Base and Prestige Classes

[dman11235]

I disagree with you on the base classes thing.  I think that all base classes should be tht 20+levels (20, then epic), but what needs to change is the design.  They should all be viable at all levels, and there needs to be less overlap and specialization in base classes.  PrCs should be the limited level ones.  And in fact, I don't think any PrC should have epic levels.  Basically, a PrC will modify some aspect of your character, and then when you finish the PrC you will remain modified and not need to take any more levels for the modifications to remain relevant.  If that makes sense.

[FlaminCows]

I disagree with you on the base classes thing.  I think that all base classes should be tht 20+levels (20, then epic), but what needs to change is the design.  They should all be viable at all levels, and there needs to be less overlap and specialization in base classes.  PrCs should be the limited level ones.  And in fact, I don't think any PrC should have epic levels.  Basically, a PrC will modify some aspect of your character, and then when you finish the PrC you will remain modified and not need to take any more levels for the modifications to remain relevant.  If that makes sense.

That's fine, and that's almost how it is already, but why? Why should each base class have 20 levels? What benefit does that give?

The position I have there is based on the idea that each class is, and should be, based on a certain concept. Sometimes it is specialised, while other times it is more general, but in the end a concept is a certain thing that could be defined within parameters and mechanics chosen to fit that definition. The thing is, some concepts are more high-powered or low-powered than others. Also, some concepts are more narrow than others. Both these things lend well to classes with varying numbers of levels.

[Amechra]

The solution is simple!

Just have a buncha Prestige classes that can be accessed at level 1; they can be stuff like Paladin, or Knight, or some other broad-ish thing that you couldn't stretch to 20 levels.

Then you could have your cake and eat it too!

[dman11235]

I disagree with you on the base classes thing.  I think that all base classes should be tht 20+levels (20, then epic), but what needs to change is the design.  They should all be viable at all levels, and there needs to be less overlap and specialization in base classes.  PrCs should be the limited level ones.  And in fact, I don't think any PrC should have epic levels.  Basically, a PrC will modify some aspect of your character, and then when you finish the PrC you will remain modified and not need to take any more levels for the modifications to remain relevant.  If that makes sense.

That's fine, and that's almost how it is already, but why? Why should each base class have 20 levels? What benefit does that give?

The position I have there is based on the idea that each class is, and should be, based on a certain concept. Sometimes it is specialised, while other times it is more general, but in the end a concept is a certain thing that could be defined within parameters and mechanics chosen to fit that definition. The thing is, some concepts are more high-powered or low-powered than others. Also, some concepts are more narrow than others. Both these things lend well to classes with varying numbers of levels.

What you described are PrCs.  With the way the current system is set up, base classes should be 20 levels.  A different system would lend well to what your talking about, and it's not a bad thing, just different from the current one.  It's certainly an interesting idea, I will give you that, and one worth exploring.

The classes that you're thinking of that don't deserve many levels?  Paladin, Knight, Bard, etc.?  They all should have been PrCs.  The only base classes should be ones that have base mechanics, like Wizard, Scout, Fighter, and Warblade.  Very general mechanics that can be defined more closely by feat selection, skills, and PrCs.

[FlaminCows]

I disagree with you on the base classes thing.  I think that all base classes should be tht 20+levels (20, then epic), but what needs to change is the design.  They should all be viable at all levels, and there needs to be less overlap and specialization in base classes.  PrCs should be the limited level ones.  And in fact, I don't think any PrC should have epic levels.  Basically, a PrC will modify some aspect of your character, and then when you finish the PrC you will remain modified and not need to take any more levels for the modifications to remain relevant.  If that makes sense.

That's fine, and that's almost how it is already, but why? Why should each base class have 20 levels? What benefit does that give?

The position I have there is based on the idea that each class is, and should be, based on a certain concept. Sometimes it is specialised, while other times it is more general, but in the end a concept is a certain thing that could be defined within parameters and mechanics chosen to fit that definition. The thing is, some concepts are more high-powered or low-powered than others. Also, some concepts are more narrow than others. Both these things lend well to classes with varying numbers of levels.

What you described are PrCs.  With the way the current system is set up, base classes should be 20 levels.  A different system would lend well to what your talking about, and it's not a bad thing, just different from the current one.  It's certainly an interesting idea, I will give you that, and one worth exploring.

The classes that you're thinking of that don't deserve many levels?  Paladin, Knight, Bard, etc.?  They all should have been PrCs.  The only base classes should be ones that have base mechanics, like Wizard, Scout, Fighter, and Warblade.  Very general mechanics that can be defined more closely by feat selection, skills, and PrCs.

The trouble with that is that PrCs, by definition, are classes that can only be entered after reaching certain prerequisites. That's basically the only distinction between a base and a prestige class. I wouldn't want Bard, Paladin, Knight, etc to be PrCs because they are viable concepts for a 1st level or other low-level character.

Another reason that I think base classes shouldn't be required to have twenty levels is that, with the way the current system is set up, many base classes should not be 20 levels. In my own experience, the game is much improved when multiclassing is the norm. Take the Fighter, for example. You list it as a class that has base mechanics, but it really doesn't. What it is used for is as a dip class for proficiencies and bonus feats, or a 9 level class as a Zhent Soldier. I thought about this, and decided that is actually fine. "Fighter" is not a big concept, it is just a person who fights. Getting a couple levels for combat feats and proficiencies is all it takes to turn a character into a Fighter.

"Person who fights" is a decent first-level character concept, so Fighter isn't a PrC, but Fighter shouldn't have 20 levels because there aren't really 20 levels worth of Fighter to begin with. Having the extra levels often gets people to casually keep leveling up as a Fighter, when really the Fighter is done both as a class and as an idea and they should be adding on something else. Not only because single-classing such a class makes a character that is mechanically weak for the level, but also one that is too conceptually narrow for the same.

[bobthe6th]

Yes, multi classing is a big part of 3.5 that is oddly frowned upon. Given that is a large part of the system, and you are supposed to do it to make character concepts no class covers...

Though I would add to my list the removal of character classes. A good generic class system, like this would make mores sense. Now you don't have to be an archetype or mess with multi classing. You just make the class you want.

[Amechra]

Why does everyone ignore my posts?

It's totally possible to have prerequisites of:

Special: Must be Lawful Good

For the Paladin. Hey, lookit that! You can hit that at level 1!

Seriously, just give requirements that only require stuff that your character already has when they pick their first class level, and it's fine.

After all, Paragon classes? They are actually PrCs that have "must be X race" as their requirements.

[FlaminCows]

You seem to misunderstand how PrCs work.

Prestige classes offer a new form of multiclassing. Unlike the basic classes found in the Player’s Handbook, characters must meet requirements before they can take their first level of a prestige class.

Thus prestige classes can never be taken on 1st level. That's kind of the entire point of a "prestige" class.

Additionally, why attach prerequisites where they aren't needed? Just so all less-than-20-level classes become PrCs? Hogwash. Just make a class anyone can take, unless there is a good reason to restrict it (and there usually isn't).

Then you could have your cake and eat it too!

But... the "only prestige classes can be less than 20 levels" isn't my cake, its his cake! Why would I want to have my cake but eat somebody else's?

[Amechra]

But the order of making a character has steps before selecting your class.

Thus you can fulfill those requirements "before" taking levels in the PrC.

The thing is, I agree with you; in fact, I think that PrCs should really just have level (and maybe alignment) prerequisites, trusting that, you know, having class features that require a base class class feature to use would guide people to the right PrCs.

But eh; if you eat his cake, there is more cake for you later, since you still have your cake. That, my friend, is logic.

[FlaminCows]

But eh; if you eat his cake, there is more cake for you later, since you still have your cake. That, my friend, is logic.

But I like my cake more, and if I eat his cake then my own cake would be less satisfying because I'm already full of cake. Better to let others keep their cake and eat my own, then make more of my style of cake so I can share with others if they, too, like my style of cake. 

[dman11235]

The trouble with that is that PrCs, by definition, are classes that can only be entered after reaching certain prerequisites. That's basically the only distinction between a base and a prestige class. I wouldn't want Bard, Paladin, Knight, etc to be PrCs because they are viable concepts for a 1st level or other low-level character.

Ah, see, here's the thing.  A Paladin is a melee-oriented Cleric.  So....it's a Fighter/Cleric, with levels in Paladin.  Those concepts are covered by other classes just fine, but are specialized versions of those classes.

For the rest of your post, that's just a problem with class design.  Not the system itself.  There's a difference, and actually, a good number of people have done a lot to fix these things, myself included.  Also see: EjoThims, RobbyPants, and Prime32.  And others.

On PrCs: PrCs do note introduce wholly new mechanics.  If you're introducing a new total mechanic, it should be a base class.  PrCs can modify mechanics.  Examples: spell casting.  A base class gets the casting, while a PrC modifies it.  This can take the form of MM feats, or altered spell lists, or altered casting methods, or buffed casting in certain areas, whatever.  PrCs should have the requirements that make sense for the class, but nothing taxed.  So, a PrC focusing on unarmed strikes?  Better have IUS as a prereq.  In fact, I'm of the opinion that no PrC should be without some non-level prereq.  It needs some skill, feat, or other ability in order to be taken.  You're prestigious, for crying out loud!

[FlaminCows]

The trouble with that is that PrCs, by definition, are classes that can only be entered after reaching certain prerequisites. That's basically the only distinction between a base and a prestige class. I wouldn't want Bard, Paladin, Knight, etc to be PrCs because they are viable concepts for a 1st level or other low-level character.

Ah, see, here's the thing.  A Paladin is a melee-oriented Cleric.  So....it's a Fighter/Cleric, with levels in Paladin.  Those concepts are covered by other classes just fine, but are specialized versions of those classes.

But if you do that, you can't play a Paladin from level 1. A Fighter/Cleric with levels in Prestige Paladin can only start detecting evil at will and smiting later on. That is kinda what I already said in the very quote you're replying to, actually. There's no reason that a player should wait until a later level just to specialise in something.

For the rest of your post, that's just a problem with class design.  Not the system itself.  There's a difference, and actually, a good number of people have done a lot to fix these things, myself included.  Also see: EjoThims, RobbyPants, and Prime32.  And others.

If you read it a bit better, you would have noticed that I was saying that it isn't a problem with class design at all, because there is nothing wrong with a limited concept having a limited number of levels. I am not ignorant of this forum's homebrew, dman. However, even in the homebrew fixes I see both concepts overstretched and different concepts mixed in the same class. In short, you still haven't answered my question: why should every base class have 20 levels?

On PrCs: PrCs do note introduce wholly new mechanics.  If you're introducing a new total mechanic, it should be a base class.

Actually, quite a lot of existing PrCs do introduce new mechanics. The distinction you just purposed doesn't really exist, nor do I see any benefit in making it so.

I'm getting the feeling we're talking past each other now. You're saying that base classes should introduce mechanics and have 20+ levels and prestige classes should modify mechanics and be a limited number of levels, because... well, because you say so. I'm saying that both types of classes are in most ways the same, and the only distinction between a base class and a prestige class is that a base class can be taken at any level.

[dman11235]

But if you do that, you can't play a Paladin from level 1. A Fighter/Cleric with levels in Prestige Paladin can only start detecting evil at will and smiting later on. That is kinda what I already said in the very quote you're replying to, actually. There's no reason that a player should wait until a later level just to specialise in something.

This is wrong.  See, you may not have "paladin" written on your character sheet, but you can call yourself a paladin.  Just because a cleric is a generic divine caster does not mean you have to be a generic divine caster.  You dictate your class, your class does not dictate you. You specialize in being a paladin by acting like a paladin.  Later on, you'll start getting abilities that help further define that.  Immediately, you're using things like spells.  See: Prestige Paladin, Prestige Ranger, and Prestige Bard.

For the rest: this is what I have observed in my time working with this game.  I've been around, I've seen the way things work, and I've seen the underlying mechanics and the reasons for them.  I wrote the Guide to Homebrewing with these concepts in mind, because that is how the game works.  If you want to have the game to work the way you proposed earlier, then go ahead and make it work that way, it's a good idea.  It's just not how this game was set up and intended.  Yours would be....D&D 3.5 alt.  You can really see where they were heading with Incarnum and ToB.  The base classes were pretty general, but viable all the way through, and introduced new mechanics (soulmelds, maneuvers, respectively).  Then the PrCs modified those, and helped specialize (improved the capacity of base features, focused on specific aspects).  When I say base mechanics, by the way, it not minor things, only big things like maneuvers and such count for this.  Things like, say, in ToB, the Sun Soul Ninja's healing/harming ability does not count as a new mechanic, it's modifying an unarmed strike.

Really, in conclusion, there's no real reason why they should be 20 level base classes and shorter PrCs (only I didn't say shorter...I said limited) other than: that's the way it works right now.  That's how the classes have been designed, and anything else would just make it weird and inconsistent.  If all of them were the way you described, then I'd be taking the opposite approach.  however, there is one good thing that comes from the long base classes and short PrCs:

When you have a base class with PrCs, you can identify as a base class...specialized in a PrC.  It makes for less clutter when making a character, because you aren't searching through as many combinations (although the prolific nature of the splats kind of undermines this a bit).  With 20 level base classes (keep in mind, it's 20 levels because that's the epic limit, which is an arbitrary cut off), you have a defined track should you so choose.  Otherwise, branch out, and as you level more options become available to you.  This is a very common element in game design, where complexity increases the further along you get.  Especially in RPGs.

[FlaminCows]

But if you do that, you can't play a Paladin from level 1. A Fighter/Cleric with levels in Prestige Paladin can only start detecting evil at will and smiting later on. That is kinda what I already said in the very quote you're replying to, actually. There's no reason that a player should wait until a later level just to specialise in something.

This is wrong.  See, you may not have "paladin" written on your character sheet, but you can call yourself a paladin.  Just because a cleric is a generic divine caster does not mean you have to be a generic divine caster.  You dictate your class, your class does not dictate you. You specialize in being a paladin by acting like a paladin.  Later on, you'll start getting abilities that help further define that.  Immediately, you're using things like spells.

But that's exactly it: you'll get spells, but you won't get the paladin abilities. You focus on what the Paladin shares with other classes, I focus on what makes the Paladin separate. Yes, you can cast spells right away, but the spells are not what makes the Paladin. There are even Paladin variants that don't cast spells at all! The abilities relating to how they hunt evil are what the Paladin is really about. There is a reason the Paladin only gets spells later on, after his courage and evil-detecting spidey sense and smiting and immunity to disease. To get the abilities that most define the Paladin last would be all backwards.

For the rest: [...]  It's just not how this game was set up and intended.  Yours would be....D&D 3.5 alt.

There's no real reason why they should be 20 level base classes and shorter PrCs other than: that's the way it works right now.  That's how the classes have been designed, and anything else would just make it weird and inconsistent.  If all of them were the way you described, then I'd be taking the opposite approach.

Absolutely. I wouldn't for a moment pretend otherwise. But take a look at the subject of the thread, and the subforum that this thread is in. "D&D 3.5 alt" is what this whole thread is about!

Not all of the designer's intentions were correct. They made many assumptions as well as ordinary mistakes. Every post in this thread points out rules that were created intentionally, and are the way it works right now because that's the way they were designed. This thread lists what the designers got wrong, from each individual's perspective.

however, there is one good thing that comes from the long base classes and short PrCs:

When you have a base class with PrCs, you can identify as a base class...specialized in a PrC.  It makes for less clutter when making a character, because you aren't searching through as many combinations.  With 20 level base classes, you have a defined track should you so choose.  Otherwise, branch out, and as you level more options become available to you.  This is a very common element in game design, where complexity increases the further along you get. 

In the "short-class" system I suggested as an alternative, all the things you listed are still there save one: the defined track. However, that is not something you need an actual 20-level class for. A defined track is something you could easily do as a sidebar, with a suggested multiclass combination. That's all it takes.

With the defined track provided to those who want it, you still have the rest: you branch out as you level, and complexity increases the further along you get. Just because 20-level classes are the way the designers did it, doesn't mean that's the best way to do so.

(click to show/hide)

Indeed, give a dude enough time and he'll make a whole new edition of D&D, step by step!

[linklord231]

The idea behind 20 level base classes is that the game is 20 levels long, and people should not be forced to multiclass if they don't want to.  Ideally, taking only monk levels would be just as viable as a Wizard 3/Master Specialist 2/Incantatrix 10/Archmage 5, and would be a simpler option for newbies or for people who just don't want to make a character that complicated. 
By making a base class that is less than 20 levels, you are forcing people to engage in a certain part of the system that they didn't have to before.
Also, it offends some people's sensibilities when you say "Ooh, sorry, you can't continue taking levels in that class you identified yourself as since we started playing.  Try one of these instead."

Just throwing this out there, since the "pro 20 level classes" side is kind of under-supported right now. 

[littha]

The solution is simple!

Just have a buncha Prestige classes that can be accessed at level 1; they can be stuff like Paladin, or Knight, or some other broad-ish thing that you couldn't stretch to 20 levels.

Then you could have your cake and eat it too!

I really like this idea but I would go with ACFs rather than prestige classes. Say you have a Paladin ACF for both cleric and Fighter that give different paladin centric abilities that stack with each other. That way you are still a paladin at level 1 and you have a control over exactly where your focus is going to be.

Fighter (Paladin) would give melee bonuses and defensive bonuses (immunities, saves) whereas Cleric (Paladin) would focus on spellcasting abilities and supernatural abilities. Having them both stack for caster level and BAB or something would be nice too.

A further abstraction would be using more generic classes, Warrior, Spellcaster, Skilled are the usual three. A paladin would be a Warrior with a Spellcaster ACF whereas a Cleric would be a Spellcaster with a Warrior ACF.

A final abstraction of this idea (one that I did some work on a while back actually) is using a tree based point buy system to build your character in a classless system and then granting special bonuses if the points spent in each tree were in a specific ratio, say 10 points in the melee tree and 5 points in the divine tree gets you the paladin class bonus (probably something like divine grace).

However that is straying way too far away from the d20 system

[Unbeliever]

I can readily understand the idea of "shorter" prestige classes as opposed to base classes.  Prestige Classes are supposed to be a bit limited -- they focus on a tighter concept.  Rogue and Fighter are considerably broader concepts than Duelist.  So, the idea of a Duelist, especially with those particular abilities, can be easily played out in 5 or 10 levels.

Think about the Argent Savant or Abjurant Champion -- pretty tightly focused PrCs that don't really need more levels. 

But, really, what hangs on these distinctions?  Some people, as others have noted, hate multiclassing and prestige classes.  But, these people are kind of missing the way the game actually works. 

[FlaminCows]

The idea behind 20 level base classes is that the game is 20 levels long, and people should not be forced to multiclass if they don't want to.  Ideally, taking only monk levels would be just as viable as a Wizard 3/Master Specialist 2/Incantatrix 10/Archmage 5, and would be a simpler option for newbies or for people who just don't want to make a character that complicated. 
By making a base class that is less than 20 levels, you are forcing people to engage in a certain part of the system that they didn't have to before.
Also, it offends some people's sensibilities when you say "Ooh, sorry, you can't continue taking levels in that class you identified yourself as since we started playing.  Try one of these instead."

I don't think multiclassing is actually any more complicated than single-classing, especially if you don't have to choose your classes yourself if you don't want to. Take for example the much-discussed Paladin. It is basically a pre-built multiclass: you start off all martial with extra damage and immunities and save bonuses (among other evil-hunting abilities), then you gain spellcasting which is a whole different thing. Heck, the Fighter might be the most complicated class of all to build for, with having to choose your own features and keep track of prerequisites for them and try to get the most synergy for a style and action sequence.

I think that taking a single non-spellcaster class to 20 leads to this problem: the spellcasters become inherently both more interesting and more complicated. The thing is that even a single-classed caster does not have a single defining concept or feature but rather several. The spell selection is akin to a variety of class features. For example, a specialist necromancer isn't just a necromancer, he is also a decent diviner and transmuter and whatever other magic he chooses to learn. Even if he only chooses Necromancy spells, there is still enough variety between them to cover more than one thing. Multiple ideas, multiple concepts. A multiclass in one. The trouble with RPG game balance is that it isn't just a matter of measurable power. Perceived imbalance is just as much of a factor in making the game less enjoyable, perhaps even more. When one character is multi-faceted and another character is only one, there will be a perceived imbalance even if the one-track character is powerful.

———————————————————————————————————————————————————————————————

There was an essay I read a while ago, written by a guy who works in video game development. I've forgotten the site and his name and even the first-person shooter he worked on, but the idea he was talking about really hit me. He was talking about how the players of the game were sending a lot of complaints about the reload times of the guns, saying it was too slow. The developers took all feedback seriously especially when it is shared by a large portion of their player base, and so went about solving the problem. The thing was that the reload times were already really really fast. Not only did making it faster cut into the realism (an important part of that game), but after testing a faster reload animation they found that people were still complaining about the reload time. So they thought about it, and realised that the problem wasn't that the reload time was too slow, but that it was too boring. They redrew the animation to make the action more exaggerated and cool-looking and the complaints about slow reloads went away, even though the new animation actually took more time than the old one.

The lesson he took from this, and which was the topic of the essay, is that often what gamers think they want is not the same as what they will actually enjoy and that a designer has to both read between the lines and read behind them to find the best solution.

This goes for RPG design also, and I think it can be applied here. I'll take as a comparison another RPG: Warhammer Fantasy Role Play. In WFRP, multiclassing is an expected part of progression. Once you finish your first basic career (which happens at about 1000 xp), you then become something else. There were some people who complained about the "forced multiclassing", right from the first edition and into the third, but it remained across the editions. Why? Because it works! Yes, you cannot be a soldier or a peasant all the way through the campaign, but the thing is that as the game progresses and the characters become more heroic and important, being just one thing is no longer thematically appropriate. I think a similar thing applies to D&D. At level 20, you are no longer merely a fighter. It doesn't fit with the scope of what you are doing.

Many base classes change as you go on, whether it is the monk gaining supernatural powers or the paladin and ranger gaining spells. Separate them into their individual concepts (Martial Artist and Enlightened One, Slayer of Evil and Living Saint, Huntsman and Woodland Caster), and you open up new combinations between the sectioned classes while still making it possible to build something like the original 20-level class. As for classes that cover a single idea, they shouldn't be like the original 20-level class because at 20th level having a single thing to identify the character as will make it no longer fit in with the game.

In short: single-class is no guarantee of an easy-to-build and uncomplicated character, and at high levels identifying the character as a single thing that it was from the start may well be a bug rather than a feature.

Just throwing this out there, since the "pro 20 level classes" side is kind of under-supported right now.

T'is fine, I love debating these things. Even if nobody agrees with anybody at the end of it, the exchange of ideas is healthy for everyone involved. 

EDIT: moved over from the old thread. If you're wondering where all this came from, it was over here: http://www.minmaxboards.com/index.php?topic=8393.0
Sub-discussions ahoy!

[veekie]

Alright, I split this off to avoid derailing the other thread.

Now my 2 cents.

The structure of base classes I believe, do not necessarily need to be generic, nor intrinsically 20 levels long. Classes are a tool, a means to shape and supply capabilities. The great strength of a class system is that later abilities can safely presume that you have earlier abilities, building synergy and required secondary powers in a single convenient package, theoretically, out of the box a straight 20s build has relevant effects and works as a cohesive whole. Unfortunately most core classes don't do this.

Core base classes should be reasonably generic and cover a fair spectrum of character compositions, but such generality should not be a requirement of a base class. Specialized base classes are useful for low optimization skill players, who have a reasonably common concept that would normally require elaborate execution to pull off. Specialized base classes also lets you increase the breadth of the game with limited power creep. The Duskblade is one such, the Dread Necromancer and Beguiler others. These are tightly themed and they have abilities that are only possible with tight themes and foreknowledge of a 20 level spread to work those abilities in.

20 levels meanwhile, is nice to have, due to the particular advantage of being able to plan abilities ahead of time. Not all concepts can take adequate advantage of the full spread, but whether these should be PrCs or base classes is a matter of if another base class can support the concept until you qualify. Paladins are a nonissue, the holy knight can be built towards from either the divine or martial end, and either end supports it just fine. Dervishes and Swashbucklers meanwhile, could do with a base class of less than 20 levels because while their concepts are not quite adequate to a full 20 level stretch as envisioned, other base classes, including the generic ones, do fail to support the concept adequately, while their fighting styles are not intrinsically high level either.

[dman11235]

Absolutely. I wouldn't for a moment pretend otherwise. But take a look at the subject of the thread, and the subforum that this thread is in. "D&D 3.5 alt" is what this whole thread is about!

Okay, end thread.  See, this was all a miscommunication then.  My initial thing was just that I disagreed that the current set-up (20 level base classes, shorter PrCs) was a bad thing.  Which is, obviously, an opinion.  I was continuing the debate because I thought that you thought it wasn't the way the game was set up.

On the paladin though: a Paladin is a Cleric with persist Detect Alignment.  Sure, there's Smite and the special mount, and Divine Grace, but....is that enough for it to be its own base class?  Even if all alignments were accepted?  I submit that it would not be.  Not only that, but the major class features happen after level 1 (mount, Lay on Hands, Divine Grace).  the Detect Evil is a spell, Smite is a spell, Mount is a spell, Lay on Hands is a spell, etc.  I think that the Paladin would be better off as a Cleric PrC, and the Cleric should get more Paladin-y things from the start.  Namely, Smite Opposition, the Auras, and possibly Divine Grace-esque features.  A Paladin should take that base and focus on the combat aspects, improving Smite, the Auras for combat, and martial skills.  It should introduce the Mount as a signature class feature.  Which will not be a new mechanic, do note.  It is based off of summoning spells, which you could already do.  As it is right now, the PrC is way to specific, and this will open up that whole "divine warrior" concept to the rest of the gods.  Now, there are ways to make it better.  EjoThims is my favorite, because it makes a new concept from the combination.  There's the combat specialist, the Fighter, the Divine caster, the Cleric, and the divine combatant, the Paladin.  Also: Crusader.