Personally, I think they need to go back to the TSR era rule for dual-classing (which is what multi-classing in this context is).  I.e., you need to have a minimum attribute before you can take that class.

You're a 3rd level fighter who wants to be a mage?  You need a minimum INT of 16.

I think that would help mitigate all the F1/Rng3/MU6/C1/Bar4 builds.

Quote from: Bill;680742I am open to multiclassing as long as it is slightly 'weaker' than single classing.

My logic is that multiclassing should be for charcater concept and not for 'being uber'

Agree in principle.

But 3.0/3.5/PF has eleven classes in the core rules.  The webpage http://www.wizards.com/default.asp?x=dnd/lists/class lists 175 core classes, even if some of them are repetitive.  

Why not make a list of core classes that covers most of the multi-class options.  It may not be perfect, but it is hard to find a lot of class ideas that could not be handled by a well thought out core class.

Fighter/Mage could be a Hexblade or a Duskblade.

Is there really a lot of daylight between an Urban Ranger and a Fighter/Thief?

I love multiclass characters so its good they're in. If nothing else, they're great for when there are only one or two PCs in the group.
OTOH, 3.x is about the worst system I can think of for multi-classing. I'd like characters to be able to have 2 or 3 classes, not 20, want to be able to actually start off with two if it fits the character, and it should offer more flexibility rather than bizarre and game-breaking combinations.  

You could more or less prove mathematically that 3.x multiclassing sucked since  (from the CR system) gaining 2 levels was supposed to double a character's power. If that's right then being a 10th level wizard is equal to being a 8th level fighter + an 8th level wizard (more like what the levels would've been for an AD&D fighter/magic-user) rather than being 5th/5th. Splitting the levels directly meant they had to invent a bunch of weird bandaid PrCs like Eldritch Knight or Mystic Theurge.

But the CR system was... not very good.

I like multiclassing, I like 3e multiclassing (yes, its holes are well-documented, but its main virtue is that it allows characters to have careers that cover a wide ground, which I like a great deal), and I care little for the sanctity of archetypes, but I was under the impression that the 5e backgrounds pretty much covered the ground that multiclassing does, which I am also fine with if that's true. So, whatever.

Quote from: Piestrio;680403How about we just not have multi-classing?

That's generally how I ran 3.x and it worked just fine.

Sure the system wanks and the like hated it but I didn't want them infecting my game anyway.

Win-win.

I agree with this. They could do some Specializations or whatever they Hell they are calling them in the newest version to scratch that itch just fine. In fact, it LOOKED like they were heading in that direction instead of putting in classes like Paladin and Ranger early on.

Quote from: Bloody Stupid Johnson;681030OTOH, 3.x is about the worst system I can think of for multi-classing. I'd like characters to be able to have 2 or 3 classes, not 20, want to be able to actually start off with two if it fits the character, and it should offer more flexibility rather than bizarre and game-breaking combinations.

Just so. 3.X multiclassing was a godawful system where you had to wait until 6th level-- or later-- for the mechanics to support your character concept and where you were rewarded for building illogical character combinations.  

I've been trying to fix it for years. The closest I've been able to come are systems where you get class features (but not numerical benefits) as class level + 1/2 non-class levels, with feat taxes, or using Gestalt rules with fixed progression and Level Adjustment per number of classes. Neither option is remotely satisfying.

The fact that D&D Next is going back to 3.X multiclassing-- variant or not-- instead of either AD&D multiclassing or some refinement of the 4e system is one of my major disappointments with it.