Multiclassing and resource management systems.

[TarkisFlux]

That's not quite how it works Frank.

Each track consists of 7 distinct power. Each class is built out of 3 track "slots" or whatever (which we will call A, B, and C for ease), which grant the 7 powers of a track at different levels. The A slot gives you the track things at 1, 3, 6, 12, 15, and 18; the B slot gives you things at 1, 4, 7, 10, 13, 16, and 19; and the C slot gives you things at 2, 5, 8, 11, 14, 17, and 20. And you "multiclass" by dropping one of the tracks that they give you and gain a different track to fill those seven missing powers. So you basically get something else's class power every 3 levels.

It's an interesting idea, but it's not particularly revolutionary or anything.

[Edit] Ninja'd

[Mask_De_H]

So for this hypothetical system, would you set up other class types under the same power schedule to start with or would you want to keep inter and intra class niche protection?

[OgreBattle]

It's an interesting idea, but it's not particularly revolutionary or anything.

Who has done it already?

Sounds like Final Fantasy XI's main/sub job system, where you have a primary Job (say, Warrior), and your /subjob gains abilities at 1/2 level rate. So a level 30 Warrior/Monk has lvl 30 warrior abilities and lvl 15 monk abilities.

[Lokathor]

but as a 10th level character you meet the 10th-level pre-req for 5th-level powers.

Why the fucking hell are you sticking with this terminology.

If you're 10th level you should be using "10th level" powers.

If you're building from the ground up, don't keep the shitty spell level system or a "powers up to half your level" system or anything like that.

Jeegus Fuck.

[TarkisFlux]

It's an interesting idea, but it's not particularly revolutionary or anything.

Who has done it already?

The fact that I'm not aware of anyone else doing it does not make it any better or more special. It's really nothing more than a very formalized power replacement scheme of the sort that you could find in a "pick a power of the appropriate level from this shared list" setup. Any game with a subclass bit, or the thing I said in the last sentence, is already doing something similar.

They do slightly more interesting things with it in the racial and guild tracks. The racial ones are a way to strip out some class based advancement and build in some racial advancement, and seems a pretty good way to keep race important over your entire advancement. The guild tracks are the rough equivalent of prestige classes, since you replace an existing class track when you take a guild track. Instead of getting new wizard powers or whatever, you get new arcane order powers (and some old ones get over-written).

Is it an improvement over 3.x multiclassing? Maybe, since it's largely functional (in so far as I've cared to look). It would fill OgreBattle's multiclassing desires. But if you wanted to change anything other than 1/3 of your powers (BAB, saves, skills, your mom) you'd be out of luck.

[Nebuchadnezzar]

It's been a while since I've checked, but as of my last reading magic items counted effectively as a fourth track, ostensibly so as to not have to give npcs full loads of equipment. In that the available magic items weren't particularly fleshed out or effective it seemed a somewhat reasonable trade.

[ModelCitizen]

On open multiclassing, I don't see any other way to make playable monsters work in D&D. If your monsters look like D&D monsters then the progression for a Succubus Cleric pretty much has to go:

Succubus 6
Succubus 6 / Cleric 1
Succubus 6 / Cleric 2
...

Otherwise you're stuck either trying to write a special 20-level class for every single monster someone might want to play, with the added constraint that taking a hybrid option can't lock you out of any iconic monster abilities.

Personally I'd rather play a game with playable monsters than one with 13 different resource systems.

[Username17]

On open multiclassing, I don't see any other way to make playable monsters work in D&D. If your monsters look like D&D monsters then the progression for a Succubus Cleric pretty much has to go:

Succubus 6
Succubus 6 / Cleric 1
Succubus 6 / Cleric 2
...

Otherwise you're stuck either trying to write a special 20-level class for every single monster someone might want to play, with the added constraint that taking a hybrid option can't lock you out of any iconic monster abilities.

Personally I'd rather play a game with playable monsters than one with 13 different resource systems.

That isn't necessarily true. Just at the very basic level, many of the "monsters" people want are actually fine as normal first level characters. Nobody really cares if you don't start out with a couple hit dice of "Gnoll" or "Bugbear" with your starting Gnoll or Bugbear.

Secondly, if you have Paragon Pathing, you aren't going to have open multiclassing, but you can still have transitional classes. In such a case the Succubus would start as a Controller 6 and then gamely march up to Controller 7, Controller 8, Controller 9, and finally Controller 10 before they paragon pathed out to Hierophant or Morning Lord like a Cleric would.

Thirdly, since you're going to have some kind of optional character development sideline such as feats or development points or something, the playable monsters could eat those optional bits up until their monster powers were "paid off". And then being a full monster would have a level minimum equal to what level would be required to get enough player's option points (or whatever) to buy all the monster powers. For something with few interesting abilities (like a giant), that might even be lower level than the monster in the wild; while for a monster with a long list of cool powerz like a Succubus it could be considerably higher (you are getting a character class in this model). So you might be a Succubus Cleric 9.

Fourthly, if you're going to do subclasses and main classes, monsters can just be one of those slots. Our Gnoll might be an Artificer/sub-Ninja while our Frost Giant is a Paladin/sub-Is a Fucking Frost Giant. For really big monster ability lists like the Mind Flayer, Mind Flayer might be your main class, so you only get to be a Mind Flayer/sub-Marshal. Or whatever.

Fifthly, the "playable" monsters could be actually made out of PC classes in the first place. The Hill Giant could just be a Berserker of whatever level. Ethergaunts could just be Wizards. The Succubus out of the book could game mechanically be a Tiefling Beguiler with some premade selections.

---

Bottom line: there are a lot of ways to make playable monsters work. So claiming that we need to keep open multiclassing so that playable monsters can work is short sighted. After all: open multiclassing has not delivered us playable monsters that work! Seriously: Savage Species was a terrible book and it did not work.

-Username17

[ModelCitizen]

Bottom line: there are a lot of ways to make playable monsters work. So claiming that we need to keep open multiclassing so that playable monsters can work is short sighted. After all: open multiclassing has not delivered us playable monsters that work! Seriously: Savage Species was a terrible book and it did not work.

-Username17

Savage Species was grafted onto a monster system that was already in place, had to live with poor decisions that had already been made (standard vs elite array, CR < HD, gross lack of parity between beatstick monsters and beatstick classes, and so on), and was written by people who made it clear basically everywhere that they weren't trying very hard. These are the writers who thought the Anthropomorphic Bat was a good idea and that's LA +0, no classes involved at all. If you're going to judge design based on SKR's inability to implement it (with the deck stacked against him, no less) then you inevitably arrive at "all possible mechanics are shit."

Most of your solutions essentially lock every monster into a racial class forever. That's better than not being able to play monsters at all, but it's still not very good. I'll grant that if you want a tiering system that lets players change classes at certain breakpoints, and that doesn't count as "open multiclassing," then there are alternatives to "open multiclassing."

[Username17]

Savage Species was grafted onto a monster system that was already in place, had to live with poor decisions that had already been made (standard vs elite array, CR < HD, gross lack of parity between beatstick monsters and beatstick classes, and so on), and was written by people who made it clear basically everywhere that they weren't trying very hard. These are the writers who thought the Anthropomorphic Bat was a good idea and that's LA +0, no classes involved at all. If you're going to judge design based on SKR's inability to implement it (with the deck stacked against him, no less) then you inevitably arrive at "all possible mechanics are shit."

You're still arguing that the "only way" to get playable monsters to work is to pursue a method which has historically never worked. I'm open to the possibility that someone might make it work some time in the future, but I'm certainly not going to hold my breath. And to argue that it's the only thing that could work is arguing without evidence.

Most of your solutions essentially lock every monster into a racial class forever. That's better than not being able to play monsters at all, but it's still not very good. I'll grant that if you want a tiering system that lets players change classes at certain breakpoints, and that doesn't count as "open multiclassing," then there are alternatives to "open multiclassing."

Let's say for the moment that you want Dragon Breath to be on a cooldown timer. It's simple, it's reasonable, it's traditional. If you have open multiclassing, your Dragon's next level could be Sorcerer or Assassin or Swashbuckler or something. And now you have two choices:

• Have all classes everywhere be on cool-down timers like they were WoW classes.
• Solve the problem of characters on different resource management systems having weird synergy.

And that's before we get into the simple multicaster problems where somehow fist level of Swashbuckler has to give some sort of abilities that automatically scale so that an 8th level Dragon gives a fuck about them even though actual 1st level Swashbucklers like in an essentially mundane world solving mundane problems and the Dragon is already a flying firebreathing death machine.

The problems of open multiclassing are titanic. It has never worked and I am openly contemptuous of anybody ever getting it to work. I mean, the closest I think it ever came to working was This. And that wasn't really "take whatever class you want and layer it on top of any other class or monster you want", that was "we'll make specific PrCs that convert the monster you're looking at into a character of the rough type you're looking for", which isn't the same thing at all.

Seriously, it's just stupid. How can you write the first level of Swashbuckler so that it gives something level appropriate when taken as the next level of:

• A 6th level Rogue.
• A 1st level Illusionist.
• An 8th level Dragon.
• A 15th level Shadow Lord.
• A 10th level Samurai.
• A 3rd level Ranger/2nd level Druid.

It has never been done. It will never be done. If we hold our breath and write D&D content for the next 30 years it still won't be done. It's not a reasonable aspiration to even have.

-Username17

[ModelCitizen]

You're still arguing that the "only way" to get playable monsters to work is to pursue a method which has historically never worked.

It has been attempted once, by a blithering idiot working under unfavorable conditions. Yes, 0 / 1 is "never" but it's hardly grounds for dismissing the idea out of hand.

How can you write the first level of Swashbuckler so that it gives something level appropriate when taken as the next level of:

[examples]

You have to grant abilities based on character level rather than class level. I've said before I wasn't a fan of that because WotC's two attempts at it sucked, but they were either written in a big-ass hurry or by people who didn't give a shit or both. (We know ToB was rushed and I don't think anyone is going to defend CPsi.) I can post an example of a less abusable scheme sometime probably tomorrow.

[Username17]

You're still arguing that the "only way" to get playable monsters to work is to pursue a method which has historically never worked.

It has been attempted once, by a blithering idiot working under unfavorable conditions. Yes, 0 / 1 is "never" but it's hardly grounds for dismissing the idea out of hand.

No. It's been repeatedly tried and the tries come up failure every time. Skip Williams had a set of ECL rules. Andy Collins did an ECL overhaul. Pathfinder did a set of monsters as players rules. World of fucking Warcraft d20 has a set of monsters as players rules. There's that soulforge book that tried to do monsters as players with XP costs. If you search google for peoples' homebrew rules you find lots of them.

None of them has gotten even remotely decent. Some of them are minor overhauls, some of them are total ground-up rewrites. But it never fucking works. Because the groundstate assumptions are fundamentally wrong. The whole idea that it is particularly plausible to be able to layer a "level 1" ability set onto a character without knowing whether they are an archer or a lancer or a magician or a dragon, or even knowing whether they are 1st level or 17th and have that ability set be valid in all cases is a totally insane pipe dream.

How can you write the first level of Swashbuckler so that it gives something level appropriate when taken as the next level of:

[examples]

You have to grant abilities based on character level rather than class level. I've said before I wasn't a fan of that because WotC's two attempts at it sucked, but they were either written in a big-ass hurry or by people who didn't give a shit or both. (We know ToB was rushed and I don't think anyone is going to defend CPsi.) I can post an example of a less abusable scheme sometime probably tomorrow.

Then you don't have an open multiclassing idea. If you are seriously just going to let people take "a 12th level Swashbuckler ability set" at 12th level regardless of whether their last 11 levels had any swashbuckler in them, you don't have classes. At all. What you have is a set of abilities that have levels attached to them. Congratulations: you have one class and it is "Wizard" (from a 3e D&D standpoint).

But of course, even then the track record of that kind of thing is not particularly good from a balance POV. Lightning Bolt is a lot better if you already Fly, Cone of Cold is a whole lot better if you have Solid Fog than if you don't. And so on. Self synergy is a really big deal. So big a deal that I am openly contemptuous of people who claim that they can sort it all out and have people take whatever bundle of abilities they want and have that come out even roughly balanced most of the time.

Basically, your defense of open multiclassing sounds like the libertarians who claim that true capitalism would be totally awesome but it's never been tried. It fucking has been tried. Repeatedly. And it has been awful.

-Username17

[Mistborn]

If you were to actually do that, you'd not only have to balance all the C tracks against each other, you'd also have to balance all the C tracks as they synergized with all the possible A and B tracks. Let's say you had 8 base classes and ten levels. You'd then be declaring that you had balanced 512 class track combinations at ten distinct power levels. There are ways to speed that up, but I'm not exactly holding my breath on it ever being accomplished. Adding base classes increases your work cubically and adding levels multiplies your work only once.

-Username17

Well if your going by that logic aren't Feats an invalid concept since you need to test every conceivable combination of feats. Doubly so if your do Tome style scaling feats that are each practically their own progression.

[Mask_De_H]

If you were to actually do that, you'd not only have to balance all the C tracks against each other, you'd also have to balance all the C tracks as they synergized with all the possible A and B tracks. Let's say you had 8 base classes and ten levels. You'd then be declaring that you had balanced 512 class track combinations at ten distinct power levels. There are ways to speed that up, but I'm not exactly holding my breath on it ever being accomplished. Adding base classes increases your work cubically and adding levels multiplies your work only once.

-Username17

Well if your going by that logic aren't Feats an invalid concept since you need to test every conceivable combination of feats. Doubly so if your do Tome style scaling feats that are each practically their own progression.

Not really, since there would be less feats that need to be produced than abilities to fill up a class to ten levels. And you're balancing against broad ability sets (a character at level X) instead of between discreet abilities or a series of abilities (choosing the maneuvers that will make up a class and character at level X).

[Emerald]

Most of your solutions essentially lock every monster into a racial class forever. That's better than not being able to play monsters at all, but it's still not very good.

The problems of open multiclassing are titanic. It has never worked and I am openly contemptuous of anybody ever getting it to work. I mean, the closest I think it ever came to working was This. And that wasn't really "take whatever class you want and layer it on top of any other class or monster you want", that was "we'll make specific PrCs that convert the monster you're looking at into a character of the rough type you're looking for", which isn't the same thing at all.

To detour from open multiclassing to racial classes for a minute, I like the idea of having versions of different classes for monsters as well as PCs so you can play a sneaky succubus or a casting ogre without sucking, but I don't think they really need to be customized on a per-monster basis; you don't need different fighters for different PC races, so you probably don't need more than one "big tough bruiser" monster class if it's general enough. Arcane and divine casters have spells, fighter types have feats, and rogue types could expand rogue talents into another select-your-own-stuff system, so could you do the same for monsters?

One could make monsters classes to fit whatever monster roles you want to have, similar to the Tome of Fiends monster classes, so you have a Brute class that gives you DR, regeneration, knockback, and other "big tough monster" stuff, a Channeler that gets you the oodles of random SLAs, etc., with plenty of other classes like Deceiver and Infiltrator and such to cover all the needed roles; basic stuff like improved senses would be selectable by anyone from a common list. Then you go with a hybrid mutlclassing approach for PCs where you advance two classes like AD&D multiclassing or 3e gestalt, instead of the much-reviled open multiclassing, and also use that to hybridize monster//monster or monster//PC class combos for the more powerful versions of monsters: if an ogre is a Brute and a vrock is a Channeler, for instance, then an ogre mage might be a Brute//Sorcerer, a balor might be a Brute//Channeler, and so forth.

Making a succubus as a PC (for instance) would then be fairly easy. It starts out as a Deceiver (or whatever monster class you feel best fits sexy demon chicks), and you can either continue on with that if you want to just play "a succubus," or you can make it a Deceiver//Hexblade or Deceiver//Beguiler or whatever if you want to mix things up a bit, and to do that you use the normal hybrid rules like any other character. Monsters that come with inherent casting or class features like dryads or rakshasa are actually X monster class//Y PC class, and you can trade out Y for [whatever] if you don't want to play a Y and don't mind the slight refluffin that results.

Does that sound like a somewhat-reasonable system that it might be worth it to fiddle around with, or am I too drunk after Christmas dinner to come up with good ideas?

[Mask_De_H]

Gestalt is a clusterfuck, but your basic idea is the "monsters as class or subclass" that's been bandied about for a couple heartbreaker attempts.

[ModelCitizen]

Basically, your defense of open multiclassing sounds like the libertarians who claim that true capitalism would be totally awesome but it's never been tried. It fucking has been tried. Repeatedly. And it has been awful.

-Username17

Oh really? Well, if you're going to resort to sophistries like this, it's obvious this has nothing to do with RPG mechanics and everything to do with your fear of having anyone challenge your ideas. I don't think there's anything I can gain by discussing this with you, because the only thing you really have to say is "WAAAAHH BUT BUT MY EGO!!!"

[...You Lost Me]

ModelCitizen, to discount everything in an entire argument except the last part which is basically not even part of the argument, and then say "obviously you can't be reasoned with" is whiny, passive-aggressive, and pointless. If your argument in favor of open multiclassing is "lol frank is a bitch" then you can do us all a favor and go away.

[MGuy]

I hate to say this but I believe that Frank is very right on this subject. Lacking his insight and recorded attempts I thought long and hard about having multiclassing be a thing in the game I'm still trying to make. I looked at it in three ways all which I had to abandon.

1) Was where I tried making certain abilities come online at certain levels. So the idea that was presented earlier. Now I never gave this to other players but when I was writing up some samples it I ended up noticing I was basically just cherry picking abilities. At which point I had to ask "What's the point of the classes?"

2) So then I tried to give all classes all scaling abilities. Then people could just pick and choose what abilities they wanted and when. Then you could multiclass and get access to a different ability list. This had me run face first into two issues. It was a repeat of the first one AND making abilities that all scale for every class is fucking hard and I could not get it effectively balanced enough for me to give it a pass. Some things indeed scale better than others. There were other problems but they are unrelated.

3) So now I've been making abilities that are "appropriate" to the level players are at. Flying/teleporting comes online at a certain point.The means any abilities that might grant (like turning into a bird) all come online at the same level. Now I pondered about how multiclassing might work with this but, as Frank pointed out, what Level 1 abilities would be worth taking when you have 10 levels in another class/es?

As far as I can tell it just is not worth the trade off. The logic actually follows the way Frank laid it out. I'm all ears if you have another way of doing it but over the past year of me wrestling around with it I really don't know what other way you could attempt to make it work.

[Username17]

I'm all ears if you have another way of doing it but over the past year of me wrestling around with it I really don't know what other way you could attempt to make it work.

If you insist on having open ability gain and classes, then what K and I discovered was that you had to having the abilities be prerequisites for the classes and not the other way around. That is: you can take whatever abilities you want (whether by level or bought with dev points or something in between), and those abilities are prerequisites for "being a class". If you want to be a Barbarian, you need 3 abilities off the "Warrior" list and 1 ability off the "wilderness" list, while being a Ranger might require 2 "Warrior" abilities and 2 "Wilderness" abilities and being a Druid might require 2 "Priest" abilities and 2 "Wilderness" abilities and so on and so forth. As an individual character you could end up qualifying for lots of classes, but each session you'd only have one (taking some amount of time to change your wardrobe and respec to become a different class).

The classes themselves would provide some sort of basic things like give you base defenses and some sort of fundamental class function. Prestige Classes would just be high level class templates that had long and possibly weird lists of prerequisites. Basically, we'd be talking about "classes" in the Final Fantasy Tactics sense. Where they literally are a thing which you change completely by changing your hat after you've picked up the prerequisite ability requirements.

---

To detour from open multiclassing to racial classes for a minute, I like the idea of having versions of different classes for monsters as well as PCs so you can play a sneaky succubus or a casting ogre without sucking, but I don't think they really need to be customized on a per-monster basis; you don't need different fighters for different PC races, so you probably don't need more than one "big tough bruiser" monster class if it's general enough. Arcane and divine casters have spells, fighter types have feats, and rogue types could expand rogue talents into another select-your-own-stuff system, so could you do the same for monsters?

4e actually had a decent setup of monster classes. They didn't go anywhere with it, but "Artillery, Brute, Controller, Lurker, Minion, Skirmisher, Soldier" is a much more (potentially) helpful set of monster classes than "Animal, Construct, Dragon, Fey, Giant, Humanoid, Magical Beast, monstrous Humanoid, Outsider, Plant, Undead, I'm probably missing one". I mean for fuck's sake: they wrote up Fey that were bruisers. The 3e Monster classes were meaningless.

Still, while you could certainly make classes like "Controller" that handed out abilities in levels such that you could make a Succubus, a Vampire, an Assassin Vine, or a Nymph with the same class, I have fundamental problems with this as regards make Monsters into playable archetypes. The most basic condition is that if you tie a Niefel Giant's abilities to the Brute (or Soldier, or whatever) class, and then you hand it over to player control and they decide they want to be a Cleric instead they won't end up having the Niefel Giant's signature abilities. And while you could say "just open multiclass it: be a Controller until you are the level of a Succubus and then multiclass into something else", that requires you to get open multiclassing to work properly which I don't think is ever going to happen.

But 4e had another idea with its monster classes that was I think pretty helpful to think about: which is that the "Lurker" class is way less powerful than the Rogue or Assassin classes. That is to say that the expectation is that players are going to fight equal numbers of "monsters of their level" and mop the floor with them. Now 4e never codified that in any way, but I could see how you might be able to. If you did have a subclassing system, the monstrous classes could simply be sub-classes in terms of the power they bring to the table. In such a case, one character might be "Knight/sub-Ninja" while another character would be a Wu Jen/Artillery because the "Artillery" monster class is already a sub-class. You could also do it the other way, where the monster classes were just as good as PC classes, but the PCs got sub classes and the monsters didn't (unless of course, they were monster PCs). in such a setup, the Succubus would get to be a "Controller/sub-Scout" or whatever.

But regardless, and back to the original point of the thread: the moment you have a sub-classing system you're going to have to handle characters juggling multiple resource management systems. And that is, as Lago pointed out: not particularly easy. Some systems merge easily: spell prep and spell points are almost the same thing, so as long as you get enough of either to use level appropriate abilities it's just like having more of whichever one you normally use. Other systems don't merge well at all: warmup and drain step all over each others' toes. Other systems are inherently synergistic: spell prep supercharges rage bars super well. There are things you can do about it, but it's hard.

-Username17

[Kaelik]

set of monster classes than "Animal, Construct, Dragon, Fey, Giant, Humanoid, Magical Beast, monstrous Humanoid, Outsider, Plant, Undead, I'm probably missing one".

Aberration, Elemental, Ooze, Vermin.

Because you know, Elementals are clearly not Outsiders, and Oozes, Vermin, Aberration, Monstrous Humanoid, and Magical Beast are all completely different and could not possibly be combined into one or two types.

[Libertad]

Multi-classing has always been unequal, so trying to design "equal" monster classes isn't going to work either. That doesn't mean that monster classes should be ditched entirely, given that every Edition of D&D has terribad class combinations. Don't throw the baby out with the bathwater.

Having generic monster roles as classes, like "Brute" and "Sneak," are definitely valid options if they perform well in their roles. I also think that LA 0 variants (with the serial numbers filed off) are a great way to get players interested, like Trolls being Trollkin in Iron Kingdoms, Golems being Warforged in Eberron, etc. Not monster "classes," but it's a quick and popular fix.

I'm also fond of Oslecamo's Monster Classes, in that they go for the Level=Challenge Rating=Hit Dice of the monster and have scaling abilities which don't become irrelevant at higher levels (sleep becomes deep slumber, DR improves with HD, etc).

Hell, if you want playable monsters right out of the box with the least amount of work, pick the LA 0 option. Eberron did it, Frank and K did it with Drow in the Tomes, Warforged are popular, and people still want to play as Drow despite their +2 LA.

It's like an Occam's Razor for choices.

[Mask_De_H]

For subclassing, couldn't you just make the subclass give out passive abilities and in-flavor at-will abilities? Or would that raise problems with the Rogue (new at-wills aren't contingent) and Assassin (better options than sitting with thumb up ass)?

And I assume that unless you place the abilities that require specific resource allocation inside the class hat you get, using the FFT/Black Forest method of character generation would not work for the resource allocation class setup we have (for the same reasons we took DrP to task for), right?

[Username17]

For subclassing, couldn't you just make the subclass give out passive abilities and in-flavor at-will abilities? Or would that raise problems with the Rogue (new at-wills aren't contingent) and Assassin (better options than sitting with thumb up ass)?

You get certain effects with Rage Bar with the Berserker as well. And with Drain on the Warlock. Anything you can do in an "always available" fashion makes risking taking rounds where your resource management system won't give you anything a better deal. Heck, an always-available blast even incentivizes a "big risks, big rewards" style WoF over a deck which is more even keel.

For example: let's say our Paladin works like a Crusader. Normally, Counters are really awesome, but drawing two Counters sucks because it leaves you with a standard attack action. That acts as a powerful disincentive for a Paladin to have more than one or two Counters in their ready pile. But if the Paladin can sub-class and get a decent enough attack to use every turn that a strike doesn't come up in their hand - filling up on a lot of counters looks a whole lot sexier.

But I wouldn't say that's insurmountable. After all, if literally every sub-class is handing out at-will abilities, then you can build in the fact of sub-class availability to every single resource management system you write up. So while in abstract it is a considerable powerup to the Rage Bar that the Berserker can use whatever their sub-class' abilities are until they build up enough Fury to turn into a bear and do a whirlwind of death - you'd actually be building the Berserker class with that in mind because every PC would have a sub-class (starting at whatever level the subclass comes online). The fact that the Hero's abilities are already always at-will and thus he doesn't care nearly as much about whatever his sub-class contribution is would also be taken into account when you made the class.

In short: yes there are effects where resource management system X is going to be a better match for resource system Y than it is for Z (or vice versa) for every single combination you can imagine. But that's not actually a problem if everyone who sub-classes gets the same resource management system for their subclass. The Berserker is going to rely on their sub-class contribution at the beginning of most fights, the Hero is going to rarely rely on their sub-class contribution at all, and you just build that expectation into the two classes.

And I assume that unless you place the abilities that require specific resource allocation inside the class hat you get, using the FFT/Black Forest method of character generation would not work for the resource allocation class setup we have (for the same reasons we took DrP to task for), right?

Yeah, if you're going to do FFT/Black Forest chargen, I don't think you can jumble resource management systems around. Everyone has to be on mana points or cooldown or something. Having 13 or 15 resource management systems is a luxury that class railroads can take advantage of. I really don't see that working in an ability grab bag system of any kind.

-Username17

[Mask_De_H]

Hmmm, why not add the railroads in later, then?

As in, you start out all at-will (or whatever is decided for subclasses) for a few levels with the ability grab bag, then once you've claimed enough abilities, you go onto the Railroad Tier and your previous abilities/hat becomes your subclass? It allows for a period of organic growth, gives you the railroad, and creates a hard shift in how the mechanics are viewed (in which you could phase out the accumulation of lower tier abilities). That then acclimates players to mechanical paradigm shifts, which you'd do again in the Logistics and Dragons tier (in expansion material)