DnD 5E and FATE Accelerated

[jack]

DnD 5e

A while back I bought the 5e DnD ("DnD Next" or "DnD") player's handbook and just now have been reading through it. I actually really like it.

It reminds me of 3.5 but streamlined, with a few of the good aspects of earlier editions and 4e. That's about what I wanted out of DnD!

Many of the combat rules are simplified a bit, but look about equally balanced. Progression is simplified -- feats are more powerful, but optional, you can take them instead of a stat increase. Thus they do more to define your character and less to "here's a feat-tree you have to take".

There's no separate saves, you make a "dex save" or "con save". Your character has a single proficiency bonus which scales with level from +2 to about +5, which is added to everything you character is good at (weapons they're proficient with, skills they're trained with, etc).

They've added some fluff to the front page of the character sheet (personality trait, ideal, bond, flaw) and a suggesting for getting temporary mechanical advantage when your flaw comes into play. I have ideas for those bits up, to focus people further on the bits that actually come up in play (whether they matter mechanically or not).

The classes and races are similar to 3e -- there's the classic races (human, elf, dwarf, halfling) and further races (tiefling, dragonborn, gnome, half-orc) which don't automatically exist in all settings.

Like 4e, all spellcasters have a few infinite use cantrips which function as their standard attack options. I like that all characters have something specific to do in combat. And like 4e, fighter has some abilities beyond "hit it with my axe" to bring into play in combat -- although not many, I think that could be beefed up.

It reverts to generally winging the exact physical layout rather than using a battlemap. Which I like because combat is simpler and faster. Although I admit, it does remove some of the good effects in 4e, that there were many more tactical options for the party to work together, other than "we all hit it repeatedly".

The general power level is flatter between 1st level and 20th level, even more so than 4e. I think this is probably good, since it's almost impossible to balance things at both ends, but it does potentially mean less variation. But it has good effects that a character a few level higher than you" feels like "an adventurer like you, but more experienced" not "a demigod". And that there's less artificial scaling where every PC gets regular stat boosts to increase to-hit and damage-per-second and armour-class -- as does every monster.

It seems like, 1st level is really a tutorial level (although actually, I'd like an EVEN SIMPLER introduction for some newbies) where characters all have stuff they can do, but some of the key class features kick in at second level (eg. rogue has backstab damage at first level, but gets a free disengage/hide action from second which is nearly as class-defining). 4th or 5th feels like a typical point for experienced 3.5e players.

In addition to flattening the power level, the magic-item economy is gone. The classes are designed to be balanced mostly as-is, with a minimum amount of gold and almost no magic items. So you can run a low-magic campaign where the only magic is PC and NPC spellcasters, and add a magic sword for effect when it seems dramatic, not assume that everyone is carting around cartloads of +1 stuff else they're unplayable.

I think it could sensibly by used to run either an old-school "kick in the door and take as much treasure as you can before you die" session or a "mostly about roleplaying with some combat" session which are the sorts I enjoy the most.

4e is probably better for tactical combat -- I like that in theory, but never find it works well for me in practice.

Has anyone actually tried 5e?

FATE core and FATE accelerated

I've also been following a couple of people's suggestions and reading about FATE. IIUC it's based on ideas from FUDGE, based on a very freeform mechanics-light structure. Ideal for "here's a wacky idea about X" or "here's an existing setting (Dresden Files) with clear flavour but vague on specifics, can we adapt that to a game" and producing setting and character sheets with minimal write-up and no need to spend ages trying to balance PC activities.

Basically it sounds really fun if you want an adventure without tactical combat at all (there's still some tactics, but not based primarily on characters specific abilities).

Although some people apparently flounder if they're used to DnD -- there's definitely a "everyone should choose things that are appropriate, not always what would be most effective for the character". (Like Dogs-in-the-Vineyard, it seems it's more fun to pick character traits which come up about half the time -- but some people find it hard to resist arguing that they ALWAYS apply.)

Has anyone actually tried any of the editions of FATE?

Comments

[rysmiel]

Progression is simplified -- feats are more powerful, but optional, you can take them instead of a stat increase. Thus they do more to define your character and less to "here's a feat-tree you have to take".

The coarser granularity here is one of the things I found less appealing about it, fwiw.

There's no separate saves, you make a "dex save" or "con save".

I like that being official, though; that was one of the things I house-ruled leaning on a lot very early on in my time as a DM.

The general power level is flatter between 1st level and 20th level, even more so than 4e. I think this is probably good, since it's almost impossible to balance things at both ends, but it does potentially mean less variation. But it has good effects that a character a few level higher than you" feels like "an adventurer like you, but more experienced" not "a demigod".

That's kind of the least appealing thing about what I have heard of 5e to me. Perhaps it is overweening ambition, but I have always most naturally thought in terms of campaign running from "novices just starting out who would likely lose a fight with a log" through "veteran adventurers" and "significant political figures who still handle the high-end quests personally" to "legends who will be remembered for millennia", though pretty much every D&D-type system goes off the rails when you get into the last third to quarter of that arc anyway. Though that said, the part of my brain that does that is the same part of my brain that does fiction and writing fiction's a sight more important to me now and for the foreseeable future.

[jack]

Thanks!

The coarser granularity here is one of the things I found less appealing about it, fwiw.

Hm. I think it might be something like, I'd like the process of making a character unique by specific minor abilities that they have on tap, different to other characters with the same sort of mechanical build.

But a lot of the time in 3.5, there's a LOT of "these feats you must take if your group is optimised at all or your character just won't be effective" that don't have a lot of flavour justification, like "Power attack -- as you become a more skillful swordfighter, you can do more damage to people" which seems like it should apply to everyone. And a LOT of feats which provide small mechanical bonuses which you have to remember at various times. Like a board game. And I like doing that if I've time to get good at it -- if we've juuuuuust defeated some goblins, and I want to see, next session, can we improve our techniques and do better? But most of the time I DON'T have time to become good at it, it's just an endless vista of "here's 100 feats from the source book and 500 feats from other books you don't have and you can't wing it or it might be unbalanced, choose some, and if you choose wrong you may look greedy or stupid". Like, I might like to work up to that, but I've honestly not played enough to do so.

And I know some people don't have that problem because they play a tactics-heavy game and do know the pros and cons of various minor advantages. And some people don't have that problem because they don't focus a lot on combat and it doesn't matter if characters are optimised. But it felt more accessible to me.

I guess maybe, it's not the granularity, I'd like it if there WERE smaller feats that provided a _specific_ minor ability, but I liked that feats were less of a gotcha.

[rysmiel]

I think it might be something like, I'd like the process of making a character unique by specific minor abilities that they have on tap, different to other characters with the same sort of mechanical build.

I can certainly see the point of that, but (at risk of being grammatically irritating) I'd query the relevant scale of "unique". I mean, it would seem to me that how much needs to go into that for making a fighter feel unique in a party of four fighters of the same general build is rather different from how much needs to go into that for making a fighter distinct in your classic party of one fighter-type, one cleric-type, one sorceror-type and one skill-monkey. I have tended to favour the latter both for story reasons and... for all that I dislike excessive minmaxing, the choice between four characters who can each hit for on average ten points and one character who with a bit of thought and synergies with the other very different characters' skills can hit for an average of sixty points does seem to come out good on a pure combat-mechanistic front too.

But a lot of the time in 3.5, there's a LOT of "these feats you must take if your group is optimised at all or your character just won't be effective" that don't have a lot of flavour justification, like "Power attack -- as you become a more skillful swordfighter, you can do more damage to people" which seems like it should apply to everyone.

Indeed, and some of that feels to me very much like throwing all sorts of different mechanistic fixes at the very fundamental 2e issue with combat-types scaling linearly and wizards scaling quadratically such that they are way weaker for the first few levels and ridiculously much stronger at high level. (Like "give wizards lots of fairly low-level buff spells so that when they're at a point that they are tossing big destructive evocations around in combat that's not in competition with resources for them to power up fighters", and "for Desna's sake have all fighters take one level in an arcane spellcaster class at some point so they have true strike available for key moments.")

And a LOT of feats which provide small mechanical bonuses which you have to remember at various times. Like a board game. And I like doing that if I've time to get good at it -- if we've juuuuuust defeated some goblins, and I want to see, next session, can we improve our techniques and do better? But most of the time I DON'T have time to become good at it, it's just an endless vista of "here's 100 feats from the source book and 500 feats from other books you don't have and you can't wing it or it might be unbalanced, choose some, and if you choose wrong you may look greedy or stupid". Like, I might like to work up to that, but I've honestly not played enough to do so.

That makes a lot of sense, and I am sure I am thinking of this much more from a DM than player perspective; I definitely do have the drive for squeezing rulesets for interesting and effective outcomes, but that's far more likely to lead me to nuclear terraforming in Civ III than to thinking about RPGs.

(Incidentally, I am now 224 years into kittengame thanks primarily to the mention of it on your idle-games post a bit ago, and enjoying exploring the ruleset interactions and optimisation space of that immensely.)

And I know some people don't have that problem because they play a tactics-heavy game and do know the pros and cons of various minor advantages. And some people don't have that problem because they don't focus a lot on combat and it doesn't matter if characters are optimised. But it felt more accessible to me.

It certainly sounds so, yes. I shall be interested to see how your experience playing it works out.

[jack]

That's really interesting. I guess something like, potential reasons for feats include:

#1. Balance classes at high level by giving fighters essentially obligatory level-ups.
#2. Create more variety in characters -- a reason to play a whip-fighter rather than a trident-fighter.
#3. More options for optimisation and synergy.

I agree #1 was necessary, but it's not desirable it was necessary. I know *about* it but never played consistently enough to recognise the problem of my own experience. Likewise, I quite like the idea of "wiz, rogue, fighter, cleric", so I think having different sorts of fighter is superior but not necessary to enjoy the game.

I thought that's what _you_ were saying -- if not, how did you use feats, why do you care if feats are granular (or if they exist at all)?

I definitely do have the drive for squeezing rulesets for interesting and effective outcomes, but that's far more likely to lead me to nuclear terraforming in Civ III than to thinking about RPGs.

Likewise -- I love exploiting systems, but most of the time doing that isn't the most enjoyable part of a multi-player game. It's noticeable in magic -- I love the IDEA of discovering a deck which combines lots of previously-ignored cards into something surprisingly unbeatable. But when I actually play, having a tense back-and-forth where both decks do interesting things is usually more fun.

I shall be interested to see how your experience playing it works out.

Thank you, I'll report back if I actually manage to get a game going :)

[rysmiel]

That's really interesting. I guess something like, potential reasons for feats include:

#1. Balance classes at high level by giving fighters essentially obligatory level-ups.
#2. Create more variety in characters -- a reason to play a whip-fighter rather than a trident-fighter.
#3. More options for optimisation and synergy.

I agree #1 was necessary, but it's not desirable it was necessary.

Is there a word missing there ?

#2 and #3 strike me as positives worth aiming for. #1 strikes me as a problem that needed fixing, but possibly not at that particular scale.

Likewise, I quite like the idea of "wiz, rogue, fighter, cleric", so I think having different sorts of fighter is superior but not necessary to enjoy the game.

I like the idea of having different kinds of classes that do those different things, and I certainly think adventures that give synergistic challenges are more fun; not to my mind absolutely necessary to split the talents in those directions an even four ways - splitting the "clericness" of a party up by having a paladin who also gets part of the "fighterness" works for me if they add up more or less competent in all that set of skills.

I thought that's what _you_ were saying -- if not, how did you use feats, why do you care if feats are granular (or if they exist at all)?

In an ideal world, what I would want is a system which well supports ability to make fighter-types distinct, character and flavour-wise, from the other major types and to an extent among themselves. I'd like to be able to play Conan as distinct from Lancelot as distinct from d'Artagnan and have rules that mechanistically supported that well without making any one of those archetypes drastically superior to any other at the same level (Which probably necessitates taking away d'Artagnan's musket and pistols, but so be it). Whether the way the rules support doing this is through different base character classes or different prestige classes or different selection of feats is immaterial to me at a personal preference level; at an approachability level, I can see that a reasonably small number of base classes is a good thing, and that without some sort of careful higher-level rules about multiclassing, checking prestige classes for balance becomes a combinatorial nightmare, so optional feats appear to be the scale at which most of this could be done within something recognisable as the paradigm of D&D.

On one level my philosophical ideal is "lots of granularity, lots of options with interestingly different flavour, all equally good". On the other, I can see that enough playtesting to confirm no hidden gamebreakers is a hard task, and that in a finite world it's probably better to have fewer feats if they are ensuingly better-tested for balance. I'd like to imagine a model where anything that's as nigh-essential in practice as the Power Attack->Cleave->Great Cleave feat tree to keeping a fighter up to the challenges should become a fixed feature of the class and leave the optional choices for ones that are more interesting. Then again, I'm a long way from sure how mechanistically different fighters can be within the basic role of "the character that hits things", so maybe that's an argument for a finite range of well-defined feats too.

Same arguments apply, I think, to distinguishing a cleric from an oracle from an inquisitor, though well-defined game settings seem to have an edge up in making the divine-spellcaster options distinct by making the relevant gods distinct, flavourwise. Likewise, with the range of stuff covered by the full set of more-or-less traditional D&D wizard spells, it's fairly easy to split them up into a handful of distinct schools and get flavour out of saying "OK, you are particularly into/good at/trained in X but you don't get to do Y and Z. Or summon things."

I'm interested in the potential for entirely different paradigms that might work well with those classic four, and I don't know enough about things people have tried in that direction; never been in games where anyone played a martial-artist monk type, for example, and that seems to be a fairly enduring character class.

[jack]

Is there a word missing there ?

I mean, given the DnD rules at the time, it was necessary to have some way of giving fighters power progression that keeps up with wizards. But not every game needs that, it's better not to have that problem in the first place :)

not to my mind absolutely necessary to split the talents in those directions an even four ways

Yeah, I didn't necessarily mean _exactly_ those four, just that, characters with a range of specialisations ideally all timeshare time in a spotlight, and it's more forgiving to balance if you can be 10% better or worse and still the obvious "fighting" character, whereas if several characters are both optimised for raw damage output (or any other use), then being 10% better or worse can immediately make some of them feel superfluous.

I assume you can mix and match other combinations, although I don't know which work best.

I'd like to be able to play Conan as distinct from Lancelot as distinct from d'Artagnan and have rules that mechanistically supported that well

Yeah, definitely. It's blatant in 5e that wizards ideally don't know all spells for their level, so "enchantment wizard" and "blaster wizard" can be quite different characters. And fighters have SOME choices, but comparatively few, and only "archer or brawler or a couple of others", not the mix-and-match choose different spells can give wizards.

I think what I'd do is take a cue from 4e and from wizard design, and give fighters other abilities that are thematically appropriate, and function mechanically as spells, but are just dramatically appropriate. Either "as often as they like instead of a regular attack, but not always tactically appropriate" or "limited to N times per fight/session".

Eg. "Challenge, target makes a will save, else attempts to duel you in some way."
Eg. "Stunning blow, target makes fort save else is knocked unconscious (any ally may wake)" -- basically "fighter's sleep spell"
Eg. "Scything sweep, attack two adjacent enemy" there was lots of this in 4e, I'm not sure if it's in 5e or not, though hopefully it will show up if it hasn't already.
Eg. "Duellist's persistence" an attack where you engage for multiple turns doing minimal damage, but then get off a nova attack at the end that can function as a killing blow.

In fact, now I think of it, you could do the same between classes -- as magical classes share spells, but have different lists of class spells, martial classes could have a list of manouvers, and which ones you can choose for your class.

But this is all speculation, I've not tried it, I don't know if it would work.

optional feats appear to be the scale at which most of this could be done within something recognisable as the paradigm of D&D.

Right, I agree, SOMETHING should take the role of mechanical differentiation, and if so, it needs more granularity, but it seemed feats weren't really doing that anyway so I wasn't thinking in those terms.

If they were, you can add in some granularity, eg. "take a feat which gives you any three abilities from the mini-feats list" or something.

[rysmiel]

I think I am pretty much in agreement with everything you say here. Though I'm sure two of your four suggested distinctly flavoured attack notions are supported as feats in Pathfinder, and I suspect near equivalents for the others are too, so the 3e paradigm already enables that with a little bit of reworking. (Gosh it is helpful that they put all their rules stuff freely online and concentrate on charging money for the setting and adventure details.)

[jack]

Thank you. And yes, I love the open gaming approach.

That's good to know, although I think it's partly a matter of focus -- it's not so much that you CAN have the feats to do that, as being explicitly guided towards it in character creation, as with wizards choosing spells. Or that, if you decide feats need a "once/encounter" limit or similar, there's an existing framework for it, each feat doesn't need to describe it differently.

[jack]

FWIW, I was looking at the 5e fighter and one of the paths, martial, did do quite a bit of this -- you get to choose 3 from a list of manouvers, and get to apply them so many times per combat, borrowing some of the better fighter-features from 4e. And I liked it, although I felt there was room for a lot more.

[seekingferret]

You might also be interested in Legend, a 4Eified remix of 3.5 designed to solve this problem of 3.5 by creating a bunch of balanced, mix-and-matchable level progressions. http://www.ruleofcool.com/ It's a free download on that page.

[rysmiel]

I have visitors for the next week or so but will likely check that out shortly thereafter; thanks for the recommendation.

[jack]

I like that being official, though; that was one of the things I house-ruled leaning on a lot very early on in my time as a DM.

Oh yes, I forgot how it worked in 3.5, there was a one-to-one mapping from con/dex/wis to fort/reflex/will save -- but they were called something else. In 4e, they made each save depend on the better of two abilities, which stopped being a one-to-one mapping. That was good in some ways (it gave you more flexibility to depend on different core attributes) but bad in others (that basically everyone was reasonably ok, there was rarely a drawback of having _bad_ saves).

Having a con/dex/wis save seems to make more sense. You can still get bonuses to it. And IIRC they also have an str/int/cha save (which is used by some things but not much yet). That seems more rounded, albeit more complicated.

[jack]

That's kind of the least appealing thing about what I have heard of 5e to me.

I would basically never have time for a lev-1 to lev-20 campaign, although I wish I had been able to at some point. I like the possibility.

But someone pointed out in 3.5e, it's not so much like "level 1 is Pippin, level 20 is Aragorn", it's more like "level 1 is Pippin, level 5 is Aragorn, level 20 is Manwe wrestling Morgoth personally".

But Pippin->Aragorn seems plenty for 95% of epic journeys. Pippin->Manwe is beyond most people's conceptual scope, difficult to cram into a setting without breaking the settting, and the mechanics which are fun for modelling "Manwe wrestles Morgoth" are often _different_ to the mechanics that are fun for modelling "Pippin wrestles gollum", not just "the same, but everyone does +200 damage".

I've not looked at the higher levels, because I want to ease myself in with low-levels first, so I don't know if it works out mechanically, but the intent seems compatible with "epic journey from never-been-on-an-adventure to hero-of-the-ages", especially if the DM knows to stock the world so the players stick out about the right amount.

[rysmiel]

I would basically never have time for a lev-1 to lev-20 campaign, although I wish I had been able to at some point. I like the possibility.

I came pretty close in four and a bit years in secondary school, and while that was nominally weekly during term time, it was interspersed with other people's games in other generic fantasy settings, bits of Judge Dredd, quite a lot of Talisman and embarrassingly much teenage drama, so it probably took a lot less than half the sessions. And I know I'd manage it better now, I could probably have whacked more than 10% off the running time just by banning summoning spells. (Summoning a couple of small earth elementals to help fight a band of goblins slows things down a little. "This round I will summon a warrior angel with half a dozen special abilities, and next round a brass dragon with half a dozen other special abilities, and next round a titan with more special abilities again, to fight the Big Bad and his set of henchmen with all their special abilities" just turns into a quagmire.)

The first couple of Pathfinder Adventure Paths run level 1 to 20, at the fastest advancement rate 3.5 supports or the even faster default Pathfinder rate; the next few run 1 to 17 or so on the grounds that the very top end has relatively few players interested in it and also tendency to turn into a morass. I have looked through some of those and thought about them in detail, as both fomorian and zorinth are playing and/or DMing them with the sort of modifications that need a lot of thinking through.

But someone pointed out in 3.5e, it's not so much like "level 1 is Pippin, level 20 is Aragorn", it's more like "level 1 is Pippin, level 5 is Aragorn, level 20 is Manwe wrestling Morgoth personally".

That feels like a bit of an exaggeration to me; certainly the latter would feel more like the kind of thing you'd want to be a fair way into the old 2e Immortals rules (which I never got to DM) than your regular characters capped at level 36.

But Pippin->Aragorn seems plenty for 95% of epic journeys.

I suppose so. I'd like to be able to cap out at a Hercules or Cuchulainn scale, myself. Have rules that reasonably support "nigh-demigod cutting through army of mooks to reach arch-villain" without having to roll for each and every orc along the way, sort of thing.

Pippin->Manwe is beyond most people's conceptual scope, difficult to cram into a setting without breaking the settting, and the mechanics which are fun for modelling "Manwe wrestles Morgoth" are often _different_ to the mechanics that are fun for modelling "Pippin wrestles gollum", not just "the same, but everyone does +200 damage".

Agreed entirely with your third point, and I've yet to see a system that felt to handle that scale really well at all, let alone in ways that flow organically from lower levels; Eternity Publishing (which is one guy, who unfortunately seems to have named his company after how long it takes him to get anything done) has some very high-level stuff that feels to me just from reading like 70% of an excellent game. (The other 30% being either incomplete, or really really strange.) As for your first, I don't feel I have a sample size to judge either way; as for characters that powerful breaking the setting, that to my mind is what Outer Planes are for.

[jack]

Oh cool. I'm glad you had the chance to do a full-distance campaign (what was it like?) I'd certainly _like_ to, just not enough to devote more than a year to it, especially with the chance of people having to drop in and out.

Sorry, I think I am exaggerating a bit with the range of 3e, I think what you say is basically reasonable, but I think it made sense to narrow the range, and it's possible they've gone too far, but I still like that they have.

[rysmiel]

What it was like at the time was mostly a lot of fun, plus learning to balance campaign arcs that went somewhere with giving people enough to be interested in in the moment, which had some problems when the group fissioned and I got left with all the wizards.

In retrospect, I wince some at how derivative it was and how very much better I could do today, given world enough and time.

How often are you hoping to be able to play ?

[jack]

Cool.

I wince some at how derivative it was and how very much better I could do today, given world enough and time.

I symapthise with that a lot. But I think that's basically inevitable -- no-one is born knowing how to GM (or any other skill) and you generally learn by trying and improving. You only know better because you DID try. If you didn't wince it would mean you never tried, or never improved :)

How often are you hoping to be able to play?

I really don't know, most probably my interest will flame out in a little bit. I think the most realistic goal is to establish a simple one-shot I can use with people who have little experience, so I can practice GM'ing, and have an evening introducing people to roleplaying which is worthwhile even if it doesn't go further.

If it did go further, I'd like a regular play session (maybe once a month? possibly once a week but probably not), maybe as a world scattered with one-shots that can naturally be chained together ("Most of your party were satisfied with their loot from the ruined temple and didn't want to set out so soon, but the Gifa Redaxe and Sorilea gathered some more companions in town and vowed to beard their ertwhile antagonist in her lair..." :)) but I don't know what would actually work well.

[seekingferret]

Have you seen West Marches? http://arsludi.lamemage.com/index.php/78/grand-experiments-west-marches/ It struck me as a pretty smart approach to the problems of an inconsistent player pool.

[jack]

Oh, cool! No, I hadn't seen it, but was drawing from a very similar concept at a different GM's blog, http://thealexandrian.net/wordpress/1223/roleplaying-games/opening-your-game-table

I wasn't sure if I could do that, but it was in my mind as something to potentially transition into if an occasional one-shot works. I imagined having a world where players _could_ just go and explore, filled with background politics, mini dungeons, etc. And organising sessions implicitly in that shared world, at first just "lets do the mini-plot for this mini-dungeon", but that if it becomes a regular thing, players can express their own preferences for exploring things, or I can fill out simple plots with encounters stumbled into on the way, etc.

[jack]

I'm interested two people ended up with a similar approach here, it suggests it's a useful one for people with a regular desire to play but difficult schedules. And a nice mix of old-school "go into the dungeon and maybe it'll just be too tough for you" play with a more modern "here's a world that makes some kind of sense" play.

[seekingferret]

Yeah... someone in the comments to the post at the Alexandrian mentions West Marches, so I think it's likely that there's an overlap in gaming aesthetic as well as in scheduling needs, but it seems like a smart idea.

[rysmiel]

That sounds like a lot of fun and I hope it works out.

Are you planning on players having consistent characters, or getting to have new sets of characters in different adventures ? The latter can be fun if people want to explore different types of characters, and there's nothing like the expression on the face of a player when they realise the powerful wizard whose tower they are clearing out is the character they moved on from playing a few games ago.

[jack]

Thank you, I'll update livejournal when there's any news :)

Are you planning on players having consistent characters, or getting to have new sets of characters in different adventures ?

I haven't thought about it, but similar campaigns sound to have used a mix, encouraging players to develop a character so there's some continuity of plot between different parties, but allowing different characters when it seems interesting (or to play with different level parties).