Li Shenron
Legend
I think this is an interesting idea. It's not actually that different from the 3e approach... or is it?
D&D 5E Multiclass in 5E
- Thread starter Stormonu
- Start date
I think this is an interesting idea. It's not actually that different from the 3e approach... or is it?
Hmmm. There are many options, and each has its flaws.
Here's a potential solution which hasn't been mentioned yet:
Mix certain AD&D dual class rules with 3E multi-class rules, as follows:
* You must have greater than X in a primary stat to gain levels in a new class
* Your XP is only awarded to one class at a time. (This could mean you assign XP from each combat to one class or the other, or it could mean that each time you level up, you choose which class will gain XP next.)
* You can alternate class each level, just dabble in one then stop, or have a complete change of heart. e.g. frfrfrfrfr, or ffffffrrfffff, or fffffffffrrrrrrrr
* You gain everything that class would normally give at that level, except HP which are only awarded to the class of the highest level.
This provides a lot more flexibility than AD&D multiclassing, without being as harsh as 3E multiclassing could be. It means that a fighter who picks up magic late can become a competent mage in a much more reasonable amount of time. i.e. spell caster classes mix better with non-spell casters than in 3E.
However, this method has problems too, particularly around front-loaded class abilities. I suggest including some implementation of Crazy Jerome's idea (5000XP to start multiclassing in the first place), to mitigate that problem. The more heavily front-loaded the class, the more XP you need to earn before you can pick up its first level. It would also result in low HP for evenly distributed classes, especially if they are higher level in their lower HD class.
Another alternative would be AD&D style multiclassing, but where you divide XP by one extra class.
Fighter/Mage divides XP by 3.
Fighter/Mage/Thief divides XP by 4.
This keeps them deliberately further behind the other characters to mitigate the high-level power issues, but does nothing to alleviate the low level power issues. HP would be divided by the number of classes (fighter wizard rolls both d10 and d4, adds con twice, then divides by 2)
Yet another option would be to simply create multiple XP tables and go with AD&D style multiclassing.
Single class = 1000, 3000, 6000, 10000, etc as per 3E
Two classes = 2000, 6600, 13200, 22000, etc (2200 * level)
Three classes = 3300, 9900, 19800, 33000, etc (3300 * level)
The point being that this works almost identically to AD&D where you divide xp evenly between your classes, but includes a bit of a penalty without being as drastic as my previous example. The exact numbers would obviously be tweaked with further thought and play testing. This set of numbers would result in a level 8 fighter, adventuring with a level 6/6 Cleric/Wizard, and a 5/5/5 Fighter/mage/thief.
A comparison for those not familiar; in AD&D 1E, an 8th level fighter would be with a Fgt6/M-U6 and a Clr6/M-U6/Thf6. So you can see that the 2200* and the 3300* has a similar or more significant impact than the exponential XP charts.
HP as per previous example (divide by number of classes).
All of these methods have a common flaw - you don't level up at the same time as the other players. Individual tastes will determine if that is a problem or not. For my games; it wouldn't matter in the slightest.
Another random thought; what if you skipped 1st level when picking up a new class? That way the 'front loading' of proficiencies etc can be ignored, while still allowing you to gain features from the new class. Hmmmm.
Here's a potential solution which hasn't been mentioned yet:
Mix certain AD&D dual class rules with 3E multi-class rules, as follows:
* You must have greater than X in a primary stat to gain levels in a new class
* Your XP is only awarded to one class at a time. (This could mean you assign XP from each combat to one class or the other, or it could mean that each time you level up, you choose which class will gain XP next.)
* You can alternate class each level, just dabble in one then stop, or have a complete change of heart. e.g. frfrfrfrfr, or ffffffrrfffff, or fffffffffrrrrrrrr
* You gain everything that class would normally give at that level, except HP which are only awarded to the class of the highest level.
This provides a lot more flexibility than AD&D multiclassing, without being as harsh as 3E multiclassing could be. It means that a fighter who picks up magic late can become a competent mage in a much more reasonable amount of time. i.e. spell caster classes mix better with non-spell casters than in 3E.
However, this method has problems too, particularly around front-loaded class abilities. I suggest including some implementation of Crazy Jerome's idea (5000XP to start multiclassing in the first place), to mitigate that problem. The more heavily front-loaded the class, the more XP you need to earn before you can pick up its first level. It would also result in low HP for evenly distributed classes, especially if they are higher level in their lower HD class.
Another alternative would be AD&D style multiclassing, but where you divide XP by one extra class.
Fighter/Mage divides XP by 3.
Fighter/Mage/Thief divides XP by 4.
This keeps them deliberately further behind the other characters to mitigate the high-level power issues, but does nothing to alleviate the low level power issues. HP would be divided by the number of classes (fighter wizard rolls both d10 and d4, adds con twice, then divides by 2)
Yet another option would be to simply create multiple XP tables and go with AD&D style multiclassing.
Single class = 1000, 3000, 6000, 10000, etc as per 3E
Two classes = 2000, 6600, 13200, 22000, etc (2200 * level)
Three classes = 3300, 9900, 19800, 33000, etc (3300 * level)
The point being that this works almost identically to AD&D where you divide xp evenly between your classes, but includes a bit of a penalty without being as drastic as my previous example. The exact numbers would obviously be tweaked with further thought and play testing. This set of numbers would result in a level 8 fighter, adventuring with a level 6/6 Cleric/Wizard, and a 5/5/5 Fighter/mage/thief.
A comparison for those not familiar; in AD&D 1E, an 8th level fighter would be with a Fgt6/M-U6 and a Clr6/M-U6/Thf6. So you can see that the 2200* and the 3300* has a similar or more significant impact than the exponential XP charts.
HP as per previous example (divide by number of classes).
All of these methods have a common flaw - you don't level up at the same time as the other players. Individual tastes will determine if that is a problem or not. For my games; it wouldn't matter in the slightest.
Another random thought; what if you skipped 1st level when picking up a new class? That way the 'front loading' of proficiencies etc can be ignored, while still allowing you to gain features from the new class. Hmmmm.
MoonSong
Rules-lawyering drama queen but not a munchkin
that is moot, from what I can read on the class descriptions, you don't get any proficiencies from multiclassing. Skipping first level if anything makes classes more frontloaded
Crazy Jerome
First Post
It's the "progresses independently" part that makes it unlike 3E. So if a fighter pays the 5,000 premium to be able to start getting wizard XP, he can advance as fighter or wizard, each using the same chart, but not stacking the levels.
I didn't talk about it in the original, but I was also thinking that the players/table would have some leeway in how they split the points. Perhaps you must put at least 10%, 20%, 25% in each class every time you get points, but the remainder you can split or put all into one class, as you choose. How high you set this minimum percentage effectively caps how many classes you can take--and reduces your flexibility the more you take. Then the table can also adjust the 5,000 XP premium. By adjusting those two variables, you dictate a lot about how much multiclassing is worth it, and how far it can be pushed.
One of the nice things about 3E multiclassing is that every player gets to set the mix of classes how they want, but the 3E way isn't the only way to handle that. And like the 1E/2E way, the 3E way suffers in that it isn't easy to adjust, as each level in any class makes the next one harder. Using my proposed system, if you want to put a 50/50, 60/40, or 75/25 mix in two classes, you'll get results that about match what a casual player would expect.
BTW, the big disadvantage of my proposal for traditional D&D play is that once you get near the level cap, it becomes inordinately advantagous to spread out. This becomes especially true (and happens sooner), if there are a lot of levels to take. The more levels, the less the premium means at the upper end. Now, this doesn't bother me that much, because if some heretofore single-class Fighter, at level 15, decides it is worth it to knock out five levels of wizard, even with the premium, then I see that as more rounding out the character than adding raw power--especially since there is a limit on how many classes he can so take. But I must acknowledge that such a character growth dynamic is a bit off in traditional D&D, and more like what you would get with Hero or GURPS or Runequest or such. It's nothing that can't be house ruled around (e.g. force early multiclassing for any character that wants it), but the house rules are not very elegant.
Edit: Now reading Zustier's post above, I see he anticipates me. But I'll let my stuff stand as a different wording, slightly different take on some of the same concepts and issues.
Certainly I would treat stacking elements as gestalt--i.e. you get the hit points, weapon attack bonus, etc.--each of any single class you choose. It has to be that way in this system, because of how the levels progress. If taking levels in a wizard is knocking a bit off your fighter XP, and thus costing you a level or three in your best class for hit points, then the effect needs teeth. Otherwise, there is no cost. Not having exponential character advancement tables, a 50/50 character will fall further behind than 1 level versus a single class. In fact, this system works better in Next than it would in any previous D&D edition, because the bounded accuracy means that such a character might still only be penalized a point of attack bonus. The slow advancement of numbers means that you can let such level differences stand.
Last edited:
I will say that any multiclass system they create should not use XP.
5e is meant to be modular, and XP is a commonly altered or thrown away mechanic.
If you look at 3e, you technically had xp losses for multiclassing outside of favored classes, but I've never used or seen that restriction used in all the 3e games I played....but I'm sure other people did.
So there solution has to be more core to the game, XP won't do the job.
5e is meant to be modular, and XP is a commonly altered or thrown away mechanic.
If you look at 3e, you technically had xp losses for multiclassing outside of favored classes, but I've never used or seen that restriction used in all the 3e games I played....but I'm sure other people did.
So there solution has to be more core to the game, XP won't do the job.
Tovec
Explorer
Not that I disagree Stalker0 but the 3e XP penalty was relatively minor. I know some games that used it - those are the same games that used XP at all - but most didn't. BUT, at least with the [MENTION=1544]Zustiur[/MENTION] 's last suggestion the penalties are not minor. The normal way is 1000 * level, which more than doubles to 2200 * level and triples for 3300 * level. Something this extreme can be accounted for even when you don't use XP - delay the character advancement by that same rate.
(I'm going to round off the 100's) A 2 class character with 12000 has 3 levels, a single class character has 4. The problems with this of course is that it makes multiclassing too valuable. If you only a level behind* per extra class, that is a lot of extra power for the sacrifice. Not to mention this assumes the classes are both the same level, instead of off-set in any number of ways.
Either way I think if you want to multiclass with XP that is fine, it can even translate over to non-XP games, but the multiclassing has to be costly enough to avoid those two problems I just said.
* I realize that you're not necessarily only ever a level behind per class, I was just making an example - one that applies more readily at non-XP tables.
Crazy Jerome
First Post
Do you mean here avoiding varying XP by class, ala 1E/2E as does it? Or do you mean using XP at all to balance multiclassing, even if the XP chart is consistent? The latter is a tough requirement, and I wonder how one would even begin to approach it in an elegant manner.
Stacie GmrGrl
Adventurer
My preferred method would be to go classless but that will never happen with D&D... it would take care of the multiclassing challenge though.
To summarize it to a more fundamental level: Multiclass balance should not adjust the rate at which a character levels.
Someone give this man a cigar. I couldnt, apparently I have given you too many.
This is exactly right. For our purposes, we scrapped XP tracking when 3e first came out, and good ridence. If XP tracking is required to enable multiclassing, Im going to cry.
Regardless, I distinctly dislike varying rates of advancement being the balancing/enabling factor for multiclassing just as much as I dislike it for handling class balance (remember the XP charts for the 2e- theif).
Crazy Jerome
First Post
My problem with that is that it appears to lock in the 3E fundamental mistake that every level is just as good as another any other level. That is, it equates taking a 1st level of barbarian (or sorcerer or whatever) as equally good for a 10th level character of any class mix, which is clearly false. (I don't think that the designers of 3E thought they were equally good. They knew some were traps. But the system pretends that the levels are equally good, because the character pays the same for it.)
If you know a way out of that which isn't a big mess to handle, I'd like to hear it.
Ridley's Cohort
First Post
My problem with that is that it appears to lock in the 3E fundamental mistake that every level is just as good as another any other level. That is, it equates taking a 1st level of barbarian (or sorcerer or whatever) as equally good for a 10th level character of any class mix, which is clearly false. (I don't think that the designers of 3E thought they were equally good. They knew some were traps. But the system pretends that the levels are equally good, because the character pays the same for it.)
If you know a way out of that which isn't a big mess to handle, I'd like to hear it.
The way out: the default should be class abilities key to character level rather than class level.
Therefore:
* A Fighter9/Paladin1 can Smite Evil once per day for +Cha/+10.
* A Fighter9/Wizard1 can cast 5 Magic Missiles with his 1st level spell slot.
* A Wizard5/Fighter5 can cast a 10d6 Fireball.
Now there may be tweaks necessary to soften the frontloading, but the basic model holds.
The fundamental design error in 3e multiclassing is that many (but not all) class abilities are tied to the class level. We should simply erase that factor. A 10th level character needs class abilities that a highly useful to CR 10 threats -- everything else is just flavor. I do not see a problem with a Wizard5/Fighter5 having one or two spells per day that every bit as good as the Wizard10 second string options.
Celestian
Explorer
Thanks for the link.
That is very unfortunate. Multi-class system in 3e was the worst version for me. I actually preferred the hybrid system introduced in 4e to the 3e system.
I will keep my fingers crossed and hope they go with the AD&D after all. 3e system is "dual class", not multi-class.
Just minor points, but
a. 3e had a variant where character could start at first level as 0/0 level multiclass, but you didn't get full benefits of both until second level
b. 1e also had training costs (although not everyone used them) and you had to spend time and gold to train in each separately for leveling (if you failed to train in a class when it was time to level, you could not level in it until you did).
My problem with that is that it appears to lock in the 3E fundamental mistake that every level is just as good as another any other level. That is, it equates taking a 1st level of barbarian (or sorcerer or whatever) as equally good for a 10th level character of any class mix, which is clearly false. (I don't think that the designers of 3E thought they were equally good. They knew some were traps. But the system pretends that the levels are equally good, because the character pays the same for it.)
If you know a way out of that which isn't a big mess to handle, I'd like to hear it.
Well, the impression that I got from something Stan! wrote in a d20M product was that the inequality you talk about was intentional and explained the difference between multiclassing in D&D and d20M. d20M was designed assuming characters would multiclass, because in modern times it is easier pick up new professions, take classes and pick up new training due to transportation, communication methods, proliferation of colleges, extension courses, seminars and other schools etc.
This is how I viewed it. Personally, I think ease of multiclassing is too easy in 3e. I think the training rules variant from the DMG should have been the default with longer training times.
Old martial artists in our own world talked about needing letters of introduction to go train with someone and it was rare to be accepted by a second instructor. There had to be special circumstances. Bouncing from instructor to instructor would not be accepted. Heck caucasians learning Chinese martial arts was frowned upon until the late 60's or early 70's here in the US. Now, add to all of that that it takes 3-10 years to get a black belt (and most likely your first level D&D monk is the equivalent of 3rd-5th level in one that teaches chi kung or its equivalents in other styles and you are talking a long period to gain first level as a monk.
If you know a way out of that which isn't a big mess to handle, I'd like to hear it.
Some kind of out of class progression could help. They did this with Bo9S where levels in other classes counted as half levels in Bo9S classes.
For 5e they could do something similar. For example, when you take your first level Fighter you get a fighting style and a CS die. Out of class levels could allow access to that fighting styles higher techniques, but no more CS dice.
As the presented classes so far all share (or in the case of the Wizard will share) a kind of subclass (Combat Style, Scheme, Pact, Domain, Origin, Wizard School), you could go with this method for all classes.
Tovec
Explorer
My problem with that is that it appears to lock in the 3E fundamental mistake that every level is just as good as another any other level. That is, it equates taking a 1st level of barbarian (or sorcerer or whatever) as equally good for a 10th level character of any class mix, which is clearly false. (I don't think that the designers of 3E thought they were equally good. They knew some were traps. But the system pretends that the levels are equally good, because the character pays the same for it.)
If you know a way out of that which isn't a big mess to handle, I'd like to hear it.
This brings up a good point. How good or effective should a multiclass character be? How much of a new class should get added to an existing class?
Should a level 5 fighter/level 1 wizard be the same as a level 3 fighter/level 3 wizard?
I don't have any answers here. I think that we need to look into what we want from multiclassing before we start getting bogged down in the details of what is wrong with previous versions.
I said it before, but I prefer multiclassing to be something that represents additional training or refocusing later in a character's life. I look for multiclassing to bring new abilities or additional focus to an existing character. But I'm not necessarily looking to become a spell-blade. It is usually going with similar classes or at least ones that add a very specific effect. Fighter-Ranger or Paladin-Monk or whatever would often apply here before Fighter-Wizard.
I know others of you want something more akin to dual-classing or gestalt or hybrids, where you advance as both a fighter and a wizard at the same time. I like this too but it isn't necessarily what I'm looking for primarily with multiclassing. If for no other reasons than this, I think we need to look at 2 different versions of multiclassing.
Crazy Jerome
First Post
When I've seen this tried before, it typically comes out as that "big mess to handle" I mentioned. However, I haven't seen the Bo9S version, and thus don't know how clean it is. Like half ranks in skill ranks, this is something I prefer to see cleaned out of a new system from scratch.
I suppose in your 5E suggested variant, if the list of "subclasses" is small enough, you can simply show progression in each one in a kind of master class. Or more likely, show the "off level" progression in each class. Hmm, still not sure that comes out clean ...
Crazy Jerome
First Post
I take it from this that d20M (odern?) has a cleaner version of the original? How does it work that makes it cleaner?
Similar Threads
