In the final piece in our "A Short History of..." series, Eric-Jon Rössel Waugh details the journey of id Software from a moolighting programming outfit to a leader in 3D gaming technology.

Though id is most remembered, and sometimes cursed, for kicking off the 3D era and what it's done to and for the industry (particularly on the PC end), id's first real impact lay not in the graphics card wars or online techno-jock matches, but rather – like Activision and EA before it – in a new way of doing business. That, and bringing a new perspective to American PC software – one that perhaps the Stamper Brothers would have appreciated.

Though when John Romero, John Carmack, and Tom Hall were all employed at Softdisk – sort of a digital magazine company – they were ostensibly there to write PC games. To them, videogames meant Nintendo: products with action, smooth control, and a tactile world. The problem was, at that time, the PC was primarily a business platform; although games certainly existed, and faster processors, improved displays, and sound cards were all conspiring to make those games more viscerally appealing, in general the PC was still more of a hobbyist platform, geared primarily toward the Dungeons & Dragons audience. Though technically more powerful than the consoles of the time, the hardware wasn't designed or dedicated to run Altered Beast: it was built to run WordStar. Perfectly fine if you want to be the next Richard Garriott; frustrating for the midwestern kid who breaks into the game industry, hoping to be the next Miyamoto.

As it so happens, the trio in question was clever enough, and their workplace atmosphere oppressive enough, that they spent nearly all of their free time subverting the system: working on their own experiments on company time, conspiring together explicitly because they weren't supposed to collaborate openly. Out of his doodling, John Carmack managed to cobble together a routine for a smoothly-scrolling game level – much as in nearly every console game after Super Mario Bros. Staying up all night, Carmack and Hall reproduced the first level of the then-new Super Mario Bros. 3, in near pixel-perfect detail (at least as well as EGA allowed), then replaced the main character and left the game on Romero's desk as a gag. Romero flipped out and showed the game to his bosses, who passed the game under Nintendo's nose. After some consideration, Nintendo decided it wanted nothing to do with PC games and walked away. For its part, Softdisk wanted nothing to do with the graphics routine because it didn't support four-color CGA graphics (which I can attest many people were still using in 1990).

It was around this time that a guy named Scott Miller started pestering the Softdisk crew, sending reams of fan-mail praising Romero's work in particular. After some confusion, it turned out that this guy was the head of a small company called Apogee – one of the first outfits to distribute games as shareware. To date he'd had modest success, though he was hoping to lure Romero away to – perhaps – their mutual benefit. When Romero showed him the Mario demo, Miller offered to pay them whatever they needed to get the game off the ground. He had five thousand dollars in the bank; they asked for two.

So, gearing up after hours, the trio set out on their first official project together and the first game they could really call their own. Tom Hall came up with the concept of a boy genius on an intergalactic mission; Carmack and Romero set to programming and designing the game. Toward the end of production, Tom Hall realized he wasn't as good an artist as he thought – so he called on an intern named Adrian Carmack (no relation), whose art had impressed him, to do some touching up.

Commander Keen was an instant sensation; one month after they sent the discs off, Miller sent them back a royalty check for ten thousand dollars. After blinking a few times, the first thing Carmack and Romero did was invite their boss to lunch and resign in the name of all four designers. A few weeks later, in February of 1991, id Software was officially formed – though before they could really buckle down, they still had a bunch of obligations to fill for Softdisk. Over the next year, while fiddling with the follow-up episodes to Keen and planning its sequel, the id crew worked like a renegade Stamper clan, staying at work to all hours, "borrowing" office computers for the weekend then returning them on Monday morning before anyone noticed, juggling half a dozen games at once, in all genres, with all different kinds of technology, compressing years of experience into months.

Meanwhile, Apogee released its own action platformer called Duke Nukem (liberally borrowing graphics from other games), to similar success – further helping to establish Shareware as a model for game publishing. It would be id's next major release, though, that really drove the point home – and that put shareware squarely in the industry's attention.

All it took was being in the right place at the right time; in 1991, John Carmack happened to see a demonstration of Ultima Underworld, Blue Sky's upcoming 3D dungeon-crawler for Origin – and something in him popped. Although true 3D would take a tremendous machine to process – indeed an issue with Blue Sky's game, on release, perhaps if Carmack could find a neat way to fake it, to make a sort of smooth 3D scrolling to match his smooth 2D scrolling, that could run on any system... maybe he'd be onto something. After a couple of rough demos for Softdisk – one with blank, colored walls and a rattly 16-color thing with texture-mapped walls, Carmack perfected a raycasting technique (sort of like projecting imagery through a prism) that produced a convincing effect at a very low burden to the processor. Basing their next project on one of their favorite Apple II games, in 1992 id presented Apogee with Wolfenstein 3D. And suddenly the PC was the cool place to be.

A year later, Carmack perfected his raycasting engine and Romero perfected his design in Doom – reaching a level of breakthrough success unprecedented for a PC game, for a shareware game, and for a small team of kids who were told they couldn't make the kinds of games they liked. With Doom, the aesthetic of the first-person shooter was essentially stabilized – incorporating elements of adventure, exploration, horror, suspense, and dark humor with the simplistic, console-like controls and premise (qualities often overlooked, in subsequent FPSes – that is, until Half-Life). Not only was Doom possible on a PC, and not only could it run on nearly any modern system; it also was patently impossible on the consoles of the day – at least without sacrificing a good deal of quality. id had turned the clunky PC into an action powerhouse, and freely given away the keys to whoever want to hop in and test it out. The game industry has never been the same since.
 
Where it all went wrong is where id began to push the technology further than current systems knew how to handle – then started intentionally designing years into the future, just expecting the hardware would catch up with them. This started a mad rush for the fastest, most expensive systems and graphics cards, in turn creating an expectation that every game had to milk the newest hardware to its limit, creating a spiral that sent PC games back into their niche, where only the hardest of the hardcore could keep up with them and only the flashiest games had a chance of selling to this increasingly narrow audience. Today the cycle has kind of hit a lull – to the point where a halfway decent card from a few years ago is more than adequate to play the biggest games released last week, and where a bunch of subsided genres have started to make a comeback. Still, yikes.

And the other silly thing is that after the perfection of Doom, id began to pay less and less attention to the actual designs of its games, figuring its energy was better spent on the technology. Compared to Doom, Quake was a dreary and no-frills package (however impressive the engine and Nine Inch Nails were); Quake II barely had a single-player mode. Somewhere in between, John Romero got himself fired and stormed off in a huff, determined to prove the dominance of design over technology. (Tom Hall had already left, over squabbles with Romero.) In 1998 Half-Life exploded into prominence; since there was no competing with that, Quake III moved its emphasis fully over to multiplayer, resulting in criticism that it was more a ridiculously elaborate game engine than a videogame – much like, come to think of it, id's last couple of games. Even as a multiplayer sports event, the game was to some extent overshadowed by Epic's Unreal Tournament.

At this point the Carmacks were alone as the (now very rich) company's elder statesmen. John Carmack began to give increasingly strange interviews, where he spoke of his boredom with game design and of the new one-man space rocket he was building in his back yard. Longtime associates like Raven and Grey Matter began producing id's games for it – sequels to Wolfenstein and Quake. Eventually, feeling the pressure to make a "real" game again, id dialed back the clock and tried a modern-day sequel to Doom. Though again technically masterful, the game itself arrived at the same time as Half-Life 2; by comparison, id's game was criticized as clumsy, trite, and old-fashioned. Finally last year, following an attempt to sell the company to Activision, Adrian Carmack found himself fired from his own company, leaving John – a man more interested in his code than in game design or in running a company – as the sole figurehead.

id was the heavy metal band of game studios. They made their smash debut, their even bigger follow-up, they became rich and powerful overnight – and then slowly the excess and the egos ate away all they had made, as each member backed into his own corner, then vanished. And then there was one, the quietest and the brightest and the least interested of all. At their earliest and best, when all their energies were still on making the best game possible available to as many people as possible, they were  an inspiration. When they started to get distracted, they became a distraction. When they began to fragment, they became a disappointment. Now at least, with the squabbling over, I get the impression that John Carmack has been granted a kind of peace. Perhaps that's all one should hope for in the end.

Weaving the Web

The theme over the course of these recent "A Short History of..." pieces, for those who have been following, is independence: the struggle to achieve it for one's self or for others, or simply its presence in places where you'd never expect to find it. It is that cooperative struggle for individuality, for the assertion of our humanity, that leads us to all grace. And it is the abandonment of confidence, of trust for the power of the humanity in ourselves and in others, that leads to cynicism – to desolation, to loss, to potential unfulfilled, to everything that makes life disappointing today and depressing tomorrow.

These have been five tales of bold, inspired steps for the improvement of the entire game industry – for in our every action each us leaves ripples that will take years to run their course – of five blazing stars that petered out, often before their original missions had even been achieved. The purpose of these articles lies not so much in criticism of the figures in question, as in observation of what became of their ideals, and why. It's a subject worth studying, I think; the more you know of the pitfalls and what lies at the bottom, the more inclined you are to watch your step in the future.

See also:

A Short History of Activision
A Short History of Electronic Arts
A Short History of LucasArts
A Short History of Rare