If you're giving him a totally different set of classes and abilties in the heroic tier vs. the paragon tier...doesn't that kind of speak to the problem, build wise, of modelling him in 4e?
Not really. Building a character in a RPG isn't itself a simulation of the character's imagined developmental process within the fiction.
In particular, the build of a RPG character expresses the abilities/resources the character has to call on
now. The fact that, at an earlier time in his/her imagined life, the character had different abilities, doesn't matter to the modelling of the character's present abilities.
4e captures this via retraining rules. [MENTION=82106]AbdulAlhazred[/MENTION] suggested a way of using retraining to model Conan as a battle-rager fighter for the whole game, but I would start with ranger for skill purposes.
I mean you're basically saying I can model him if I break the rules between tiers.
All you have to do is rebuild him. That's not a big deal. Two of the PCs in my game were rebuilt at various points - the ranger into hybrid ranger-cleric at 6th; the wizard/invoker into an invoker/wizard at 15th. The builds themselves are all rules-legal (which is the contrast with the AD&D versions of Conan, the Knights of the Round Table, etc, which were not legal builds, involving impermissible race-and-class combinations).
My point is that, whatever version of Conan you want to play, there is a legal 4e PC build that expresses him pretty well.
Ironically, in old-school attempts at Conan, he'd be given Fighter and Thief levels and arbitrary special abilities, resulting in something you could never do as PC. Even when Gygax ranted against the Schwarzenegger Conan movie, then created a Barbarian class just to do Conan, it didn't capture him that well.
I agree on both these points - in AD&D it is very hard to actually build a legal character who will model Conan at any point in his fictional career; and the AD&D barbarian class doesn't really model Conan at all.
As a kid me and my friends loved them because they were pretty precise (we used them to settle those deeply intellectual playground arguments that kids have). Spiderman can lift a maximum of...blah blah blah.... Wolverine is definitely not as strong as the Hulk. It was the comics where things got wonky with different writers or story needs that ignored the canon in the handbooks.
When I talk about a RPG modelling a superhero, or a fictional character like Conan, I am not talking about modelling some arbitrary piece of fan-oriented product that is not itself an element of the fictional works.
In the context of Marvel heroes, the Official Handbook is not what I think of when I wonder what a character can do; I think of the comics they figure in, and particularly the most important and influential story arcs. For Conan, I don't think about someone's notional calculations as to how much he can lift, how far he can run, etc; I think of the REH stories in which he is the protagonist.
it goes for the handwavium of the comics and succeeds greatly at replicating those stories.
A RPG in which Wolverine, or War Machine, or Jean Gray, or whomever does the things that s/he does in the comics
is a RPG that has successfully modelled those characters. What other success criteria would there be?
He's not a deity, any more than any of the other "related to the gods" heroes are. He's magical, but he also fights with a sword at various times. Truthfully though, the dividing line between 'god' and 'hero' is pretty nebulous, Ahti for instance is both a warrior hero AND a sea god at the same time. This is really the problem with old myths, they just don't have the sorts of categories that RPGs do, particularly D&D.
This is an area where I think D&D can be quite flexible. Eg in 4e we have gods, primordials, slaad lords, demi-god PCs, etc - the line between god, hero and other mythic figures is pretty relaxed, and is mostly a matter of fiction rather than mechanics.
warriors who have some mystic power is a pretty common theme in legends and myths
I suspect that I am not the only person who finds D&D's spell mechanics are very particular way of handling magical abilities, that doesn't do a particularly good job of modelling mystical abilities in general. Even 5e recognises this, using a non-spell framework for druidical shapechange, monk's martial arts, etc.