather than being promoted to its own class.
Artificer. Unlike the Factotum who can specialize in nigh-magical but ultimately mundane devices, the Artificer uses actual magic in their creations. These creations are by and large more powerful than those a Ranger or Factotum might make, but are also by and large much slower to produce, and in much smaller quantities. They are also basically all the Artificer has, though each other supernatural power source has its own version of the Artificer.
Sorcerer. These guys lose a bit of their traditional uniqueness compared to Wizards with the mandatory specialization, but 5e did good by making them better at metamagic. There's plenty of room here as well for unique spell lists based on different ancestry and even power source, though care needs to be taken to avoid crippling overspecialization there, perhaps moreso than any other specialist spell lists.

The Arcane classes are in many ways the most generic, with multiple having chasses that can swap to basically any power source. But, this is high level fantasy and these classes basically define their archetypes, so that's how it goes.

Psionic Classes:
The Psion. The grand counterpart to the Wizard. It's kind of amazing what some tweaks to the casting mechanic and a few changes to how the spell list/specializations are organized can change a class.
The Ardent. The 3.5 class (/ the 5e Mystic) is actually amazingly flavorful and well-balanced with its array of powers and mantles. Here's where a lot of other more Gish-y archetypes come in, like Monks, Soulblades/bows, Lurks, and such, while also allowing full caster archetypes like the Cleric domains.

Psionics are an excellent, if underutilized, system that also happens to be well-suited to what I'm doing here.

Primal Classes:
Skilled Primal, the Monk. Aping from Incarnum with controlling the flow of internal energy through the chakras seems like the ideal here. Like an Ardent but with a different kind of discipline and spirituality.
Primal Channeller, the Druid. By which I mean the straight caster. No wildshaping and no animal companions here; those are strong features that go to other classes. While there's some room for unique builds here with different environmental specialities, by and large the Druid is a pretty boring and straightforward concept that gets by on being iconic.
Inherent Primal, the Shifter. Despite seeming like the most obvious match, it's hard to come up with anything good for an Inherent Primal; that whole notion is kind of swallowed up by the Martial power source. But a class built around Wildshaping is pretty necessary if we're going to divvy up the cool **** Druids get, and you can get some pretty cool flavour out of 'shaping the wild natures contained within oneself'. Specializing in shifting into different Types gives plenty of conceptual space to play with for a class already built around rewriting its character sheet, which is also pretty cool.

I find I'm not a huge fan of the Primal classes in standard D&D, but here, while I'm still not the biggest fan of the Ranger and Druid, they are more tolerable now, and the Shaman and Shifter are actually pretty good.

And there we go. Fifteen base classes, each being built out of about a dozen core mechanics ranging from skills to Vancian casting to Binding to Wildshaping. Just about every possible playstyle and character archetype [that's capable of believably reaching Epic levels] is addressed or can be addressed in a splatbook. Maybe at some point I'll get around to building some actual playable homebrew out of this, though I'm lazy enough I'm amazed I even finished writing this up.

My main concern with this at this point is that the spread of ability score dependencies is still pretty wonky, with only one class, the Knight, really using Str as its backbone, and that certain power sources might have other problems with build variety across different ability spreads. It also occurs at this point that the triumvirate of problem solving archetypes basically corresponds to Int (Skilled), Wis (Channeller), and Cha (Inherent), with no real regard for the physical abilities. Perhaps this can be put down to character being a mental thing, but it does kind of showcase how fantasy and D&D has grown and changed over the last thirty-five years that half the ability scores combined can't really support more than two full classes, and even there they suck because they're just asking for MAD.

2019-09-04, 03:19 PM
I sympathize. I share many, most, of your concerns/ideas and was trying to shoehorn them into 5e. But it got to the point where I was doing so much work that it made more sense to switch to another system where I could go skill-based and get exactly what I want. This also means that players can approach the game from the standpoint of 'this is something new to me' rather than 'this isn't what I'm used to playing'.

Good luck to you in your work. I like your ideas.

2019-09-04, 07:40 PM
Yeah, at this point I only actually play D&D because that's what everyone else is playing, and much prefer for example RuneQuest and HeroQuest.

But, 3.5 was my introduction to the hobby (modulo some Baldur's Gate 2 long before I even understood its connection to D&D as a kid) and remains one of the richest character creation/optimization games out there.

I guess fundamentally the reason I care is because D&D is culturally important and actually covers a pretty unique niche in fantasy, and thus deserves to be done right. And given that Pathfinder 2 looks to be shaping up as something of a mistake (in marketing, if not necessarily game design (which I've yet to look at closely enough to comment intelligently))... I think there's a level of validity here in looking at what the next edition should look like in another what? 3-4 years is the usual development cycle?

2019-09-09, 07:09 PM
One question - why can’t martials just all run off ki? If we accept that a martial has to be capable of keeping up with people who literally shape reality, and that it’s already a rather xianxia setting just masked by its presentation, why not run with it? Demon Slayer “breathing techniques” which let your sword do some amazingly impossible things, ninja/monk feats of legend, and whatnot, recognizing that no one is running on just native mortal capabilities because of that were the case, you wouldn’t have civilians who need protecting, just large poorly trained militias.

2019-09-10, 02:36 AM
I think there's a level of validity here in looking at what the next edition should look like in another what? 3-4 years is the usual development cycle?

You're in luck, we've got a whole thread (http://www.giantitp.com/forums/showthread.php?597337-What-I-hope-they-do-for-6e-DnD) for exactly that!

2019-09-10, 03:24 AM
There's a few classes that can go into Channeller Martial for me. The Binder in 3.5 can be a type of that. But in general, imagine someone who channels some external power (say, a demon possession) to basically Hulk Out. Works pretty well, I think.

2019-09-11, 11:27 AM
Divine is obviously a Channeller thing… except for the gods themselves, for whom the Divine is Inherent.The gods are often portrayed as channeling the power of collective faith. The magic flows from the faithful to the god, who redistributes some of it to the clerics.

One question - why can’t martials just all run off ki? If we accept that a martial has to be capable of keeping up with people who literally shape reality, and that it’s already a rather xianxia setting just masked by its presentation, why not run with it?I second this. There should be a level at which martial classes move past "outstanding feats of strength and dexterity" and embrace the anime.

2019-09-12, 01:18 PM
There's a few classes that can go into Channeller Martial for me. The Binder in 3.5 can be a type of that. But in general, imagine someone who channels some external power (say, a demon possession) to basically Hulk Out. Works pretty well, I think.

Another weird idea is have a martial that has a collection of NPC soldiers with them that they work with to achieve cool results. They channel the soldiers efforts into their success.

2019-09-19, 02:41 PM
When looking at magic, I'd suggest taking a glance at the pathfinder 2e, 4 tradition system, although you seem to have it mostly covered. I just think it's kind of neat, and the division between arcane and occult is pretty cool.

I also think the divisions are cool. The attitude may end up being restrictive, but I like the completionism. I particularly see this being a problem when you start to fit in things like monks, which are, to be honest, everywhere and everything from a narrative standpoint, so long as you don't count psionics, which I won't, till the mystic runs on ki. I don't think making them primal makes any sense(although I think it's a cool bit of flavour and I would love to see it run with), where it makes much more sense for the barbarian. Make sure you don't tie yourself down.

I have never seen an attempted shifter class, but would love to see one now. I had a similar sort of idea for a divine character whose basically wielded a load of channel divinity(5e at least). The idea of taking a side mechanic and making it the main thing seems a lot of fun to me(though difficult to multiclass balance).

Overall I like the ideas. Obviously this would be a serious undertaking, but I wish you good luck if you're crazy enough to try it.

2019-09-20, 12:02 AM
The gods are often portrayed as channeling the power of collective faith. The magic flows from the faithful to the god, who redistributes some of it to the clerics.

I second this. There should be a level at which martial classes move past "outstanding feats of strength and dexterity" and embrace the anime.

Fair point about the gods channelling faith. Still, I'm happy with how using the thought process I did resulted in coming up with the demigod class.

As for running everything on ki, I think there's plenty of room for Ex abilities and [what in 3.x are] Epic level skill checks as ways to achieve great power without necessarily resorting to ki. Even in xianxia novels not everything runs on ki, as a sheer overpowering physique can often do some semi-exotic stuff like creating dimensional tears. I always recommend the Tome of Prowess (https://dnd-wiki.org/wiki/Tome_of_Prowess_(3.5e_Sourcebook)) at every opportunity, as it proves it's possible to make a skill system that brings every class (except arguably the Tier 6 classes (and the Truenamer of course)) up to Tier 3.