In today's interview, we speak with Rob Heinsoo, Lead Designer for 4th Edition Dungeons & Dragons, as well as creator of the Three-Dragon Ante and Inn-Fighting games. Rob discusses early concepting for 4th Edition, approaches explored, and where the game is headed next.
Wizards of the Coast: As one of the Lead Designers for 4th Edition Dungeons & Dragons, what changes were you seeking to make in the system—that is, what elements of the previous edition were you looking forward to changing (either modifying from how things worked before, adding new to the game, or dropping from the game)?
Rob Heinsoo: My goal designing 4th Edition was to make a game that played the way I thought D&D was going to play, back before I understood the rules.
I read about D&D in1974 in Military Modeler magazine and bought the game by mail order. I'd read The Lord of the Rings, but not The Hobbit. I was ten years old and I didn't fully understand the D&D rules for another year or two, but I loved the feel of the game and its fantastic open-ended universe. I wanted epic battles and characters who could fight like Aragorn or Legolas or Gimli or Gandalf using powers that suited those characters. I wanted my 'fighting man' to be as tough and heroic as John Carter of Mars.
Some people have noticed that playing a first level character in 4E feels like playing a fourth or fifth level character in 3E. That's not an accident. First level characters in 3E could die the first time they took a hit. Worse yet, a first level 3E wizard or cleric could use up their spells in the first encounter and have nothing else to do. That might make sense if you're simulating a specific type of fantasy world where magic gets used up quickly, but it doesn't make any sense for new players who want to have fun playing the game.
So 4th Edition is meant to be played. I'd like everyone who starts playing the game now to have an amazing experience that teaches them to love roleplaying. All the initial design goals I'll mention in this interview—making games fun at first level, making sure that all characters have a cool role, giving every character fun powers—these design directions were aimed at making 4th Edition so much fun to play that 3E gamers, 2E gamers, fantasy readers, and just-plain-folks get curious about playing D&D and keep playing after trying the game once.
Expanding the Sweet Spot
Third Edition D&D is a good game; in fact, it's so good that some of its problems are easy to miss for long-term players—they're just part of how the game works. The first such foundation-problem to tackle was 3E's relatively small "sweet spot," the level band in which 3E games rocked. Before fourth level, 3E characters had few hit points or really interesting abilities. The system math didn't work out right at the lowest levels; e.g., a single critical hit from an orc with an axe could take out low level characters.
Games that are vulnerable to one-roll accidents aren't the best games, though they might fit certain narrative styles of gritty sudden-death adventuring. Once you reached fourth and fifth levels and above, 3E characters had enough hit points to hang on for awhile and enough spells for a few encounters instead of running out of their coolest abilities right away.
But 3E's sweet spot ended around 11th or 12th level. Up at 11th level, the spellcasting characters started getting 6th level spells, spells with enough power to truly alter the way the game was played. A big problem with games that included 6th level spells was that most DMs stopped being able to truly predict what their PC groups were capable of, as cunning (or maybe just brute force) magic use could short-circuit most high-level 3E encounters that seemed like they were balanced.
So for 4th Edition, we determined relatively early that our goal was make the game "all sweet spot", to give starting characters enough power and survivability that they were fun to play from the very start. 4E's philosophy is that we want to reward you for playing the game right away, not make you play a character that sounds cool, like a first level 3E wizard, and then find out that you only have one or two spells… and that if you use them too quickly, you're going to be known as the wizard who runs around shooting at things with a crossbow.
That's why 4th Edition D&D is so focused on giving characters at-will abilities, as well as abilities they can use once an encounter, and abilities that are so powerful they can only be used once a day. We want all D&D characters to have the option of feeling heroic, to keep fighting and adventuring until they are truly too beat up to continue, and not to stop as soon as they have used up their only cool powers.
All Classes Must Rock
Getting back to your original question, I hated the fact that once you started playing level 11+ in 3E, the non-spellcasting character classes didn't matter as much as the spellcasters. There was fun to be had as a fighter, or as a monk (mostly through roleplaying), but the truth was that adventures usually depended on the abilities of the wizard and cleric—where a missing wizard or cleric got some high-level 3E games I was in rescheduled. Did 3E games get rescheduled if the fighter was missing? Only if the character was central to the storyline of that session, not because the group actually depended on the fighter for survival while the wizard and the cleric were around.
The fact was that in the 3E world, wizards were the most powerful characters, heirs to a fantasy tradition from Dying Earth, Lord of the Rings, and Forgotten Realms in which the earth-shakingly powerful characters were usually wizards.
We had to change that for our game world. From the start we wanted to put 4E's character classes on more even footing. We hoped that more equal characters would help groups play games together longer instead of having 3E's problem of high-level campaigns breaking down without being certain why, when some of the players stopped having as much fun as the other players.
Of course there are places where it's OK to have uber-powerful spellcasters—from the perspective of a fantasy novelist, it can be hugely useful to have one or two character types that happen to be more powerful than all of the other characters in the world. But D&D isn't a fantasy novel, it's a shared world roleplaying game. When you're playing a cooperative game with friends, it's better if the baseline is that every character class has roughly equal potential for kicking butt and using powerful abilities that shape the game.
I shouldn't act like this was a simple decision to make or carry out. There are a lot of people who don't want to let go of the idea that the wizard should be the most powerful class. The first Player's Handbook teetered back and forth between design drafts and development drafts, and sometimes the wizard had been deliberately bumped up to be slightly better than all the other classes. I wasn't comfortable with that, and the final version of the wizard is, if anything, possibly on the slightly weak side; the wizard was all alone as the first practitioner of the controller role and we stayed cautious knowing that we could improve the class later if we needed to.
We've since acquired a better understanding of the controller role while working on Player's Handbook 2. So of all the classes, the wizard may be the class that improves the most after getting its full treatment in the book for its power source. It's not necessarily that Arcane Power makes the wizard stronger, but Arcane Power adds options like summoning and familiars that are part of the wizard's inheritance. We always knew that the wizard had "too much stuff" to be handled completely in the first Player's Handbook. With Arcane Power I think long-time wizard players will feel a bit more at-home with the class.
Powers For Everyone
So how were we going to put the classes on a more equal footing? Given how much fun 3E's spellcasting characters had choosing spells, we knew that giving all player classes interesting power selection choices was probably the way to go. I wanted a game in which playing a high-level fighter could offer interesting choices for power selection and round-by-round choices in combat. Character class options, even for characters who weren't using outright magical powers, needed to remain interesting at all levels, instead of having 3E's problem where high-level fighters might get more power from their magic items instead of their own innate abilities.
I had a personal stake in this goal. My favorite 3E character was a dwarf fighter named Sigurd in Jonathan Tweet's Elysombra campaign. But at 12th level, the point when the other PCs were hitting full stride, I was succeeding purely as a consequence of correctly guessing which magic items I should pick up before the adventure. Oil of slipperiness and a flight ring? Pure gold. Sigurd's fighter abilities? Irrelevant. So I switched to playing a tricked-out half-dragon gnome psychic warrior, thinking I'd have more fun. The story didn't suffer, given that my new character ended being the daughter of the morally ambiguous elder dragon of the campaign, but the other players never really forgave me for abandoning Sigurd.
The point of bringing in powers for every character class was to make the game fun for everyone, most all of the time. If I could replay Sigurd in 4E, I'd have a high-level fighter who had taken some multiclass cleric feats. I'd still be choosing fighter powers that mattered and I'd be protecting the rest of the party by locking down monsters when I wasn't busy killing them.
Early in design we realized that each character class needed to fill a solid role in the party. I can't take credit for understanding this from the beginning: James Wyatt and Andy Collins understood it at the start and got me on board. We all knew that character classes like 3E's bard and monk were beloved by some, but mostly for flavor reasons. In terms of being playable, neither the bard nor the monk had a clear role, they weren't as good at anything specific as the fighter, wizard, cleric and rogue were, and so the bard and the monk lagged behind the other classes when they weren't just ignored from the start.
The new 4E bard, appearing in PH2, and the upcoming monk both have clear roles, things they excel at and fun ways they're different than every other character class that fills that role. We want to reward players who think that playing a bard or a monk will be fun, not hand them a subtly poisoned time-delay capsule that will eventually wake them up to the realization that they're the weakest member of the party.
Wizards of the Coast: Were there any initial directions you pursued that were later abandoned or significantly altered? What might have waited at the end of these design alleys?
Rob: Yes, many! I'll run through a few of our abandoned notions, roughly in the order in which we abandoned them.
Power Keywords: The first draft of the power system assigned keywords to every power and had the notion that different character classes and class builds would have varying access to the keywords, so that powers were cross-class. That was a bad idea for D&D's class-based system in many ways, though I presume you could organize some other game around such a system.
Powers Every Level: Early on we somehow fixated on giving characters a power at every level, not realizing that this led to way too many powers in the game and didn't leave room for feats. We figured that out eventually.
Condition Tracks: From the start we knew that we wanted to get rid of 3E's save-or-die effects, attacks that could knock out a PC with one hit or one-die roll attacks that could cause a PC to become paralyzed, enchanted, or otherwise helpless but still alive. But our early attempts to come up with a fix used multiple condition tracks that turned out to be a lot more trouble than they were worth. The new Saga Edition Star Wars RPG ended up using a much-streamlined version of the condition-track system, but for D&D we turned toward effects that either lasted one round or ended with a saving throw.
My only fond memories from the days of the multi-track system were moments of dialogue I recorded in our quotes file. Someone said, "What's the opposite of the petrification track?"
"The liquification track. Aboleths: be very worried when they bring out the straw."
"No, we don't have a liquification track because it's part of the swallow-whole track."
A Weird Damage System: Once upon a time the game had a weapon damage system that distinguished between high impact weapons like maces and armor penetrating weapons like rapiers. It was an elegant little system, but it didn't feel right once you got away from weapons and started to account for damage from spells and powers. Worse, it didn't feel like D&D; messing with the basic way that damage was dealt with additive dice rolls wasn't something that we could pull off under D&D's umbrella.
Too Many Renewable Powers, Not Enough Attrition: Also in the 'didn't-feel-like-D&D' category, we spent a lot of time experimenting with systems in which all powers were limited use at-will powers that had recharge mechanics. I blame myself for thinking something like this could work. In truth the system didn't start feeling right until Mike Mearls and Rich Baker came up with the at-will/encounter/day split that put power attrition back into the game. For a look at what the earlier game looked like, see the 3E Book of Nine Swords which basically distilled the then-current version of 4E into a martial arts combat framework. It was fun, but not what you'd want for the whole game of D&D.
Multiple Power Acquisition Schemes: We weren't always planning to give all characters equal numbers of powers. Many times we experimented with vastly different power acquisition schemes for different classes. And when we decided against those approaches, there were people in R&D, including myself, who sometimes balked and felt like giving different classes different numbers and types of power might be a good way of differentiating between classes. But sentiment didn't pan out. All of our actual experiments with different power-distribution schemes didn't work out, so we moved ahead with the notion that a richer understanding of our system might give us room to experiment in the future.
The 'Traveller-style' Character Generation System: The original Traveller sci-fi RPG had a unique character generation system that allowed you to roll out your character's entire backstory as you were generating his or her abilities. Back in those days, back before we had computers, we rolled up a lot of characters that we never got around to playing, and Traveller was one of the most fun games to roll characters for. Some of my co-workers disagree with me, but I'm certain that part of the fun was the risk—if you pushed your character too far and got greedy for more and more skills, your character might finally fail a survival roll and you'd end up dying before you ever entered play. So creating a character was a game of its own.
I regale you with tales from the days before computers because a couple of times during the 4E process, Mike Donais and Dave Noonan and I worked away to see if we could generate amusing and worthwhile character generation systems that would be as much fun as a rainy-day activity as the old-style Traveller character generation system was.
The answer, so far at least, has been no. Not and stay simple and make sense for D&D. Some of the thinking that went into the project eventually contributed to the backgrounds systems used in the Forgotten Realms Player's Guide and PH2.
But the great news that I did not expect is that you can more or less "play" the D&DI Character Builder the way we used to roll up characters in Traveller! Go into the Character Builder and choose "Quick Character." You'll get a random 1st level character, complete with backgrounds, and you can level them up as you choose or let the Character Builder level them up. I didn't know it would be so fun. Maybe I should have known this would work out, since Mike Donais has been working on getting the Character Builder programmed with sensible choices.
Wizards of the Coast: You mentioned games that are vulnerable to one-roll accidents. Aside from D&D, are there other games you've played that you've experienced this in?
Rob: I had to think about this for a couple minutes. I've read lots of games that are vulnerable to one-roll accidents. But the RPGs I've played often, other than various incarnations of D&D, usually avoid the one-roll problem. The games I've played most other than various versions of D&D are Feng Shui, the Chaosium/AH versions of Runequest, Heroquest, and Everway, then followed by Champions and Over the Edge as games I thought about all the time but hardly played.
Feng Shui characters are action heroes who can take a terrible pounding and keep fighting. There are some balance problems with specific powers, but as Game Master I always fixed/ignored those powers, no problem. Likewise, Everway is about mythic heroes, ditto for Heroquest. A single unfortunate die roll doesn't settle your hash in these systems. Of these games, only Runequest had a marked one-roll problem, so it's possible that I've unconsciously migrated to games that allow you to keep playing instead of losing characters due to entirely abrupt randomness.
Runequest had one-roll accidents because of various character and monster powers, but the biggest quick-death factor in RQ was a weapon critical system that could eliminate even the strongest character if they got hit in the head or the chest with an impaling weapon. Once you became a powerful RQ character, you received Divine Intervention abilities that were supposed to give you a chance of escaping death… but such escapes were both dicey and costly, as you lost a random amount of power even if you survived. The D&D death saves system offers some of the same dramatic chance for a quick recovery but is a lot more generous than RQ and doesn't suck away your power points when you manage to stay alive.
But it's not fair to address this issue without acknowledging that high lethality and extremely risky combat is exactly what many roleplaying games want to portray. For example, one of the Legends of the Five Rings systems was pretty deadly, but the sense of the game was that fighting with swords was risky, and people died easily. Old time Traveller had a bit of that—you could get in an ill-advised gunfight against the wrong type of gun and kiss your vaporized character goodbye.
I think that some of the problems with D&D's earlier save-or-die effects came from similar thinking; that is, default approaches to portraying particular types of dangerous powers. If you're creating a fantasy game in which medusas can turn people to stone, wizards can cast disintegrate spells, and enchanters can dominate people's minds, the obvious first approach is to make such abilities deadly serious, nasty magic that has sudden and permanent effects the way they appear in fantasy novels.
But of course fantasy novels don't tend to turn their protagonists to stone or paralyze them through crucial scenes, not as a rule. So one of the changes from 3E to 4E was to get away from the sense that we had to simulate precisely how specific magical effects would hit everyone in a fantasy world and concentrate on portraying the experience of exceptional sword and sorcery heroes who nearly always have a chance of fighting through petrification and disintegration and domination and blindness in a couple of rounds—all the effects that could ruin an entire 3E session if you were unlucky.
Now that I think of it, Champions also had a one-roll accident problem when the game started out in the 80's. Champions is a point-buy superhero RPG adored by power-gamers who design uber-characters, efficiency experts who love elegant solutions, and players who want perfect control over their characters. Most Champions combat follows the example of multi-panel four-color comic book slugfests: long and detailed and full of many different maneuvers and attacks. But in the early days of the game, the Killing Attack powers broke that pattern. They weren't costed right, and even when Killing Attacks didn't manage to kill people outright by hitting the "Body" stat (a low number representing exactly how much damage you could really suffer before death) instead of having to whittle away at people's Stun (a very high number, more like D&D hit points), Killing Attacks dealt a huge amount of Stun damage because the game was simulating the idea that attacking people with guns and swords was really deadly. But it turned out that wasn't really what the superhero genre wanted; most people didn't want to have their superfights cut short by the killer superhero or villain who was using nothing but Killing Attacks. Simulating guns didn't fit the genre. So they changed the rules and eliminated the Stun modifiers for Killing Attacks (among other tweaks) and then Champions combat played out more the way people wanted it to. The Champions story somewhat parallels our decision to make sure that D&D PCs don't have their sessions cut short by save-or-die effects and that certain higher-level PCs don't eliminate threats too quickly by using 3E save-or-die spells against the monsters.
Wizards of the Coast: As a designer, where do you look for inspiration—is every game session a chance to playtest new material? Do you house-rule constantly? Do you alter the rules of other games you play?
Rob: Inspiration for creative work comes from all over. I read a lot, mostly novels (science fiction/fantasy novels) and short stories and magazines of all types, ranging from Archeology and Scientific American to The New York Review of Books. I don't say much about such interests in WotC D&D work, but my personal blog at robheinsoo.livejournal.com has more to say about how these interests intersect with gaming.
Playing new games is always important. I'm interested in everything from WWII squad-level games to ancients miniatures to card driven games to indie RPGs. There are many design ideas and insights in games within these somewhat separate genres that haven't been taken advantage of by designers in the other genres. Dominion is the favorite game in my family at the moment, and it's a great example of a design that applied new ideas from one genre (M:tG's draft format) to create something new in a different style of game (parallel deck-building).
I'm not sure my usual method of coming up with a new game design qualifies as a method of inspiration: I think about a type of experience I want to have while playing a game. If there isn't already a game that provides that experience, I start thinking about how I could create it.
As far as house rules are concerned, I tend not to house rule competitive games unless it's as simple as saying: "Look, Knightmare Chess is a great game, but we have to play without the Fireball card—it's broken and it ruins the game."
I don't have such qualms when I'm running roleplaying games. Then I'm almost certain to be using house rules, partly to fix problems and party to engineer the flavor of game or session I'm aiming for. Running 4E by the book, to make sure I understand how the game is working for everyone else… that's weird and different!
Wizards of the Coast: Beyond the pages of the RPG, you also helmed the design for Three-Dragon Ante and Inn-Fighting; how did these projects develop? In a past Design & Development article, you've discussed the genesis for Three-Dragon Ante, but what of Inn-Fighting? How did the game go from concept to finished product?
Inn-Fighting was a surprise. Bill Slavicsek asked the RPG R&D department to come up with some possible designs for a dice game. Then it turned out that we couldn't produce any of the submissions because of budget constraints. I inquired about what we could actually produce, then took an idea that had been kicking around in my head as an in-world D&D dice game played in taverns, and turned it into a tavern brawl dice game for gamers. Compared to what was originally suggested, I cheated immediately by adding cards—with the number of dice I'd been told were possible, I figured I could add cards to the design if I took out a bunch of dice, so I made a game with a very limited number of dice and a few cards.
The first version of Inn-Fighting kind of sucked. I think it was Chris Youngs who made the most helpful suggestion right away, about standardizing what the attack dice meant. The second and third versions of Inn-Fighting were better.
As you of all people know, Bart, I'm not entirely finished with Inn-Fighting. I'm working on finally getting the character conversion system done so that people can turn their D&D characters into Inn-Fighting characters. We promised that when we published the game and then I let 4E design get in the way. Logan Bonner worked out a better format for the conversion than I had figured out and now I'm trying to find the time to write it up so the programmers can create it.
Wizards of the Coast: Are there other such games you have in mind, at least in concept?
Rob: Yes, I always have a game design project going. This is the perfect moment to ask, since I've just finished design work on the sequel to Three-Dragon Ante. It's a deck of 70 entirely new cards using the same rules as the first game. You can either play this deck alone, or mix pieces of it with the earlier cards since the cards will have the same card backs. The goal is to make the sequel even more fun that the original 3DA. I've had great help from people at WotC and the sequel is now the version of 3DA my family wants to play. I can't tell you exactly when it will be published, but it's moving into production and we're figuring out the schedule. I'm happy.
Wizards of the Coast:How did you get your professional start in the gaming industry—can you tell us of your early design work, with Wizards of the Coast and even before (at Atlas or Daedalus Games)?
Rob: Everyone starts as a fan. As a high-schooler in the early 80's I started contributing to Alarums & Excursions, an RPG fanzine put out by Lee Gold. (A&E is still going, it just had its 400th monthly issue.) I wasn't trying to get into professional game design while writing for A&E, I was just having fun. A&E had many professional game designer contributors and a high number of contributors who went on to become professional game designers or game publishers: I met Robin Laws, Jonathan Tweet, and John Nephew through the fanzine. Then I started as a freelancer picking up small jobs that Robin didn't have time for. And then Magic: the Gathering came out. It was good timing for me. Left-brain/right-brain ambidexterity turned out to be a good trait for playtesting and designing trading card games.
I helped playtest Jonathan Tweet's On the Edge for Atlas Games and Robin Laws'/Jose Garcia's Shadowfist for Daedalus. On the Edge was one of the first wave of TCGs published after Magic. It stood out because it had something new to offer: interesting rank-building and influence mechanics that weren't just a restatement of Magic. In the long run, On the Edge's pacing wasn't right due to insufficient card flow. It was too hard to come back once you fell behind, and the game was too solidly designed around its intended pace to have an opening for an easy fix.
Playtesting is a common avenue toward working professionally. The trick of successful playtesting is to remember that you're not the designer. That is, your job isn't to revise the game and prove that you can design. The playtester's job is to tell the designers what's actually going on with their system, a somewhat scientific analysis and report of probable consequences. It's functioning as part of a team. If you play the role well and don't try to step on other people's roles, you demonstrate that you can work on a team, and that demonstration can go a long way.
For me, playtesting turned into editing work on Shadowfist, which turned into a job as the netrep at Daedalus, which turned into co-designing the first couple of Shadowfist expansions and helping on the Feng Shui RPG line. Unlike On the Edge, Shadowfist had legs. It's still a relevant design. Zev Schlasinger and his Z-Man Publishing company put out something like eight or nine Shadowfist sets over the years, and from what I can tell the game is still fun and its multiplayer dynamics work well.
My professional game design career seems to be about creating or flowing into unexpected opportunities. I signed on at Daedalus as a netrep and ended up helping design Shadowfist sets. I joined Chaosium to work on Greg Stafford's wonderful game world of Glorantha and that led to a job doing quality assurance for a Gloranthan computer game King of Dragon Pass, which then led to contributing to that game's design. I joined Wizards of the Coast to work as a writer on the D&D Worlds team, helped with 3E's Forgotten Realms, and ended up writing the story and flavor text for the Gold Edition of Legend of the Five Rings. Then I helped with a soccer trading card game for Europe called Football Champions and ended up running the design for a few years.
Playing the Chainmail minis game at lunch got me a job on the Chainmail team, and that later turned into working as lead designer on D&D Miniatures. Designing Three-Dragon Ante helped demonstrate game design skill, which allowed me to work with Jonathan Tweet designing Dreamblade. And all that bumped me toward leading the design of 4th Edition D&D.
I think of this path as a careen that has been maintained long enough to become a career.
Wizards of the Coast:Any other advice you'd have to impart for aspiring designers eager to work in the industry—courses they might take in school, games or gaming habits they might consider, projects they might design on their own for the experience?
Rob: If you are one of the people who immediately starts doing what you love, you're lucky. But if not, my advice is to stay open to opportunities that weren't your first plan. Don't be frugal with your creativity. You're going to have interesting ideas the rest of your life. Exerting yourself on the ideas you have now, even if they crash and burn, will help train you to make better use of other good ideas later on.
If you're going to work on roleplaying games, you'll need to improve your writing. The best writing advice I ever read was from Stephen King, who noted that anyone who writes three hours a day is going to be a good writer ten years later. If you can't manage three hours a day, even one or two hours a day makes a big difference.
If you're looking to be involved in a part of the game industry where you're helping imagine worlds and cultures, it helps greatly to have an understanding of how history has played out on this world and how cultures are put together. No matter what you study in school, omnivorous reading is going to help. Very few game designers get to have the luxury of a narrow focus.
The contacts I made by paying money to publish a few pages in the Alarums & Excursions fanzine led to some of my best friends and the people who helped start my professional career. The internet makes interacting with communities of people who share your enthusiasm even easier. But I have to admit that I don't know precisely how I'd use it if I was trying to break into gaming now. Aside from D&DI, I'd probably spend time on the Forge, the indie RPG design site. And I'd interact with various publishers on their blogs and livejournal pages.
Obviously if you want to work on specific games, you should interact with the companies and designers you're most interested in. For WotC and D&D, you should aim to write freelance for Dragon and Dungeon Magazines. Then take the designer/developer tests whenever they're available, because even if you don't get the job, a great showing on a design test may get you freelance work some day.