Posted on

A Gay and Magnificent Revel: Postmortem

Last September I decided that I wanted to learn how to build a VRChat world.  I had never made a VR project before, but I’ve been working in and around interactive entertainment since before the turn of the millennium, so I have Opinions, and plenty of them.

I wasn’t sure what sort of world I wanted to build; I have notes and outlines for various ideas I’ve had over time, but none of them seemed quite right.  And then, by happenstance, I reread The Masque of the Red Death, by Edgar Allan Poe, in particular his description of the prince’s imperial suite.  The image of that series of color-themed rooms, lit by giant braziers, was arresting to me. 

Also, at this particular moment in history, what could be a more appropriate inspiration for a VR world than a story about locking yourself up to hide from a plague and then throwing a big party with all your friends?

So I made a world. And having made one, I have some thoughts about what I tried to do, what went well, what did not go well, and what I regret. I thought I’d share them.

Aesthetic Principles

There were four primary things that I wanted to accomplish with my design:

  • First, I wanted to create a social space suitable for a party, with interesting places to hang out, and a mix of large spaces for the main party with smaller spaces for more private conversation. 
  • Second, as I’ve discussed elsewhere, I believe in the power of size in VR, and I wanted to use scale as a tool to create a certain gravity. 
  • Third, I wanted to employ the principle I identified in my favorite 3DOF experiences that there should be nowhere wrong to look. 
  • And fourth, I wanted to build it out of free assets, which was partly about an affection for free culture and partly about me being a cheapskate.

Things That Went Well

To begin with, I finished it, and it works, neither of which was a given.  It is sort of stunning looking back just how little I knew about anything.  I’d done some Unity tutorials, but I’d never shipped anything real.  I’m proud of myself for sticking with it through … a variety of setbacks.  About which more anon.

I also think overall, Revel is a pretty neat world.  There are lots of spaces to explore, lots of places to hang out with friends, and lots of odd things to look at.  (I want to take a moment to note that the Inexorable Orb *predates* the “orb contemplation” meme.  I got there first, damn it.)

I think I had some successes with scale.  The frescoes on the ceilings of the imperial suite are really cool, and the aforementioned Inexorable Orb works pretty well.  And I think I achieved some good effects with various objects that are a little uncanny on account of being unduly large.

I succeeded in using only free assets, which was at times an interesting challenge.  I had to search in all sorts of places to find things I wanted, which was an adventure in and of itself.

Things That Went Less Well

My biggest misstep was that I had not fully internalized the difference between assets that are designed for standalone PC games, and assets that are optimized for VR.  As a result, a lot of the assets I picked turned out to be enormous, and I wound up having to pare a lot of things back to get the world size down to something manageable.  My initial build was two and a half times as big as the final version is, and the final version is still larger than I would like it to be.

For this reason, I basically failed in my goal to fill the visual field with interesting things; I just didn’t have the polygon budget for it.  (It was also a challenge finding enough free and interesting assets to fill the space to the extent that I did.)  However, on reflection, I think filling the visual field is a less meaningful aesthetic principle in a six-degrees-of-freedom setting.  In a 360-degree film, things are happening on a fixed timeline; every moment the audience is looking at something uninteresting is a wasted moment.  But a VRChat world has as much time as the audience is willing to give it; as long as there’s enough interesting stuff in the space to keep people looking around, it’s OK if there are some relatively blank spaces.  (This may be less true of VRChat experiences on rails, like Magic Heist.)

Relatively sparsely filled areas were helpful in another respect, namely my objective of creating spaces for private conversations.  I had not realized in advance that VRChat does not have audio occlusion, and so being on the other side of a wall from someone makes absolutely no difference in terms of being able to hear their conversation.  However, the abbey is large enough that the storerooms are sufficiently far from the rooms of the imperial suite that normal falloff will actually give you some privacy.

Although I had some successes with scale, it was much harder than I anticipated, for a couple of reasons.  First, scale turns out to be very hard to judge in the Unity editor.  Things regularly seem much larger or smaller in VR than they looked in the editor window.  I had to do a fairly significant rebuild when I realized I had made scale decisions that looked completely ridiculous in headset. (Another lesson: test in headset early and often.)  Second, my commitment to free assets (and my limited ability to create assets from scratch) made some scale efforts difficult; scaling objects up from what they were supposed to be sometimes works really well, but sometimes it just looks weird.

Regrets…I’ve Had A Few

A principle I believe is very useful in almost all parts of life is “Begin with the end in mind.”  I…did not employ that here.  I did a lot by the seat of my pants, in ways that were often ill-planned, and that caused me a lot of extra work.  (If we ever happen to be in the same room after I’ve had a few, ask me about the floors, and why I am a dumbass.)

I also did not think through some of the implications of my own ideas, and so despite my belief in the power of sound, I searched for music late in the process.  I wish I had started that early and allowed my soundtrack to inform more of my decisions along the way.

That said, both of these issues stem from the fact that I just did not know what I was doing, and was not in a position to solve problems before I knew I had them.  So I feel OK about it.  But I’ll do it differently next time.

Posted on

The Implicit Empire

There are a lot of essays I want to write within the field of How D&D Is And Why, but before I even begin, framing it that way has some problems.  The concepts and patterns of thought I want to talk about aren’t limited to D&D; they appear throughout tabletop RPG, and even out into books, digital games, movies, and beyond.  At the same time, you can play D&D without most of these elements, and many people do.  It’s tricky to clearly articulate the idea space.

The reason for this, I think, is because while D&D is theoretically setting-agnostic, from its earliest editions it has incorporated a lot of very particular, and often sort of weird, assumptions about how the game setting works.  These assumptions range from the mechanical — e.g., memorized spells forgotten upon use — to the cosmological — e.g., the Inner and Outer Planes.  All of those assumptions are technically optional, but they’re incorporated by reference throughout the rest of the game.  Abandoning them requires a certain amount of work to follow out all the chains of influence.  [This makes those elements a fairly powerful default; people will change elements that are important to the game they want to play, but leave alone elements they just feel neutral about.]

This makes D&D in many ways an inelegant design, but I think that baroqueness was actually an important element of D&D’s success.  More cleanly generic RPGs have sometimes struggled to gain traction, because it’s hard to provide a hook for players’ imaginations to grab on to.  D&D’s idiosyncrasies are a sort of canonical adventure of setting, a baseline understanding of what RPG worlds look like to fall back on.  (In 4th grade, I was accosted by a new classmate who was affronted that the cover of the book I was reading featured a spellcaster wielding a sword.  We became best friends.)

For this reason, I think it’s meaningful to talk about a default D&D setting, even though you can’t go buy a boxed set for it.  There is an ur-setting that lies behind and ties together all the tens of thousands of campaigns that implement D&D’s setting assumptions, and the scope of its influence makes it useful to talk about.  I call it the Implicit Empire.

Its broad familiarity — and frankly, its internal contradictions — makes it wildly generative.  Every weird quirk of how D&D does things is a site for exploration, elaboration, or subversion that will be legible to anyone who knows the game. At the same time, the fact that it emerged haphazardly from a gumbo of what a not especially diverse pool of contributors in the late 70s thought was cool means that a lot of those tropes and premises are often real problematic in real deep ways. There’s a lot going on in that tension that I want to explore.

Posted on

Aesthetics of Play: Sports and Spectacles

I recently re-read Roland Barthes’ essay “The World of Wrestling,” in which he draws a distinction between a sport, which is an event based on the “demonstration of excellence,” and a spectacle, which is a ritualized narrative embodying a struggle of moral values.  He identifies professional wrestling, in which the show is compelling despite its outcome never being in doubt, as a spectacle.  Its counterpart, boxing, is a contest of strength and skill, and thus its outcome cannot be predetermined without obviating the whole point of the match.

I think this distinction illuminates a dyad of aesthetics of play: an RPG can be either sport or spectacle (or neither), and which it is to be has profound effects at every level from rules design to actual play.

The sportsman aesthetic perceives the realm of play as a field for achievement.  Players pit their characters against challenges, test their skills, and win victories (or suffer defeats).  This may be an adversarial competition between the players and the GM, or even between players, but it need not be: RPG as sport can equally easily be a contest of each player against themself.  In every case, however, the point is to play well: to build an optimally designed character, to plan the best heist, to select the best tactics.

Now consider, by comparison, games in which the player characters are never really in jeopardy, and their ultimate victory is never in doubt.  These games confuse and disgust the sportsman; one might wonder why they even bother to roll dice if it doesn’t actually matter.

To the lover of spectacle, however, the point is the experience.  The players come together not to test their skill, but to participate in a ritual narrative.  You don’t worry about whether Robin Hood is actually going to get captured by the Sheriff of Nottingham; Wile E. Coyote is not going to catch the Road Runner by optimizing his ACME purchases.  In this mode, randomizers aren’t a factor to struggle against, but an infusion of the unexpected which permits the players to be simultaneously actor and audience.  They know, more or less, what will happen, but they don’t know how.  The essence of spectacle is the satisfaction of seeing events unfold as you knew they would, leavened with the joy of surprise at exactly how they did.

Adapted from an essay originally published on LiveJournal
Posted on

Le fraise sans merci

I often get bogged down in the process of game design shortly before playtesting. In part, this is often because I get bogged down in revisions and edits. However, often I simply get daunted by the process of making the components.

I’m mainly talking here about board and card games; to playtest an RPG, you generally need to write some adventures, but what the heck, no one ever playtests RPGs anymore anyway. But at some point when developing a board or card game, you have to actually make boards and cards.

Aside from the physical making — index or business cards get you a long way in game prototyping — I often need to make up a whole lot of relatively arbitrary game tokens. If I’m writing a game about the secret politics of a restaurant kitchen after all the humans have gone home, I’m going to have to stat up a lot of creatures with Influence, Ruthlessness, and Deliciousness. And how do I know what a rutabaga’s Deliciousness is relative to a chanterelle? Presumably less, but how much less? How about a parsnip? Is a strawberry more or less ruthless than a ripe Camembert?

Worse, I know that the vast majority of these I will get wrong, because the main problem of early-phase playtesting is getting the asset distributions sufficiently right that you can figure out whether the core mechanics are worth saving. The task of spending a ton of time producing components that will probably survive only a single playtest is a daunting one, and one that often confounds me for a long time.

Posted on

Experiences That Have Shaped My Thinking: The National Security Decision Making Game

Back in 1994, I went to the Origins game convention, which was in San Jose that year. One of the things I did was to play a game called the National Security Decision Making game, which was a simulation run by a couple of guys who had taught at the Naval War College. It was intended to model, in abstract form, most of the major players in international politics and their important interactions. I was all set to mix it up international relations style. However, upon drawing my role, I got to be a region of the United States.

This was 13 years ago, so I don’t remember the game’s details terribly well. I do, however, remember the basic dynamics of the US’s domestic politics in the game, because that was what I mostly had to deal with. There were, I believe, five regions of the US — New England and the Mid-Atlantic, the South, the Midwest, the Plains States, and the West Coast. I was New England. The main objective of the regions was to secure national resources for themselves, in the form of a share of the national budget, which was refigured regularly. A region could accomplish this by lobbying the President, who determined the budget.

There were also three politicians, whose base condition was to be a Senator, but one of whom would be elected President by the regions every so often. I don’t remember the politicians’ names, but let’s call them Senator Gravitas, Senator Unctuous, and Senator Nonentity.

At the beginning of the game, we had a choice to make, and the senators made their pitches. Senator Gravitas seemed intelligent, trustworthy, and possessed of good plans for the nation. Senator Unctuous, meanwhile, mostly seemed ambitious. He said the right things, but his eyes were a little too clearly on the prize, and he just seemed a little sleazy. Senator Nonentity I don’t remember at all; I merely assume he must have existed because I’m pretty sure there were three senators, and we shall not speak of him again. Instead, let us assume he retreated to the ranks of those elder statesmen who are always discussed as potential presidential candidates, and whose chances always seem quite good except for their inability to excite either donors or voters. Needless to say, President Gravitas was elected, and it was morning in America.

The Gravitas administration was probably quite successful; he threw himself into foreign affairs with a will, and things seemed to be mostly going his way. I, however, was not paying that much attention, because I wasn’t allowed into some of the most important stuff, and I was mostly concerned with the fact that my share of the federal budget was not what it could be.  I managed to wheedle some concessions out of the President, but the other regions were pushing hard too, and he had a lot to do.

Shortly before the election rolled around, Senator Unctuous asked if he could have a word with me, the Midwest, and the West Coast. “I have a proposition,” he said. “If you three vote for me, I will give you the entire federal budget.” We were startled. We were a little scandalized. We could do the math. Thus began the Unctuous Administration.

Sen. Gravitas was really pissed off. Here he’d been doing a good job, getting things done, treating everyone fairly, and we had straight up stabbed him in the back. I felt a little bad about it, but I was getting a much bigger slice of the pie, and pie is a wonderful cure for guilt. The South and the Plains States were pretty ticked off too, but there wasn’t a whole lot they could do about it. Unctuous wasn’t about to throw them a bone, because if he annoyed one of the regions in his coalition enough to lose it, he was going down for good.

I don’t remember how the game went from there; it had been running a long time, it was late, and I think I went to bed before we got through another term. Still, I think about that experience a lot when I think about national politics.

originally published on LiveJournal

Failure Modes in LARP
(July 21, 2005)

I am, in some respects, a grim person. One of these respects is that I tend to respond to the conclusion of a project by introspecting a while on what went wrong. (I try to stay positive, but this generally means an occasional ray of light among the grim. (This is why I never publish my postmortems of anything; they’re depressing. And vulgar.))

Anyway, last weekend I was on staff for a LARP, and I’ve been mulling over places I erred. Out of the internal discussion, I’ve noticed a few distinct types of failure in assembling a LARP. And never one to resist a taxonomy, I share them here.

Failures of Conception: Sometimes, an idea that seems clever just isn’t. Or sometimes there are bits of gameplay that would never have worked, but you didn’t manage to put two and two together. Case in point from the game just past: in game, there were two ways to learn a new skill. You could brew a potion using the alchemy system, generally having to pass through three to four intermediate steps and using up consumable resources. Or you could find a tutor and spend ten minutes getting taught the skill. Individually, both mechanics seemed reasonable. Side by side, I feel like a dope.

Failures of conception are frustrating, because you can’t really explain them away — you screwed up, no two ways about it. At the same time, they’re the easiest to learn from. The error is usually clear, and often you can draw some nice principle to blog about, which is pleasant.

Failures of Implementation: Sometimes bits of gameplay didn’t work because they didn’t work like they were supposed to. You ran out of time, someone else on staff misunderstood the packet, the crucial prop broke, whatever. It didn’t work right. The classic example, in my experience, is the political mechanics for The Bear in Winter, the Victorian steampunk game a crowd of us ran in January of 2004. I got caught up in making really badass maps, ran out the clock, and failed to finish the asset lists. It was basically unplayable. It’s possible that the mechanic would have proven a terrible failure of conception — it was a little avant-garde — but I don’t know; it never got its chance. (To this day, I want to run a game where I can take it for a spin. If you’d like to play in a small-scale experimental diplomatic simulation, bug me about it and maybe that can happen sometime.)

Another example, to my mind, is the combat mechanic from The Camel in Summer (the game from last weekend — yeah, there’s an Animal in Season motif going on there (huh, now I want to run a 1920s expatriates game called Coq au Vin)), which I realize some people feel was a failure of conception. The original idea was an attempt to make hand-to-hand combat, which was not really appropriate to the Arabian Nights bazaar setting, vaguely embarrassing. It was akin to your generic non-boffer LARP combat mechanic, but in place of rock-paper-scissors, we were going to use Rockem Sockem Robots. At the last moment, we couldn’t find Rockem Sockem Robots, and used a travel set of Hungry Hungry Hippos instead. Unfortunately, the table we used didn’t stay level, which made it sort of pointless; Hungry Hungry Hippos is very easy when you have a slope in your favor. It’s possible the mechanic would have been problematic in any event, but we don’t really know; the tilt pretty much spoiled it.

Failures of implementation drive me nuts, because there’s not much you can draw from them other than “shit happens”. They also cause arguments; I get most defensive when someone is critiquing a failure of implementation as if it were a failure of conception. I can’t really defend the gameplay in question — it didn’t work. At the same time, I don’t feel like we can say much useful about it other than on a theoretical level based on what we instinctively feel ought to work, and purely theoretical arguments about LARP are best resolved with nerve gas. (See? Depressing and vulgar. Well, violent, anyway.)

Emergent Failures: Intrinsic to the form, emergent failures happen when the players behave in unanticipated ways that cause gameplay elements to collapse. Arguably, this is a subtype of failures of conception, because in theory you should have been able to anticipate that the behavior at issue might happen, but if you can anticipate everything the players might do, there’s not much point from where I’m sitting in running the game at all.

An example of emergent failure is the underutilization of the alchemy system in The Camel in Summer. Many of the potions you could make affected the combat system — healing, powerups, etc. The dysfunction of the combat system, therefore, meant there was no demand for a lot of what you could do with alchemy.

Emergent failures are one of the things that I realize really aggravate me about the essentially interactive-literature-style games we’ve been running. I used to run boffer games, with large staffs and an essentially players-vs.-staff narrative structure. When things went wrong in unexpected ways, it was possible to roll with it. The biggest game I ran, a weekend-long piece called Mysteries of the Sunken Hundred, went horribly awry on the second day. The players had failed badly at one plot thread, and were basically unwilling to do what had to be done to avoid catastrophe. So my co-GM and I sat down (after cursing a bit), brainstormed a few minutes, and drafted an entire new slab of gameplay to get the game back on track. We slotted it in, and none of the players ever realized we hadn’t meant to do it that way from the get-go. I can’t do that in an IL game. I don’t have the staff to make more than subtle changes to the flow of the game, and even if I did, the PvP narrative structure of the style makes it hard to help one player out of a hole they’ve gotten into without possibly screwing over another player.

Failures of Aesthetic: I almost hesitate to put this last category in, but I’ve seen (be honest; I’ve done) things that made the players absolutely miserable, but worked *exactly* the way the GM intended them to. I hesitate because I think the first three categories are more or less objective, while this one depends on the common but not universal proposition that if a player didn’t have fun, that is by definition a gameplay failure. Now, there are certainly times when players create their own unfun, but I find it useful to distinguish between gameplay that didn’t work and gameplay that did work but wasn’t fun. Case in point: I wrote a game once where the entire player group was killed by fiat halfway through and spent the rest of the game on the run from the spirits who collect the dead. There were a few players who felt, either in or out of character, that all good people ought to embrace their eternal reward, and were pretty upset about the whole thing. I think, on the whole, the conceit was worth it, but I still consider it a failure in part for that reason.

Posted on

Playtesting is hard

My friend Brian came by yesterday evening to test out a board game I’m toying with. It went pretty well, all things considered (which is to say it successfully illustrated several ways in which I failed dismally), but it reminded why it’s so easy for a game design project to bog down.

It’s easy to design a game — logistically, anyway. All I need is my brain and something to write on. I may need to spend some quality time crunching numbers, but that’s a process that I can fit into my life without too much difficulty. To playtest, however, is another process entirely. I have to find someone interested in playing a game that is, as I like to put it, “not guaranteed to be in any way fun”; we need to find an adequate slice of time that is available for both of us; we need to avoid the temptation to gab on about whatever comes to mind.

And when all is said and done, all that’s accomplished is a handful of data points. With card games, an evening usually gets two or three sessions in; last night we didn’t even finish a full play-through. This is OK for a first prototype, as even a partial play-through reveals all sorts of stuff that needs to be tinkered with that will change the gameplay pretty radically, but as you move closer to a final version, it’s damn hard to get in enough play hours to really put a design through its paces.

Originally published on LiveJournal
Posted on

Aesthetics of Play: Catharsis

Many players bring an aesthetic that I think of as cathartic play to their relationships with their characters . The cathartic playstyle approaches roleplaying as a venue for players to take risks or indulge impulses in ways that might have unpleasant consequences in real life. The satisfaction of cathartic play is the chance to blow off steam.

Consequently, cathartic characters are often similar to their players, but with certain traits amplified and certain inhibitions muted. In a more extreme form of the aesthetic, these characters may be wholly designed around the traits to be amplified. I’ve known people who used short-tempered characters to work through their anger issues; I like to play impulsive people from time to time as a break from my usual overthinkery.

More commonly, however, a cathartic character is simply a version of the player who kicks more ass and isn’t afraid to be a jerk. This milder form of the aesthetic is extremely popular — at its root, traditional “hack and slash” gaming boils down to “It’s us, but we’re killin’ orcs and takin’ no guff from nobody.”

Posted on

Dinosaur Mind

At the Game Developers Conference this year, I went to the Casual Games Summit, which wound up discussing a variety of markets underserved by the hardcore-centric status quo. One of the speakers noted, “People think that games for children have to be so simple. Have you ever looked at Pokemon? I can’t figure all those critters out.”

There’s an interesting issue there about the nature of complexity.  Pokemon is intimidating to the uninitiated — there are several hundred of those little critters, with weird names and subtle distinctions between them.  But it’s not really that *complex*; mechanically, it’s pretty simple (a little baroque, maybe, but not complex). Its complexity lies in the *diversity* of the game assets, and the fact that children can master this complexity is basically the same phenomenon as when six-year-olds memorize every dinosaur of the Jurassic through Cretaceous periods; kids are good at absorbing massive swathes of systematized trivia.

“Dinosaur mind” is a talent that fades for most people over time; I know I find my brain less willing to hold on to information I’m not going to need later, or that I can look up if I need to, as the years go by. Interestingly, I think gamers as a class tend to hang on to their dinosaur minds longer; I can cost out a GURPS 3e character without a book, and I know people who can debate the differences between the spell lists in the first and second editions of AD&D from memory. Probably it’s a matter of practice.

Dinosaur mind has implications for designing games for non-standard audiences. When designing for non-gamer adults, you can assume that your audience can tolerate at least moderate complexity, but not that they’re willing to memorize all sorts of fiddly bits. When designing for children, the opposite is true; it’s probably wise to place the complexity of a children’s game in the data, not the algorithm, to use a computing analogy.

Posted on

Reflections on Complexity in Game Design

I’d like to share some personal terms of art I use when talking about game design, because I will probably want to use them in the future, and it would be handy to be able to simply hyperlink to what I mean.

There are three dyads I want to talk about today. The first comprises brittle and robust; these terms discuss the scope of things that a game system can do. A robust system can handle a wide variety of issues and situations without breaking down. The HERO System, for example, is an RPG which places a high value on robustness; the implicit design goal is to be able to handle any concept within the game’s mechanics. Original D&D, conversely, is a brittle system; it’s pretty good for going into dungeons and killing things, but anything outside that scope requires the players to expand the rules somehow. (Arguably, this was a good thing for RPGs as a whole, by demanding large-scale rules innovation and ferment from the get-go, but that’s a different topic.)

The second pair of concepts is simple and complex, which cover, in essence, how much stuff you have to remember or reference in order to play the game. A system in which you have to roll a die and exceed a certain number to succeed is simple; one where you have to roll a die, apply a raft of modifiers, cross-reference with the difficulty of the task on a table, then roll another die, apply a different set of modifiers, and check another table, depending on the results on the first table, is complex. (Some of you may know what I’m talking about here.)

Finally, we have elegant and baroque, which refer to the relationship between the other two quantities. An elegant system has high robustness relative to its complexity; it is no more complex than it needs to be. A baroque system, on the other hand, is more complicated than it needs to be for its expressive power. This isn’t necessarily a bad thing; some systems are deliberately baroque in order to convey a certain flavor. Original Deadlands, for example, is extremely baroque, using all the major polyhedral dice, playing cards, and poker chips in its resolution system; some of this baroqueness, however, pays dividends in setting a tone for the game.