Which Multiclassing Power Paradigm is Best? ToB? 4E? Else?

[Bill Bisco: Isometric Imp]

In 3.5E PHB, if you have 18 Levels of Fighter, and take 1 level of Wizard. You cast spells as a 1st level wizard and have a caster level of 1. With Tome of Battle, you can take 18 Levels of Wizard and take 1 level of Warblade. You can use maneuvers as a 9th level Warblade and have an initiator level of 9 (the only exception is stances which by RaW the first stance chosen must be 1st level, but that aside).

4E made it such that multiclassing is done by grabbing a power from another class. Thus a Fighter could take Wall of Fire (from the Wizard power list) at the appropriate level.

Another way could be that, multiclassing would give you 1st level powers but at the competency of your char level. Thus for example, a Fire Bolt Type spell would deal 1d6+Char Level Damage rather than 1d6+ Caster Level Damage. So, a fighter could multiclass into a wizard and have a level of competency appropriate for his character level, but not being able to choose the more powerful higher level spells.

So, my question is, what is the best way to adjucate multiclassing and competency for player characters?

[malak]

I like 4e multi-classing in that it removes the worst abuses of 3E multi-classing. I think my answer would be:

4e multi-classing, but any power from the non-base-class use the highest ability modifier for attack and damage.

[hogarth]

I don't really have a problem with the 3.5E version of multiclassing; it doesn't inherently offend me that making a terrible character is possible.

It's possible that I might approve of 4E multiclassing/hybrid classing/whatever if I didn't think that all 4E abilities were pretty much equally boring.

Personally, I'm not crazy about the idea of totally unlimited multiclassing because then all of my characters tend to blur together into the same amorphous blob of useful abilities.

[Winnah]

I like AD&D multiclassing. While the single classed character has more powerful abilities, the multiclassed character typically has less powerful abilities due to lower level. It evens out due to a more diverse array of options, far superior to anything I have seen in a 3rd or 4th edition game.

The system is not without issues though, no doubt contributing to the elf fappery that still plagues D&D. It tend to work best where class combinations compliment each other, such as a mage/thief that cast spells to boost their stealth abilities, or a fighter/priest that can self buff.

Combinations involving mage and priest can be limited. A longer work-day, especially at lower levels, but the character can still only cast one spell a round, and spellcasting is a lot easier to shut down in AD&D.

I can't comment on dual classing, as I have never seen a coherent implementation of that system in any game I have played.

[talozin]

AD&D multiclassing is okay as far as it goes, but like much of AD&D, it suffers from the requirement that you figure out what you're doing at character creation. The lack of any option to, later on in your career, say "you know, I'd like to pick up some thieving abilities", is problematic. Yes, I know you can dual class, IF you're human, IF you have crazy high stats, and IF you don't mind dealing with the "give up your old class until you reach higher level" nonsense. These aren't always possible, especially if you want to change course late in life -- how many fighters who weren't planning to dual-class already put a 17 into, say, Wisdom?

I like the concept of 3E multiclassing, but the execution was shit. It's not so much that making a terrible character is possible -- just about any interesting game system allows that -- as that making good characters out of some combinations is difficult or impossible. But the idea that you can pick up abilities from any class whenever you gain a level, and get something appropriate -- that part I like. The part where multiclassing spellcasters is like hammering nails into your own hands I don't care for.

4E multiclassing just seems odd. Considering how samey most abilities are, setting feats on fire to get access to slightly different flavor text feels more like a trap option than anything you'd really want to do. Unless you're doing it because you've discovered a combo that leads to Real Ultimate Power, in which case OK, but is there no middle ground there?

[Username17]

I'm surprised no one has given shout outs to the Final Fantasy method of everyone having subclasses: limited selections that can and must come from different lists.

Like AD&D Dual Classing, the actual implementation requires way too much grinding to be acceptable in a table top format, but the idea seems fundamentally sound.

-Username17

[echoVanguard]

I'm surprised no one has given shout outs to the Final Fantasy method of everyone having subclasses: limited selections that can and must come from different lists.

Are you talking about stuff like Final Fantasy Tactics, where you had Chemist -> Black/White Mage -> Oracle/Time Mage/Summoner ?

echo

[Username17]

I'm surprised no one has given shout outs to the Final Fantasy method of everyone having subclasses: limited selections that can and must come from different lists.

Are you talking about stuff like Final Fantasy Tactics, where you had Chemist -> Black/White Mage -> Oracle/Time Mage/Summoner ?

echo

No. Final Fantasy XI, where you were a Black Mage (sub White Mage). Your sub class gave you some White Mage doohickies, and you would often do that because there was some black white synergy.

The actual system leaves much to be desired. The number of effective combinations is very much smaller than the number of combinations. Thieves almost always sub Warrior or Ninja and never sub Summoner. Black Mages do not sub Paladin and Monks do not sub Ranger. Also, you have to level up your job and your subjob separately, which means that you have to do a lot of fucking grinding.

But in a Table Top game, where your subjob class abilities were simply declared at the level they came online rather than being grinding rewards, that would be OK. Also, you'd want to make it so that being a Warrior/sub Mage was a viable life choice. But that's a complaint common to RPGs in general.

-Username17

[echoVanguard]

Honestly, I think the 4E model is intrinsically ok, where you have guaranteed powers from your class and can progress in other classes with your optional advancement units (feats, nwps, talents, etc). 4E's implementation of it, on the other hand, was...well, not very good. But I think the principle is sound, and that's pretty much how it works in our system (married with 5-level class blocks). Regardless of which class you start out as, you can freely use your Talents (which are basically feats) and Techniques (which are basically powers) to pick up any ability you qualify for, even if it's from another class's tree. The difference is that class probably gets handed all the prereqs for that ability, while you have to pick them up a la carte. Coupled with the fact that any redundant ability or trait gets refunded (ie, you get a free pick of anything you qualify for if a level of your class would give you a trait, talent, or technique you already have), it's easy to mix and match abilities to suit your character concept - you can have an Agent with a dash of Sorcerer, or an Agent with twice as many sneaky tricks as a standard Agent, or a full-on mix between Agent and Sorcerer, even though all three of the aforementioned characters are Agent5.

echo

[hogarth]

I like AD&D multiclassing. While the single classed character has more powerful abilities, the multiclassed character typically has less powerful abilities due to lower level. It evens out due to a more diverse array of options, far superior to anything I have seen in a 3rd or 4th edition game.

Actually, AD&D multiclassing is the type of thing I meant by having characters become indistinguishable blobs.

For instance, if I were making an AD&D party from scratch, I'd be hard-pressed to choose a fighter over a fighter/cleric, or a thief over a fighter/thief (or cleric/thief or magic-user/thief), since in the medium to long run you're just one level behind.

So all my characters turn into "spellcasting plus whatever". That's fine, as far as it goes, but then why have fighter and thief classes in the first place?

[crasskris]

AD&D multiclassing was stupid from day one. Human multiclassing was simply a veiled punishment for daring to have meaningful character development (of a kind), and the meta-human multiclassing was arbitrary and stupid. I still don't see the difference between a fighter/cleric and a paladin (or a full cleric, for that matter), aside from the fact that no evil paladins were allowed and the fighter/cleric always sucked in comparison.

3e/3.5e had, like in so many cases, a good idea, but was too married to it's legacy to tap into it. 3e kinda multiclassing works best, imho, with short-leveled classes with a distinctive schtick. A seventh-level character can be fighter 3 / bard 4 or fighter 2 / bard 5; in both cases, he'll be a skald, only with varyiing emphasis on (synergizing) abilities. He can't be fighter 7 though, because fighter only has 6 levels before the concept of 'puts swords in people' needs to be diversified with another class.

4e multiclassing works well with the 'choose your destiny at point zero' approach of a rigid class system. Getting a (usually trimmed) standardized combat power that is almost indistinguishable from your usual standardized combat powers is of course one of the worst implementations of this idea, but well: 4e.
It would be far more interesting if it allowed a mage to command an army by multiclassing into soldier, or a ranger to summon demons by multiclassing into warlock. Which kind of plays into the direction of the Final Fantasy implementation, I suppose.

[talozin]

For instance, if I were making an AD&D party from scratch, I'd be hard-pressed to choose a fighter over a fighter/cleric, or a thief over a fighter/thief (or cleric/thief or magic-user/thief), since in the medium to long run you're just one level behind.

This is much more true in 2nd edition than in 1st; the demi-human level limits were much harsher originally, to the point where if you were even thinking of playing your character past 5th level, you had to think carefully about whether going multiclass was worth it versus the unrestricted advancement of a single-classed human. By the same token, if you were bound and determined to play a demihuman, there was almost no reason not to play a multiclass character, assuming you qualified for one -- not always a given in those pre-point-buy days.

There were also other subtler drawbacks to certain types of multiclassing. Playing an elf, for instance, prevented you from taking advantage of Raise Dead, meaning you had to come up with a much pricier Resurrection if you wanted to be brought back from the dead.

But generally, yes, this was very much in keeping with the idiom of AD&D. If you rolled well on your stats, you were rewarded with a better class, whether that was a multiclassed dwarven fighter/thief or a paladin or a dual-classed fighter/magic-user. Multiclassing was right in line with that.

[crasskris]

Speaking of 4e, one of the actually worthwhile multiclassing ideas in that heap of...um...something negative was the option to choose different paragon paths / epic destinies based on your multiclassing choice. That one's probably a keeper, as long as your class system is tiered.

For example, Mighty Knight might be a paragon path to both the heroic classes soldier and noble, while Dread Necromancer might be an option for both wizards and cultists. Both paths allow to become a Death Knight in epic stage.



Edith has corrected some grammar issues.

[ModelCitizen]

I like the 3e/Final Fantasy 5 system. 3e's implementation isn't great though. To make it work low level spells need to scale fully with level, progressions like 3/4 BAB need to not exist, martial classes need to be subdivided less. But it's a sound idea and it's the only scheme I can think of that lets a character change directions mid-campaign.

But the ToB/Ardent system is 100% grade-A shit. The fact that a Fighter 4 / Warblade 1 can grab more 3rd level maneuvers than a Warblade 5 (and Warblade 1 / Fighter 4 can't) means the system is fucked. If you want to make dipping a caster class worthwhile, the low level spells should stay good. Don't tie spell access to character rather than class level and then make people wade through prereq trees to figure out what spells they can take. That's just asking for people to break something.

[Seerow]

But the ToB/Ardent system is 100% grade-A shit. The fact that a Fighter 4 / Warblade 1 can grab more 3rd level maneuvers than a Warblade 5 (and Warblade 1 / Fighter 4 can't) means the system is fucked.

The Fighter 4/Warblade 1 has more second level maneuvers, because his initiator level is 3. It is kind of annoying the warblade gets a whole bunch of higher level maneuvers at the start, but he's permanently a full level of maneuvers behind the straight warblade, so it's not as bad as you imply. The only bad part I can see is that the Warblade 1/Fighter4 doesn't have any 2nd level maneuvers, where the other way around has all 2nd level.


I mean the Fighter 4/Warblade 1 has 3 2nd level maneuvers. A Warblade 5 has 3 1st level, 2 2nd level and 1 3rd level maneuver known. Plus one extra maneuver readied. The straight warblade is pretty much completely superior here, as it should be.

[tussock]

No multiclassing, which is to say Leadership is the best power paradigm. That way, even if you have a whole 'nother set of actions, your sidekick can't outshine anyone else, and your primary shtick still works just fine.

Now, if you want to support "Yeh, I was a Fighter, but then I apologised to everyone and I've been off the sword for 60 days now" you need to make Prestige Classes for that, where a 6th-8th level Fighter can start being a Wizard or Cleric and not suck for it. Maybe a themed prerequisite feat to represent a quick dip that need not be followed up on. Or like Menter Basic, where a "Paladin" is just a 9th+ level Fighter who grabs 1/3 his level in Cleric spellcasting.

[ModelCitizen]

I mean the Fighter 4/Warblade 1 has 3 2nd level maneuvers. A Warblade 5 has 3 1st level, 2 2nd level and 1 3rd level maneuver known. Plus one extra maneuver readied. The straight warblade is pretty much completely superior here, as it should be.

I constructed the example wrong. The problem I was going for is dipping a ToB class late so you can spend all your extra slots at 1st level right as you get access to a bubble of really good maneuvers. I was going for 3rd level maneuvers, but you're right, that's like Crusader 4 / Warblade 1, not Fighter.

[Krusk]

Best system. 3e with no standard length on classes. A wizard is a 20level class, but a fighter is 6. Maybe a ranger goes out to 7 or 8. Prestige classes can exist as normal.

If I remember right, in 4e everyone took the first multiclass feat. None of the feats are paticularly good and once you take feat taxes you then snag one multiclass feat and then improved init or something. I also remember every party being leader striker striker and defender who strikes hard. Popular multiclass feats were in order: something that let you in an abusive PP you can't otherwise get in. leader if your leader did it wrong. strikers.

[Seerow]

I mean the Fighter 4/Warblade 1 has 3 2nd level maneuvers. A Warblade 5 has 3 1st level, 2 2nd level and 1 3rd level maneuver known. Plus one extra maneuver readied. The straight warblade is pretty much completely superior here, as it should be.

I constructed the example wrong. The problem I was going for is dipping a ToB class late so you can spend all your extra slots at 1st level right as you get access to a bubble of really good maneuvers. I was going for 3rd level maneuvers, but you're right, that's like Crusader 4 / Warblade 1, not Fighter.

Even then, initiating classes still only count as half for other initiating classes. Each class's IL is tracked separately. In your case you'd have a Crusader IL of 4 and a Warblade IL of 3. So the warblade still only gets 2nd level maneuvers.

There is no possible way you can dip into a ToB class and get top level maneuvers. You can of course dip in at later level for better than 2 (dipping in around level 9 for 3rd level maneuvers is popular, especially since that's when some of the nicer stances like thicket of blades come online), but you'll never be able to multiclass and have higher level maneuvers known or more of your highest level maneuver known than a straight classed character.

[Swordslinger]

4E is probably the best in concept (though the execution sucked). The concept of 4E is that you're always getting a level appropriate power, and that's a good thing. The idea that, you're 17th level and you're getting a 17th level power is a good one and something designers should keep in mind.

3E's system arguably wasn't bad in some places, it's just that it didn't work for level based powers at all like the vancian casting system. It worked fine for fighters and rogues. Toss anything vancian in the mix and it sucked ass. Getting sleep at 7th level just isn't a level appropriate power, no matter how you try to disguise it.

The ToB system was certainly an improvement but more of a bandaid fix.

[Username17]

Getting away from 4e's incredibly shitty implementation, there is a salvageable core. It looks like this:

• At level X and level Y, you get a new class from a higher tier. Like 4e Paragon Paths and Epic Destinies except they would all be selectable by any class like Epic Destinies and they would presumably actually matter.
• Within a tier, you get a limited number of selections off other lists.

The big problem with this kind of multiclassing is that in order to make your cross list selections, you have to read every single list in the game. And that's hundreds of pages just to determine that you want a cross-class stone skin or doubledge. And by extension, it is very hard to balance. Because it's hundreds of pages to wade through to select each cross-class power selection, it is unreasonable to expect that playtesters will have thought of even a small amount of the total possibilities. Cross-class vampiric touch is going to be OP on Berserkers, rampage is going to be OP on Scouts, true strike is going to be OP on the Beguiler, and so on and so on.

-Username17

[LR]

But in a Table Top game, where your subjob class abilities were simply declared at the level they came online rather than being grinding rewards, that would be OK. Also, you'd want to make it so that being a Warrior/sub Mage was a viable life choice. But that's a complaint common to RPGs in general.

-Username17

So, more like Guild Wars where horizontal advancement reigns and Warriors have taken skills from most caster professions?

[Seerow]

The big problem with this kind of multiclassing is that in order to make your cross list selections, you have to read every single list in the game. And that's hundreds of pages just to determine that you want a cross-class stone skin or doubledge. And by extension, it is very hard to balance. Because it's hundreds of pages to wade through to select each cross-class power selection, it is unreasonable to expect that playtesters will have thought of even a small amount of the total possibilities. Cross-class vampiric touch is going to be OP on Berserkers, rampage is going to be OP on Scouts, true strike is going to be OP on the Beguiler, and so on and so on.

Well what if the powers were brought closer to how 3e organized them. ie instead of having a completely separate power list for each class, you have a big central list of spells that most arcane users draw from. A big list of techniques that martial characters draw from. A big list of prayers divine casters draw from. Etc.

Now each class still has their sublist from this list, so they are different from each other, but there's some overlap there so there's less material to have to memorize. (ie instead of needing to know 10 different power lists, you pretty much need to be familiar with like 3), and picking up powers from classes that draw from the same list is easier.

Say you're a Barbarian and want to take that Rogue Power that lets you teleport through shadows at a short range, which a Barbarian couldn't normally get. You can do that relatively easily. Trying to pick stuff up from a totally different class, like a Wizard, is going to be harder (require more investment, or get you something a couple levels behind the curve, or the thing you get might key off a resource that you have less of, something along those lines).

[Winnah]

Class and role can become irrelevant if multiclassing is taken to an extreme. You have to account for players cherry picking the best abilities at every level, irrespective of class. At which point, whether they started the campaign as a barbarian or wizard becomes meaningless.

So you have to think about whether you even want to preserve the concept of Class as a characteristic that defines a character over the course of their career.

[Username17]

Well what if the powers were brought closer to how 3e organized them. ie instead of having a completely separate power list for each class, you have a big central list of spells that most arcane users draw from. A big list of techniques that martial characters draw from. A big list of prayers divine casters draw from. Etc.

Then every Arcane character is going to play really similarly to every other Arcane character or character creation is going to be a nightmare and balance is going to be hard to achieve. The "primal list" or whatever is going to be either short enough to wade through and evaluate quickly, in which case all the primal characters are going to end up very similar; or it's going to be long enough that you can't - in which case chargen is a nightmare and playtesters won't be able to cover more than a small fraction of the play space.

Now I'm emphatically not saying that repeating powers is a bad idea. It's a very good idea. The backstab of the Ninja and the Assassin should use the same mechanics. Firstly because it makes learning the game easier, secondly because it makes writing the game easier, and thirdly because it doesn't really harm anything to have similar ability concepts actually be the same abilities.

But if you're making a small number of choices from a big list in order to do your multiclassing thing, you're making the players and playtesters do a lot of work for relatively little gain. Contrast with a Final Fantasy Subjob system, where what you get for subjobbing Knight or Shaman is fixed, and then you only have to check a number of entries equal to the number of possible classes to subjob rather than the entire list of everything in the game. For the original PHB, that would be like reading 10 domains instead of the hundred plus pages of spells.

-Username17