D&DNext: Playtest Review

[Username17]

I for one would be totally OK with bolas being useful out to 10th level and beyond. If bolas are good enough for He Man and the Masters of the Universe, I think they are good enough for mid level characters to use.

My problem is that this isn't the result of things scaling well, it's the result of things scaling shitty. So bolas aren't going to be used by Man At Arms and Lockjaw, but by groups of Halfling warriors that you pay to follow behind you and throw bolas at enemies while you clean up in melee. I'm not even opposed to their being an Incan-inspired Halfling empire that you hire troop detachments from, but to have that show up emergently implies that the designers aren't thinking about the interaction of things.

Man At Arms, Sportsmaster, Cat Woman, these are all characters of about 6th level who use bolas as a go-to weapon. And that's fine. The game should do that. As far as I know, no one hires detachments of Incan Imperial Skirmishers to hunt manticore with.

-Username17

[RobbyPants]

Towards the end of 3.5 they actually were pretty good at making Tier 3 classes with a wide variety of mechanics. Which means they had a really nice selection of versatile and powerful classes that avoided the rocket-tag game. This of course created balanced martial and caster classes without them being remotely the same. To me this is the biggest missed lesson.

Tier 3 is fighter tier right?
I prefer classes with more options myself.

Assuming you're talking about JaronK's tiers, tier 5 is the fighter tier. Tier 3 is where he puts a lot of Tome of Battle and non-core casters. Do note that he also puts the factotum in this tier and the rogue in tier 4, so take that for what it's worth.

Tome of battle was created to make martial spell casters. And could really use some refinement. It was actually the basis for 4e too.

My understanding is that ToB is actually a hasty 3.5 port of an initial test version of 4E, not the other way around. Basically, they tried to make a 4E of it, decided against it, and figured they could make money off of it during 3.5's run.

[nockermensch]

My understanding is that ToB is actually a hasty 3.5 port of an initial test version of 4E, not the other way around. Basically, they tried to make a 4E of it, decided against it, and figured they could make money off of it during 3.5's run.

Also known as "the greatest tragedy of RPG history".

[ishy]

My understanding is that ToB is actually a hasty 3.5 port of an initial test version of 4E, not the other way around. Basically, they tried to make a 4E of it, decided against it, and figured they could make money off of it during 3.5's run.

For example, encounter powers were something we were already experimenting with in the tail end of 3rd Edition. Design work on Tome of Battle: Book of Nine Swords explored this space independently of the early design work on 4th Edition.

[zugschef]

Tier 3 is fighter tier right?
I prefer classes with more options myself.

Assuming you're talking about JaronK's tiers, tier 5 is the fighter tier. Tier 3 is where he puts a lot of Tome of Battle and non-core casters. Do note that he also puts the factotum in this tier and the rogue in tier 4, so take that for what it's worth.

That's because he is an idiot and seriously claims that beguilers are a tier below the sorcerer.

I'd say a real practicable power tier is based on spellcasting ability:

1. Full casters /unlimited spells known
2. Full casters /limited spells known
3. Partial casters and UMD classes
4. Non-casters

[Username17]

My understanding is that ToB is actually a hasty 3.5 port of an initial test version of 4E, not the other way around. Basically, they tried to make a 4E of it, decided against it, and figured they could make money off of it during 3.5's run.

For example, encounter powers were something we were already experimenting with in the tail end of 3rd Edition. Design work on Tome of Battle: Book of Nine Swords explored this space independently of the early design work on 4th Edition.

Baker, Donais, and Mearls translated current versions of the Orcus I mechanics into a last-minute revision of Tome of Battle: Book of Nine Swords.

The quote you have is only technically true. In that Tome of Battle was Orcus I, which had been intended to be 4th edition, but instead they scrapped it. The month later, the entire Orcus project was canned because Rob Heinsoo, Bruce Cordell, and James Wyatt decided that it didn't have enough daily resources to track. Seriously.

-Username17

[Drachasor]

It's true that JaronK's classification of each class is far from perfect. However, I think the general principle of his tier system is pretty solid.

Tier 5 can overcome challenges with a toolset that's almost cripplingly specialized
Tier 4 can have that highly specialized capability, but are a bit more flexible.
Tier 3 overcome challenges in a variety of ways very effectively (e.g. quite flexible).
Tier 2 can pick a few ways to "break" the game (often build dependent) and do the above.
Tier 1 can "break" the game tons of ways and do the above.

Jaron himself tends to classify this sort of thing someone poorly. "Breaking" should be considered as "game changers". Teleportation is an example of a game changer, as it is something the DM has to be ready for specifically. The same with lots of divination abilities, etc. Basically, they have the capability to redefine what a lot or most of the challenges are.

Personally I think that's why Tier 3 is a pretty nice place for class design. It allows for flexibility and versatility, without totally overturning the game.

I'm not 100% sure on the Beguiler and Dread Necromancer here, but I do know they do have less access, generally speaking, to game changers do to their limited spell lists. That said, they have a high optimization floor like the Tome of Battle classes. A Sorcerer, for instance, has a really low optimization floor, so making a bad sorcerer is very easy. So we should bear in mind that optimization floors are completely ignored by this approach.

I'd say that going with Full Casters / Partial Casters / Non-Casters doesn't capture things the same way, because that's highly dependant on the spell lists. A Warmage, for instance, is a Full Caster, but they have a very limited spelllist that makes them far worse than any other full caster, and as such they have far fewer ways of dealing with problems.

And to avoid arguing about the design quality at the end of 3.5, let me just say that 3.5 did solidly demonstrate you could have a huge variety of class abilities, builds, and classes in that sweet spot in the middle. 4E failed to follow up on this and instead was built around everyone largely using the same mechanic.

[RobbyPants]

Jaron himself tends to classify this sort of thing someone poorly. "Breaking" should be considered as "game changers". Teleportation is an example of a game changer, as it is something the DM has to be ready for specifically. The same with lots of divination abilities, etc. Basically, they have the capability to redefine what a lot or most of the challenges are.

Except for he did partition it by figuring on what breaks games. That's the difference between tiers 2 and 3.

I'm not 100% sure on the Beguiler and Dread Necromancer here, but I do know they do have less access, generally speaking, to game changers do to their limited spell lists. That said, they have a high optimization floor like the Tome of Battle classes. A Sorcerer, for instance, has a really low optimization floor, so making a bad sorcerer is very easy. So we should bear in mind that optimization floors are completely ignored by this approach.

Well, he puts them both in tier 3, but they can totally break a game with their own powers. His criteria aren't specific. They should be tier 2, at least, in that they can break the game, but in fewer ways than wizards.

[Drachasor]

Jaron himself tends to classify this sort of thing someone poorly. "Breaking" should be considered as "game changers". Teleportation is an example of a game changer, as it is something the DM has to be ready for specifically. The same with lots of divination abilities, etc. Basically, they have the capability to redefine what a lot or most of the challenges are.

Except for he did partition it by figuring on what breaks games. That's the difference between tiers 2 and 3.

Yes, but my problem is how he goes about defining/explaining/showing it. He's so focused on combos that are obviously broken and creative rule reading that he misses the more essential and straightforward game-changing abilities these classes have. A well-played Tier 1 or Tier 2 will require an immense amount of work for the players and DM to figure out what to allow and what to stop. And where to draw the line is completely unclear. Add to that the fact the game tends to not even think of having counters to half the things high level casters do.

I'm not 100% sure on the Beguiler and Dread Necromancer here, but I do know they do have less access, generally speaking, to game changers do to their limited spell lists. That said, they have a high optimization floor like the Tome of Battle classes. A Sorcerer, for instance, has a really low optimization floor, so making a bad sorcerer is very easy. So we should bear in mind that optimization floors are completely ignored by this approach.

Well, he puts them both in tier 3, but they can totally break a game with their own powers. His criteria aren't specific. They should be tier 2, at least, in that they can break the game, but in fewer ways than wizards.

I'd say that compared to a min-maxed Sorcerer they are both better designed classes, but also easier to stop from wrecking/breaking/game-changing the campaign. Though I agree that taken as written they are certainly more powerful than the Tier 3s. It's just that when I last looked at them, it isn't too hard to tone them down to a T3 level. At least that's how it seemed to me.

[schpeelah]

Yes, but my problem is how he goes about defining/explaining/showing it. He's so focused on combos that are obviously broken and creative rule reading that he misses the more essential and straightforward game-changing abilities these classes have.

Basically this, except for the rest of your post where you seem to think those tiers have any worth anyway. Rogue is tier 4, factotum is 3 and artificer is 1 not because of power level - rogue has a fair amount of staying power while artificer is feeble because his core class features do not work more often than they do for a large chunk of his career and factotum sucks at life under most circumstances - but because JaronK personally does allow artificers doing a large number of broken item combos at level 20, does allow weird factotum shenanigans, and does not allow rogues using flasks and Rings of Blink.

[infected slut princess]

I am 99% sure JaronK has never actually played a real game of D&D with a real DM. I wish I could go to his house and smash his computer.

[Kaelik]

However, I think the general principle of his tier system is pretty solid.

The general principle of his tier system is how powerful is a class with only the things he allows it to do which is entirely based on his subjective bullshit.

For example, Dread Necromancers get good spells, infinite undead armies, and the ability to Wish for a Ring of Infinite Wishes, and keep doing that forever. But JaronK personally doesn't let people use infinite undead armies effectively, or Wish for items, or use Arcane Disciple and Expanded Learning effectively. He does however let Factotums use his houseruled version of an online feat, a 3.0 skill that doesn't exist in 3.5 and is setting specific, and Item Familiar. So they are "really effective."

There is no possible method of calling a Dread Necromancer Tier 3, because it gets multiple class features that instantly and irrevocably break the game. To any extent ability to break the game at all is a factor, Dread Necromancers are "Tier 2."

But fundamentally, even if his list wasn't stupid, the idea of making a Tier list that ranks classes in the high tiers for having things that break the game, and therefore, definitionally will not come up while playing the game, is retarded.

I am 99% sure JaronK has never actually played a real game of D&D with a real DM. I wish I could go to his house and smash his computer.

His recent thing has been posting about how he plays through completely unedited modules with no changes to account for players in a party that includes some other people, and him playing a Tainted Sorcerer with DC infinity spells.

Yes really, he is playing a character with DC infinity spells, and he thinks this counts as really playing a game of D&D.

[Drachasor]

Yes, but my problem is how he goes about defining/explaining/showing it. He's so focused on combos that are obviously broken and creative rule reading that he misses the more essential and straightforward game-changing abilities these classes have.

Basically this, except for the rest of your post where you seem to think those tiers have any worth anyway. Rogue is tier 4, factotum is 3 and artificer is 1 not because of power level - rogue has a fair amount of staying power while artificer is feeble because his core class features do not work more often than they do for a large chunk of his career and factotum sucks at life under most circumstances - but because JaronK personally does allow artificers doing a large number of broken item combos at level 20, does allow weird factotum shenanigans, and does not allow rogues using flasks and Rings of Blink.

I do think the Tier-system is useful, but like I said I don't agree with all of his classifications.

I'd argue the Rogue is generally tier 4, because they do have some pretty sharp limits and narrow specialization. Sneak Attack is great when it works, but getting it to work isn't trivial and some creatures are immune. With certain feats/magical items/UMD/etc the rogue becomes a rather complicated case. Which shows, imho, that it isn't a well-designed class (like most of the PHB classes).

The Factotum certainly isn't as good as JAronK thinks. From what I've seen Artificers honestly are quite good due to their insane crafting capability.

That said, perhaps I should find a different terminology since speaking about Tiers brings a lot of JaronK's insane baggage.

Here's is why I think it is useful though:

Tiers:
5. Crippled classes. Too inflexible to be able to handle a wide variety of challenges, too narrowly focues.

4. Semi-crippled classes. Still very inflexible. They can handle a larger variety of challenges with difficulty.

3. Good classes. They can handle a variety of challenges competently with at worst a handful of DM intervention to handle the rare broken mechanic.

4. Overpowered classes. They have a lot of ways to build them that require a ton of DM-intervention and forethought to handle a lot of specific abilities. Often they can outdo other classes at their own schtick.

5. Really overpowered classes. Like 4's, but without the build limitation, one of these can access a huge spectrum of abilities that require a ton of DM work to deal with.


That part is useful.

However, I think the general principle of his tier system is pretty solid.

The general principle of his tier system is how powerful is a class with only the things he allows it to do which is entirely based on his subjective bullshit.

For example, Dread Necromancers get good spells, infinite undead armies, and the ability to Wish for a Ring of Infinite Wishes, and keep doing that forever. But JaronK personally doesn't let people use infinite undead armies effectively, or Wish for items, or use Arcane Disciple and Expanded Learning effectively. He does however let Factotums use his houseruled version of an online feat, a 3.0 skill that doesn't exist in 3.5 and is setting specific, and Item Familiar. So they are "really effective."

There is no possible method of calling a Dread Necromancer Tier 3, because it gets multiple class features that instantly and irrevocably break the game. To any extent ability to break the game at all is a factor, Dread Necromancers are "Tier 2."

But fundamentally, even if his list wasn't stupid, the idea of making a Tier list that ranks classes in the high tiers for having things that break the game, and therefore, definitionally will not come up while playing the game, is retarded.

Those are the only two ways a Dread Necromancer (DN) can be a problem. It's pretty easily avoided by a DM though.

1. Limit the unlimited undead bit (quite a few trivial ways to do that while still allowing the Dread Necromancer to have his undead).

2. Don't allow broken crap with the Planar Binding spells.

I grant that JaronK's view on things is deeply flawed, but when you look at the DN as a whole it is much weaker than normal full casters. Take a look at its 7th-9th level spells:

7th: Control Undead, Destruction, Finger of Death, Greater Harm, Mass Inflict Serious Wounds, Song of Discord, Vile Death
8th: Create Greater Undead, Horrid Wilting, Mass Inflict Critical Wounds, Symbol of Death
9th: Energy Drain, Imprison Soul, Mass Harm, Plague of Undead, Wail of the Banshee.

Nothing that bad.

They also get 5 Necromancy spells of their choice over 20 levels to add to their list. Not a big problem there (any potential problem is easily managed).

Overall, they're a very well-design class imho, with only two problems that ANY Sorcerer, Wizard, or Cleric could duplicate. Both of those problems are the sort of thing most DMs can stop in their sleep. So not a big concern, generally speaking.

Remove his "subjective bullshit" and there's a use to the classification of the tiers if not always his personal assignment of classes to each tier.

[Kaelik]

And once again, if you are going to divide the classes into Tiers there are many possible things to base the Tiers on. You could rank classes based on power. You could rank classes based on easy of use. You could rank classes based on versatility, but then you run into the problem that you have to define versatility, and you could rank classes based on amount of game breaking and DM intervention to fix that.

But here is the important thing: Those are apples and oranges and pears and bananas. You can't have one ranking reflect all of those facets.

For example, if you ranked things on power, Frenzied Berserkers and Rogues and Wizards would all be in the same place. And if you ranked them on ease of use Sorcerers would be the hardest and Berserkers would be the easiest.

And if you ranked them on versatility, you would have to specify a minimum level of competence at each thing before it was counted, because being able to versatily deal with level 5 challenges at level 20 is meaningless, and in that set up, the Bard and Factotum are the least versatile, because they are not competent at fucking anything.

And finally, if you are going to rank things on how often the DM has to intervene to address broken shit, that is literally the last thing you should ever fucking touch.

Now, your attempted Tier rankings pretend to be able to compare the Rogues difficulty of build to play well with the Warblades... ease? Is the Warblade even easier? It seems like it is only easier because it comes in one book, which while true, doesn't change that now days people can just ask about Rogue builds on the internet, and Warblades are harder to play than Rogues after being built.

So now even ease of play has to be divided into ease of build and ease of play.

So if you want to talk about a Tier system, the first goddam thing you need to do is determine which of the five different things you are actually ranking, and then not fucking include elements outside the one you are ranking to differentiate classes.

The Rogue is not less versatile or less powerful than the Warblade because it is harder to build. And it is easier to play once it has been built for you by the resident minmaxer. That isn't a compelling reason to change it in versatility or power tier.

[Drachasor]

Agreed. Certainly the big thing the Tier System fails to address is optimization floors and simplicity in play.

It's focused on versatility and power (this is helped by the fact that AFAIK, you don't have classes they don't fit into the following categories).

Tier 6 has neither.
Tier 5 has power, but no versatility.
Tier 4 has power with some less powerful versatility.
Tier 3 has power and decent versatility.
Tier 2 has Power the DM must design against(!) and decent versatility.
Tier 1 has Power the DM must design against(!) and a ton of versatility.

Where "versatility" is perhaps a bit vaguely defined as both different methods of overcoming an encounter as well as a variety of tools that allow one method to work. It's good enough for a general idea though.

Of course, it uses Potential Power for classes, which completely ignores the optimization floor. Because you CAN make a Sorcerer that is Tier 6 and sucks at everything. Whereas you can play a Beguiler that sucks at everything, you can't make one that does (without being stupid with a low main stat). A Tier of Crappiness system might take this into account.

Something like: Tier X: Classes with a variety of builds that hit an optimization floor of Tier X, but it's hard to go lower. Starred if the class appears elsewhere because you can change them without gaining levels to operate at a new tier (e.g. Wizards acquiring/preparing/using different spells).

That might give something like:
Tier 6: Fighter, Sorcerer, Barbarian,* Rogue, Ranger, Wizard*
Tier 5: Cleric*, Barbarian*
Tier 4: Warblade, Crusader, Swordsage
Tier 3: Dread Necromancer, Beguiler, Druid*
Tier 1/2: Wizards*, Clerics*, Druid*
(Obviously I'm didn't go over every class).

Again, such a thing is rough, but useful. The above is not meant to be flawless, just a rough guess. It does get hard to measure around Tier 1/2.

Though all this information is probably more easily expressed with a picture. Say a bar graph showing the range of power an individual class has.

Which classes are simpler to play is something that all the above still leaves out, of course. This is also a complex topic since simple can often mean a lower tier (e.g. you just hit things with your pointy stick), but that's not something that has to be true (e.g. you can break walls, break spells, and rip holes in reality by hitting it with your pointy stick).

Kaelik, I do think you undersell the Bard quite a bit. They're far from a bad class. Especially when you take into account a lot of the splatbooks.

As for ranking taking into consideration DM intervention, well, that's unfortunately part of how 3.X works, because there's quite a lot that casters can do that the DM must take into account.

[Ice9]

The floor/ceiling thing is something you can't ignore - it's not uncommon to see Wizards that are the least powerful member of the team, because they picked damage spells and weak SoDs when the party already includes a damage-monster or two. Also the reason ToB is seen as uber-powerful - it's a lot more "idiot proof" than other melee.

Another thing is breaking the game - not all ways of doing that are equal, and even which ones are worse depends on the game environment. IME, things that are actually infinite aren't much of a problem in practice.
Player: I do these steps and get infinite wishes / solars.
MC: Ok, you win D&D. But if you want to keep playing, then no.

Things that are broken, but should legitimately be powerful, just not as much as they are - and there's no agreement on where that line should be drawn - are more of a problem. So is stuff that's pervasively overpowered without one component being the issue (*cough* Druid). And of course, there's the difference between overshadowing the party and smashing all the MC's stuff - it's often not the same classes.

Also, you can't discount non-versatile power completely. The PF Gunslinger, for example, can fairly easily reach a high level of power without much versatility. Now sure, in a solo campaign, this wouldn't matter; you just mentally tag normal-tactic-using foes as "easy mode" and attack the weak spots when you want some challenge. However, in a group, it's a non-trivial endeavor to provide foes that don't go down instantly in a hail of bullets, don't render the other party members useless, and aren't noticeably contrived.

[Drachasor]

Good points on all counts, Ice9.

So we might have a graph with two axes of power and versatility. Each class would be some sort of solid shape on that graph representing the range of power and range of versatility they could have.

I would note that if you have a lot of non-versatile power, the problem isn't so much that you aren't effective or can't hurt some parts of the game. It is more that you can't interact much at all with a host of other parts of the game. Which isn't very fun for the player. It is true that the difference between a great damage, a ton of damage, and insane damage is not represented well by the Tier system since it has a single scale that includes versatility. Much of the time this isn't an issue, but sometimes it is.

I do view this sort of thing as something that helps the DM and players look at the classes. Ideally provides some guidance when considering house rules and the like.

[Kaelik]

Where "versatility" is perhaps a bit vaguely defined as both different methods of overcoming an encounter as well as a variety of tools that allow one method to work. It's good enough for a general idea though.

No it isn't. Versatility is vaguely defined as "stuff JaronK likes" and you can either give a real definition, or shut up, because "different methods of overcoming an encounter" doesn't justify being higher on a tier list that measures power, and "a variety of tools that allow one method to work" would rank Rogues and Barbarians as some Tier that does have versatility, because they are very Versatile in that.

Of course, it uses Potential Power for classes, which completely ignores the optimization floor.

No it doesn't. It uses JaronK's ass. And if you want to actually Tier the classes you have to define your own optimization point, and make it noticeably less complete arbitrary than JaronK's ass.

By Potential Power, Commoner is Tier 1, because a Commoner that buys a Candle of Invocation, Gates in a Genie, Wishes for two Rings of Free Wishes and a Candle of Invocation, and repeats the process is infinitely powerful and infinitely versatile. And this trick applies to literally everyone.

A Dread Necromancer with Gated in Djinns using Wish to emulate literally any spell up to 8th level is clearly Tier 1.

So right off the bat, you literally cannot use "potential power" as the balance point, and you have to make something else up.

Kaelik, I do think you undersell the Bard quite a bit. They're far from a bad class. Especially when you take into account a lot of the splatbooks.

It is possible to optimize a Bard to do some things in a way that is vastly better than a Factotum, certainly, but the traditional Bard class is basically not a part of that and all those builds are much better having one level of Bard. Feats give you the +8 bonus to Inspire Courage and DFI, and anything with one level of Bard going into Sublime Chord is better than anything with more than one level.

It is merely an example of one class that is "versatile" by sucking at everything, and losing to level appropriate challenges. It can be built to specialize, but then it loses the Versatility it pretended to have.

As for ranking taking into consideration DM intervention, well, that's unfortunately part of how 3.X works, because there's quite a lot that casters can do that the DM must take into account.

So? That doesn't mean you try to jam it into the same tier system that rates versatility and power, because every single class instantly breaks the game at level 3.

[fectin]

Why at level 3?

[DSMatticus]

JaronK's tiers are gibberish. Their explicit purpose is to help DM's figure out what sorts of classes can feasibly murderhobo together and who needs buffs/nerfs if you're going to try and get multiple tiers to murderhobo together. But his basis for deciding what tier a class goes in is "how powerful is the most powerful build of that class that I would allow at my table."

The first problem is that "whether or not JaronK would allow it at his table" means exactly fuckall. He doesn't ban broken things for being broken, but he will ban broken things for being too simple. So if you combine five pieces over three sources to break D&D, JaronK approves. If you use one ability in core as written to break D&D, JaronK bans.

The second problem is that "how powerful is the most powerful build of a class" tells me absolutely nothing about the build of the class in front of me. Sure, you can build a sorcerer that grabs a delicious combination of utilities and fuckyou's and have it come out really powerful. But you could also build something that is even shittier than the warmage, which puts it... low T4? T5?

His approach is fundamentally bad. It does not do what it sets out to do. It can't.

[Kaelik]

Why at level 3?

A Candle of Invocation is 8,400gp, presuming WBL and a party of 4, the party can first buy one at level 3 when they have 10,800 gold between them, as opposed to level 2 when they have 3,600 between them.

Now in reality, you don't always have four party members, you don't always have WBL, the accumulation occurs over the course of the level, so you probably get it somewhere at 2, and you could always elven profession check your way to it at level 1, but in reality no one ever uses the goddam candle, so hypothetically, a party of four level 3 PCs would be guaranteed to do it.

[NineInchNall]

I've never understood how versatility really matters that much in making something overpowered. I mean, how often have you read something along the lines of, "It's not that Druids/Wizards/Clerics can be awesome, it's that they can do it any number of ways!"

If you can overcome an encounter with tactic A, why should we even care that you can overcome the same encounter with tactic B? If the objection is that you can overcome encounter X with tactic A, encounter Y with tactic B, and encounter Z with tactic C, I am equally confused. After all, we want characters to be useful and competent in all encounters we throw at them so that no one is useless and sits playing Smash Bros. rather than D&D.

[Username17]

If you can overcome an encounter with tactic A, why should we even care that you can overcome the same encounter with tactic B?

Because it means that encounters that require Tactic A and encounters that require Tactic B are both your bitch. The MC can sort of fake intra-party balance by adjusting encounter composition as long as there is anything the weaker characters can do that the stronger characters can't.

The Cleric Archer is a huge problem for balance in actual play. Not because a Cleric isn't generally better off saving his spell slots to summon angels and drop doom tides on enemies; but because as long as CoDzilla can bust out a spikes enhanced ironwood morningstar and out-melee the beat stick classes, then the CoDzilla character is outshining the Fighter even in contrived scenarios where the players have to do melee damage. Most player power discrepancies are paved over with heavily weighted encounter selection and copious amounts of mind caulk. But when one of the characters is versatile enough to actually be able to sing "Everything you can do I can do better", that doesn't work.

A versatile and overpowered character is much more of a problem in actual play than a narrow and overpowered character is. Enchanters are broken, but if you really push the limits of charm monster the MC can and will send you through the gauntlet of undead, golems, and horde creatures until no one notices that the Enchanter character is still better than the other PCs. But when the Druid slaps down an animal companion that meat shields better than the Fighter, the jig is up.

Similarly, abilities that are very powerful like charm monster are more of a problem than abilities that are straight game destroying like wishing for items. A truly broken rule like wish is straight up instantly banned, modified, or stealth errataed. Something that is simply considerably more powerful than other things of its level has no such defense and will just run roughshod over the game.

-Username17

[Sashi]

The first problem is that "whether or not JaronK would allow it at his table" means exactly fuckall. He doesn't ban broken things for being broken, but he will ban broken things for being too simple. So if you combine five pieces over three sources to break D&D, JaronK approves. If you use one ability in core as written to break D&D, JaronK bans.

This comes directly from fighting games. Where one character is banned because he sucks but has a glitched unblockable sweep that's easy to deal with, and another character is Tier1 because he has a glitched supercombo that instamurders your opponent but is hard to do.

Which, for DnD, is pants-on-head retarded. It doesn't matter how complicated the steps are, because once you distil the combo into a character sheet you're still just hitting the "I win" button over and over.

[NineInchNall]

The MC can sort of fake intra-party balance by adjusting encounter composition as long as there is anything the weaker characters can do that the stronger characters can't.

...

... then the CoDzilla character is outshining the Fighter even in contrived scenarios where the players have to do melee damage.

...

But when the Druid slaps down an animal companion that meat shields better than the Fighter, the jig is up.

All of what you say I agree with, to some extent. I'm evidently drawing a different conclusion. To wit, the problem is the weaker character that needs contrived scenarios in order to (falsely) appear balanced with the rest of the party.

EDIT: But yes, I understand what you're getting at and why the versatile character "causes" problems in actual groups.