The Gaming Den

Multiclassing in D&D is basically rubbish on the systemic level, because it works on the system of "trade the 6th level of Fighter for the 1st level of Wizard", so if you ever get anything level-appropriate from it it's completely by accident. Various feats and PRCs, like Tashalatora and the Mystic Theurge, have been created to help with this, but most of them have their own issues and anyway they're all patches.

I'm trying to find a solution for this, and the one that looks best right now is some variety of partial gestalt, but that concept gets a bit screwy when you take more than two classes. If anyone has any ideas on that subject, I'd love to hear them.

That's not really the topic though, my main question is this: What is the appropriate balance point?

When you're a Fighter/Wizard, obviously you can't be allowed to be as good a Fighter as the single-classed Fighter and as good a Wizard as the single-classed Wizard in your party. (Pretend for a moment that Wizards aren't already better Fighters than the Fighters, and that Fighter and Wizard are equally valuable roles.) In fact you can't be allowed to be either, because if you give up too little in any area of competence compared to what you gain, then even if you're not strictly superior in all ways to your single-classed cousins multiclassing is basically free power and everyone will do it. Conversely if you lose too much from each area of competence, you won't be level-appropriate at anything, which means multiclassing is pointless. Obviously, we don't want either.

Ideally, as a Fighter X/Wizard X you'd be a slightly worse Fighter than the Fighter 2X, and a slightly worse Wizard than the Wizard 2X, making up the difference and becoming an equal contributor partly with versatility and partly with synergy between your two skill sets. Also ideally, it should be possible to emphasise one or the other in your build (Fighter X+Y/Wizard X-Y) without that being a trap option. Realistically, that can't possibly be a valid choice for all values of X and Y, but it should be for some.

So how much is "slightly", do you think? How should a Fighter/Wizard stack up in Fighting and Wizardry class features compared to his single-classed friends in order to be neither a trap nor a no-brainer option?

My initial thought is that he should be two levels behind. (In class features of course, not HD.) Two guys at level X-2 = one guy at level X, right? So a Fighter 18//Wizard 18 is theoretically equal to a Wizard 20 or a Fighter 20. Which, true, the implementation does not bear out in the least. But is there any reason why the theory doesn't hold?

What you mean by "Multiclassing" is "Multiclassing in and out of caster classes". The value of multiclassing is cut entirely along caster lines; Casters never have a reason to leave their caster class (Barring situations like the aforementioned Mystic Theurge, and even then you get heated debate over that), whereas from a power standpoint, non-caster classes don't really have much reason NOT to multiclass. Hell, one of the problems with the fighter is the fact that it doesn't really give you much reason at all to stick around, a level or two of just about any non-caster class makes the Fighter a better fighter with virtually no sacrifice to their own flavor.

As to your actual question, "How do casters mutliclass without presenting their nuts for kicking?", that's a fight that goes back all the way to 3.0. Lots of solutions have been forthcoming, and none of them are very good. Dig around the history of this place and you'll probably find one or two.

Well, that's true, but only because, as you say, noncaster classes very rarely get class features at high levels that aren't just more numbers on some class feature they got at low levels (usually not even level-appropriate numbers, at that.) And of course, you can get numbers everywhere and they mostly stack, so multiclassing works out for them if you do it with a little thought.

But I wasn't thinking only about casters, except insofar as "casters" is synonymous with "people who get meaningful class features". Think of Binders or Totemists or something, rather than Fighters.

What exactly do we gain by having an ultra-modular system where people can choose the exact ratio of how much Class X and how much Class Y their character is instead of writing up a bunch of Final Fantasy XI class-subclass pairings and writing fixed advancements for each of those? I mean, as it is, there is a difference between Fighter 6 Rogue 4 and Fighter 7 Rogue 3, and especially Rogue 1 Fighter 7 Rogue 2, but are those the kinds of differences we actually care about preserving?

Well, people complain about the Mystic Theruge for this exact reason. An MT will be 3 levels behind in both classes (barring Ur Priest shenanigans). This keeps them at least one spell level behind at all times, which is a pretty big deal.

So, that being said, being two levels behind keeps you one spell level behind. Until you hit 19th level (17//17), you will be feeling the effects of this.

The idea seems nice, but it will probably never work in practice without a full rewrite of the system.

Multi-classers would need to get abilities which are level-appropriate. If the shit you're getting is not meaningful at your level, multi-classing is not going to happen. But they need to give up something, to prevent multi-classed characters from rofl-stomping everything. Thus, they must sacrifice endurance and versatility on their newly gained powers.

Binder is actually an easy example: You base "maximum vestige level" on full character level, or binder level +½ off-binder levels.

Similarly, multi-classed Incarnum users would need access to high-level chakras and boosts to essentia capacity, but the number of binds etc would stay locked to class level.

It's more difficult for casters. But something like "spells/day based on wizard levels. Then, look up [character level, or wiz level + ½ non-wiz] on the same table, and add 0 spells/day of the aditional spell levels noted in the table." Your caster level would be...something or other.

I don't think it's currently possible to make a balanced multiclassing system with DnD as is.

At least not with the premise of keeping it simple. And not with out considerably changing the power level of either mundane or arcane.

I think your best bet is to stick with pre-written prestige classes that fill the role of multiclassing. With maybe a buff-heavy Ftr/Wiz PrC for a melee gih, or a a defensive/controller heavy list for a Wiz/Ftr PrC. (Sticking with the F/W example)

A system like this would work better if every class granted something like two level appropriate abilities each level. Then, to multiclass, you just pick one ability of each class, giving you level appropriate abilities each level, but with a different mixture.

I think designing and playtesting such a system would be crazy hard, but it's about the only way I can see keeping features level appropriate for multiclassed characters. I suppose you could flat-out mandated multiclassing, so you didn't have to worry about their power level compared to straight-classed PCs.

the problem with multiclassing beyond two classes is the same problem with gestalt in some cases. adding another class add more abilities than a level of existing class may add.

the best way to do more than 2 classes would be a classless system wherein the abilities are picked for the new pseudo-class.

one of the reason MANY of the people i knew to play 2.5 was that player's options offered a "build your own class" with character points (CPs). then all those abilities roughly became feats in 3.x, when combined with the new skills that were formerly NWPs (i just call all the mess "skills")... you already had class augmentation that would give you access to things outside of your class, while letting you not have to mess with the multiclassing.

though IIRC, feats didnt offer much outside of the original class for 3.x to dip into other class functions which destroyed the idea behind the CP system.

what you might want to do to make the "slightly gestalt" is work CPs back into 3.x. turn the starting class functions into pseudo-feats, and let a new classing system let you pick a "class package" that has the most of the originating class, and then pick feats from various classes to build your own class. so a wizard gets most of his wizard stuff, maybe missing two things but gets fighters HP.

this really is how ANY class outside of: fighter, wizard, cleric; comes to be in the game anyway. the other special strange class abilities just come from need of filling in holes where you didnt give it enough of the fighter and enough of the cleric to make the paladin as full a class as either of its parents.

Something I've been Kicking around for a while: does anything actually break if xp is a currency instead of a meter? Roughly like this:
- First level of each class costs 1000 xp, second level 2000, and so on. If you're a fifth-level wizard, it still only costs you 1000 xp to grab a level of fighter.
- No multiclassing penalty.
- Most numbers overlap, not stack. Select the highest BAB among your classes, the highest will save, etc.
- remove 4x skills at level 1.

Obviously, the CR system gets broken, but that's not a change. Also obviously, everyone eventually has enough magic to do basic identification, enough rogue to find traps, and enough fighter for martial weapon proficiency and a feat or two, but I don't think that's inherently bad.

I don't know how to make feats or stat increases work. I assume they're solvable, as long as there's nothing wrong with the system /fundamentally/.

Caster level accounting becomes a nightmare.

Schieremierhereeriehcher, the fundamental fact of the matter is that a fighter18/Wizard 18 is nowhere near equal to an 18th level fighter and an 18th level wizard, due to the fundamental facts of action economy- a fighter 18 and a wizard 18 will be doing one Fighty Thing and one Wizardy thing per round, where was a fighter18/Wizard 18 will only be doing one of either- while there are passive benefits of having both, and there is synergy between the various skillsets, the fundamental fact of the matter is that the action economy is still a crucial restriction on the power level of multiclassed characters.

Your caster level is just your highest level in any caster class. Your character level is equal to whatever your highest class level is, so a Fighter 3/Wizard 3 would get an attribute bonus if he leveled either class to 4, but does not get an attribute bonus if he takes Rogue 1. If the Fighter 3/Wizard 4 then levels up to Fighter 4/Wizard 4, he doesn't get another attribute bonus, because his character level was already 4.

I have no idea if this is actually balanced (the fact that Fighters and Wizards are involved makes me assume not), but it seems simple enough to be usable.

RobbyPants wrote:A system like this would work better if every class granted something like two level appropriate abilities each level. Then, to multiclass, you just pick one ability of each class, giving you level appropriate abilities each level, but with a different mixture.

I think designing and playtesting such a system would be crazy hard, but it's about the only way I can see keeping features level appropriate for multiclassed characters. I suppose you could flat-out mandated multiclassing, so you didn't have to worry about their power level compared to straight-classed PCs.

Legend? There is an thread here somewhere.

John Magnum wrote:What exactly do we gain by having an ultra-modular system where people can choose the exact ratio of how much Class X and how much Class Y their character is instead of writing up a bunch of Final Fantasy XI class-subclass pairings and writing fixed advancements for each of those? I mean, as it is, there is a difference between Fighter 6 Rogue 4 and Fighter 7 Rogue 3, and especially Rogue 1 Fighter 7 Rogue 2, but are those the kinds of differences we actually care about preserving?

What you gain is the ability to have multiple scaling abilities from a broad range of classes. At least, that is what you gain if you have your abilities scale in power to character level.

There are broadly speaking many ways to handle "multiclassing" as a concept. And the different methods are not only different mechanically, but also conceptually. But perhaps most disappointingly, a method that handles one concept does a shit job of handling another. These different goals you might have really fuck with peoples' attempts to make working multiclass rules - as nine times out of ten the different cooks lack a consistent vision of what they mean or want when they use the word "multiclass" at each other.

Fundamentally, Multiclassing can either be for the creation of hybrid archetypes like Mage/Thief or it can be for the advancement of characters whose archetype changes as they advance through the story - like Rand Al'Thor or FitzChivalry Farseer. These are not compatible concepts. AD&D presented them distinctly as Multiclass and Dual Class, but AD&D character advancement rules were shitty, confusing, and purposefully unfair - so using that as a template would be an example of an idea that is bad.

For the hybrid archetypes, yes you want something that is like Final Fantasy subjobs. Or like 4e's "hybrid classes" only not stupid somehow. If someone wants to play a hybrid warrior mage or something, they do not generally want to wait until adventure 28 to play one. They want it now, which means that your class hybridization or subclass system has to come online all at once (if not at first level, then at least at "name level", whatever that is for your game).

But for the evolving character archetype, the transitioning from apprentice wizard to master assassin is rather the entire point. For this setup, you want to provide character advancement choices at each advancement checkpoint - and that necessarily implies that you get abilities from one list and then get abilities from another list. For this mutliclassing concept, you have to have abilities scale to level individually rather than attempt to give out class packages that are level appropriate as a whole. This is because you have essentially no control over what ability sets characters will have at any particular level.

The actual process of evolution from wilderness scout to bone knight can be done by the accumulation of class levels until knight or bone overshadows wilderness or scout. Or by the accumulation of abilities until the player decides that it is time to trade one class template for another. Or by numerous other schema. Similarly, the method by which abilities stay level appropriate could be anything from characters getting a pool of stuff to shove into their abilities to scale them (like Incarnum, but hopefully not stupid) or by simple automatic scaling, or even without literal scaling at all provided that abilities all stack together somehow.

-Username17

I think there are only a few hybrids that people care about in D&D. They want an a fighter/mage and a thief/mage. They want Fighter/Thieves, mostly because those classes are individually too limited. They want to be able to punch things with the power of weeaboo and also cast spells. Sometimes they want a cleric/thief. That's... about it.

Most hybrid archetypes exist because D&D has too much role protection on shitty but thematically distinct mundane shticks. For example, if monk was an actual good class that didn't rely on ONLY I MAY PUNCH to justify its worthless existence then you could just opt into unarmed fighting on a cleric and then no one would care about Monk/Cleric anymore. Other than that all you need is about three new classes. Fighter/Mage, Thief/Mage, Cleric/Thief.

So I think the primary focus for multiclassing needs to be on letting characters change directions mid-campaign. You can address the need for hybrid archetypes just by writing better classes, but characters who want to change their advancement path mid-campaign actually need multiclassing.


(On the other hand if you did subdivide wizard into 4-5 different classes like a lot of people here want to do, I could see how you'd still need hybrids. Writing one gish class isn't too much to ask, but having to support Elemental Warrior, Deathknight, Warshaper, Firchliss, and so on would be too much.)

ModelCitizen: that may be the case for starting character choices, but if you give people the option to hybridize they're going to take it. Players are going to want to customize at name level as well as over the course of a campaign.

@Chamomile
Can't you just gain attribute bonusses based on xp?

And if multi-classing is horizontal growth instead of vertical, you will have a very limited amount of options.

Mask_De_H wrote:ModelCitizen: that may be the case for starting character choices, but if you give people the option to hybridize they're going to take it. Players are going to want to customize at name level as well as over the course of a campaign.

I'm confused. "Name level" is a grognardism meaning roughly a high level character. For most classes Name Level was level 9. What does customizing at level 9 have to do with hybrid options you can only choose at char gen?

Schleiermacher wrote:Ideally, as a Fighter X/Wizard X you'd be a slightly worse Fighter than the Fighter 2X, and a slightly worse Wizard than the Wizard 2X, making up the difference and becoming an equal contributor partly with versatility and partly with synergy between your two skill sets.

I think that's basically incorrect. As a Gish, you actually want a character whose spells work as just well as the Wizard, and whose fighting trick works as well as the Fighter's. Like a Cleric or Druid. Otherwise you're a 3e Bard and fuck that.

Really, no one wants to play a Bard. Not in 3e. Well, basket-weavers, but ignore their opinion in a study of how shit should work.



How multiclassing should work then, in 3.5, is your Cleric/Wizard should have half spell slots for each class, your Warblade/Wizard should have half the number of Wizard slots and also half the number of Warblade slots. So the point of being a single-class Wizard is you are more Wizard and less something not-Wizard.

Note multiclassers are still totally screwed over by MAD. That's probably fine, it acts as a limited form of niche protection.

The bard should have full casting with limited spells at Src/Wiz table levels, and full BAB, and his bard and skill stuff, like a multiclass, because he is.

If you're a Warblade 8/ Wizard 18, you only lose half the number of spells you had at Wizard 8 (round to worst, so lose 2/2/2/1 and keep 2/2/2/3/4/4/3/3/2).

Tri-class, lose 2/3 your spells, Wiz 7 specialist triclass keeps 1/1/1/0 (plus bonuses for Int).

If you insist on adding Fighter or Rogue levels, lose nothing, no one cares, you already lost half your Wizard spells for that, you fool. Maybe take away some skill points or feats or some damn thing, but only if you insist on people not being Fighter/Rogues for free. Monks, Rangers, and Paladins all have slot-like things to cut in half, as do the upgrades in Bo9S.

Schleiermacher wrote:Ideally, as a Fighter X/Wizard X you'd be a slightly worse Fighter than the Fighter 2X, and a slightly worse Wizard than the Wizard 2X, making up the difference and becoming an equal contributor partly with versatility and partly with synergy between your two skill sets.

I think that's basically incorrect. As a Gish, you actually want a character whose spells work as just well as the Wizard, and whose fighting trick works as well as the Fighter's. Like a Cleric or Druid. Otherwise you're a 3e Bard and fuck that.

Really, no one wants to play a Bard. Not in 3e. Well, basket-weavers, but ignore their opinion in a study of how shit should work.



How multiclassing should work then, in 3.5, is your Cleric/Wizard should have half spell slots for each class, your Warblade/Wizard should have half the number of Wizard slots and also half the number of Warblade slots. So the point of being a single-class Wizard is you are more Wizard and less something not-Wizard.

Note multiclassers are still totally screwed over by MAD. That's probably fine, it acts as a limited form of niche protection.

The bard should have full casting with limited spells at Src/Wiz table levels, and full BAB, and his bard and skill stuff, like a multiclass, because he is.

If you're a Warblade 8/ Wizard 18, you only lose half the number of spells you had at Wizard 8 (round to worst, so lose 2/2/2/1 and keep 2/2/2/3/4/4/3/3/2).

Tri-class, lose 2/3 your spells, Wiz 7 specialist triclass keeps 1/1/1/0 (plus bonuses for Int).

If you insist on adding Fighter or Rogue levels, lose nothing, no one cares, you already lost half your Wizard spells for that, you fool. Maybe take away some skill points or feats or some damn thing, but only if you insist on people not being Fighter/Rogues for free. Monks, Rangers, and Paladins all have slot-like things to cut in half, as do the upgrades in Bo9S.

ModelCitizen wrote:

Mask_De_H wrote:ModelCitizen: that may be the case for starting character choices, but if you give people the option to hybridize they're going to take it. Players are going to want to customize at name level as well as over the course of a campaign.

I'm confused. "Name level" is a grognardism meaning roughly a high level character. For most classes Name Level was level 9. What does customizing at level 9 have to do with hybrid options you can only choose at char gen?

I was using that as a not-shit covered peasant tier character, and assumed that was first level. I don't really give a shit about grognardisms.

Here is my current thinking: let multiclass characters function at the same level as straight class characters, just with less endurance or options. So with a Wizard they might only get only half their spells per day and maybe lose a few of their class feats. A cleric gets half their spells per day, half their turning, half their domain power uses etc. I think this approach would work with Warblades but it has trouble with Fighters, but lets be honest, Fighters aren't really a functional class in 3.5. Maybe tack on some other small penalty to balance out the versatility.

So your Fighter 20//Wizard 20 gets 9th level spells and can contribute in a level appropriate manner at all levels, just not as often. To me it seems like having only 3 9th level spells vs 6 9th level spells is a pretty big restriction. A Wizard 20 // Cleric 20 can break reality about as often as a straight classed counterpart but is MAD so will be rocking lower save DCs and a few less spells per day.