Fucked in the Dark
Ports, Sequels, and Other Unholy Offspring
Between the release in late 1993 of the original version and the release of DOOM II in October 1994, id kept up a steady stream of patches and updates to the game, fixing problems and adding features. Key updates for us are 1.2 (February 1994), which added modem support and the Nightmare difficulty level that was to become a key part of the speedrunning scene; 1.666 (September 1994), which added Deathmatch 2.0 (also changing the shape of the drop-down room in E1M4 from a swastika to a more random pattern, making the game a viable release in Germany);1 and 1.9 (February 1995), the final version of the game. Also included in 1.666 was Hank Leukart's DOOM FAQ, a huge document covering details of modding software that was available and where to get it; walkthroughs and tips for play; multiplayer options; and a poem, written by Hank, called “The Night Before DOOM,” based on “The Night Before…”—oh, you probably get it. It's an amazing document, a huge amount of work, and testament to the fanatical support of DOOM's legion of fans.2
DOOM has been ported to a vast number of systems. Ledmeister's exhaustive documentation of comparisons between systems3 details eleven discrete ports, mainly to consoles: Sony Playstation, Sega Saturn, Nintendo 64, Atari Jaguar, 3DO Panasonic, Super Nintendo Entertainment System, Sega Genesis, Game Boy, Xbox and Xbox 360, and iPhone/iPad. We can also add computer ports: Win95 (an official Windows version was released in 1996, and Microsoft thought very seriously about buying id around this time), Linux, and Mac OSX. These are the official releases; once we add in community-driven ports of the game (including Mike Welsh's extraordinary browser-based Flash port), the number goes through the roof. The majority of official ports occurred after the release of both DOOM II and Ultimate DOOM, so before we dive into the former in detail, we need to quickly pass over the latter.
DOOM II: Hell on Earth was released on September 30, 1994, a mere nine months after its predecessor hit the Net. As before, the game's story was something of a loose justification of the action rather than anything of any weight, occupying itself with the liberation of Earth from Hell's invading forces. Freed from the episodic model required for shareware distribution, DOOM II instead presented a single thread of developing levels, interspersed with text screens, which did preserve something of the episodic feel. Roughly, it goes like this: escape Earth, fight to the gateway, and close it from within Hell. You may want to refer back to the DOOM Bible at this point to see how Hall's comment about id completing the arc with the sequel is largely justifiable.4 Again, levels were roughly themed, although this is far less evident and noticeable here than in the original. Sandy Petersen once again worked design, this time joined by Shawn Green and American McGee (interestingly, Romero is not credited as a designer on either game, even though he returned to design six of DOOM II's levels). In keeping with the rapid turnaround from DOOM to DOOM II, there was not a great deal of advance in the engine tech, the differentiations came down to a set of new monsters and the super shotgun, which is still possibly the most rewarding firearm to use in a computer game, ever. The new hellspawn included chaingun Zombies, missile-spewing Mancubi, and two more complex creatures: the Pain Elemental, who fired Lost Souls, and the Arch Vile, who resurrected dead creatures. Both of these added a new strategic dimension to gunfights, and the Arch Vile could be used within level design to steer the player toward particular behaviors in terms of manipulating significance within the environment. It may be relatively simplistic by today's standards, but these incremental steps toward sophisticated player manipulation are the bedrock of modern FPS design.
The game did extraordinarily well in sales, with VGChartz claiming an estimate of 3.61 million units sold5 and awards for Best Game from PC World (1995) and Best Action Game from Computer Gaming World (1995). While critics noted that it didn't massively advance the gameplay or tech of DOOM, it was nonetheless rapturously applauded. Chris Lombardi of Computer Gaming World loved it but felt he needed to point out that “it is not what DOOM is to Wolfenstein 3D, it is simply more DOOM” (Lombardi 1994). In PC Gamer, Gary Whitta hit back at this argument, snarling, “Some may complain that there's not enough new stuff here to excite weary DOOM players, that this is just a cynical cash-in, but the kinds of people who say things like that don't know anything. Real DOOM purists, in fact anyone who appreciates the game, will lap this up.” Whitta summed up by saying, “Like its predecessor, DOOM II is sick, twisted, depraved, repellant, morally bankrupt—and utterly, utterly brilliant.” His review also noted the subtle but significant design evolution from the original to the sequel.
The structural design of the new levels makes for a tougher challenge too—the puzzles are more complex, routes more maze-like and things like weapons and secret doors are more difficult to find. Perhaps the most noticeable aspect of DOOM II's new design, however, is the overall mood and tone—it's that much darker and more sinister. Remember those horrible poorly-lit areas in DOOM you stumbled into with trepidation, not knowing what could be lurking in the shadows? In DOOM II it's like no-one's paid the electric bill in a year, with many sizeable portions of many levels lit just enough for you to find your way around, but not quite enough to feel safe. (Whitta 1994)
Even while working on Quake, id was smart enough to milk the satanic cow, and 1995 saw the release of Final DOOM, two new self-contained DOOM II episodes: Evilution and The Plutonia Experiment. Finally, DOOM was also given a retail release, packaged as Ultimate DOOM (April 30, 1995) and containing an additional episode, Thy Flesh Consumed. TFC is generally regarded as by far the most difficult of all the episodes (including DOOM II), and the focus on high-octane, supertough combat perhaps reflects the fact that Sandy Petersen went nowhere near it. Instead, design credits go to Romero, McGee, and Green, with E4M7 And Hell Followed designed by modder John “Dr. Sleep” Anderson and E4M5 and E4M9 by Tim Willits, who joined id in March 1995. TFC is a bit sporadic and messy as an episode: this is not a criticism of its levels but, rather, acknowledgment of an inevitable result of the episode being thrown together for added value, rather than as a coherent piece. About it, Romero has commented, “I thought that we should create an extra episode and give the buyers more value, so Episode 4 was born. I had several people creating maps for the episode, which is why the consistency isn't there like Episode 1.”6 According to Willits, even that may be overstating the planning that went into things.
We knocked that out in like a week or two. It was crazy. It was like a week, two's work. Our beta testing back then was just to play it through on all four skill levels, and if you could make it through, then ship it. And a lot of it was fragments of other maps that had never got finished. We were not that clever to have any purpose or organizational skill. Then I think Romero just went through it and saw which ones fell in whichever direction. (TW)
In terms of console ports, there's a general slowdown in gameplay (a perennial complaint about console shooters that gets shooter geeks like me all sneery, Halo notwithstanding). Other common features are compromised or missing multiplayer aspects, which is understandable really, given the era of consoles we are talking about; the odd thing like lack of crushing ceilings in PlayStation and Jaguar among others; an increase in bugs (on the Super Nintendo Entertainment System particularly); and, in most cases, a shift to the map order. Sometimes this was due to memory restrictions, as in the case of the Sega 32X port, which contained “a mélange of levels from all three episodes, all lumped into one continuous game” (Young 1994). The really interesting commonality about all of this is how remarkably the tables are turned from the development context that drove id in the first place: console ports tend to be more sluggish and buggy than the PC original. Obviously, more modern ports (e.g., to Xbox) lack fewer features and, interestingly, offer experiences much closer to the original in terms of gameplay and maps. This includes the touchscreen port to Apple products created by John Carmack and the PrBoom team. PrBoom had been running as a source code port for years, following id's release of the DOOM source code in 1997. This, in itself, deserves mention, given the huge adrenaline shot it gave to unofficial source ports in the community. The code, when released, only ran on Linux, but it didn't take long to migrate. Team TNT (who gave the world Evilution on Final DOOM) created Boom, a cleaned-up port that then led, in turn, to Marine's Best Friend, created by Lee Killough (whose name is familiar from a huge DOOM archive now hosted on Romero's own site).7 PrBoom grew out of these two ports and formed the backbone of Carmack's code work on the iPhone port of DOOM. In an undated progress report, he gave his reasons for staying close to port code as it stood.
I am trying to not be very disruptive in the main codebase, because I want it to stay a part of prBoom instead of being another codebase fork. While I can certainly add a bunch of new features fairly quickly, iterating through a lot of user testing and checking for problems across the >100 commercial DOOM levels would take a lot longer. There really is value in “classic” in this case, and there would be some degree of negative backlash to almost any “improvements” I made.8
Four novels also came out of the world of DOOM. In 1995 and 1996, Dafydd ab Hugh and Brad Linaweaver published Knee-Deep in the Dead, Hell on Earth, Infernal Sky, and Endgame, which take the premise of the game as a starting point before spiraling off into quite a radically different take on events. They were not particularly highly thought of by id, who have always taken a hands-off attitude toward transmedia activities. After DOOM 3, there were two novels by the game's writer Matt Costello (DOOM 3: Worlds on Fire in 2008 and DOOM 3: Maelstrom in 2009), a board game, and a movie tie-in, but discussion of these falls outside the scope of this book.9 DOOM 3 however, deserves a more detailed examination.
On June 1, 2000, via one of John Carmack's plan file updates, id announced they were developing a reboot of DOOM using the most advanced engine yet thrown at a game.
It wasn't planned to announce this soon, but here it is: We are working on a new DOOM game, focusing on the single player game experience, and using brand new technology in almost every aspect of it. That is all we are prepared to say about the game for quite some time, so don't push for interviews. We will talk about it when things are actually built, to avoid giving misleading comments.10
Although technically titled as the third game in the franchise, DOOM 3 is really the spiritual reimagining of DOOM for a more contemporary engine. As such, considering how and why it differs from the original and how and why it remains true is perhaps the best groundwork to lay in terms of understanding DOOM's legacy for FPS games in general. Coming in at 785,000 lines of source code (DOOM had 54,000),11 id Tech 4 offered real-time dynamic lighting, which, for geeks like me, was just extraordinary. Essentially, rather than working out light paths and bounces in advance and then dumping sprites into the mix afterward, DOOM 3's monsters would cast shadows as they moved. This is not just a software feat in itself but another instance where a technological advance opens up the potential for some truly extraordinary gameplay and experience design. The most striking, memorable moment in DOOM 3, for me, is near the beginning of Delta Labs: on entering a lobby area, a huge Imp appears to scuttle across the room in silhouette, a shadow cast by the creature actually running across the glass skylight far above (fig. 22). It's very difficult to convey in print the sense of awe this moment inspires, and the fact that it was all being calculated in real-time, rather than a prerendered animated sequence, remains one of those moments that characterizes for me just how far games have come in a short span of time. Like I said, I'm a geek, and my childhood was taken up with being amazed that if I hit play on a cassette recorder, then I could play Horace Goes Skiing (Beam Software 1982) twenty minutes later on the telly. But some things transcend their historical implementation, and the Delta Labs Imp is a really amazing moment regardless of whether you are a child of Clive Sinclair, Bill Gates, or Gabe Newell.
DOOM 3 basically took the central premise of DOOM but augmented it with some additional bits and pieces. Cloud describes it as “a desire to maintain the spirit of DOOM and revive this gameplay and this universe in a modern time frame” (KC). For example, in DOOM 3, we get to be present for the actual invasion itself; the question of UAC's culpability in the disaster is foregrounded, and we get a couple of actual nemesis figures in Malcolm Betruger (English, of course, and with a gammy eye) and Sarge. We have lost the episodic structure, and action remains on Mars Base pretty much throughout, with a quick detour to Hell to pick up the Soul Cube and a conclusion at an alien archeological site, where we close the main portal to Hell. How much all the added shenanigans are actually necessary is questionable really: while DOOM 3 is a hell of a lot of fun to play, it wasn't ever going to win any awards for deep or complex content, and it does leave you wondering if any of the additions are actually necessary. The original DOOM still shows that if things are moving fast enough, story is largely disposable.
The thing about id Tech 4 is that although it was a truly extraordinary piece of technology, it pushed so hard in one direction that something had to give elsewhere. The trade-off for real-time dynamically rendered lighting and the atmosphere this enabled was speed and the number of agents that could be present in any environment at any given time. In real terms, DOOM 3 is very, very dark, mostly taking place in cramped corridors and lab rooms, and it's certainly much more sparsely populated than its predecessor. In keeping with the direction FPS games had taken since DOOM, it does, however, fulfill Hall's original desire to create spaces that felt somehow like the real world. There is no place for Petersenesque splashes of primary color, scrolling headwalls, and vast Bavarian keeps (although the gradual swamping of the base with intestines and gore gives Delta Labs a suitably gross finish). DOOM 3 does lack a little in terms of immediate visual and environmental diversity, even with some standout set-piece moments, but that's compensated for by a level of detail and subtlety that was never possible in the original. Willits argues that if DOOM was a technical feat in 1993, DOOM 3 was challenging to the point of madness in 2004: “When it started we had no earthly clue how we were going to build [it]” (TW). He describes it as a “paradigm shift” in terms of technology, art, and design. Under those circumstances, compromise in some directions was inevitable.
We've changed from the lean-to-the-point-of-starved “story” of DOOM (Fight your way out of Phobos Base…. Oh, hang on. Fight your way out of Deimos Base…. No, wait. Bear with me. Fight your way out of Hell). Now we're regrouping with our marine squad, trying to upload a call for help, realizing UAC is corporately liable, discovering that (shock, horror) the creepy one-eyed English gimp is actually in league with Satan, being betrayed by Sarge, and cleaning house by being sent to recover a special artifact that can close the portal. This is all done by following a logical sequence of travel gradually deeper and deeper into the base complex (and thus closer and closer to the epicenter of evil). This requires a convincing and logical flow of travel, a reason to be in each location that joins the dots along a longer arc. The bottom line is that the original DOOM just didn't require any of these, because there was basically no linking, running story. So if we went from a warehouse to a castle to the inside of a lung within the same twenty yards, that was fine because it didn't really have to make any sense, and there was certainly very little need for continuity (and as I've mentioned previously, we have to be a little careful about projecting too much of this retrospectively onto the level sequencing). In other words, DOOM 3, like other FPS games that use their levels to bridge the beginning to the end, had to evolve its environments more subtly. Furthermore, because it was trying to be “realistic,” it had a limited palette of environments from which to choose. This placed pressure on level design in terms of dramatic visual reward, and perhaps the most disappointing part of DOOM 3 is that any exterior spaces are extremely pressurized runs from one oxygen canister to another, meaning you don't ever really get the same chance to just stop and admire the scale, one of the things that made DOOM so special.
Using light as a primary gameplay mechanism was a similarly risky strategy that split players and critics down the middle. Flashlights were, of course, quite well established in the genre by this point. They had the great gameplay value of reducing the player's ability to see and predict what was out there in the dark, and they had given designers and artists a chance to play with a whole new color palette. But id took this idea and pushed it one step further, giving the player a flashlight, but making it a choice to be used instead of a gun, rather than as well as one. You can shoot, or you can see. This meant players had to opt for a choice of vulnerability, as Perron notes: “Holding the flashlight enables us to spot the corners but leaves us defenceless, while gripping the gun gives us protection but allows for less or even no time for response” (Perron 2005). Some players had a problem with this, and Glen Murphy's 2004 Duct Tape mod quickly appeared to redress the situation.
Under the crazy presumption that a roll of duct tape has to exist somewhere on the Mars facility, the Duct Tape mod sticks flashlights to your machinegun and shotgun.12
While unpopular in some circles (Murphy's site claims eighty thousand downloads in the first twenty-four hours alone), the shoot-or-see choice certainly did change the kind of arcade-skill run-and-gun gameplay of the original.13 I'm all for anything that induces panic in a player, and it did create one of the most striking set pieces of the game, where a player must follow and protect a lantern-bearing J. Edwards through pitch-black, Imp-infested walkways in Alpha Labs 2. Having said all that, the see-or-shoot mechanism was, as Willits admits, rather less than planned.
Well, the reason you couldn't attach a flashlight to the gun was that the engine wouldn't run fast enough. I worked so hard to cover that up. “It adds this element of risk, decision making….” And yes, there were gameplay things we did once we found ourselves in that situation, but the genesis of that decision was not we're going to make this crafty experience where you have to switch back and forth; it was we're stuck and now we have to make it work. (TW)
Speaking of infestations, DOOM 3 trades off population for debris. The original DOOM was largely devoid of any furniture, with only a few plinths and the odd barrel to break the rooms up. DOOM 3 is packed with storage shelves, desks, computer terminals, and other props. Aside from adding a diegetic depth to the world, the gameplay effect of the props means things block the player from moving around, more than anything else. This is occasionally handy, for playing chase-around-the-desk with a Zombie when ammo is low, but more often, it reduces the available space to maneuver in, compounding the already generally smaller environments. Simple features, like handrails in split-level rooms, become obstacles for circle strafing and other classic DOOM approaches to combat. As Cloud comments, this drive toward or expectation of realistic environments is not always a good thing.
Realism is hot,…but I don't even know if that's a valuable goal. Believability is though. In an FPS, a person needs to be able to suspend disbelief enough to feel like he is participating in this world. Ultimately it's like cowboys and Indians. A person is, at least in DOOM, a hero, and you don't get a chance to do that in the real world very much. So it's trying to get to that point. But believability and realism are two different things, and sometimes I think we go too far in trying to create a realistic environment…. [In DOOM 3] we were really trying to make environments that satisfied what we felt were the expectations of the modern gamer at the time. (KC)
Of course, it's important to separate out your own playing experience from any claims of objective analysis, but I do find it interesting that Greg Kasavin's Gamespot review, for example, argues,
If you played the original DOOM or its sequel back in the mid ’90s (or any popular ’90s-era shooter, for that matter), you may be shocked by how similarly DOOM 3 plays to those games. The legions of id Software's true believers will celebrate this straightforwardness as being deliberately “old school,” especially since DOOM 3 is packed with direct references to its classic predecessors. (Kasavin 2004)
Rehak argues that the criticisms of DOOM 3 stem from a cultural separation “between gameplay and graphics” (Rehak 2007, 151), while arguing that it may be founded in the rise of the engine as an entity distinct from the game, which id is partly responsible for with the original DOOM. In fact, I'd argue that DOOM 3 plays very differently to its predecessor. It's still a case of run-and-gun, of course, with a straightforward, basically linear design (there's one instance of binary choice in the game, between two pathways). It features a steadily accumulating arsenal and increasing size and power of enemies. But the moment-by-moment action of DOOM 3 is quite different, if sometimes through an addition of subtle factors.
First, there are normally less monsters in any given space. This is counterbalanced by there often being much less room to move around. It's as hard to avoid one Imp's fireball in a narrow corridor as it is to avoid three in a larger space, after all, but the actual actions required to do so (in Romero's terminology, the arcade skill) are subtly, but distinctly, not the same. Second, the game tends to alternate between teleporting monsters in around the player as they progress, normally in a corridor, and forcing the player into a dark room where monsters lurk in the many shadows, neither of which are actually that like the experience of playing DOOM. For starters, as we've seen, teleporters were actually used relatively infrequently in DOOM. The same really goes for monster closets; E1M6 is distinctive for its use of them, for example. Ironically, I'm reminded here of one of Sandy Petersen's comments (although he was not talking about DOOM 3 at all).
In DOOM, you always had fair warning before monsters popped in to kill you. In later shooters, designers would often just teleport in a monster behind you without warning. Sure such a monster can kill you, but how can you prepare for it? Where is the tension? (SP)
I'm not accusing DOOM 3 of the kind of lazy teleporter usage Petersen is talking about, although it probably is a little fair to say that the room-by-room gameplay that surrounded the larger set pieces sometimes felt a little formulaic. This is possibly because id had to deliver a game that met the contemporary expectations of game length, which arguably tend to be longer than most FPS games can actually maintain satisfactorily (and it's interesting to note a recent crunch back in single-player campaign length). There certainly was warning, in real terms; it's not difficult to predict, as you move through the game, where the next demon is coming from. Because of this and an artificial intelligence system that didn't really feel too much further advanced than the original (certainly given the advances in squad tactics found in FPS games after Half-Life), DOOM 3 did pick up criticism for being too predictable. This is not unreasonable, if it's analyzed as an FPS game in the same mold as Half-Life, F.E.A.R.: First Encounter Assault Recon (Monolith Productions 2005), or, well, DOOM.
But I'm not sure that's actually the most interesting way of looking at it. DOOM 3 might not use tank controls or fixed-perspective cameras, but for me, it's right up there with Condemned: Criminal Origins (Monolith 2005), Metro 2033 (4A Games 2010), or System Shock 2 (Irrational Games 1999) as a first-person survival horror game. Particularly on high difficulty settings, ammo conservation and resource management matter as much as arcade skill. There's no particular strategic or tactical play as such, because, like Resident Evil (Capcom 1996), it's a game that wants the player to have to react constantly, not plan or strategize. In essence, every corridor is an empty room with a pillar highlighted with spotlights, just like Petersen wanted; it's just that there wasn't really that much treasure—unless you count the gorgeous, groundbreaking visuals, which are arguably a reward system all to themselves, particularly set pieces like Enpro's reactor chamber.
In fact, we could push this argument further and claim that the darkness in DOOM 3 operates along a similar principle as an old-school survival horror game's fixed camera angles, in that one of the most immediate and explicit things it does is reduce the draw distance. If you can't see as far, you can't plan ahead, and every dark corner becomes a potential threat. This changes the player's relationship with the game environment. In a lighter, more open world, like the one presented by Crysis (Crytek 2007), for example, the environment is spread out before the player, inviting them to collaborate with the designer, creating a rewarding, challenging gameplay experience that fits their own personal preference. The same can be said for FPS games like S.T.A.L.K.E.R.: Call of Pripyat that include many more horror elements in terms of diegesis but float these onto an open-world structure. The world becomes a resource, something that can be adapted and co-opted. In DOOM 3, as with survival horror games, the environment is pitted directly against the player, not simply in terms of nukage and vacuums (both of which are rare in the game), but structurally. Hardwired into gameplay is a world that forces the player through in order to complete the experience, rather than offering the capacity to be adapted for tactical advantage. Sure, there are exploding barrels in DOOM 3 but there is rarely the opportunity to use them the way a DOOM player might. They are often in tight, small rooms where the splash damage counteracts the easy kill, and there is often only one or maybe two monsters to try and take out with one hit. Again, this might be a relatively minor deviation from the original game, but I am making an argument of cumulative effect, so it's significant nonetheless. Finally, DOOM 3 is relatively slow, even for the modern FPS, and played alongside DOOM, you can see where the original's obsession with speed begins to level out in relative importance with other aspects of the final polish.
In the context of FPS games of 2004 and the context of games in the early years of the twenty-first century generally, DOOM 3 was a technological and visual marvel. Enough praise has been heaped on id for the ground that was broken in its development to last most companies a lifetime. Once again, the sound design in the game is superb. The creature design and animation redefined a generation of what Zombies and hellspawn should look like, while remaining true to the original vision of the early 1990s. If the story is hackneyed and largely forgettable, the Mars Base is stunningly realized and has a wonderful sense of place. It's bloody terrifying in places, consistently tense for the majority of the game, and includes some brilliant set pieces.
And it did really well. Despite simmering complaints about formulaic gameplay, most reviewers were ultimately won over by the game, and it scored consistently highly across the board. It sold too—over three and a half million units, according to Willits in a Eurogamer interview (August 5, 2005), although VGChartz puts it at a lower but nonetheless impressive 2.23 million copies.14 It also spawned a sequel in Resurrection of Evil (Nerve Software/id Software 2005). More to the point (at least as far as this book goes), what's interesting is not so much whether or not DOOM 3 is a good game but how it relates to the original vision: both the intellectual property and the game itself. It's here that we should probably take on board some of Tim Willits's comments from the Eurogamer interview already mentioned.
If you are any self-respecting videogame player, you love DOOM, but everyone's vision of DOOM and memories they had of playing DOOM, and what they thought DOOM should be—everyone had a different idea. When you're that popular, you will have different opinions. It was very successful for us, and I love playing it even to this day, and there are few games that look better, still, and that game came out a long time ago.
In line with this, I would argue that the criticisms of DOOM 3 derive primarily from this attempt to manage expectations of both the original and its extraordinary power to create a notional template for what an FPS game should be. DOOM 3 may be all about the invasion of Mars by the forces of Hell, and it may be a linear, bridging, accumulative-arsenal, developing-enemy, run-and-gun shooter, just like its granddaddy, but the resemblance stops there. In fact, ironically (given the complaints from some corners about it being derivative), DOOM 3 did as much to push the structural template of FPS gameplay toward survival horror as the more frequently cited System Shock 2. The advances in technology, particularly graphics, were outstanding. But it may just be remembered as quietly deviant in its gameplay too. Which nicely brings us to the question of this powerful template for FPS games, this idea of DOOM as the prototypical shooter.