You (24 page)

It was one thing to destroy Mournblade, but it didn’t have to happen right away, did it? It was hard not to think of what you could do with Mournblade’s long, black, soul-devouring weight in your hand.

It could have all kinds of uses, Lorac thought, calculating the to-hit and damage penalties he’d suffer using a class-inappropriate melee weapon. It could be a tool for redemption, or maybe for finishing the job he’d started. He could always decide when he got there.

Why not bring it back home to the folks, why not teach people a lesson, teach a lot of people lessons? Leira thought.

Brennan was in fact reasonably clear with himself that he’d think about destroying Mournblade only after he pulled it from the heart of the last son of Aerion. He thought about his sad father’s humiliation. That wouldn’t happen to him. Prendar had already thought out how many people he’d have to kill per annum to keep the thing going indefinitely—if there was one thing a game character understood, it was mechanics.

Brennan, Leira, Prendar, and Lorac were the characters, but you were the one who would decide what to do. You would come into their world, and your decisions would be the only ones that mattered. Why not take the sword, if that was allowed? Why not smash all the rules there ever were, and live forever if you could?

A
few weeks in, I sat down with the level designers to debug mission logic in the first third of the game. The question was, how do we keep the player involved in the story, and how do we make the story seem to unfold naturally around the player? As the players travel through the world, new plot developments must spring up seamlessly; nonplayer characters (NPCs) must react naturally to whatever players choose to do. A fiendishly complicated set of triggers, metrics, and tripwires would set the bits necessary to move all the scenery and cue all the NPCs in exactly the right way. Collectively, this apparatus was referred to as the plot clock.

Most of all, we focused on keeping the player from breaking the illusion of reality we were projecting. There were players out there who thought of nothing else, who took every game as a challenge to outsmart the designers and do exactly that—break our game. It didn’t take long before we developed a siege mentality. Everything became about containing players in their all-out assault on the bones of our alternate reality. They wanted, deeply and viscerally, to break our world, and we needed to make it bulletproof.

What if the player walks by and doesn’t talk to the old man? No one opens the gate until the talking takes place.

What if the player collects all the boulders in the world and makes a giant pile and climbs over the wall? Ask Lisa.

What if the player decides they don’t like the princess? Make the princess really nice so this doesn’t happen.

What if the player finds all the gloves in the world and takes them back to the store and sells them and the income is enough to buy a Sword of Nullification? A large supply of gloves depresses the local glove market, so the glove sale yields diminishing returns. Also, let’s reconsider the Sword of Nullification.

What if the player sets the store on fire, then takes everything when the owner is going into the “I’m near fire” AI behavior? The player can take the stuff, but city guards are set to hostile.

What if the player casts Genocide on all shopkeepers? Genociding any human type results in player death.

What if the player uses a wand of cold to freeze the sacred pool? Note: Sacred pool immune to cold.

What if the player casts Fireproof and walks through the flame barrier? Note: Change flame shield to force barrier.

What if the player teleports back past the doorway once it’s sealed? Teleportation requires line-of-sight.

What if the player drops the chalice into the lava? Chalice disappears, but we spawn another chalice at the altar.

What if the player does it again? There are infinite chalices.

What if the player jumps off the cliff and has so many hit points that they survive, and then they bypass the entire scene with the princess and they go on to the castle and don’t know what they’re supposed to do there, and the AI doesn’t have any kind of scripting for that? Put an automatic-death trap at the bottom of the cliff.

What if the player puts on a ring of fire resistance, casts Fireball, and the explosion hurls them over the wall, so they don’t need the key? Good for them.

What if the player summons a genie, stands on its head, wishes for another genie from a bottle, steps onto that genie’s head, and thus builds
a staircase out of the level? Add genie bottle to the list of things you can’t wish for.

So he tells you to meet him in the cellar. Can’t he just walk to the cellar? Pathfinding.

So then when you leave the room we just teleport him to the cellar, and it’s like he walked there? When you pass a certain radius, yeah.

What if you double back? He’s already gone to the cellar.

But there’s no other exit. He should have passed you, but he hasn’t. Shut up.

What if the player kills the princess? We make her immortal.

What if the player kills the lady-in-waiting? We make her immortal.

Why doesn’t the player stay home and let the immortal princess and lady-in-waiting kill every single monster in the dungeon? Because the artists didn’t make any combat animations for them.

What if the player puts a bag of holding inside a bag of holding? What if he turns it inside out? Cuts it open? Sets it on fire? Quit fucking around.

What if the EXACTLY WHAT KIND OF ASSHOLE ARE WE DEALING WITH HERE?

I
t was becoming clear that high-end game development had a bizarrely sadistic chicken-and-egg quality. During preproduction we’d all sat around and designed a game as we’d imagined it, inventing features and game mechanics and systems and telling ourselves how much fun they were going to be. And so we’d begin building levels months before the game was actually playable. When we actually began playing the game we’d discover that everything worked entirely differently from the way we thought it would, and the things we thought would be fun weren’t; the things that were fun, on the other hand, would be things we’d never even thought about. But by then the game would mostly be built and we’d have to scramble to change everything and resign ourselves to all the missed opportunities and promise to do everything correctly in the sequel, which would take another two years to build and would have an identical set of problems. The exact same thing was true for the look of the game; half the art would be built before we had a solid idea what the renderer really looked like. Not just technical specs, such as frame rate and resolution, but the intangibles—how the light fell, how solid the shadows felt, what exact register of realism or stylization it seemed to occupy. Don said it was like we had all the problems of shooting a movie while simultaneously inventing a completely new kind of movie camera and writing the story for a bunch of actors who weren’t even going to follow the script.

There was an arcade-style cabinet that sat in the corridor that ran between the library and the kitchen. It wasn’t a real arcade machine, but a PC running an emulator that let you choose from an encyclopedic menu of vintage arcade games, from
Space Invaders
to Japanese-only knockoffs of
NBA Jam
titles. It was the type of device I would have sold either of my parents for when I was nine. I was pretty sure it was illegal.

Lisa was playing an old-style vector graphics game, a world sketched in plumb-straight green and red lines. It looked like
Asteroids
but was more complicated; there was gravity and terrain. In fact, it was a distant descendant of
Lunar Lander
. She scowled as she piloted a triangular ship above a hostile landscape, dodging flak, managing the fuel supply. As I watched, she picked her way through a cave system on precisely gauged spurts of acceleration. As I watched, she bombed an enemy fuel tank and her fuel meter jumped up.

“Why would shooting their fuel give you more fuel?” I asked.

“Do you want fuel or do you not want fuel?”

She killed all the enemy bases and grabbed all the fuel, then jetted off into the void, while behind her the planet exploded into jagged, candy-colored shards.

“Why does the planet explode?” I couldn’t help asking. “Was… was that necessary?”

“Because it knows there’s a triangle out there that can take all its stuff.”

I
’d long ago noticed that there was a sort of bubble in the middle of the spring schedule not connected to anything else. This turned out to be the five weeks given over to prepping the E3 demo.

Matt and Lisa were hanging out in the Sargasso Sea of office chairs.

“What’s E3?” I asked.

“God, I’m glad Jared didn’t hear you say that,” said Lisa.

“Electronic Entertainment Expo. It’s the big industry trade show,” Matt said. “Everybody demos their next-gen games for the press. Everybody—Japan, Europe, Australia, whoever. It’s a pretty big deal.”

“It’s more than a big deal,” Lisa said. “It’s how we get funding. We need all that press to get a publisher. And we need to look like we know what we’re doing so Focus won’t shut us down. If we kick ass, somebody’s going to pay to publish our game.”

“Kick ass. You mean, if we look like we’re way, way more fun,” I said.

“Nobody really cares if a demo is fun, to be honest. It’s about whether the graphics look good.”

“So at least I’m off the hook.”

“Partly,” she said. “I think half of it is, are you going to appeal to the hard-core
Realms
fans? But the rest of it’s going to be about bells and whistles. Graphics and stuff, showing we have the next big thing that no one else has thought of.”

“You mean, your thing. The renderer.”

“Yes,” she said. “Me. I’m getting us a rough version of the graphics engine at the end of this week.”

“What does rough mean?” Matt asked.

“Well, not fully optimized, I guess, but you can load existing data into it. We can play the levels,” Lisa said. “It will probably not crash horribly every single time.”

“So, um, what does it look like?” I’d long since given up on making my questions sound informed, at least in the leads meeting. At least here, no one was under any illusions about me.

“It’s like we’ll have the same world, but faster, more detailed, prettier, I guess. Except for a hundred thousand large and small problems that I can’t explain to either of you,” Lisa added.

“We just need it to look better than everyone else,” Matt said.

“It will,” Lisa said, but she seemed to be holding something back.

“Yeah, but it’s going to have a new engine, too, right?” I said.

“Everyone will. It’s one of those years,” Matt said. “
Quake
and
Unreal,
both, and whatever Sony’s doing.”

All we had to do was put up a better game demo than everybody else, a small section of game, five minutes’ worth of gameplay, maybe, that would say everything about our game’s design, our look, our vision, and most of all demonstrate our crushing technical superiority over the opposition, which is to say everybody else in the world. Against the richest and smartest developers in the entire world, all the bearded arcade-era veterans and pissant teenagers who built their own force-feedback joysticks and all the corporate juggernauts with movie-size war chests and focus groups and market research—against them we would put Black Arts Studios, me and Lisa and Gabby and Don, and our demo.

When the new renderer came online, no one else was allowed to see it at first; Matt had it installed on Don’s computer in his office, and the four of us—Matt, Don, Lisa, Gabby, and I—sat down to look at what Lisa had made us.

The renderer is simply the part of a game’s software that displays the world; it stores all the data, all the models, all the terrain, all the textures; it knows where they are and where the point of view is, and draws them on the screen in proper perspective. A better renderer will draw more detail in less time—more complex 3-D objects, higher-resolution textures. If possible it will offer a little flash, tricks like mirrored surfaces; silvery, liquid water; translucent polygons; realistic-looking fire, showers of sparks, mists. Multiple light sources, colored lights, moving light sources. Objects that cast shadows. And always, more detail drawn faster. Every year game companies add new features that make the otherworld that much more invitingly, lusciously real. Part of it is just programmers wanting to make other programmers think, “How the fuck did he do that?” Part of it is that sensation, that “pop,” every time you see the game world drawn realer than before, that shift to sharper detail that makes everything that was the state-of-the-art ten seconds ago look dowdy, blurry, and a bit sad—it’s that “pop” that makes you that year’s new hot game and makes it more likely that retailers will stock your game instead of other people’s.

Lisa’s renderer was… odd.

It was certainly fast. It handled the gnarliest, most convoluted sections of the world without any visible slowdown. Matt panned across a broad, expansive scene of assembled warriors, distant trees and castles, a nightmarish number of polys, and Lisa’s renderer just shrugged it off without thinking. It did what we needed it to—it was fast enough to let you forget it was just drawing a bunch of data; it felt like a camera looking into the world we had built, a world you were suddenly part of, immersed in.

But it wasn’t the next-gen tech everyone was expecting. It was almost as if it didn’t want to be. The problem facing realistic real-time computer games is that the real world isn’t a bunch of polygons, it’s rounded and rough and lumpy, and computer games do their best to mimic this, even though it’s the thing they are basically the worst at doing. They’ll
use cleverly drawn textures and soft focus and tricky shading and anything else possible to make their world seem just as curvy and squashy as the real one. The world Lisa showed us was overtly angular—faceted, like crystal. The hard planes in the geometry were too apparent. It was all technology, no art. It looked a little like the graphics demos we would occasionally receive from autodidact would-be game programmers, a surprising number of whom lived in former Soviet-bloc nations. They’d have a characteristic look, garishly colored miniature jewel-toned labyrinths built solely to show off their particular arsenal of tricks—giant rotating mirrors and fountains of sparks and glistening waterfalls.