Thursday, January 19, 2012

D&D 5th Edition: Design Thoughts of My Own

The announcement of D&D Next, 5th Edition, whatever you want to call it, has unsurprisingly lit a fire under the gaming blogosphere, and now that I'm past the DtD one-day, I'm certainly up for commenting on what I'd like to see and the problems that I hope they attack. Much of this has received direct or indirect comment from Mearls, Monte Cook, or one of the pre-alpha playtesters, and I'll try to include quotes where appropriate. All quotes are lifted from the ENWorld 5e page.

They've talked a lot about modular rules, even though it's not absolutely clear what this will mean in practice.
"The new edition is being conceived of as a modular, flexible system, easily customized to individual preferences. Just like a player makes his character, the Dungeon Master can make his ruleset. He might say ‘I’m going to run a military campaign, it’s going to be a lot of fighting’… so he’d use the combat chapter, drop in miniatures rules, and include the martial arts optional rules.” - Mike Mearls 
I would like to think that this is opening the game to genre emulation of many disparate genres, without promising to emulate all genres at once. As a DM, I feel that some campaigns need cinematic wuxia, some need just enough simulationism to achieve immersion, and some need to wallow in unrelenting squalor. Some campaigns need magic items to be common, such that the game's math expects them to be there and operates according to expected gear levels; some campaigns want magic items to be rare and significant, without the game's math collapsing into constant frustration (and, ideally, without invalidating the Monster Manual).
"With fourth edition, there was a huge focus on mechanics. The story was still there, but a lot of our customers were having trouble getting to it. In some ways, it was like we told people, ‘The right way to play guitar is to play thrash metal,’ But there’s other ways to play guitar.” - Mike Mearls

Let's run through a quick example in a little more detail. A huge amount of 2e's splatbooks were genre-emulation mechanics for various historical eras (books with green covers), often including the partial or complete removal of healing magic - but often ignoring any gameplay-level need to replace it. On one hand, this made the games gritty in the extreme; any damage was going to be with you for a few weeks. On the other hand, this is further along that spectrum of realism than I necessarily want - let's face it, it's as inconvenient to the DM as it is to the players when two party members are decisively out of the action one encounter before the climax. It wasn't until later editions that they really explored alternate, non-magical justifications for healing, but that's exactly the kind of thing I look to modular rules to handle.

In a related theme, sometimes a campaign needs complex and detailed rules for sea travel, naval combat, and negotiating trade values for captured goods (for example, a Pirates of the Spanish Main style), while other campaigns would be burdened by anything more complex than some travel times and a Diplomacy check. If the DM announces the modular rules that the campaign uses at the start of the game, the PCs have a much clearer idea of what to expect, and can build their characters accordingly - perhaps picking up feats, classes, or whatever that are only available as part of that rules module. (The point I'm making here is that rules modules can affect both the frontend and the backend of the system, and that's strongly desirable.)

After the break - commentary on many other topics.
I think it is painfully obvious to everyone who reads the vague promises that the designers have made with regard to "unifying" the editions that all you have is a new standard that is, as likely as anything, only going to fragment the community one step further.
"...this sounds so crazy that you probably won't believe it right now—we're designing the game so that not every player has to choose from the same set of options. Again, imagine a game where one player has a simple character sheet that has just a few things noted on it, and the player next to him has all sorts of skills, feats, and special abilities. And yet they can still play the game together and everything remains relatively balanced. Your 1E-loving friend can play in your 3E-style game and not have to deal with all the options he or she doesn't want or need. Or vice versa. It's all up to you to decide." - Monte Cook
On its face, this is such a ludicrous and impossible claim that I am deeply entrenched in Believe It When I See It. Either that, or "relatively balanced" are the world's most evasive weasel words. Event if they are weasel words as I cannot help but suspect, I still think the designers are deceiving themselves to a great degree. A quick scan of the OSR blogosphere - a significant faction of people 5e is trying to return to the purchasing fold - reveals that many of them hold up descending AC and chart-lookup THAC0 as the cattle of the very gods, while it's a cold day in hell that your average Pathfinder or 4e player would even entertain the notion of returning to such a system. To be fair, though, the OSR faithful who fall into this category have effectively cut themselves out of WotC's potential market; I'm aware that the real target here are those tasty, money-spending Pathfinder players.

Still, this claim highlights one of the major areas where D&D has always had problems: the value of choices made during character creation is almost impossible for a new player to appreciate. When we first came to 4e, we had a table full of experienced and perceptive players, but we still made choices that would prove to be painfully suboptimal in the first few sessions of play; 4e was sufficiently different from its predecessors that our experience in 3.x and prior editions severely misled us. 3.x and 4e have pursued the goal of character customization to a very great degree: the idea of allowing players to make exactly the characters that live in their minds' eyes, and ideally do so in a mechanically compelling way. Pathfinder picked up where 3.x left off and ran with it - even as a very experienced gamer, I would say that I am daunted by all the different choice points that are available to me in customizing my character.

The thing about all of these choices is that offering them creates a chance to make suboptimal choices. A player operating with system mastery on his side may make suboptimal choices for reasons of experimentation or character appropriateness, and that's fine. A player who lacks system mastery will make a much greater number of suboptimal choices (3.x Toughness, anyone?) out of ignorance, and that choice will stick with them anywhere from a level or two (4th edition) to the rest of the character's career (all forms of the game that lack retraining mechanics). If there are varying levels of system mastery within a party, as there usually are, that player is stuck wondering why his character is so egregiously outdone by other party members. The books themselves also do a very poor job of guiding players toward good choices; the first thing every player should learn is that D&D, just like The Elder Scrolls: Oblivion, shows you default choices that are some of the most incorrect of all available choices.

To hook this back into the quote above, though, let's look at Alice, Keeper of the Old Ways, and Bob, 3.x Hypercustomizer. Imagine for a minute that the game does offer Alice the small number of decisions that a 1e fighter ought to be making. If I'm not mistaken, this boils down to race, weapon(s), armor (and shield), languages known, name, height, weight, hair color, and age. At the same time, it offers Bob at least a significant fraction of the choice points that a 3.x fighter ought to be making: race (and subrace), alternate class features, 2-3 feats (from a list of hundreds - fortunately only a few choices are worthwhile at first level), skill points (4*(2+Int bonus)), a wildly longer list of weapons, shield and armor carefully selected to fit Bob's Dex modifier, languages known, name, height, weight, hair color, and age. Possibly Background, if the DM's using one of the many different optional Background tables.

Okay, so they make a different number of choices, so what? Well, there are two ways this could go. There could be bonuses built into Alice's options that compensate her for whatever extra Stuff Bob gets from his feats, skills, and the like. This is probably, technically, the sanest way to do this, but it's pretty ugly from Bob's point of view. They aren't really playing the same class; Alice must be getting some kind of scaling bonus about as quickly as Bob gets feats, while Bob's feats must be a tradeoff in situational or otherwise limited power. The other way things could go is for Alice not to receive a compensating bonus, so Alice is just playing a 3.x fighter who doesn't get feats or skill points. This fails the "relatively balanced" test, and in a big way if WotC's design history of feats is any judge.

But I've belabored this point beyond all reason; let's move on.

As far as I'm concerned, probably the single most important thing for them to tackle in this edition, the thing that most resolutely divides 3.x and 4e in a DM's usage, is how NPCs are statted. In 3.x, there is an overriding conceit that, unless rules for race and creature type alter the playing field, NPCs are built on the same framework as PCs, with skill points, feats, and possibly character levels. Things that aren't core PC races often also have the possibility of advancing by Hit Dice, which is a lot like advancing by levels, but with radically different benefits. Therefore the DM works through all the same choice points in statting a given NPC that a player makes in constructing a PC of the same level, if not more. The DM may feel a greater freedom to ignore a lot of those options, since this NPC is only going to last as long as it takes for the players to bring about weapon-induced facial murder, but the DM does this work in high volume.

In 4e, as I've discussed in previous posts, it's not feasible to find points of solid connection between the construction of PCs and the construction of NPCs. NPCs use different stat-advancement math, NPCs ignore bonuses from ability scores, NPCs don't have feats, NPC magic doesn't use PC-side cosmology, and NPC use of magic items is strictly cosmetic. On the plus side, I can put together a whole party of fully-functional 4e NPCs (or re-skin existing NPCs) in less time than it takes to make a single 3.x NPC. That word "re-skin" - that's a 4e thing, because re-skinning fundamentally acknowledges that the underpinnings of how the NPC came to have the powers that it has is meaningless - there is no cosmological why.

The way these two editions handled NPC creation define the poles of a whole spectrum of design possibility. Original D&D on through 2e (at least pre-Skills and Powers 2e, which has from time to time been called 2.5e) had very little complication in creating a PC, and thus it was possible to create an NPC through the exact same means in a trivial amount of time. On the other hand, they mostly didn't bother, outside of spellcasters; they summed up relevant combat stats in a string of character far smaller than a tweet. The height of convenience - but that convenience came through sacrificing interest, as far as I'm concerned.

I hope WotC finds a middle way. If I were designing it, I'd focus on pre-packaging options for NPCs of various types (each of them within reach for PCs), possibly in searchable electronic content. I've gotten out of the habit of using a laptop or other electronic devices at the table, but I could get back into the habit if they remembered to make it convenient to pick things up and move them around on my computer. Oh, and for the love of the new decade, don't forget the simultaneous release of the same software to iPad and Kindle Fire. This is important: if all the tools I need to build encounters and NPCs are on my Kindle Fire and working fast, I'm not fumbling with a stack of books that are each bulkier than the machine, and I didn't have to murder a forest in my session preparation. (Also? Templates were a really neat idea in 3.0 that just became burdensome to apply. Computers are really good at the logic that goes into applying a template. Make them do the heavy lifting.)

Signing off for now, but I have a lot more to say on this topic for future posts.
Post a Comment