Grand Thieves & Tomb Raiders (31 page)

For all their developments, at the start of the new decade computer games still looked like abstractions. The graphics, no matter how detailed, were unreal, usually
‘flat’ and blocky, while the music was spartan and repetitive, and littered with angry sound effects. To the initiated a game like
Zool
may have been the height of
craftsmanship, but to outsiders it looked unmistakably electronic. Games simply didn’t compare to ‘real’ media like films or music: they appeared primitive, even infantile, aimed
at a niche audience that mainstream opinion setters often kept at bay with a barrier of scorn.

The first harbinger that games might ascend into the mainstream came in 1993, with a title called
The 7
^th
Guest
. It was a product of British publisher Virgin Games’
adventure in California, where a brilliant Scottish coder met a local cinematic games artist, and they were let loose to experiment with delivering games using a brand-new medium: the compact
disc.

Both the Scot, Graeme Devine, and the Californian, Rob Landeros, were neophiles – eager consumers and explorers of new technology. Devine was a former bedroom coder who had encountered
little sympathy for his hobby as a schoolboy. When he bunked off to finish programming his first game, a port of the arcade racer
Pole Position
for Atari, he was naively honest, as he
recounted in a documentary that accompanied the
7
^th
Guest
reissue. ‘I went back to school with a note saying I’d taken a week off to finish up this game.
Didn’t lie, didn’t say I had the flu – which is what I should have done. I took it into school, and everyone said, “OK, you’re expelled.”’

Landeros was seventeen years older than Devine, and it was his art background that led him into the US computer industry. He worked for Cinemaware, a company famed for squeezing brief
pseudo-cinematic experiences on the Amiga and ST – they were momentarily impressive, showing for instance a detailed jousting match, but were often adjuncts to more
mediocre games. While earning respect for his polished artwork, Landeros was unhappy: ‘Long hours, cranking stuff out,’ he recalls. ‘I was dissatisfied with the management at
Cinemaware, to put it delicately.’ When he heard that Virgin Games had acquired the budget label Mastertronic and was looking for staff in its Orange County office, Landeros didn’t
hesitate to join them. He found a warm welcome there – he met Devine, who sported long hair and Scooby Doo T-shirts, and they quickly formed a partnership. ‘Graeme was head of
programming at Virgin Mastertronic, and I was head of the art department,’ he says. ‘Graeme was fairly new to the States, a boy-wonder programmer from Britain, enamoured with America.
We hit it off.’

Like every other developer and publisher, Virgin Mastertronic was producing games for consoles and home computers. They were delivered on cartridges or floppy discs, subtly different
propositions for developers, but sharing a key constraint: size. Floppy discs fared better, as they were cheap and games could be spread across several of them. The Cinemaware games Landeros had
worked on needed at least two, but even then the limits were visible on screen, with repeated sequences and static backgrounds.

But a new medium was emerging. Personal computers, prohibitively pricey for all but the wealthiest hardcore gamers, could now be fitted with ‘CD-ROM drives’ – compact disc
readers that used the music CD technology for storage. Costing hundreds of dollars, the drives were expensive, and a top spec computer was needed, but a single CD-ROM could hold the same data as
hundreds of floppy discs.

CD-ROMs fascinated Devine and Landeros. There were a handful of games available in the new format, but they were conventional floppy disc titles with additional bells and whistles – music,
or perhaps a longer introduction – and the pair suspected that the potential of CDs had barely been touched. Each had large collections of
laserdiscs, and was used
to the ‘random access’ of finding any scene at any time – could there be game ideas here?

The pair’s boss at Virgin was Martin Alper, who years earlier had shepherded the
Chiller
game through negotiations with Michael Jackson’s lawyers. He liked their ambition,
and with his blessing, Devine and Landeros flew to a string of conferences on the topic, learning all they could, meeting programmers and absorbing ideas. They enjoyed their research, perhaps too
much: after the fifth junket, they had what Landeros describes as ‘an attack of conscience’, and started developing a game design.

Although CD-ROMs provided an abundance of storage for content, the pair’s ideas always revolved around a ‘capsule’ environment for the game. They were initially inspired by the
claustrophobic settings of movies like
The Shining
and
Die Hard
, but the atmosphere was drawn from the TV phenomenon of that era, David Lynch’s
Twin Peaks
, and its
offbeat, relentlessly escalating mystery. ‘“Who killed Laura Palmer? Who killed Laura Palmer?” We wanted to create that sort of intrigue,’ says Landeros. Their ideas
coalesced around a murder mystery set in a haunted house, a dramatic, cinematic story that was also a game. Their proposal was called
Guest
, a play on the 1990 movie
Ghost
.

Alper was more than keen. They submitted their idea at nine in the morning and by lunchtime he had agreed not only to fund the game, but also to allow them to leave to set it up as part of a new
company. ‘Graeme and I returned from lunch in a state of semi-shock,’ recalls Landeros. ‘Graeme said, “Have we just been fired?”’

They hadn’t quite; they had been released to assemble everything that a game in the new medium would need. The technology was still uncharted – for most developers the challenge was
to fill the storage space, and Devine and Landeros weren’t entirely clear how they could either. They founded their new company, Trilobyte, in south Oregon, and there they found a large
mansion house to serve as the setting of their game. They set up a camera in each room, and filmed a 360-degree panorama around it. If nothing else, digitising this would generate a lot of
data.

The results were dispiritingly poor. The footage looked pedestrian, but worse, it was juddery and blocky. CD-ROMs could hold a wealth of information, but they
weren’t designed for video. The CD drives simply couldn’t read the data fast enough. The output had to be at an ugly, low resolution, and even then was subject to any mechanical pause
from the CD-ROM drive.

One of Trilobyte’s artists, Robert Stein, suggested a solution: he used 3D modelling software to create a virtual room with furniture floating about it in ghostly ways. The execution was
visibly better than the homemade footage, and it would be far easier to add special effects. This alone might not have been enough to overcome the technical issues without an inspired innovation by
Devine: a compression/decompression routine that took the vast amounts of video data and compacted it into a far smaller size. Once compressed, it would take up less space on the disc and, vitally,
transfer to the computer quickly enough to allow high-resolution images. But Devine’s real triumph was in the decompression. The data was still squashed when it arrived, but with a fast
enough chip – Intel’s top-of-the-range 486 – it could be converted back into its original form in real time. He had created a way to make ‘full motion video’ stream
off a humble data CD. ‘No one thought it could be done,’ says Landeros, ‘but Graeme figured it out. That was a real technological breakthrough.’

Making the game started to feel closer to making a film. Devine and Landeros’s core idea was for a ‘branched’ series of videos, where the story would advance by showing
different scenes as the player progressed, like selecting chapters on a laserdisc, with the order determined by the player. But the scenes needed a plot, dialogue and actors. So they took on a
professional writer, horror novelist Matthew Costello, to script the game for them. This alone was a mark of creeping professionalism; designers had previously tended to treat players to their own
attempts at dialogue. The writer was supplemented with a full roster of talent, demonstrating that Trilobyte had a different order of ambititon. They hired directors, actors, and a musician –
a full film-making crew.

The live acting, computer-generated mansion and Hollywood script were brought together, and the
Guest
concept became
The 7
^th
Guest
game.
The plot concerned Henry Stauf, a rich toymaker who summoned visitors to his eerie mansion, where fiendish puzzles awaited. When each was solved, a small clip would play, illuminating more of the
story – an early example of ‘cut scenes’. These vignettes showcased the game’s groundbreaking technology, but also the strange effect that live, un-interactive footage could
have: sometimes it was chillingly immersive, but sometimes so cheesy the atmosphere evaporated. And it was a stop-start gaming experience with puzzles that were more like brainteasers, barely
interactive and almost wholly divorced from their setting. One of them, in which letters written on soup tins had to be rearranged to form a sentence, became infamous for its lack of relevance to
the plot.

But it didn’t matter.
The 7
^th
Guest
would go on to sell more than two million copies, and it didn’t just sell itself – it was such a phenomenon that it also
pushed the CD-ROM drives and PCs with 486 chips required to play it. Manufacturers saw their sales quadruple in the wake of the game’s release.

The 7
^th
Guest
wasn’t anywhere near the league of professional film-making, but it moved games into the same sphere – a non-gamer could look at
The
7
^th
Guest
and understand it, even if they were barely impressed. And it showed something else: CD-ROMs might be a specialist market, but there was a strong demand for media-rich
gaming and only large teams of well funded developers would be able to meet it. For the sequel,
The 11
^th
Hour
, Devine and Landeros were given a budget of four million dollars,
an unprecedented sum at the time. The financial stakes had risen and the pool of people who could afford to gamble so much was tiny. No eager amateur could fund this kind of game; there was no
place for a bedroom coder here.

Geoff Crammond could be thought of as the last of the lone coders. As the eighties became the nineties, the home developer model was still plausible, certainly for computers.
The Amiga and the Atari ST
were at their peak and the PC had joined them, and although games were often polished by specialist artists and musicians, a coder writing on
spec could, at a pinch, hold his own against a development team commissioned by a publisher. After all, everyone used the same hardware. And if the challenge was to stretch it further, then
Crammond was surely already in the lead.

After
Revs
, Crammond had capped off his 8-bit career with a haunting 3D strategy game called
The Sentinel
. It was another technical triumph, allowing the player to pan around
static yet immensely detailed landscapes, with gameplay so strange that it remains in a genre of one, even now. But Crammond had been bitten by the racing bug, and found himself drawn back to the
genre. ‘By the time I had finished doing
Revs
I had become a racing fan and followed F1 avidly,’ he says. ‘I thought I would probably do an F1 game eventually.’

It would be a while until he had the chance, though. In the meantime, still working solo, he was experimenting with simulating a vehicle driving over a randomly undulating landscape, and
discovered that, with the car’s suspension ratcheted up, finding ramps and jumping off them was a tremendous thrill. He decided to abandon the curviness of the landscape and instead
concentrated only on the ramps, distilling the fun to its pure essence on a series of short, increasingly absurd tracks. Crammond called the game
Stunt Car Racer
, and it was a sharp
departure from his previous, rather more serious creations. The physics was as carefully modelled as ever, but it was used in the service of death-defying leaps, which, unless perfectly timed,
would cause cars to plummet hundreds of feet to the ground. It was terrifically popular – another number one for Crammond – but writing the game entirely alone took its toll.
‘
Stunt Car Racer
took three years,’ he recalls, ‘and that seemed a long time to be working on something that
might
be a success.’

About two years into its development, his regular publisher Firebird was acquired by MicroProse. ‘I had always liked the brand image of MicroProse, with its roots in simulation,’ he
says, ‘so for me it was a good situation.’ Ideal, in fact, as after
Stunt Car Racer
was
published, the publisher entered negotiations with the McLaren
racing team about a possible Formula 1 project.

‘The games industry was changing, with more licensed product and sequels,’ he says. ‘It was getting harder to know how completely original product would fare in that market
place.’ With a Formula 1 game, the concept was widely known and a market was sure to be there. ‘That was the perfect moment to start the F1 game that I had been wanting to do. As it
turned out, McLaren’s involvement got stuck over the terms of a deal, but as the game was progressing nicely, we decided to press ahead without them.’

Crammond applied his usual, perhaps obsessive, attention to detail. As he had with
Revs
, he simulated real-world tracks for the game, but this time with the power of the Amiga
generation of computers, and the cars and tracks were reproduced to unheard of specifications. The kerbs, for instance, weren’t purely ornamental – they were raised, and his physics
engine was minutely tailored to mimic the way real Formula 1 cars ran over them. ‘The fact is,’ he says, ‘I am not an F1 driver, so the way to know that the simulation is
authentic is to not “cheat”, but to model every effect that could be perceptible and then compare simulated performance with real performance data.’ It was faithful enough that
professional Formula 1 drivers have given it the nod of approval.