Friday, November 5, 2010

Open Friday: Class, Sub-Class, Split Class

In my post earlier this week, I began thinking out loud about the concept of the "split class" as a means of adding specialized character classes without having to introduce whole new classes (or sub-classes). Of course, this raises the question of just what the differences are between a "class," a "sub-class," and a "split class."

I have some thoughts on this myself, some of which run counter to the Gygaxian usage, but I'll save those for a post this weekend. For now, I'd like to open the floor to discussing both the implicit distinctions between the three types of classes in AD&D and, assuming one disagrees with those distinctions, how they might be better presented -- up to and including eliminating them entirely -- in order to distinguish more adequately between them. What interests me here is seeing where philosophy and game mechanics intersect in people's minds: what is a character class and to what extent can/should it be multiplied and subdivided?

I'll likely be online today, despite its being Friday. The response to the Petty Gods open call has been remarkable thus far and I need to clarify some points about it for the many writers and artists who've already stepped up to help. So, expect at least a post about that later today.


  1. Class:
    A prepackaged collection of traits and abilities representing an archetype. Classes typically represent the core player roles in this RPG.

    A class that shares many of the traits and abilities of another class, with some unique variations that are found no where else.

    It exists because it provides variety while making rules simpler; in the cases where it is the same as the parent class, we can just reuse the same rules without having to define new ones. In this way, it is very much the same as a subclass in OO programming.

    However, the sub-class is (largely) expected to function in the same role as the original class. Paladins fight, thief acrobats steal, and so on.

    A class that combines traits and abilities from multiple (core) archetypes. It can function simultaneously in each of these roles within a game session.

  2. I think you're mistaking "split class" and "multi-class." Split-class, as defined earlier on Grognardia, is the thief-acrobat: you start as one class, reach a certain point, and then rather than continuing to progress in that class you start gaining levels in a related but different--usually more specialized--way.

    The bard class from 1e is similar, though not precisely the same. The "prestige" classes from BECMI (druid, avenger (I think--it's been a while), etc. are very much in the same vein.

    Personally I'm not sure you need them, or even sub-classes. The original intent, IMO, was to have "Fightman Man" equally apply to Sir Lancelot, a Samurai, and an eskimo hunter. It's up to the player to role-play the differences.

    I'm against adding classes that don't have a specific, concrete purpose unserved by some other class. To me even Paladins are just giving exceptional fighters some powers they would have had to work to achieve, just because they rolled high at chargen.

  3. On two seperate occasions I've had a character that evntualy became a Halfling Master (from the Five shires.) The master split class gives halflings Druidic abilities and increases a Shireling halflings Denial ability to insane levels. Although I've taken the option to become a Master on two occasions, I've come to loath the split-class. There is almost no draw back to taking the Master path. Alot of people feel the BECMI halfling needs a boost, I understand that (and respectfully disagree,) but I stand firm that the "Five shires" went far to far with the power up

    A bad experience with a Neutral fighter who wanted to become an Avenger left an equaly bad taste in my mouth. The Choatic and Lawfull Fighter split-classes give the characters some potent new abilities at the cost of some loyalty promises, but the neutral path offers only responsibility with very little in the way of power. When the player read his options he was sorely disappointed (to the point that he almost quit the game) even after i offered to let him bend the rules and become an avenger or a paladin (which he refused to do.)

    So all in all I don't care for split classes. They either are to restricitive or to potent in m,y experience.

    I've had much better experiences with BECMI's Multi-only-classes. Basicaly these classes are added to an exsisting class for an XP penalty (or in some cases they use a completely seperate XP progression.) I've seen more then a few Humanioid Shamen and Wiccas through the years, and I'm not opposed to characters taking the merchant or Lycanthrope options. i feel these classes are much better balanced then the split classes in BECMI, although not as elegant as subclasses in other editions.

  4. I'd say that a "class" is more like a literary archetype. On a basic level, Lancelot and Conan carry out the same function (protagonists who fight evil, monsters or what not), though their descriptions differ significantly. I see this as a great advantage of the class system, one that is very threatened by the inclusion of sub-classes.

    However I also value the amount of player options in the form of powers and abilities, that the sub/split-class systems bring to the table.

    Perhaps an inclusion of special abilities (each with it's own limitations) available to each class would be an answer to this. This way the players would have a larger palette of "moves' or special abilities to chose from, without obscuring the initial archetype concept of classes.

  5. I go for the simple alternative myself. They are all classes. This probably derives from the very early era of D&D when a lot of new character classes were introduced in many of the fanzines, APAzines, and magazines such as White Dwarf and even The Dragon. This meant the general approach for a derivative class was to construct the new class in it's entirity. You often used one of the primary classes as the template, but there was no real connection between, say a thief and a ninja or assassin. They were seperate classes, even if they made use of related abilities. And yes, I used a lot of classes in my game, some of which were very specialised and culturally determined. Added to the spice of life.

    "Split class" was only ever used by us to describe a character who was multiclassed. In that they had to split their experience between multiple classes. Then again I don't think I ever used Unearthed Arcana so the official definition must have passed me by (I must have read it at some point, since it is on my shelf, but I mustn't have been very impressed with any of it).

  6. I don't think that there's any substantive difference between classes and subclasses. The old AD&D subclasses seemed like they were grouped merely by theme. I always treated them, and they behaved exactly in all respects, as if they were completely different classes altogether.

    I think maybe my take on classes might be different than a lot of other D&Ders... ideally I prefer classes that are sufficiently flexible in their design that they can allow multiple mechanical interpretations of the archetype that they're meant to represent.

    I don't think many classes in any edition of D&D really do that, though, so the alternative is having either lots of classes, so I can pick the one that most closely adheres to the interpretation of the archetype that I'm going for, or at least a GM that's flexible enough to work with me on tweaking the classes to make it more what I want.

    I don't have any opinion on split classes, really. That always seemed like a bit of a strange concept to me, and I've never used one.

  7. @squidman

    It sounds like you'd probably enjoy Pathfinder's approach to class options, then (even though they have a huge number of extra classes also showing up in their expanded material), in that they have fairly archetypal class descriptions and then a ridiculous amount of options that let you tweak the "subtype" of wizard, rogue, fighter, cleric, etc. that you want to play with an a la carte sensibility.

    Power-creep extraordinaire, but as a design principle it's interesting. A bit of AD&D, a bit of GURPS.

  8. "The original intent, IMO, was to have "Fightman Man" equally apply to Sir Lancelot, a Samurai, and an eskimo hunter."

    Perhaps but the D&D and AD&D game system itself does not promote this. There is every reason for a fighter to adopt Plate (AC2) and a Bardiche (d12 damage) and no advantage to being an swashbuckling rogue like Conan/Fafhrd/Mouser let alone an aboriginal hunter.

  9. "Perhaps but the D&D and AD&D game system itself does not promote this. There is every reason for a fighter to adopt Plate (AC2) and a Bardiche (d12 damage) and no advantage to being an swashbuckling rogue like Conan/Fafhrd/Mouser let alone an aboriginal hunter."

    Exactly. I was just thinking about that this morning. Part of the Fighter's "benefits" is the ability to use any armor and any weapon.

    But, yes, if you want to play a "Fighter" who is lightly armored, or who doesn't know how to use every weapon but is really good with a bow, then the Fighter class "as is" doesn't really work. That's why I think little modifications are fine - lose this set of abilities and gain this other set.

    And, yes, you could argue that you can play a lightly armored swashbuckler type with the standard Fighter class and just use leather armor or no armor, but you're giving up a major benefit of the class. Later on in the campaign, you're going to come across +2 plate mail, and it doesn't fit your character concept to use it, so you're penalizing yourself.

    The other classes don't have this problem as much, I don't think, but the Fighter one was always one that I felt was open to expansion into other areas. But, I am biased because I like having a lot of class options, unlike a lot of people who read this particular blog.

  10. If I were to set up classes from scratch for an rpg, I would use 3, and only 3.

    Fighter is any and all types of characters that use heavy armor and a wide range of weapons.
    Rogue is any and all types who use light armor, more finesse based weapons.
    Spellcasters are anyone who casts spells.

    I would give spellcasters different starting spell lists based on whether their background was aiming them toward fighting magic, support magic or healing magic, but none would be restricted from learning any type. Many sorts of different casting techniques would be possible. Everything from memorizing spells from a book, to getting them from divinity, to manipulating ley lines. Any one caster would start play knowing one method of getting their spells, and it would be exceeding difficult (not impossible) to learn other means.

    Rogues could be thieves, rangers, bards, or a variety of other things, all depending on their imagined background.

    Fighters could be paladins, rangers, bards, soldiers, etc, again depending on their background.

    Starting from the basic foundation of the above triumvirate, specific activities the player wanted to attempt would be adjudicated by the DM much in the same way as a character from the LBB's wanting to try to follow goblin tracks.

  11. I prefer the modern more flexible class set up to the older methods.

    Basic RPG does this pretty effectively I think

  12. Clearly the difference between "sub-class" and AD&D "split-class" (or 3E "prestige class") is that the latter presumes an advanced level in some basic class.

    I'm at a point where I don't like sub/split/prestige classes; they open the door to an overwhelming flood of new classes, and keep campaign play fundamentally unstable as long as new ones are being introduced.

    I prefer sticking to 3 or 4 basic classes and having some customization available within a known, structured system for each. Wizards being distinguished by spellbook selection (or perhaps college availability), for example. Fighters distinguished by a Feat-like system (though fewer than in 3E). Thieves I've never found a good analog for; skill points suck hard.

  13. migellito said: "If I were to set up classes from scratch for an rpg, I would use 3, and only 3. Fighter, Rogue, Spellcaster..."

    Overall I very much agree, and it's what I've done for some time now. Only point of disagreement is that for game design, I say pick one magic "story" and stick to that for your campaign. Proliferating magic-explanations are unnecessary complication. For me, Vancian book-magic is all I think necessary.

    In other words, standard D&D with the Cleric discarded.

  14. I never really pondered "sub-classes" much back in the day. I guess I still haven't (until this post anyway). They were always just sort of accepted as a more specialized offshoot of an existing class and just played as such.

    I don't recall ever utilizing a split-class, nor am I fully sure why I would.

    How different is a split-class option ultimately than a traditional sub-class? Couldn't one argue that a sub-class, say Paladin, is simply a path a "level 0" figher chose to take, rather than having to wait until level 5? They're still bound by the same initial ability score requirments and whatnot.

    Plus, things like tightrope walking or tumbling aren't something you just pick up one weekend while you're exploring the dungeon. Usually would take years of training to use effectively. I'm not sure if it makes sense to suddenly develop skills in such things upon reaching level 5 while raiding the mummy's tomb.

    It makes more sense to me that you've been training in some manner or another in the applicable skills prior to your days as an adventurer before becoming level 1 Acrobat, Ranger, Assassin or whatever.

    Unless split-class abilities are something like Jedi skills that you can pick up in an afternoon training with Yoda on Dagobah a la "Empire Strikes Back".

  15. Random extra thought -- Granted my 1st preference now is a Feats or Spellbook-like system (specific ability picks within class structure).

    My 2nd preference is actually split-classes, as you call them (a.k.a. "prestige classes", et. al.) In my 3E era I actually really liked the UA idea of making bard/paladin/ranger all prestige classes (splittable at around 5th level). Link. One great advantage is that it pushes back choice complexity, so brand-new players still only have to parse the 4 most basic choices to start with.

    Last place would be full sub-classes starting from 1st level. The more you add, the more complex it becomes, without any upper bound.

  16. In my understanding (which is ADD1E-centric):

    One of the basic D&D classes with the traditional powers/disadvantages

    Variant on one of the basic D&D classes, hopefully balanced, power-wise.

    I think of them more a a "splinter-class" in that they break off from the basic class and go off on another path after developing basic-class functionality.

    While I'm at it, might as well hit the other class-types.

    Melding of 2+ other classes. Hopefully balanced to some extent, in that there is some payment to be made for the broad array of abilities. Such as, "fighter/MUs can not wear armor and cast spells."

    Taking up another class after first advancing in another.

    NPC Class:
    Odd, under- or over-powered classes unsuitable for players.

    I prefer variety, so I am open to all of them provided they are balanced in some fashion.

    Unlike others, I liked to use the many non-weapon proficiencies to customize and/or flesh out a particular character. As a DM, I did not let NWP turn everything into a skill/stat roll, but they were explicit markers as to the background or priorities of a PC.