Computer games look more like movies every year - the ones that make it to the shelves, that is. Will their creators start looking like filmmakers?

SILICON HOLLYWOOD

by Allen Varney

[Published in Amazing Science Fiction #1 (new series), June 1998]

Suppose magazines worked like computer games.

YOU: "Hey, here's an overview of computer games! Wonder if I can handle it. (reading) 'Article requirements: 20/15 corrected vision with augmented color sensitivity, English 3.30 or higher, Vocabulary 97, minimum IQ 120 (140 recommended), reading speed 120 wpm (600 recommended), 244 cc free cranial capacity (560 for full comprehension).' Man, just last year I upgraded my brain, and now I'm lame!"

Computer games today push the envelope like Chuck Yeager on benzedrine. Next time you see those fearsomely high-end PC games infesting the front shelves at CompUSA or Computer City -- Unreal, the latest Ultima installment, SimWhatever -- think about it. When each game started production, maybe two years ago, the designers had to aim at a moving target: the tech level most customers would command when the game hit the market. Imagine their reasoning.

DESIGNERS: "Hmmm. Most game buyers have bought their computer in the last six months. They want a game that convinces them it was the right purchase, not some Tetris clone that runs on a 386 DOS fossil. All our competitors will be designing for that high end. If we aim lower, our game looks puny. If we aim high, our game will do cool stuff theirs can't."

And so game publishers stampede up the power curve. Enchanted by their breakthroughs -- Riven's breathtaking beauty, Tomb Raider's swooping camera angles, the intriguing community of Ultima Online -- we hardly remember that history is repeating itself.

GAMES AS FILM

The computer game industry is replaying the early days of movie-making. In the 1970s computer games came from a programmer's lonely bedroom; in the early '80s, from frantic little companies with eight guys and a disk duplicator. To film historian and computer game producer Warren Spector (System Shock, Ultima Underworld I and II, Wings of Glory), that sounds like film in the early 1900s.

"Anyone who understood the technology of how a camera worked was making movies," says Spector, who holds a masters' degree in film studies. "The camera was the processing unit, and you had to develop your own film. There was a lot of experimentation; they hadn't agreed on the standards, like film size or even sprocket-hole size and shape. They didn't know how to deal with sound. People were making it up as they went along."

Filmmakers got lucky: Their platform stabilized. "The fundamental technology of film hasn't changed since the mid-'30s," says Spector. "There have been detail differences -- ooh, widescreen, wow -- but by the early '20s, people pretty much knew how sound and color would work. You can go back to The Great Train Robbery [1903] and say, 'Okay, I see how we got to Star Wars from here.'" But good luck hunting for Quake's ancestry in, say, Zork (1976).

Will game technology ever stabilize? No one foresees that happy day. Yet as the computer game business grows, it's still looking more like Hollywood every year. A New York analyst firm, Access Media International, measured America's 1997 computer and home video game market at $9 billion. That's about where American film was in 1986. Games are catching up.

As they grow, game publishers are facing the same problems film studios did. They're starting to solve them the same way -- but the solutions make life hard for publishers who can't adapt.

THE MAD LIBS SYNDROME

Remember the old party game Mad Libs, where the host asks guests for random nouns and verbs, then plugs them into a story? Over the years, every article on computer games has inevitably included some sentence like, "Soon computer games will seem more realistic than ever because of exciting new technology like [NOUN], [ACRONYM], and [JARGON]." Here, Mad Libs-style, are the words for this year's blanks.

DVD-ROM: Consulting firm InfoTech says this next-generation optical format should reach mainstream desktops by next year, becoming the principal software format by 2003. Game publishers are moving up to DVD as soon as possible, because a first-generation DVD holds ten times as much data as a CD-ROM, and later standards should double or quadruple that.

3Dfx: The 3D accelerator card wars are over and 3Dfx won. [1999 UPDATE: Sorry, I was a touch premature here.] To see its Voodoo2 chip's beautiful lighting and shadow effects, get a 3Dfx card and check out Quake II or the Tomb Raider series.

Positional sound: What's that behind you? Is someone creeping away to your left? Get your headphones ready for sound that puts the game's monsters right in your room.

Windows 95 rules OK: Live with it.

More sophisticated AI: Action game designers have finally realized that games play better when the monsters aren't stupid. Soon adversaries will size up threats, retreat when wounded, call for reinforcements, and steal your health kits.

Cinematic camera angles: In Tomb Raider you saw heroine Lara Croft (gaming's first sex symbol) from dramatic angles as she crept through caves and ruins. Expect future third-person games to move the camera as smoothly as Alfred Hitchcock.

So far as technology is concerned, the parallels with movies are clear. Want to know what games will eventually look like? See Toy Story. Know what they'll sound like? Go to a Stallone movie and listen.

But what will these games play like? Better polish up your crystal ball, because no one really has a clue.

It's unlikely they'll be interactive movies, a category that produced a few good games (Gabriel Knight II, the third and fourth Wing Commander games) and way too many duds. Most successful computer games don't really resemble movies, nor even stories. They're more like immersive environments, what Steven Johnson calls "digital architecture." Johnson, editor of the funky online zine Feed (www.feed.com), believes that though (say) Myst and Riven offered stories full of mystery, people played them primarily to explore their novel worlds.

What's more, when games reach the Internet, they need not be about violent conflict, Johnson says; they could create community instead. Networked Quake, for instance, "offers a perfectly good environment for people to meet in virtual reality," he says. "There's no reason you couldn't use the Quake engine to stage virtual poetry readings."

Maybe so, but so far no publisher has announced a 3D Poetry Slam. In fact, aside from Kesmai (Air Warrior), not many game companies are making money on the Internet. Eric Goldberg, president of the online gaming company Crossover Technologies (MadMaze, President '96), jokes, "We're not a smoking ruin, so by the standards of the industry we're doing well." Origin's Ultima Online has scored a rare success, but it's mainly for a core audience willing to live their lives online. In recent years at least two dozen other companies have announced online games, but few have made it to the market.

Design trends today point toward reruns of past ideas. Look at some high-profile titles out now or coming up, almost all slight variations on established categories.

Action: Lots of these first-person shooters are standard killfests, but Half-Life (Valve) may draw interest for its glitzy settings, good AI, and storyline by sf writer Marc Laidlaw (Dad's Nuke, The Neon Lotus). Shiny Entertainment's Messiah casts you as Bob the cherub, just down from Heaven to find the seven seals from Revelations. Ion Storm's Daikatana, a handsome time-travel adventure with inventory control and strong character interactions, will draw a million fans just because it's designed by John Romero (DOOM, Quake). [2004 UPDATE: Okay, now it reads like an embarrassing gaffe, but who knew Daikatana would be THAT lame?] Interplay's Klingon Honor Guard should sell to Trekkers. But the hottest buzz accompanies Epic MegaGames' Unreal, with its horrific monsters, breakthrough environments, and amazing on-the-fly level editor.

Strategy: Among two dozen Command & Conquer clones and another dozen Warcraft copies, LucasArts has its powerhouse Star Wars license to distinguish its Master of Orion clone, Rebellion. Interplay also takes the licensed route with Star Trek: Birth of the Federation. One offbeat imitator, Activision's Battlezone, superimposes a top-down radar strategy map over a first-person 3D shooter view. Bullfrog's long-awaited Populous III updates the venerable "god game" for real-time action. Sid Meier (Civilization) is working on Alpha Centauri, a turn-based step beyond Civ, at his new company, Firaxis.

Simulation: Pick your era: Red Baron II Deluxe (Sierra), Panzer '44 (Interactive Magic), iF-22 Persian Gulf (I-Magic), or Comanche Gold (Nova Logic). Pacifists will delight to hear that Will Wright (SimCity) is preparing The Sims, where you guide a single SimCity family to build the best home in their neighborhood.

Games for bubba: The unexpected success of Trophy Bass and Deer Hunter revealed an untapped market for trailer-park computer games played with a mouse in one hand and a beer in the other. Wizardworks is doing Deer Hunter 2 (they don't use "II" because the target audience presumably can't understand Roman numerals), and Interplay plans Redneck Rampage Rides Again.

Sequels: Expect plenty of Roman numerals on store shelves: Command & Conquer II, MechWarrior III, Final Fantasy VII, Ultima IX, Interstate '77, Duke Nukem Forever, and dozens more.

Odd stuff: Only a few products ring the novelty bell. Haptek's Virtual Friend lets you converse with a startlingly realistic alien, who speaks responses aloud when you type in text messages. Sanctum (Digital Addiction) is a new "virtual trading card game" like Chron X or the online "ManaLink" version of Magic: The Gathering. History of Glamour (Nicholson New York) tells a quirky farm-girl-turned-rock-star story that helps adolescent girls forge their own identities.

Sounds conservative, doesn't it? You'd expect more companies to try something new -- because, as in early Hollywood, no one knows what not to do.

"Try something new"? Not in today's market.

THE BUSINESS END

Gamers don't often care about business details, but it determines what games they can buy, where, and when. Look at some of the factors currently shaping the market.

Development costs: Today a publishable game keeps at least one designer, three programmers, and a whole team of artists busy for upwards of a year or two. Hardware costs, overhead, all those cans of Jolt cola.... You can't buy in for under $1.5 million. [1999 UPDATE: Now it's $2 million and rising. 2004 UPDATE: What is it now, $6 million? Eight? Ten?] Origin's Wing Commander IV cost $14 million. Many companies can't afford the publishing business. Now they're studios, development houses, for a few metastasizing survivors.

Vaporware: As you read this, has Falcon 4.0 shipped yet? If it's late (again), it has plenty of company. Bullfrog's Populous III is, at this writing, one year late. Blizzard's Starcraft finally arrived in April, about 18 months behind schedule. Late ship dates have become a painfully routine joke in the industry.

Even so, Falcon 4 is pushing it. In 1993 Microprose planned to release the game on floppies; Falcon supposedly hits this spring on DVD, having missed the intervening CD-ROM era. [1999 UPDATE: The game shipped in summer 1998 on CD-ROMs.]

Games slip for lots of reasons. A complicated design has a lot of bugs; some new tech arrives, and adapting always takes longer than the developer thinks; publishers take forever to review a developer's game in progress. Still, Stephen Beeman, vice-president of the Illusion Machines development group in Austin, Texas, notes, "At some point you have to say, 'This is inexcusable.' You have to ship a game." In hitting 15 monthly production milestones for its recent real-time strategy game Dawn of War, Illusion Machines lost a grand total of two days.

[1999 UPDATE: This didn't keep Dawn from bombing big-time, and Illusion Machines is now defunct. Is this a lesson? Beats me. 2004 UPDATE: A correction from Illusion Machines programmer Bryce E. Maryott: "Dawn of War didn't bomb, it never hit the streets. The publisher, Virgin Interactive Entertainment, went under financially just before the game was completed. [...] The rights to the game were sold, re-sold, and re-sold again. By the time it stopped being shuffled around, I imagine the technology wasn't up-to-date enough to make it worth while to market." Sorry for the mistake, Bryce! --AV]

Why don't publishers wait until the game is done before announcing it? Because by the time you ramp up your publicity, technology has passed you by.

Payola: Once you've spent millions developing the game, you pay through the nose to get it on the shelves. Prepare to spend 10-15% of your likely revenues on legalized bribes -- "shelving fees" -- for a month of shelf space in the big chain stores. If you don't pay, your customers won't get to play. Lots of good games aren't selling today, because they can't get displayed. (By the way, the same systematic extortion prevails everywhere in chain retailing.)

Fear: "There's lots of chaos out there, a lot of uncertainty," says Beeman. "Publishers get anxious at the idea of a million-dollar budget. Their catchphrase is 'revolutionary, not evolutionary,' but if you pitch them a revolutionary game, they have no incentive to decide. If they decide and they're wrong, they're fired. If they're right, the rewards aren't big enough to justify the risk. If they don't decide, they look prudent."

Bob Jacob, a self-described "failed developer" who has left game production after a 14-year career at Cinemaware and Acme/Malibu Interactive, expresses similar frustration. "The environment militates against innovation. Most concepts come out of the publisher's group, and they assign the concept to a developer. The developer is really selling production capacity; the odds of selling an original concept are very small. Not a publisher I know of would have funded DOOM."

Publishers are not only reluctant to fund, they're eager to cancel. Though the pre-release fatality rate in computer games doesn't approach the movie industry's 99.9%, well over half the projects that game companies initiate never hit the shelves, even after years of development. Some big-time talents have worked in computer games for years -- any company would hire them on four hours' notice -- but through no fault of their own, they've never actually shipped a game. Over time, these people can become tense and jumpy.

It's one more weird symptom of the game industry's headlong pace, its cutthroat competition, and its screwy structure. Hundreds, even thousands of games actually do ship each year, but almost all of them, products of one or two (or more!) years' work by dozens of talented people, die ignominious deaths of neglect. Something has to change. Everyone keeps hoping, or fearing, it'll change soon.

WAYS OUT?

Possible solutions come from - wait for it - the film industry.

Warren Spector continues his comparison between computer games today and the early days of movie-making: "We're still consolidating. Call it the 1920s, when the big studios controlled the mechanism of filmmaking. But inevitably the people who are making games are going to take over. Eventually somebody will figure out, 'Hey, without me they're nothing.' If g.o.d. gets funding, that would knock 20 years off movie history right there."

"G.o.d." is the audacious abbreviation for Mike Wilson's proposed Gathering of Developers. Wilson, former marketing chief at id Software, wants to create (in the words of g.o.d.'s mission statement at www.godgames.com) "a banding of outstanding development teams publishing their titles through the company.... G.o.d. is the 'joint venture' of the development industry, a role that other publishers could never fill." Crazy? Hardly. In movies, the directors and stars control their projects. Why not in computer games too?

The difference is, film people have had seven or eight decades to clean up their act. They make deals with a handshake, show up on time, work horrendous hours, and finish the job on schedule -- or they don't get another. Reliable filmmakers get funding through "completion bond financing," where a bank or bond company guarantees their deadlines and thereby reduces a studio's anxiety. The same completion-bond idea could help smooth relations between game developers and publishers -- if the developers start acting like professionals and meeting deadlines.

Okay, computer games don't work quite like movies. Even low-budget indie filmmakers enjoy a fundamentally stable platform; game developers, given their crazily soaring technology, are forced to surf. But every creator, in every medium, must cultivate discipline. If you have to surf, then surf rigorously.

As the gaming market grows and matures, its rules are changing. Creators who aim to win can learn a big lesson from film, an older and larger game: Shape up, or you lose.

Freelance writer and game designer Allen Varney wouldn't have missed this deadline by five days if he'd had a $1.5 million completion bond.