D&D 5E Multi-class support in 5E

[Dannyalcatraz]

But it's not "irreducibly bundled." True20 has greater irreducibly bundled initial entry points for classes, so we can't say that the D&D classes are "irreducibly bundled."

They are- for D&D. How mechanics in other RPG systems- however closely related they may be to D&D- is irrelevant to the irreducibility of a class within D&D.

So what? Why is that a bad thing?

if you think you've scored some kind of point here, you don't know me.

I didn't say it was bad, I said it was inefficient and slow. That is not a value judgement, it is a plain vanilla mechanical assessment. If that is how you want to build your PC, more power to you.

I have a PC (in a 3.5Ed campaign on hiatus) who is a PHB Sorcerer who wears Scale Mail and uses Maul, and has done so since level 1- efficiency is not my motto in PC design.

And one level of fighter basically means that he got it practically all at once. Still don't see the problem yet?

Nope- he got some intensive training at one point in his life, and mastered the basics. His BAB still won't increase like a dedicated warrior, though.

The basics of many things can be learned and retained for a lifetime. I know enough of a few languages to keep myself fed, sheltered, and ask for help- but mastery takes a while. (There are only 2 languages I know well enough to read a novel.)

Even though I only played organized football for 1 season in HS, decades later, I still know how to properly don the pads- which vary by position- do a form tackle, and a few other tricks. But even if I were currently in shape, my skills would pale in comparison to someone who played 4 years in HS or anything more than that.

Please get back to me once you tell me how that is actually achieved in the 3e system that's not better achieved by a more reducible complexity model, such as True20's class system.

As I've said elsewhere, my favorite system of all time is HERO, and I can build any D&D PC from any version in it. Much the same could probably be said of GURPS or M&M's W&W sourcebook.

...because they are classless systems.

Like the game though I do, IMHO, True20's classes are virtually devoid of flavor, to the point that the game might as well have gone the way of W&W. True20's classes are nothing to write home about.

You want 100% character cconcept modeling, don't look for it in a system with classes.

 

[Dannyalcatraz]

Combine that with the fact that you have to pay more feats to swap your limited number of class powers...

That ticks me off to this day.

And did you notice that some of the multiclassing feats don't actually qualify you for the new class' Paragon Paths? What were they thinking?

 

[Aldarc]

They are- for D&D. How mechanics in other RPG systems- however closely related they may be to D&D- is irrelevant to the irreducibility of a class within D&D.

Doesn't this make "irreducible complexity" a meaningless concept?

if you think you've scored some kind of point here, you don't know me.

I didn't say it was bad, I said it was inefficient and slow. That is not a value judgement, it is a plain vanilla mechanical assessment. If that is how you want to build your PC, more power to you.

I have a PC (in a 3.5Ed campaign on hiatus) who is a PHB Sorcerer who wears Scale Mail and uses Maul, and has done so since level 1- efficiency is not my motto in PC design.

The problem is that the 3e system rewards you for dipping into fighter, as that is far more efficient, for those abilities (and more) rather than through feats and the like.

Nope- he got some intensive training at one point in his life, and mastered the basics. His BAB still won't increase like a dedicated warrior, though.

I think you are severely underestimating how much you get by briefly dipping into "the basics" of some classes due to how front-loaded a number of these classes were.

As I've said elsewhere, my favorite system of all time is HERO, and I can build any D&D PC from any version in it. Much the same could probably be said of GURPS or M&M's W&W sourcebook.

...because they are classless systems.

Like the game though i do, IMHO, True20's classes are virtually devoid of flavor, to the point that the game might as well have gone the way of W&W. True20's classes are nothing to write home about.

You want 100% character cconcept modeling, don't look for it in a system with classes.

Now you mistake me. We agree here, and that is largely the point that I am making. I want a system that doesn't have 100% character modeling. I have repeatedly stated that want classes[ to be meaningful as classes and not as "what I dip into for particular abilities" to model my character. The latter approach is far better serviced with classless or generic class systems. Unfortunately, I see a tendency in the treatment of 3e-style classes and multiclassing to gravitate more towards the latter, which is why I am opposed to it.

 

[Dannyalcatraz]

Doesn't this make "irreducible complexity" a meaningless concept?

No- any concept has to be examined within its proper context. An RPG class is irreducibly complex only within the context of the context of that RPG. Compared to a classless system from another game, you can break any class into its component parts. But within its native system, the class cannot be broken down any further than its initial bundle.

The problem is that the 3e system rewards you for dipping into fighter, as that is far more efficient, for those abilities (and more) rather than through feats and the like.

So what?

I think you are severely underestimating how much you get by briefly dipping into "the basics" of some classes due to how front-loaded a number of these classes were.

Nope. Since I first picked up the game in 1977, 80%+ of my D&D PCs have been multiclassed.

So I'm pretty clear on what you get- or don't get- when you multiclass a PC. Some classes- mostly the martial ones- you get quite a bit: a HP boost, a BAB bump, a host of weapon proficiencies and one or more armor & shield proficiencies, all of which are evergreen in their utility, even though they may be overshadowed by supernatural abilities.

In contrast, the casters don't offer quite as much with a dip: a useful spell or two (which may not be useful as the party levels), possible special abilities like Turning (which only improve with class levels) and so forth. You will be able to use certain magic items, though, the one benefit that is truly an unending font. But its value depends upon what gets handed out in the campaign, since a dipper won't be a crafter caster.

I want a system that doesn't have 100% character modeling. I have repeatedly stated that want classes[ to be meaningful as classes and not as "what I dip into for particular abilities" to model my character.

Assuming multiclassing exists, dipping will be an unavoidable consequence unless the actual multiclassing system is fairly restrictive.

I said that I thought Hybridizing was a close second to 3.5E's take on things, but even with that respect, it fails to model many of the PC concepts I come up with. It won't even handle 1Ed stuff, where I favored things like Ftr/MU/Th characters. So while I think it's superior in many ways, Hybrid multiclassing fails as a way to support my preferences. It's simply too constrained.

 

[UngeheuerLich]

3rd edition multiclass is fine. But not in the context of 3rd edition math.
A simple change like:
- base caster level introduction
- the +2 bonus for the good save is only applied once
- fractions are counted in
immediately made it work a lot smoother!

When math and class design is done well, 3rd edition multiclass is very good indeed.
The only thing which needs to be considered is how you start with a multiclass character on level 1!
3rd edition had multiclass rules for that (level 1/2 in the DMG) I´d like to see that too! It was also useful for starting below level 1

 

[Viktyr Gehrig]

And did you notice that some of the multiclassing feats don't actually qualify you for the new class' Paragon Paths? What were they thinking?

My theory? Third Edition and their "Back to the Dungeon" aesthetic was an attempt to go back to Gygax AD&D from the campaign setting and metaplot days of Williams AD&D... notice how all your Clerics had the same spell list again, Demons and Devils in the Monster Manual, Barbarians and Monks in the PHB, no kits (until really, really late in the cycle), and so on and so on.

4e was their attempt to go back to Classic. Simplified rules, simplified archetypes-- four types of what you do and three types of how you do it-- simplified skills. They were trying to get back to the "pick up and play" simplicity of Classic. They kept the distinction between races and classes and they kept feats, because those had become iconic, but they really pared down on a lot of the things they felt had become excessive in 3.5e.

Which really makes me curious about where they're going to go with 5e because aside from the obvious need to come up with some kind of "new hotness", it seems like they're going to have to swing back to some point between the two.

 

[Aldarc]

No- any concept has to be examined within its proper context. An RPG class is irreducibly complex only within the context of the context of that RPG. Compared to a classless system from another game, you can break any class into its component parts. But within its native system, the class cannot be broken down any further than its initial bundle.

That's what I thought, and that's what makes it a meaningless term, as the point in which a class is "irreducibly complex" "within the context of the context of that RPG" is always debatable and usually ends up meaning "whatever the designers printed in the book."

So what?

It rewards class-dipping, which devalues the meaningfulness of class selection and classes as archetypes.

Nope. Since I first picked up the game in 1977, 80%+ of my D&D PCs have been multiclassed.

That's fantastic, but 3e has only been around since 2000, and that's the only system I'm concerned with right now. Out of curiosity, what problems do you have, if any, with the 3.5 system of classes and multiclassing?

Assuming multiclassing exists, dipping will be an unavoidable consequence unless the actual multiclassing system is fairly restrictive.

I said that I thought Hybridizing was a close second to 3.5E's take on things, but even with that respect, it fails to model many of the PC concepts I come up with. It won't even handle 1Ed stuff, where I favored things like Ftr/MU/Th characters. So while I think it's superior in many ways, Hybrid multiclassing fails as a way to support my preferences. It's simply too constrained.

What if you made multiclassing fairly restrictive, but made the classes broad and flexible enough to create a wide array of character concepts? Especially consider how D&D Next is incorporating the use of "themes" that provide a lot of the flavor concepts that often tied in with classes.

 

[Dannyalcatraz]

That's what I thought, and that's what makes it a meaningless term, as the point in which a class is "irreducibly complex" "within the context of the context of that RPG" is always debatable and usually ends up meaning "whatever the designers printed in the book."

Its not meaningless at all.

In some languages, the base unit is a broad concept, reflected in the way the language is written. In other languages, the base written unit may be a word or syllable.

In the first kind of language, trying to reduce a written symbol to a single word in another language is essentially impossible. Within the context of that language, that written symbol is irreducibly complex. It cannot be broken down further. Some things are thus not expressable within that language without creating new words & corresponding symbols or combining old symbols into increasingly complex ones.

In contrast, languages of the latter kinds are able to break concepts down more into component parts, allowing for greater precision in communication. Advantages of that sort are big reasons why Latin, then French were the languages of diplomacy before English became so ubiquitous.

It rewards class-dipping, which devalues the meaningfulness of class selection and classes as archetypes.

There is nothing in the game forcing you to multiclass- the archetypes remain solidly simulated. You want to play an archetypal barbarian, don't multiclass out of barbarian.

Besides, how many literary or mythological figures would be universally statted out as single-classed in D&D terms, anyway?

Out of curiosity, what problems do you have, if any, with the 3.5 system of classes and multiclassing?

Within the context of D&D? Nothing major, really. The only blips are:

1) not being able to start off as a multiclassed PC at level 1 without going to the gestalt rules (which seem a bit gonzo to me). Pretty minor, really.

2) too many PrCls that are mechanically some kind of Base Class+. While this is a nasty problem, it is not the multiclassing system itself that causes it, but rather poor PrCl design.

3) too many PrCls that are "fixes" for casters losing casting potency by multiclassing, which I think is a feature of the system. If you're a caster dabbling in other classes, you shouldn't be as good a caster as your more single-minded contemporaries. Flexibility comes at the price of excellence.

Compared to Fantasy HERO? There is a certain lack of flexibility in PC design, surprise, surprise.

What if you made multiclassing fairly restrictive, but made the classes broad and flexible enough to create a wide array of character concepts?

As with anything, the devil is in the details. I can discuss past mechanics because I've seen exactly what I can and cannot model in 1Ed, 2Ed, 3Ed, 3.5Ed and 4Ed. I won't know what I can model in 5Ed until I see the rules.

If I can't model PCs that interest me, 5Ed will not interest me.

Especially consider how D&D Next is incorporating the use of "themes" that provide a lot of the flavor concepts that often tied in with classes.

Can't speak to themes since we're not using 'em in 4Ed and I haven't seen anything about their 5Ed incarnation.

 

[Primal]

The latest chat transcript includes a quote that they're "shooting for" 3e multiclassing.

Conceptually, 3e multiclassing is ideal. The math was problematic, but entirely fixable.

I don't like system mastery and I didn't like 3E multiclassing; and for me it wasn't just about math. With 3E-style multiclassing, your PCs end up usually being either "gimped" or monsters with horribly broken combos (in my 3E/PF games I've even denied multiclassing).

Since themes are already strongly tied to your background, IMO Pathfinder archetypes or 4E hybrid PCs might be the best way to do it (although I wouldn't mind 4E-style feat-based multiclassing, either). Return to playing Wiz 4/Rog 4/Fgh 4 in a group of 12th level PCs? No thanks...

 

[Hassassin]

Class selection is not meaningful if your first level is rogue because that provides you with 24 more skill points at first level than starting out as a fighter.

This is one of the things PF fixed with their skill system. I use it in 3.5.

Full hp for first level is also annoying. Maybe something like roll + 3 would work, but I haven't tried it.

 

[Aldarc]

Its not meaningless at all.

Now apply it to D&D. Your idea of "irreducible complexity" is far more vacuous of a term and subject to debate. Let's take standard PHB1 3e or Pathfinder classes, for example. Do you see any case where the classes are not "irreducibly complex"?

There is nothing in the game forcing you to multiclass- the archetypes remain solidly simulated. You want to play an archetypal barbarian, don't multiclass out of barbarian.

But do they? What makes an archetypal "barbarian" different from say a fighter with a barbarian theme package and rage as an optional combat feature/feat?

Besides, how many literary or mythological figures would be universally statted out as single-classed in D&D terms, anyway?

It depends on how many classes were available and how flexible character customization is in a given system.

If I can't model PCs that interest me, 5Ed will not interest me.

Then you clearly hate D&D and you should just play some other system like HERO instead. Joking.

 

[BobTheNob]

I thought multiclassing was at its best in 2e (and thats not saying much). It wasnt perfect, but it worked.

3e's idea was interesting, and when I first read it I almost fell out of my chair thinking how awesome it was. Then I played 3e, tried to multiclass and regretted trying. I now think 3e's was a great concept, but didnt work in practice.

Then 4e...what multiclassing? A feat that lets you swap a power...big deal. A multiclass paragon that let you pick power from another class and was strictly inferior to other paragons? Really? Hybrids as a last minute after thought which had trouble with Synergies and MAD for a large number of combinations? Eeep. IMHO 4e multiclassing was s singular disaster.

I really do want them to come up with something cool for multi-classing. I really, really do. It just means so much to me that after all these years they can find a way of modelling the guy that walks multiple paths and fits nicely into the game. Who knows, it might happen

 

[Dannyalcatraz]

Now apply it to D&D. Your idea of "irreducible complexity" is far more vacuous of a term and subject to debate. Let's take standard PHB1 3e or Pathfinder classes, for example. Do you see any case where the classes are not "irreducibly complex"?

It's not vacuous at all.

A 1st level Ftr has things that simply cannot be gained via feats or other means short of multiclassing- besides the freebie feats, they get a bonus feat, a BAB higher and more accellerated in rate of increase than all but other warrior classes, larger HD. The Cleric gets Turn Undead & full divine casting. The wizard & Sorc have full arcane casting. The Rogue has the highest rate of accumulating skill points. Monks get special treatment for their unarmed attacks. Other classes have abilities that are likewise available by taking levels in those classes (some of which are similar to or identical to those in the PHB classes).

None of which are available any other way than via multiclassing. The class is thus the base, irreducible unit for gaining those abilities in D&D's system.

(Don't know PF well enough to comment.)

But do they? What makes an archetypal "barbarian" different from say a fighter with a barbarian theme package and rage as an optional combat feature/feat?

Just because there are two roads to a destination does not mean one is less good or beautiful- they're just different.

As I said before, I don't know jack about themes, and AFAIK, there is no Feat that grants rage.

If there is an ACF that grants rage, that still makes it a feature of a class.

And the fact that there may be more than one way in a game to make a "barbarian" doesn't mean the archetype in the Barbarian class is dead. It just means you can now make more than one kind of barbarian.

Then you clearly hate D&D and you should just play some other system like HERO instead. Joking.

Taking your joke seriously for a second, this actually gets to part of what I'm talking about when I say that 4Ed doesn't feel like D&D to me, but rather, feels like a completely different FRPG.

Whenever I play a FRPG, I try to make my PCs fit within the game. Even though I'm a "concept first" PC designer, I don't design my PCs the same way for D&D as I do for Fantasy HERO, which is different from Earthdawn, which is different from Talisantha, which is different from Harn, which is different from GURPS, etc.

It's not that I have a strict set of guidelines that I refer to and say, "Nope, that's a Harn PC, I can't make her in D&D." It's a bit more subconscious than that...but the results are plain as the character sheets. The PCs I play in the various FRPGs generally have a distinctive feel, like they were made by different people.

And the 4Ed PCs just look...different...from everything that went before.

 

[LordArchaon]

I'll try to state my opinion. It's difficult, because it's a "liquid" opinion, even if Multiclassing is one of my top priorities for the game (all my characters have been MCs more or less), and I usually have very clear ideas on game design...

I'm going to say 4e Hybrid is my favorite type of MC, especially when coupled with additional MC feats.
The good thing about 4e Hybrid, that I think could be maintained in 5e, is that you can choose at any level if you want to get your power from class A or class B. The bad thing is that many classes lost too many features and you could buy just one or two with Hybrid Talent, but that's fixable IMO.

I'm a bit worried about "shooting for 3e-style multiclassing", because even if they maintain 4e-style MC-like feats to get some minor features of other classes, if you make the true multiclassing system in the form of "take a level in X", there will always be the problem of having to loose something.
True, the math of saves and attack bonuses is fixable.
True, you should lose something when multiclassing.
But still, I can't see 3e-style multiclassing being as flexible as 4e-style hybridizing where at any given level you go "Ok, should I get a Fighter exploit or a Sorcerer spell?"... If they manage to solve this problem, with a solution that I'd call "Level Equivalence" or something, then I'm ok with 3e-style. But then again, if they do solve it, will they be able to retain the difference between, say, a Fighter 4th/Sorcerer 7th and a Sorcerer 4th/Fighter 7th? My guess is that the level should just put a cap on the number of different features and available ones, not on the power of each of them. This, for a spellcaster, would mean that your caster level or equivalent mechanic would still be based on your level, and damage of your spells as well. It could also mean that maximum spell level could be also based on level. What you'd loose is number of spell slots, spells known, special features.

Practical example of my ideal 5e multiclassing:

Fighter
1 Basic Maneuvers, Challenge, Favored Weapon
2 Battle Stances
3 Fighter Tactics (Defender, Striker, Controller, Leader)
4 Weapon Specialization, New Maneuver
Features = approx. 2x class level

Sorcerer (thread about my ideal 5e sorcerer)
1 First Spell-seed, Basic Forms, Arcane Fundamentals
2 Elemental Halo
3 Arcane Surge
4 Second Spell-seed, New Form
Features = approx. 2x class level

Fighter 3 / Sorcerer 1:you get 2 features from either Sorcerer or Fighter, up to level 4.
Choosing Weapon Specialization and Elemental Halo gives same power in melee as Fighter 4 plus additional synergistic feature; choosing First spell-seed and Basic Forms gives nearly full-fledged Sorcerer casting, but a Sorcerer of the same level would have two spell-seeds (basic elements to form spells), and more forms, and the encounter Arcane Surge to cast powerful spells. And you'd loose Weapon Specialization.
-> You always loose something, but you can choose what to loose and what to gain, and the total character level would be the cap.

Now example with Wizard, which is more problematic since it has spell-slots. (There would still be classes that are more or less difficult to multiclass, I guess. Wizard could be a difficult one)

Wizard
1 Cantrips, Basic Spell, Level 1 Spells (2)
2 Level 1 Spells (3), Ritual Casting
3 Level 2 Spells (3/1)
4 Second Basic Spell, Level 2 Spells (4/2)

Now, each slot should be considered a different feature, given the power... or not? Let's see what happens following the previous model: we should get approximately 8 features. But we end up having 10 if we count each slot as different. Let's see the different results with the previous example:

Fighter 3/Wizard 1: you get 2 features from either Fighter or Wizard, up to level 4.
Single Slots are features -> Level 2 Spells (2) [these are two features, maximum level]
Multiple Slots from one level are single features -> Level 2 Spells (0/1), Level 2 Spells (1/2)

So I guess we should count single slots as class features, or you'd end up getting spell slots pretty fast multiclassing as Wizard.
An Ideal Fighter 3/Wizard 1 could get a single Level 2 slot and a Basic Spell (at-will).

Now another problem: class proficiencies. I'd say a single proficiency per level taken into second or third class. Only your first class gets all the proficiencies. What does this mean? That the Fighter 3/Wizard 1 would only learn how to use a Wand, not an Orb, if they're different implements. Or could need to get the "Spellbook proficiency" first, just to be able to use his spell slots. On the other hand, a Wizard 3/Fighter 1 looking for some better armor, would only get Leather armor at first, and would get Chain (or Hide) only when Wizard 3/Fighter 2. Note that he/she would likely advance spell slots of his/her highest level only. So the choice wouldn't limit actual power, but a lack of 1st level spell slots could limit versatility and durability, if wanting to take Fighter features at all (could always want Fighter levels just for HPs and proficiencies...)

EDIT: Another possibility to balance out and decide what features you can take by multiclassing would be giving some things requirements. If we go "Weapon Specialization I" and "Weapon Specialization II", it would mean that to get WS II you'd need WS I first. You could still get both with just one level of Fighter if your total level is right, but you'd have to invest everything on it, you couldn't just jump to II simply because your total level allows it. Same goes for spell slots but always counting just 1 of them at a time. We have "Spells I (3 slots)" and "Spells II (1 slot)" and to take that level 2 slot we'd need to take a level 1 slot too, making the choice a bit more difficult. The simple act of adding numbers to a class feature would imply that to take "number II" you'd need "number I" first. If wanting to be creative, we'd simply add multiclassing requirements under each class feature, and state that the Sorcerer's Arcane Surge needs a Spell-seed and Basic Forms, while the Elemental Halo would require just a Spell-seed. Things as Ritual Casting, Basic Spell or Arcane Fundamentals would be poach-able without requirements...

EDIT 2: To make things really granular and controlled with this multiclassing system, we could also state that you can always only have as many class features in a given class as (class level in that class x 2). You could always retrain in order to keep your power level in row with your character level. A Fighter 6/Wizard 1 could still have access to the spell slot level of a Wizard 7, but it would be only one slot, with the other slot being of the inferior level. The character would have attained this by retraining first level slot into second level and second level into third level and so on. You'd still lack the ability to cast many spells. You'd get that only by advancing Wizard more.

 

[AlioTheFool]

Problematic is an understatement. IMHO, 3e multiclassing devalued classes, turning them into packages you dipped into for min-maxing abilities instead of actual class archetypes.

This is my biggest issue with the anti-multiclassing argument. Not everyone likes playing archetypes. Some people want to create their own stories, rather than trying to replicate Drizz't, Raistlin, Gandalf, or whoever.

Personally, I never play archetypes. I'm no fan of "typical" wizards. My bards don't sing, they're more jester-like. My fighters are deep thinkers, my rogues don't steal.

There is room in the game for both archetypes and for those who reject them. I wrote a blog post last week about this. Basically, you make classes so that specific features are only available if you continue to grow in that specific class. If you multiclass, you can acquire varied features, but not "advanced" ones. I believe this is the direction WotC is going to take. I wasn't at DDXP, so I can't say for sure, but this is my gut feeling.

4e turned classes into straightjackets, mandating that if you chose to be a fighter, you would advance as a fighter every level you gained, forever. Not that I don't see your point, but the 3e mentality is a better place to start, even if it does need improvement.

Agreed.

Personally, I found 3.5Ed's version of multiclassing to be the best incarnation of the overall mechanic in D&D. Still my fave.

I found 4Ed's Feat-based system quite the straightjacket as the sole form of multiclassing. As one option among others, though, I thought it was a good idea. Still, though, I see room for improvement. The overall form of the multiclassing feats should be standardized: gaining a skill and an AW power as an E power- both chosen by the player, rather than the designers- seems a pretty solid choice for a dabbler. In addition, because of my personal penchant for playing Jacks of all Trades ('cause I am one), I found the "single MC feat" restriction for all but the Bard to be overly restrictive as well. Let those who want to dabble do so freely- consider what they're giving up in the context of the game.

I found Hybrids to be decent, a close second ton3.5Ed's take. I didn't necessarily care for which particular class abilities they chose for the classes' Hybrid forms, though.

I agree with the majority of this post. I'm pretty anti-4E style. I despised Hybrids. I agree that multiclassing shouldn't make you "better" than everyone else, but in order to do that, 4E makes you feel like you're "worse" than everyone else, which isn't good either.

I am confident the team will make multiclassing in D&DNext balanced enough that whether you multiclass or not, you feel like your character is worth playing in the group.

I don't like system mastery and I didn't like 3E multiclassing; and for me it wasn't just about math. With 3E-style multiclassing, your PCs end up usually being either "gimped" or monsters with horribly broken combos (in my 3E/PF games I've even denied multiclassing).

Since themes are already strongly tied to your background, IMO Pathfinder archetypes or 4E hybrid PCs might be the best way to do it (although I wouldn't mind 4E-style feat-based multiclassing, either). Return to playing Wiz 4/Rog 4/Fgh 4 in a group of 12th level PCs? No thanks...

I wouldn't mind seeing a system that allowed themes to be layered upon your character to round it out. It could be done very well, and I touched on that myself, though very briefly. Themes were one of my favorite additions to 4E, as they allowed a better opportunity to "multiclass" (though that wasn't their intended purpose) than feat-based or hybrid multiclassing. Their one downfall was "you can take one" which was a continuation of the disappointment of hybrids and feat-based style.

 

[Aldarc]

This is my biggest issue with the anti-multiclassing argument. Not everyone likes playing archetypes. Some people want to create their own stories, rather than trying to replicate Drizz't, Raistlin, Gandalf, or whoever.

I'm not against multiclassing; I'm against 3e-style multiclassing that all too easily encouraged class-dipping that provided large benefits with little downsides, thereby causing what I felt was class dilution. If you just want to create a free-form character with little regard for class, there are better systems out there for that.

 

[LordArchaon]

I compiled and (I hope) polished my ideas for a good Next multiclassing, leaving it 3e-styled like the devs want... It's getting good feedback on twitter, hope you can comment too!
Blog at Wizards Community: "My Ideal Next Multiclassing"

 

[Gryph]

Nope- he got some intensive training at one point in his life, and mastered the basics. His BAB still won't increase like a dedicated warrior, though.

The basics of many things can be learned and retained for a lifetime. I know enough of a few languages to keep myself fed, sheltered, and ask for help- but mastery takes a while. (There are only 2 languages I know well enough to read a novel.)

Even though I only played organized football for 1 season in HS, decades later, I still know how to properly don the pads- which vary by position- do a form tackle, and a few other tricks. But even if I were currently in shape, my skills would pale in comparison to someone who played 4 years in HS or anything more than that.

This is why I prefer 1e/2e or 4e's Hybrid classes. If I am trying to model that kind of early, lifetime training I want it reflected in my character from level 1. If I have to dip into multiple classes while levelling just to vaguely arrive at my initial character concept, it feels like a frustrating kludge.

When 3e multi-classing worked for me was when, for in story reasons, a character changed his focus and switched classes. Much like the old 1e dual classing rules. Five levels of Fighter followed by levels of Cleric was an ejoyable part of 3e, for me. Fighter2/Rogue4/Wizard3/Prc4 gives me the same kind of reaction that Aldarc has posted in this thread.


As an aside, I totally agree about feat based mutliclassing, too restrictive and too expensive for the one concept it covered reasonably well (representing lifetime training).

 

[Dannyalcatraz]

This is why I prefer 1e/2e or 4e's Hybrid classes. If I am trying to model that kind of early, lifetime training I want it reflected in my character from level 1. If I have to dip into multiple classes while levelling just to vaguely arrive at my initial character concept, it feels like a frustrating kludge.

OTOH, like it though I do, i have to say where Hybridizing fails utterly is one place I spend a lot of time: PCs who are not dabblers or 50/50 guys, but (non-Bardic) Jacks of all trades.

Which is why I like it...but not as much as 3.5Ed's rules.

 

[AlioTheFool]

I'm not against multiclassing; I'm against 3e-style multiclassing that all too easily encouraged class-dipping that provided large benefits with little downsides, thereby causing what I felt was class dilution. If you just want to create a free-form character with little regard for class, there are better systems out there for that.

It's not about free-form classes. I understand your concern, but I firmly believe that with proper planning and effort, the team has design space to make 3E-style multiclassing work. Obviously it couldn't actually be 3E-style, but that idea of picking and choosing pieces to make your whole character.

I compiled and (I hope) polished my ideas for a good Next multiclassing, leaving it 3e-styled like the devs want... It's getting good feedback on twitter, hope you can comment too!
Blog at Wizards Community: "My Ideal Next Multiclassing"

Great post! I like that system quite a bit. It's along my own lines of thinking, but you put more detail into the specifics. Well done.

OTOH, like it though I do, i have to say where Hybridizing fails utterly is one place I spend a lot of time: PCs who are not dabblers or 50/50 guys, but (non-Bardic) Jacks of all trades.

Which is why I like it...but not as much as 3.5Ed's rules.

I'm big on playing bards, but under 4E, the need to burn feats just to begin to feel like you were a Jack-of-all-Trades was crippling. Playing a straight-up bard gave no feeling of varied ability.

I don't know how close they truly are in play, but my favorite 3.x derivative of multiclassing was that found in Neverwinter Nights (the video game). It was a lot along lines of the system LordArchaon posted, where you need to take multiple levels of a class to gain access to various features of the class.