Conceptual value of Prestige Classes?

[Luccan]

In a level based game I don't think there's a substitute for actually taking a level in a class with a cool, longer and more exotic name than a base class could have. I don't think you can really replicate PRCs with subclasses (kit modifications) either, since those are tied at some level to hitting the same notes as the base class or else they become something completely different. Prestige classes and subclasses/variant classes/whatever you call them fulfill different roles and there's space for both.

Having specialized classes that are shorter than the main base classes and have certain requirements to enter is a very simple and common sense idea.

But there are two ways that 3e, especially in the early part of its run, did not execute well on that idea:
- Base classes with bad and few class features, making PRCs feel mandatory
- PRCs with piddling and over-specific class features, especially ones that were setting-specific and/or could have easily been replicated with minor "organization" bonuses.

Most of the objections in this thread seem to be about those execution failures, not about the core concept, which IMO is very sound. There were a lot of good PRCs in 3e as well.

Honestly, PRCs are such a simple idea that the only reason not to include something like them in a D&D-type game is if you want it to be very minimal and standardized. That's understandable for something like an indie game where you want it to be self-contained. Less so for the flagship of the genre.

I don't think PRCs are a necessary extension of a class and level system, whether or not subclasses and ACFs can fill the roll. It over-burdens the system by demanding short progression classes that can be entered at a variety of levels and are at least close to on par with any level they can be entered from. It's too easy to fall short or shoot past your goal. You could do it, but you need a tight grip on progression and every PRC needs to be considered with all possible base classes and other PRCs. The more you add of either, the more likely you are to fail. PRCs are a nice option in the right games, and one I'm glad to have in 3.5 (and its direct spin offs) where it's necessary. But you shouldn't need it any more than you need multiclassing; it's a class based game, if you can't persuade enough people to stay in a single class for every level to begin with, something has gone wrong.