Second time's the charm; procedural NPC dialogue in Nostrum

Last time I tried some type of "procedural narrative" thing, my hubris got the better of me -- naming the system after one of the most famous and influential writers of all-time was, perhaps, just a little arrogant.

Despite my attempt to scope it properly, that system suffered greatly from trying to do too much stuff... It was so much stuff that it was difficult for me to write anything with it. So with the procedurally generated NPCs in Nostrum, I'm developing a much simpler system which will hopefully work better, to solve a smaller problem...

The basis is still the same: Elan Ruskin's GDC 2012 presentation on AI-driven dynamic dialogue in Left 4 Dead 2.

This is what I'm working on, March 2014

What are you working on? This is what I'm working on:

"Get Better Soon" is a VR-powered gay clubbing simulator haiku. Imagine a universe where EA invests heavily in sexualizing men using the latest in DirectX technologies... throbbing, pounding, pulsing bodies -- a perpetual shower. Nothing in the voice of the cicada intimates how soon it will die. A commission for Different Games, made possible with funding from the National Endowment for the Arts.

"Charity" is a procedurally-aided Thief-like set in Ciudad, a vast 13th century Moorish boomtown slowly sinking into the ground. You're a "placer", a freelance thug, an alchemist -- you beat people up and turn blood into money. Will you side with the environmentalist royals, the all-consuming corporation, or the industrial workers of the world? Underground we fought the earth together. Inspired heavily by the high-profile failure of Thief 4.

"Nostrum" is a VR-ish roguelikelikelike life simulator about just war theory. You're a freelance pilot based in the Mediterranean Sea circa 1936... well, you would be, if the Fascists would just quit killing your business with all these silly airspace regulations. Over several years you will befriend several islands' worth of alligators, corgis, giraffes, zebras, and more -- and then watch their homes burn. It's Animal Crossing meets Animal Farm, and you're just the small business owner caught in the middle? The first video game ever made about World War II.

"Radiator" is... I don't even know anymore.

"Get Better Soon", dev diary 2: character art and production value.

This is a development diary series for "Get Better Soon", a commissioned game I'm making for Different Games 2014. If you want to see it and play it, then sign-up to attend Different Games in April in NYC (for free!)

Bodies, much like video games, are routinely commodified -- there are "cheap"-looking and "expensive"-looking bodies. Society devalues and discriminates against certain body types, while affording privileges to other body types. We read video games in much the same way, based on the shape of the game's body... the packaging and production values, and/or "paratext", of a game. Production values are a relatively quantifiable way to impress people and convince them to pay $60 USD for a set of mechanics that have remained virtually unchanged for decades.

What if "queer games" weren't popularly characterized by the do-it-yourself gumption of personal stories, expressed predominantly through webpage text, by artists with few resources? What if Electronic Arts directed their next-gen AAAAA commando-developer divisions to build big budget romantic comedies about time-travelling transgender witches who critique Foucault?...

How a "Last of Us" art dump thread teaches "vision."

At Polycount, Rogelio Olguin posted an art dump of some environments from The Last of Us, as well as some notes on his environment construction process. All of his posts are worth reading, but I'm going to copy the part that really stuck out to me:

"One thing I hold true is that one texture will not make things look awesome. Imo well balanced shaders will. I think we are starting to move away from this one texture looks sick. Like the below textures are kind of boring alone but together it looks sweet. I really do not think anything we did in ND is "special" it just well balanced from good forms and composition (Modeling) to materials treatments (texturing)"

This shift in AAA art workflows, from focusing on individual art assets to thinking more holistically about how shaders, lighting, and art direction work together -- I think most AAA-affiliated 3D game artists agree with that in concept, but a cursory look around the Polycount forums shows that many of them still focus on honing one perfect asset for one screenshot.

Part of the problem is a games industry that demands certain types of presentation, style, and specialization in portfolios. ("To work here, you must have one normal-mapped sci-fi crate prop, with perfect topology and no wasted UV space.") But I think most of this problem stems from a lack of vision.

Design futures: AutoBrushes, levels that build themselves, and the politics of procedurality.

If Bethesda's detailed GDC presentation about modular level design kits for Fallout / Skyrim showed me anything, it's that modularity is actually a huge pain in the ass -- and not the good kind of ass pain either. Why should we keep building 3D levels in this slow, totally unnecessary way, with a workflow that's at least a decade old?

I remember a time when level design was slightly faster, and that time was the time of the brush. What if we could combine the benefits of modularity (variety / adaptability / abstract detail out of design) with the benefits of a brush-based workflow (simplicity / speed / focus on "platonic forms" of design)?

Game development as drawing; gesture, iteration, and practice.

(NOTE: There are sketches of nude human figures in this post, with their anatomy intact.)

If you ask any great AAA game artist about the single-most important thing you can do to get better at art, they'll probably start mumbling about "foundation." Photoshop, Maya -- these are just newfangled versions of pencils or paintbrushes or clay. They won't really teach you how to paint, or how to sculpt, or how to look at things and represent them. In this way, a bit of traditional, non-digital fine arts education can be an extremely useful tool sometimes.

In the pretty casual 12 week, 2 hours a week drawing class I took, the teacher presented two ways of thinking about drawing:

Game narrative as improvisational theater / negotiation.

The current narrative systems prototype Shakespeare has been somewhat disappointing so far: the director switches, seemingly erratically, between 5-6 different plot threads, and nothing seems coherent. I need a way of (a) allowing the player to influence story pacing / scope, and (b) a way for the system to push back, to try to force some story pacing / scope.

For this, I'm looking at how improvisational comedy generates and upholds structure. You might've heard that improv is about "always saying yes," but there's a lot more to it, apparently.

Specifically, longform improv comedy involves actors cooperating to "find the game" -- to find the core of a joke. Each actor makes "offers" to expand upon a premise and move action forward, hopefully toward a funny destination, and usually, actors err on always accepting offers ("saying yes") and building upon it since "blocking" offers frustrates your scene partners. However, it's very possible to "say yes" to a premise while still "blocking" the "game."

Here's an explanation from an NYC improv comedy personality, Will Hines:

A smoother triplanar shader for Unity.

To review: procedural UVs are amazing and you should consider using them in your games. Now, the old triplanar shader I posted was great at hard-edged cubes, but it didn't handle the transitions between textures very gracefully; curved surface like cylinders and spheres were forbidden.

So I took a look at how James "@farfarer" O'Hare handled the blending in his triplanar terrain shader, and how Tom "@quickfingerz" Jackson grabbed normals in his own triplanar shader (but the blending in his shader would "blow-out" a lot, I found) and I combined their respective strengths. I also added different handling for top vs. bottom textures, since grass rarely grows on ceilings. (Textures in the shot above are from Farfarer's pack.) One last change: I let Unity's built-in surface struct calculate world normals instead of calculating my own.

So far, I've been unable to get normal maps working with it, so if any enterprising blog readers would like to instruct me how to do it, and share that technique, then I'd be much obliged.

Here's my shader so far. Do what you will with it:

Narrative systems workflow; using Fourier analysis and level design metaphors to systemize stories.

This assumes familiarity with Shakespeare, a procedurally-branching narrative system that I'm designing. For an overview / introduction, read "More talk, more rock."

I started by arguing that interactive fiction's narrative systems expose too much complexity and detail to its authors and players, or at least more than most people need or want. With Shakespeare, I hope to achieve just a fraction of that functionality, and I think that fraction is enough to be very compelling while facilitating a writer's work.

In engineering Shakespeare, I think of the system in four parts:
a) The real-time system that runs algorithms, interfaces with the game as the player plays.
b) The data / format of narrative itself, how it's structured.
c) The Unity editor interface for generating, editing, or creating the narrative data.
d) The suggested workflow / instructions for using that interface.

Now that I have enough of a base implemented, I'm starting to think more about that last part, the operations design. Roughly, I think the tool could work like this:

More talk, more rock: on algorithmic game narratives, speculative narrative design futures, and "Shakespeare."

Last time, I wrote about procedural narrative in the context of "process intensity." Here, I expand more on designing the procedural / process part.

Back in an expertly-conducted 2011 Rock Paper Shotgun interview, Dan Pinchbeck argued that game development culture unnecessarily separates narrative from the rest of a video game:

"I just want story to be talked about as a gameplay element that sometimes isn’t there. It’s part of the set of tools that a game designer uses to create an experience – and it should be thought of along the same lines, as physics or AI or something more mechanical."

We have physics engines or texture libraries, so why don't we think of narrative as a modular "asset" or "engine" or "library" to be swapped around as well? Why can't narrative be more "mechanical." Where's all the narrative middleware? (Storybricks doesn't seem to be doing too well, unfortunately. I also don't agree with them, that proc narrative is mainly an AI problem...)

Radiator Blog: Three Year Anniversary

Wow, I've been blogging here for about 3 years now. This blog is now approaching the end of its toddler years. Much like last year, and the year before, here's a "greatest hits" compilation of this past year's posts:

(Oh, and feel free to have some cake. Forks and plates are over there, on the table.)

GAME ARCHITECTURE CRITICISM

• Why game architecture matters and the reality / unreality of de_dust. If digital life is real-life, then digital places are real places. QED.
• The shape of crime and escape. Notes from a BLDGBLOG panel / talk on heists and bank design, and me trying to apply it to game architecture. Sort of a sequel to last year's "On the first person military manshooter and the shape of modern warfare" which takes off from another BLDGBLOG talk on military infrastructure as war.
• Infinite Omaha. I went to the real-world Omaha Beach, then compared it to various video game portrayals of Omaha. 
• The shadow of the white cloud: architecture criticism at the 1893 World’s Fair and BioShock Infinite. These days, I'm more reluctant to credit the upcoming BioShock Infinite with intellectual subtlety -- it's a game where you shoot men while riding ziplines -- but still, "City Beautiful" is one of the most influential and distinctly traceable trends in architecture ever. It's also a forerunner to thought on Disneyland, theme parks, and "artificial" architecture.
• A short history of non-monoplanar first person movement. The biggest con of non-monoplanar movement is that it's hard to make giant impressive sweeping landscape levels: you're usually either stuck inside a padded room, or you're running outside on an asteroid, and neither offers a grand vista because you're either inside or outside but rarely just off to the side. The horizon is always shifting.
• What makes good writing on level design? I think writing is best when it's specific, and here I try to articulate what I like about certain writers' styles.
• On Joiner and the cult of greeble. If Sleep No More spams readables to show how much money went into it, then TF2 maps spam meticulous func_detail brushwork to imply the mastery and skill of the level designer. What happens when you automate the writing of readables, or automate the placement of detail? I think indie developers are more conscious of the cult of detail than AAA devs, mostly because detail is expensive to add.

COMMISSIONS

• Level With Me, a post-mortem. A Portal 2 mod I did for Rock Paper Shotgun. The level design is some of my better work, and I like the idea of game journalism in the form of games, but it seemed somewhat cooly received. I have to conclude that it must simply be not as good as I think it is... or that Portal 2 players are super lame.
• The Future of the FPS, written for PC Gamer UK in issue 240. A short essay and list of really cool indie FPS games and how they're changing the genre, kind of the basis for my later RPS series. Thanks Graham!
• A People's History of the FPS. A three-part essay series for Rock Paper Shotgun that argues mods are transcending their video game bodies, becoming genuine culture that is increasingly independent of the products that they're meant to be "modding" and adding value to.

ON GAME NARRATIVE

• The myth of psychological realism in narrative. Argues that thinking of fictional characters as "people" is meaningless for a writer. It is much more useful to write by thinking of a character as a vehicle for plot, and let the player fill-in character for themselves.
• Dishonored fails as an immersive sim in its first minute. The simulation should be "immersive" -- meaning, the scope of it should be consistent and everywhere. Scripting special cases goes against this genre dogma.
• Dishonored uses the Heart to lie to you. You'd expect the Heart to be an unreliable narrator of some sort, but it doesn't lie to you with narrative -- it lies to you through gameplay and psychological framing.
• "Stair K": architecture criticism, Thief, and a coffee maker. Situates Thief as dialog on social class and urban architecture. (e.g. stairs are invisible to rich people who take taxis, not subways, and frequent buildings with abundances of elevators) It argues that in Thief, stealing is framed as an ethical act because the rich deny the truth and infrastructure of cities.
• Thief 1's "Assassins" and its environmental storytelling. I really hate the type of analysis that just thinks of game narrative as a static text that you read -- game narrative is also a game design tool, a way to make the game better to play. Games tell stories, yes, but those stories tell games too.
• What do simulations simulate? Argues that a simulation gap is important for framing a narrative.
• The structure of Sleep No More (part 1, no spoilers) and (part 2, detailed and spoilery). You paid a lot to see this damn show everyone's raving about and now you're inside, on a timer. Are you going to spend your valuable time (a) reading faint scribbles on random pieces of paper under a dim flickering light-bulb or (b) follow the crazy naked people who have an interpretive dance orgy in a blood-smeared disco?
  
  I still think a lot of "game critics on Sleep No More" like the idea of it more than how people actually consume it -- unfortunately, reading is boring and performance is captivating. So I argue the readables function as set dressing to assure you of the production's expense, not to serve as barely coherent narrative in a familiar plot that's hundreds of years old. Of course, the dancing's fantastic, but I guess it's hard to argue for the value of dance to gamer culture.
• Rule Databases for Contextual Narrative. On modding Valve's dynamic self-branching conversation system and using it to author dynamic self-branching narrative, and how Emily Short's already doing something like that, naturally. I think it's one of the more promising directions toward a holy grail of procedural narrative.
• Balls and conversation: let's narrativize the sports genre. I really love baseball movies, but I'm really bored by the focus on statistics, which is probably why Moneyball sucked. There's a rich tradition of sports narratives in film and literature, but in video games it's conspicuously absent. Let's change that.
• "Do you think shooters take themselves too seriously?" We watch blockbusters in a special way, I think, but the gulf between action films and action games is this: the films are structured to be human and sympathetic, but games are sociopathic and mean. This is a game narrative writing problem.

ON GAME CULTURE

• Frog Fractions should really win something at the IGF.
• On appreciating the UV texture flat as fine art. Here, I propose three aesthetic modes for enjoying texture flats on their own merits and glorifying them as authentic game art, rather than the silly concept art we parade as game art. I later re-wrote this piece for Game Developer magazine, as "Loving the Bones."
• Desperate Gods and rules-forcing in games. Pretty recent, but I think it's a good summary of current thought on the issue -- if you can play a game of Starcraft in your head, and Starcraft exists fundamentally more as a mental construct than a product, then why can't we just argue the rules of Starcraft in the same way we interpret and amend the laws of board games.
• On grad school for games / what studying at Parsons was like. Imagine a cohort of game developers from all around the world, and 50% are women, and 10% aren't straight people. Parsons is like the rainforest: diverse, beautiful, and vital to the global ecosystem -- but it's also humid, with lots of insects everywhere, and it's constantly in danger of deforestation. It's not for some people, while others will really grow to love it.

GAME CONFERENCE / FESTIVAL NOTES

• Why Indiecade is the best games conference / festival I've ever been to. It might sound like hyperbole but it really isn't.
• I spoke at Games for Change this past year, on LGBTQ attitudes and developers in games. It went great. I began with "I'm Robert Yang, and I'm a practicing homosexual" -- and the entire auditorium erupted in applause and cheering. It was an amazing feeling.
• Notes on the Games for Change industry. Fun fact: I got into an argument with a G4C speaker in the comments. His stance -- yeah the games suck, but people want to put a lot of money into this, so just accept it. My stance -- art should be a free or reasonably available public good, not a product.
• How the worst part of the game industry uses PAX East to teabag your entire face with its cancerous scrotum. I encourage everyone to go to at least one big mass market game convention, because that's when you will know what "indie" really means and you'll realize how small, puny, and insignificant we "video game intelligentsia" really are. The sheer amount of money being thrown around in this industry is insane -- the money spent on a 20-foot tall Blops booth-complex, blaring out noise at a regular interval, is a huge contrast to the humility and humanity of indie game culture.
• What were the main trends of GDC 2012? A look-back on what happened and what stuck out as significant.

UNITY TUTORIALS / RESOURCES

• Shader-based worldspace UVs ("triplanar") in Unity. The worst thing about BioShock's environments is the cookie-cutter feel of the game architecture, the result of modular building in game engines today. The scale and proportions don't feel human or plausible. To me, one answer is to embrace old school BSP construction techniques with procedural UVs so that you can scale your primitives to arbitrary sizes without texture stretching.
• How to integrate Unity and Twine. Notes on Unity's web player JS hooks, and how that can feed into Twine's JS, or any webpage's JS, really.
• How to dig holes in Unity terrains. How to use depth mask meshes to selectively mask geometry, then disable the terrain collider temporarily.
• The best Unity tutorial writer in the world. He really is. I'd pay him to write a book, in fact, but unfortunately I'm poor.

On why Convo is now a WW2 spy romance, and the myth of psychological realism in fiction.

Short version: I've chickened out, a bit. Long version?

To make some sort of procedural "anything", you have to have an idea of what the building blocks of that "anything" are, or at least what you'll argue they are -- and then either frame your game in those terms or expressly simulate those terms. So if Convo is a game about narrative instead of people, then what's a unit of narrative?

From there, my thinking goes like this...

"Loving the Bones" in Game Developer magazine, October 2012

I adapted my blog post into a short article for the October 2012 issue of Game Developer magazine, promoting the art of the humble texture flat as its own art-form and mode of appreciation. The three masterpieces discussed are: Rob Laro's tankbuster sheet, Thomas Varoux's palace lightmap, and Anna Anthropy's miner spritesheet. Together, I thought they represented a good cross-section of non-photorealistic / desktop / mobile / 3D / 2D / environment / character art going on today. Pick up an issue of GD mag at the game convention nearest you, or squint at this low-res but somewhat legible clipping to the left.

What I'm Working On Right Now

So after coming back from Europe last month, I thought about why I haven't been good at finishing things these past few months, and I've decided to try making something different in a different way, and to see how well that goes. That means no first person, no reliance on written narrative, no vast architectural worlds, and no "art + writing first" approach, which is what I've been doing (with varying degrees of success) for the better part of this year.

Convo is a squad tactics game about linguistics and binge drinking, and I'm working on systems / interface first... I've never paid so much attention to UI before. I have a few ideas as to how to structure the narrative and such, but I'm avoiding anything resembling implementation right now. Roughly, I'd say it's based on the notion that a game narrative's job is to emphasize and/or problematize a game's simulation gap.

Someplace Else: Source is a Source remastering of Adam Foster's Someplace Else, in anticipation of Black Mesa Source's impending release and the 5th anniversary of Minerva: Metastasis' release. If you're interested in partnering for Minerva Day in some form (fan art? photography? sculpture? design criticism? fan fiction? etc.) then please get in touch.

I'm also in the middle of writing a three-part series of essays for Rock Paper Shotgun. It'll probably appear in about 2-3 weeks, and the whole thing will get published across a single week -- so lucky you, not much waiting involved.

Stay frosty.

How to integrate Unity and Twine.

EDIT, 16 May 2013: Unity has changed their Browser to Web Player communications, but just a little. Basically, you don't use "getObjectById()" anymore, you just use "getUnity()" to get a reference to the web player -- more details here.

*****

Okay, I'm one of those people who thinks the problem with interactive fiction is that it's not sexy enough. However, I think IF, as a mode of interaction, is extremely powerful and is quite possibly light years ahead of whatever we're doing with narrative / meaning in the latest 3D whiz-bang video games.

Then one day I realized -- I could combine the CYOA tool Twine with the web player export of Unity, and the two could possibly hook into each other through Javascript. Turns out, they can.

Unfortunately, the project I used it for -- well, it didn't really work out -- but maybe someone else can use it?

The Unity Web Player has a useful method Application.ExternalCall() that can call Javascript methods on the web page. Similarly, you can call SendMessage() from a page script to call a method on a specific GameObject and even pass strings into Unity. That's the gist. If you need more help / my code snippets, a more detailed guide is here:

Rule Databases for Contextual Narrative... and spelling bees.

Valve's Elan Ruskin gave a fantastic talk at GDC 2012 on using "Rule Databases for Contextual Dialog and Game Logic" -- basically, the implementation behind the dialogue response system in Source games, most recently used in Left 4 Dead 2 and DOTA 2. I'm surprised more people haven't picked up on it because I think it presents some really effective research on procedural narrative systems.

A lot of game logic / narrative resembles a flowchart, especially with the advent of visual scripting systems like Unreal's Kismet or Twine -- resulting in this deeply entrenched concept of branching structure. Authoring and changing these individual branches is usually very expensive.

France Diaries: sketches + a 3 sentence review of "Game Feel"

I'm in France right now. This is part of a series of game architecture diaries about France.

If anyone thinks I'm qualified to offer advice to beginning level designers, then here's that advice: draw, even if you're awful at it.