John Carmack is, perhaps, the nearest thing to a demigod the gaming world can claim to have. Half man, half deity, touched by the hand of the eternal, wandering the earth, solving really hard math problems, Sisyphean, DOOMed forever like Leonardo (or any of the other ninja turtles) to eternally strive for a goal that can never be reached — to capture reality.
He’s a hero to gamers and open source software advocates alike. He’s been consistent in his support of open code — he released the source for Wolfenstein 3D in 1995 and the Doom source code in 1997. When the source for Quake was leaked and a developer ported it to Linux and sent the patches to Carmack, he and id Software didn’t sue the miscreant, they based their official Linux port on the new code. id Software has since released the source code to Quake, Quake 2 and Quake 3, all under the GNU General Public License (GPL). The id Tech 4 Engine (used for Doom 3) will also be open source, although nobody much wants it. These have all been Good Things (well, excepting Doom 3 and Quake 4, where things started to go sideways, but more on that later).
Because of the nature of Moore’s law, anything that an extremely clever graphics programmer can do at one point can be replicated by a merely competent programmer some number of years later.
Arguably, no single person has done as much as John Carmack for PC gaming since it began. His geek credentials are beyond reproach.
Literally tens of thousands of maps and mods and add-ons and a new generation of source ports to new platforms and even entirely new games have been put together over the years using Carmack’s technology — a nearly endless cornucopia spilling fun out all over the internet — and none of it would have existed, none of the massive ecosystem of office-chair enthusiasts and developers would have developed without id and Carmack and the openness of their approach. It is not a stretch to say that the last 15 years of PC gaming would have been much-diminished without them.
But there have been some worrying developments in recent times.
id’s games have pretty much always been kind of crap. Yeah, I said it: but let me qualify it. They’ve been amazingly fun games, certainly, they’ve been built on engines that, for the times they were released, were groundbreaking. Without exception, they were gorgeous, fast, and had unparalleled, rock-solid ‘feel’ and physicality, and launched the entire e-sports phenomenon as a result.
But the games themselves were more technology demos than actual games. They were fun seemingly in spite of themselves. id, collectively, even in the heady days of Romero’s Hair, couldn’t design an innovative bit of gameplay if their lives depended on it.
Without Carmack crapping out these groundbreaking 3D engines at ever-increasing intervals, id Software would be little-known or regarded.
I say this as an old-man gamer for whom Doom and Doom 2 and Quake 1 and 2 and most especially 3 were transformative experiences, and massive milestones in a gaming career that goes back to 1997 and my TRS-80 Model III. Hell, I’ve written more than 10,000 words on their games right here on this site in the last couple of months. I’m actually older than Carmack himself. Now get the hell off my lawn.
Until id Tech 4 — the Doom 3 engine — none of that slightly-dodgy game design mattered very much, though. The engines and the toolsets were simple enough that modders and mapmakers could dive in and, if they had the determination and that peculiar warping of the mind that lets people do mapping for 3D games without pounding pencils into their eyesockets, get stuff done relatively quickly. The ecosystems around the games flourished, dependably. The fact that the core games themselves were kind of limited, slapped-together and, in a way, traditionalist (where tradition was rooted only a year or three in the past) didn’t matter. A new id game meant a massive groundswell of creativity, every time. The fact that id, other than Carmack himself and his Mad Math Skills, weren’t all that creative — well, that didn’t matter much. We — the fans and enthusiasts and gamers — happily took up the slack and made each successive generation of technology into a phenomenon.
A million servers bloomed, and a billion maps were played.
Until Doom 3 and Quake 4, built using id Tech 4. Then everything pretty much slowly ground to a halt.
It’s not that those games were bad, or that the engine was all that terrible, although it didn’t seem that there were an overabundance of good decisions being made.
A few things were working together in concert. The tech and toolsets had gotten so complex that casually picking them up and making some maps just wasn’t possible, unless you’d been doing it for years already.
Doom 3 was DARK LOL. Which is to say that Carmack’s decisions to focus on lighting and shadows — which seems in retrospect to have been a blind alley — dictated some distinctly uninspired gameplay design decisions — the most notorious being the ‘monster closets’ that just felt lame in a 2004 game, and the no-duct-tape-on-Mars decision that the game wouldn’t allow the player to use a flashlight and gun at the same time, an artificiality created to exploit the shadowing technology, and a glaring and annoying one that detracted from gameplay rather than enhancing it. It was a clear unfortunate example of id letting technology dictate gameplay design decisions.
It was a fine and enjoyable game, though, in the end. It was beautiful for the time, and it was appropriately scary. But it also lacked decent netcode, and the movement, after the smooth, fluid, visceral, natural feel of movement in the id Tech 3 engine that powered literally dozens of single player and multiplayer games and mods, well it made the player feel as if gravity was dragging them into the dirt. Realistic perhaps, appropriate for survival horror, but not much else. And not much fun.
The larger problem was that it was product. It came and went, a few diehards made maps for it, and it faded from memory.
Quake 4 saw some tweaks to the engine, with a little more outdoor scenery (and the most embarrassingly bad skyboxes in any AAA game ever), a little less of the DARKNESS LOL, but it was created, apparently deliberately, for single player first and foremost, a decision, after the incredible success of the Quake 3 generation of games that almost singlehandedly brought e-sports into the spotlight, that continues to bewilder. Hell, Quake Live, which literally came out of beta last week, uses a tweaked Quake 3 engine — id Tech Minus One — a full decade later.
What the fuck could they have been thinking? Was it that they knew that the netcode was crap? Was it that they realized the tools and technology had become so complicated that the community creativity we saw with previous generations would not flower in the same way, and thus be an embarrassment? Was it a money decision somehow? Why would they have said “I know! Let’s just remake a couple of our old games, using a new engine, but make the engine suited exclusively towards single-player games! That sounds precisely like what our audience is clamoring for! And hell, even if it flops, it’ll only take Carmack most of a decade to come up with the next generation, provided he doesn’t run out of steam or get distracted by shiny things…“? What on earth were they thinking?
Perhaps it was arrogance, or greed, or incompetence. You — by which I mean me — don’t want to think so, but we have to wonder. Maybe they actually thought people had flocked to their earlier games, that massive gaming communities had developed and swelled and turned into something amazing because the games themselves were just so perfect.
Maybe they thought that all they needed to do was repeat themselves, but with more graphical juice and a gravity gun copied shamelessly from Half Life 2, maybe more specular goo poured over everything, more realtime PC-killing shadows swinging around, and all would be well.
They were catastrophically wrong. It was always all about the community, and the maps and modding and online play, and the actual out-of-the-box product a distant second. It wasn’t that they were the greatest game designers ever to walk the Earth — no, they had amazing engines to work with, and a leader committed to open source, but they were just bad enough at it to inspire ordinary guys in dorm rooms and basements all over the world to try to do better.
So, yeah, anyway, then we got Quake 4. A couple of patches long after the game was released made online play reasonable if not very satisfying, fixing utterly essential things like running up stairs (seriously, that was broken in online play in Quake 4 — from the company that invented deathmatch), but by that time, projects like Rocket Arena 4 were abandoned, and the community had moved on from id, to greener, more deathmatchy pastures.
It was the beginning of the end.
TL;DR, guys? Executive Summary: it’s always been about the community, and forgetting that means Bad Things.
Compare: id Tech 3 games, id Tech 4 games. All but one of the latter were at least partially inhouse. id Tech 4 was, to all intents and purposes, a failure, and though it’s certain that id made a lot of money from the games that used it, it marked a sea change in the way the company approached games. Other changes, like the growth of console gaming, that cancer on our collective liver, were pushing hard as well.
Things are changing, and recent news on id’s upcoming Rage, a full 7 years after Doom 3, do not paint an encouraging picture. They should have learned their lessons from the failure of id Tech 4, but it seems, based on recent pronouncements, that they definitely haven’t, and id Software may well be on the way into obscurity. I wish it weren’t so.
Stay tuned for the pulse-pounding Part II, in which we decide whether Our Father Who Art In Carmack has turned to the Dark Side or not.
stavros thewonderchicken has written 125 FGEC articles.