Multiclassing via Feats. Thoughts?

[Puggins]

To me, that implies that Half-Elves may get a single "free multiclass ability" as a racial power.

... or they might just get a bonus feat, which sorta helps with multiclassing.

As an aside, I'd be really bummed if most of the list of 4e feats turns out to be crammed with "Wizard Training: Ray of Frost" and what-not. This isn't me speculating, mind you- I have nary a clue.

 

[Mercule]

I think converting 3E characters to 4E will be an exercice in futility as in close to impossible, especially multiclassed 3E chars and chars with PrCs.

lvl 1-20 vs lvl 1-30, totally different multiclassing, no PrCs anymore, totally different magic system, mostly different magic items. Combine all of those together and it becomes pretty much mission impossible, especially when you think many (would most be too strong here?) 3E characters are multiclassed.

Which is probably why they said, "We ain't even gonna try to publish guidelines for conversion."

My original stance was that I'd finish up my current 3.5 game, then switch. I'm starting to think eyeballing a conversion to 4e might breathe new life into the game, ease my DMing workload, and better fit with my campaign goals (which include god-slaying, epic tier stuff).

Everything looks like it'd translate fine except for the gnomish druid and the soulknife. The gnome will be in the MM. Our druid is mainly a healer and summoner, so cleric w/ ranger feats might work. And the soulknife might work as a rogue with a grandfathered weapon or a warlock or wizard ability (touch-only eldritch blast?).

 

[katahn]

Sure, its reminiscent of a problem with 3e multiclassing, but its not 3e multiclassing. It doesn't work the same way, and more importantly, what you're producing isn't the same thing as it was before. The idea of 90/10 doesn't necessarily make sense anymore because you're not mixing levels like you were before. The outcome of "8 levels wizard, 2 levels fighter" isn't what you're getting. You're adding class abilities to a different class.

I wasn't talking in levels, I was talking in terms of % of abilities. If one looks at one's character sheet and 90% of the abilities come from class A and 10% of them come from class B then one is a 90/10 A/B multiclass. Hopefully that's a bit clearer.

The outcomes, "fighter with 5 abilities from a wizard" or "wizard with 5 abilities from a fighter" are inherently different from one another.

The "virtual feat" insight is important for people who want to multiclass, but it doesn't have strict game balance implications.

It's more like this:

Class A starts with the equivilent of 10 virtual feats representing its level 1 class abilities. For a fighter these would include abilities like marking, armor proficiencies, weapon proficiencies, number of starting skills and so on. If Class B doesn't start with the same number of virtual feats, then going for an A/B multiclass that is close to 50/50 requires starting off as A, otherwise the A/B that starts off as B will be comparitively underpowered at any given level.

Now maybe no one in their right mind will want to do a 50/50 A/B (or close to it) multiclass, but the principle remains pertinent. If a feat/ability is a measure of relative power, and all abilities can be taken 1:1 with any feat, then the implication is each individual ability is intended to be roughly equivilent in the amount of power/effectiveness it adds to the character. If this is the case, and one class starts off with more of these abilities, then class balance is already broken at level 1 and that isn't what I'd call a good thing.

 

[Plane Sailing]

Um, what about when you want to play a concept that isn't covered by any existing single class? Like gishes?

Well, they've already identified that there are some variations which don't work so well, hence the planned 'arcane defender' class.

To be honest, I'd be saying to someone who couldn't make their character concept with the feat based multiclassing "come up with another concept then".

Of course, I started D&D in the days when your character concept was "do you want to play a fighter, a cleric, a magic user or a thief" and didn't get any further than that, mechanically

 

[katahn]

Well, they've already identified that there are some variations which don't work so well, hence the planned 'arcane defender' class.

To be honest, I'd be saying to someone who couldn't make their character concept with the feat based multiclassing "come up with another concept then".

Of course, I started D&D in the days when your character concept was "do you want to play a fighter, a cleric, a magic user or a thief" and didn't get any further than that, mechanically

AFAIK the "arcane defender" will be in the FRCS and will be called a "Swordmage". You would also predate me in D&D... I started when "elf" was a character class.

 

[Lacyon]

Class A starts with the equivilent of 10 virtual feats representing its level 1 class abilities. For a fighter these would include abilities like marking, armor proficiencies, weapon proficiencies, number of starting skills and so on. If Class B doesn't start with the same number of virtual feats, then going for an A/B multiclass that is close to 50/50 requires starting off as A, otherwise the A/B that starts off as B will be comparitively underpowered at any given level.

One thing to keep in mind - it might appear that class A is getting a bunch more virtual feats than class B because class A gets more abilities in total, but in terms of actual power level they may still be getting the same amount.

The key is to make going "all-in" cost the same number of feats regardless of which way you do it. This may mean granting several of Class A's abilities in one feat as a package deal, for instance.

 

[Jer]

Um, what about when you want to play a concept that isn't covered by any existing single class? Like gishes?

I think it comes down to what your concept of a "fighter/wizard" is. There are two concepts I've seen played since the early days of D&D that fit fine with this type of multiclassing. One is the "fighter who knows some magic" - or as we used to call it back in the day "The Elf". Primarily a melee combatant who takes some "utility spells" to take some pressure off the wizard. The other is the "wizard who can use a sword" - or "Gandalf". Primarily a spellslinger who likes to be able to wear a bit of armor and use an actual weapon in a crunch. Both of these concepts of a "fighter/wizard" would be fine with this kind of multiclassing. (Plus the reduced gap in hit points and combat bonuses between classes will make that wizard who knows a bit about fighting much less likely to end up a smear on the rug when he decides to pull out his sword. And has there been a reveal yet on whether and how much a wizard/warlock is penalized for wearing armor and casting spells in this edition - or if there's even a penalty at all anymore?)

OTOH - I personally like the idea of a fighter/wizard that uses his magic to supplement his fighting style, and for that concept this kind of multiclassing probably isn't going to work well. That's an argument for a completely different class - one that balances the fighter/wizard aspects and makes the powers available more focused on the concept. And given what I've seen that's what I would do - come up with a new class with powers specifically attuned to the concept I had (or wait for the Swordmage to come out and see if it matches what I want in said concept).

So what's your concept for a "fighter/wizard" that you don't think this kind of multiclassing would cover? And is it actually easier to model that concept in 3e than the way that's been suggested here?

 

[Cadfan]

Class A starts with the equivilent of 10 virtual feats representing its level 1 class abilities. For a fighter these would include abilities like marking, armor proficiencies, weapon proficiencies, number of starting skills and so on. If Class B doesn't start with the same number of virtual feats, then going for an A/B multiclass that is close to 50/50 requires starting off as A, otherwise the A/B that starts off as B will be comparitively underpowered at any given level.

But it doesn't necessarily work that way. Look, if a Fighter has 7 "virtual feats" that a Wizard would need to take to get a Fighter's stat block, and you want to multiclass your Wizard to a fighter in that style, then you're going to be at level 12 when you get to the 50/50 mark not counting powers or efficiency losses due to stat allocation. If a complete 50/50 split isn't practical, then its a poor metric for judging the balance of a level 1 character.

Now maybe no one in their right mind will want to do a 50/50 A/B (or close to it) multiclass, but the principle remains pertinent. If a feat/ability is a measure of relative power, and all abilities can be taken 1:1 with any feat, then the implication is each individual ability is intended to be roughly equivilent in the amount of power/effectiveness it adds to the character. If this is the case, and one class starts off with more of these abilities, then class balance is already broken at level 1 and that isn't what I'd call a good thing.

Again, that's not necessarily true.

For one, not all abilities add the same amount to every character who takes them. Easy example- the Fighter has a "virtual feat" over the Ranger in that the Fighter can wear heavy armor. But an archery based Ranger who takes Armor Proficiency: Heavy Armor is actually worse afterwards, because heavy armor doesn't permit the use of his high dexterity bonus, and provides overall less protection than light armor + dex. In that case, the fact that the Fighter can wear heavy armor and the Ranger cannot doesn't mean the Ranger is worse. Now that's an extreme example, but it is also one that applies to a Wizard with a high Int and a Fighter in plate armor.

Second, even when the numerical bonuses are identical, characters don't automatically get the same use out of an ability. For example, a Wizard with lots of ranged attacks doesn't get the same value out of a high AC and high HP in comparison to a Fighter who fights in melee reach.

The real metric for the balance of a multiclassed character is in what is actually possible and practical under the ruleset. In this case, the real test looks to be this: "Is a [class] with a [different class] trick or three a viable character in comparison to other possible characters, including non multiclassed characters?

 

[katahn]

In an A/B multiclass if A has more virtual feats but we group a bunch of them together so the collection of little feats is buyable with a single feat, and if in doing so we've achieved a parity where it takes X number of feats to mimic a given level of A or B, then all that's happened is we reach the point I said we should be at... Namely that there is no advantage to starting off as A or as B in an A/B multiclass.

Regarding the point of abilities not being desirable in a particular A/B combination, that shouldn't alter the point that A and B should be equally viable starting points for an A/B multiclass that strives to be 50/50 (or close to it). Sure a fighter/ranger might not *want* to wear heavy armor under most circumstances, but if a player wants to have a fighter/ranger character they shouldn't have to spend an extra feat if they started off as a ranger and not a fighter versus starting off as a fighter and adding ranger.

All of this might be moot if it is basically impossible to multiclass to the extent of 50/50 just via feats, and the closest one can get is something like 60/40 or 75/25. Under that circumstance someone who has a concept of a 50/50 A/B type looks to be out of luck until they can select their paragon paths and elects to take the B class then and retrain their prior feats possibly.

 

[Cadfan]

All of this might be moot if it is basically impossible to multiclass to the extent of 50/50 just via feats, and the closest one can get is something like 60/40 or 75/25. Under that circumstance someone who has a concept of a 50/50 A/B type looks to be out of luck until they can select their paragon paths and elects to take the B class then and retrain their prior feats possibly.

The bold portion would be the key quote. And it goes further than that. Every bit of information we have so far indicates to us that 60/40 is not a good way to measure 4e multiclassing.

In short, Multiclass A/B is NOT the same as Multiclass B/A in 4e. Not in the sense that one is disadvantaged in comparison to the other, or that either A or B is a better place to start, but rather in the sense that they are simply not the same thing at all and comparing them is an apple to oranges comparison.