So, one bit at a time:
[sblock=RE: Alrough]
Arlough said:
You are defending a city that is being assaulted by the Daelkyr and their forces due to a dimensional rift. It perfect sense that they would replenish when you rest, given they have that rift available. The problem is, when do you rest? They control the timing. They control the terrain. You have nowhere to retreat to as this is your home as well.
If they are the ones that are attacking, then how does the adventure end and your adventure resources replenish?
For this scenario, the party closing the rift would probably be the focus of the "adventure." To save the city, they assault the daelkyr clustered around the rift and pull it shut. The four-person party above might have to find the item that can seal the rift (exploration/interaction), and then fight off the daelkyr between them and the rift (combat) until they reach the goal, seal the rift, and get to take an uninterrupted rest. If the party rests before then, the daelkyr replenish, or the artifact goes missing, or they move the rift (or open a second one!).
You could also break it into smaller bits, if you wanted a longer event in the campaign. You could have them secure a part of the city, to defeat enough daelkyr and build enough fortifications and save enough townsfolk that they can mount a counterattack -- or at least get a night's sleep. If they sleep before they've secured the zone, they get attacked, or the daelkyr replenish.
Arlough said:
You are walking somewhere (because all the horses died of plague, or because you spent too much money on that awesome armor, or whatever) and on your way to the next major city, you run into trouble. It isn't that surprising given the Points of Light nature of D&D, but this is a hostile land and you keep getting attacked. What is the end of the adventure here? Is it when you finally reach your destination, or is it when you finally reach a destination, or is it when the DM feels sorry for you and meta-games a break for you guys?
If it is just the DM giving meta-game replenishment, then how is this substantially different than the DM making sure that nobody gets an extended rest?
For this one, you can do it a few ways, too. The first one that jumps to my mind is to say that the adventure ends when they reach a point at which they can rest easily for the night (like a fortified town), and then put the encounters in between them and their destination (6 bandits, 4 starving wolves, and a vicious thunderstorm!).
If the party tries to retire early, the adventure recharges by more wild creatures or dangerous highwayfolk finding the adventurers, either while they sleep, or during the next day.
You could extend the journey out for longer by putting multiple "adventures" in between the party and their destination. If rests are on the days/weeks timescale, you even can end up with something very much paced like LotR, with long periods of wilderness travel punctuated by occasional civilization where the party can recover for the next leg of their journey.
It's not meta-game -- a DM in designing an adventure probably should have goals for the party already (or the party should have given the DM goals, like "We want to go to the Free City of Greyhawk,"). All this is doing is saying that there's going to be a set of challenges (encounters) between you and your goal.
[/sblock]
[sblock=RE: Hussar]
The trick with your example is that none of them could do all of the above. Luke studied hard and could do charm person and force punch. Circe had one spell. Achilles had one superpower. Gandalf spends much more time in LotR swinging a sword than he does casting spells. On and on.
It's D&D that gives us casters who can do every single thing that every single spell caster in myth and legend can do, and do it again tomorrow.
That, right there, is the crux of the problem.
When your one superpower is "I am completely invincible," I think you only really need one.
But your argument seems to be "They can do too much!"
To which my response is: not when they're properly balanced.
The paradigm I mentioned above gave wizards 3 spells per day. That's it.
I'd imagine a Vancian wizard having many more spells in their books than they can use in a day, but potential variety doesn't matter in play, only actual capability, and their actual capability might be, "I can only do three awesome things, and then I'm spent." That's vertical vs. Horizontal advancement, that is: they're limited to be on par with everyone else vertically (in raw quantities of win). Their horizontal power doesn't matter in that equation.
We can talk about the horizontal power, too -- the versatility -- but I think most folks would agree that divine casters in D&D have had a bigger problem with too much of that than the wizard ever did (spellbook costs and learned spells kept a lot of it down).
Hussar said:
We had a fairly bog standard group of adventurers including a rogue. We also stumbled across a Chime of Opening during our adventures. So, what happened after that was that the rogue got to open the locks first. But, if we found any locks that were too difficult for the rogue, we pulled out the Chime and popped the lock.
So, effectively, the rogue got to handle the pedestrian stuff and whenever it really mattered - because who's going to put a really difficult lock on something that isn't important? -, out came the magic and the rogue got to sit back down.
And that's just wrong. It should be the other way around. The rogue should be the one dealing with the really difficult stuff and the wizard can handle the run of the mill, everday stuff. The wizard (or in our case, any character, since a Chime of Opening isn't class specific) should never be able to do stuff in other people's sandboxes better than the class whose sandbox it is.
That sounds more like a specific problem with the
Chime of Opening being an at-will
knock effect than it is a problem with magic in general. There's a lot of other ways to design a
Chime of Opening.
Yes, as I've pointed out above, when your big-boom effect gets used more frequently, things become unbalanced, because the limit you assume on the big boom is that you can't use it frequently.
Hussar said:
But, the wizard, if he can do all these things, shouldn't be BETTER at it. I should not be able to consistently outdamage the fighter. I should not be able to consistently out problem solve the rogue. I should not be able to consistently out heal the cleric. ((granted, that last one's usually not a problem ))
The thing to note there is
consistently. The vancian wizard in my model above can't
consistently do jack.
Sometimes in the history of D&D, that limitation, of limited-use, has been forgotten. It doesn't need to be. And you don't need to overhaul the magic system or dramatically change the D&D wizard for that limitation to be remembered and enforced.
[/sblock]
[sblock=RE: pemerton]
pemerton said:
My concern with this is that it makes a certain minimum number of encounters between recharges crucial to balance.
Encounter-based design makes a certain minimum number of monsters in the encounter crucial to balance. I presume that doesn't concern you. Why does this?
Just like in encounter-based design, you can play with that number. There can be single-roll "minions" and entire-party "solos" and even entire-adventure "champions" (or whatever) in this, too.
pemerton said:
The game doesn't depend on resolving this one way or another - if the PCs rest up and return, and the encounter hasn't recharged, the encounter is now easier - but this doesn't effect the effectiveness of the PCs relative to one another.
I don't understand why you'd
want to use static encounters that do not change, if only because it creates a lot of bookkeeping (X dead critters, critter Y has Z damage, critter B has a (save ends) effect that is dealing ongoing 5 damage so here's a few d20 rolls to see what happens to it when no one's around...), and jacks down the challenge of the game to cakewalk levels (blow all our dailies and encounters, run back to town, rinse, repeat).
pemerton said:
But on your adventure-based paradigm, the GM has to worry not only about pacing, but about intra-party balance.
Not within an adventure, she doesn't.
Using unreactive encounters in general is a pretty "advanced DM" move, and if a DM wants to do that for some reason, and is still very concerned about this strategy, they should be advanced enough to think to ban Vancian spellcasters as inappropriate for their campaign. I think the DMG should absolutely talk about what is assumed in an adventure, and what happens when you change those assumptions. There'd be a lot fewer annoyed rogues in 3e if the DMG noted that a campaign full of undead and constructs will make them much weaker in combat.
pemerton said:
The closest that D&D has every had to a system in which a rogue can convince a king with 5 Diplomacy checks, and succeed at that task even if there are some failed checks in the meantime, is the skill challenge. Which almost certainly will not be part of D&Dnext.
"Checks which require multiple successes" is not the same as "skill challenges." It would be very odd if they decided that one check does everything possible -- and result in a binary game which I believe few would find very satisfying. You can have the former without the latter.
pemerton said:
I have grave doubts about the likelihood of D&Dnext incorporating a skill resolution system that makes having 10 goes, only 5 of which need to succeed, as effective as having a single auto-successful go (via a spell). Any such system that I can conceive of would have to have as much metagaming to it as skill challenges, and therefore would not be suitable for inclusion.
The nature of a "flatter math" system is that it relies on more die rolls to accomplish more difficult tasks, rather than just jacking up the DC.
The 3e DMG, under Diplomacy, just put "Make your enemy your trusted ally!" at a high DC. You could do it in one check, you just needed a VERY HIGH check.
The 5e DMG, under "Influencing NPC's," might include an "attitude track" where each success moves your target a little closer to being your friend. You could even take the 3e Hitting You In The Face-Hostile-Unfriendly-Neutral-Friendly-Helpful-Free Hugs continuum. The DC to change one to an adjacent one might not be that high by itself, but a trained character will hit it more often than an untrained character even with a modest bonus.
So 5 successful Charisma checks move the target from Hitting You to Helpful.
A
Charm Person spell instantly puts the target at Helpful.
You can also divide it up amongst creatures (which REALLY puts the wizard at a strategic disadvantage). If every NPC is effectively a social minion requiring only one success to make them friendly (something like the 4 kobold slaves I referenced above), the wizard's spell is still very effective, but it's a lousy use of power.
And the numbers aren't exactly set in stone. You can futz with them in any direction to get a granularity that's right for you. The numbers I gave are just the numbers you come up with using 4e's existing mechanics as a base -- which hit a pretty useful zone, I think.
pemerton said:
Without metagamed pacing and resolution management of a skill challenge variety, having the rogue use 10 checks on the king will require actually playing out 10 checks, some of which fail and have the king send the rogue away, and then involve the rogue coming back later to keep at it, etc
Each check can be more granular. I can imagine a rogue in 5e convincing a king by having the DM say: "Okay, tell me what you say and make a Charisma check" having the rogue's player give some statement, and roll a dice, and the DM describing the reaction to that particular statement, using the DMG's attitude track to describe how the king reacts to each check.
If you get sent away, that's a sign you've failed that particular challenge.
pemerton said:
Game systems that do make this sort of thing viable, such as 4e (to an extent, via skill challenges), Burning Wheel and HeroWars/Quest do so precisely by incorporating scene framing and resolution techniques of the sort that make 4e so widely unpopular.
All I know is that it's possible to have balanced Vancian casting a la classic D&D in an adventure-based model without making non-spellcasting classes feel like second fiddles all the time. I've tried to show that fairly empirically with all sorts of maths.
I think that's a reasonable goal for D&D, and since it's possible, I think it's reasonable that they would do that. I'm certain they're smarter than me and probably figured this out before 4e was even launched.
That doesn't mean they will, of course. Clearly I am not on the design team for 5e (just givin' out awesome ideas for free on ENWorld

). The concept that Vancian spellcasting is bad and wizards are too powerful and impulse to make everything homogenous is clearly a strong one.
But I don't believe it needs to be that way. You can have your cake and eat it, too. You can have your powerful, limited magic and your reliable, effective skills, accomplishing the same thing, via different methods.
I want that to happen.
[/sblock]
*whew*
Time for a drink.
