What follows is a two-part post, combining lesson plans from two separate days of my course “Comparative Media Poetics: Cinema and Videogames.” The learning objectives for the course centered around three main analytical questions, which animated the course and which students were expected to respond to in their written assignments. When looking at a given text, the course asked: 1) How is this particular film or particular game put together? 2) What effects and functions are engendered by its specific construction? 3) How has the historical development of the medium shaped this construction?
All three of these questions come together in a particularly potent way during this week, where we took a close look at game developers who developed a visual style out of technological necessity, but then paired that style with a genre that worked well with its specific effects. On the agenda: 1990s-era survival horror games.
Lesson One — Constraints and Inspirations
I start my lecture of the first day of this sequence with a question that puts students in the minds of programmers creating games in the 1970s, 1980s, and early 1990s: How do you achieve the most impressive visuals possible, while working within the constraints of limited 3D rendering capabilities?
Just how limited these capabilities are depends on the decade. You can see the full, animated version of my presentation here, but just as a quick visual reference, it unfolds like this:
Here, I offer a quick history of 3D rendering in games, animated by answers to the question I begin with. We start in the 1970s, with two possible answers.
1) Don’t attempt polygonal rendering. Creatively adjust the behavior of sprites in a 3D game, to provide the illusion of depth.
This is the trick used by Night Driver (Atari, 1976), one of the very first racing games to be presented from a first-person perspective. The car was simply a plastic overlay on the arcade machine’s CRT monitor. The “road” was represented by white “reflectors” (it’s much easier to represent a road stretching forward in three dimensions at night). These are rectangular sprites, which adopt a “swaying” motion as the player turns corners, hurtling further “into” the screen.
2) Forgo redering surfaces. Use wireframe images instead of filled-in polygons.
This was the technique used in Tail Gunner (Vectorbeam/Cinematronics, 1979), the first arcade game to be rendered in true 3D. It used its vector display monitor to render wireframe representations of fighting spaceships.
Now, we can jump into the 1980s, a decade known for such landmarks as Tron (Steven Lisberger, 1982) and the Dire Strait’s “Money for Nothing” music video (Steve Barron, 1985).
3) Use polygons, but forgo texture mapping
This was something of the “house style” of the 1980s, as the above-mentioned examples make clear. And, in 1983, it showed up in arcades, in the form of I, Robot (Atari, 1983), the first game to feature filled polygonal graphics, with simple flat colors.
Now let’s jump ahead to another decade, the 1990s, in which the possibilities of 3D-rendered games exploded thanks to innovative uses of the home computer, alongside new, much more powerful home consoles.
4) Use creative programming to achieve the look of fully-rendered 3D, without actually using the resources.
This was the technique pioneered by John Carmack at id Software, which debuted in Wolfenstein 3D (id, 1992) and launched to stratospheric popularity in DOOM (id, 1993). Much as the earlier Ultima Underworld: The Stygian Abyss (Looking Glass Studios, 1992) had done, id’s engine used a type of pseudo-3D made possible by ray casting.
5) Prioritize the elements of your game, and allocate polygonal rendering accordingly.
Finally, we arrive at the specific technique that this this week of class is most interested in. The first developers to go with this approach were the French team Infogrames, with their Alone in the Dark (1992). With Alone in the Dark, Infogrames chose to allocate resources such that only characters and interactive objects were actually rendered in 3D. The backgrounds were just static pictures—in fact, they were hand-drawn pictures, scanned into bitmap images.
This cut down on rendering costs considerably, allowing the game to run on 1992-era home computers. However, it arrives with a severe tradeoff. Unlike Wolfenstein 3D, which re-draws the room the player is in as they look around from a first-person perspective, Alone in the Dark can only offer static angles onto its action. The game’s camera bounces back and forth between different static shots as the player moves their character, each of which is completely dictated by the hand-drawn background images. Here is an example of what the game looks like in motion:
The static backgrounds provide a severe limitation. But, as Orson Welles purportedly once said, “the enemy of art is the absence of limitations.” Infogrames may have been limited to static frames, but they took this as an opportunity to craft some very memorable framing—framing that would not have been possible for game developers working with a mobile camera tied to the player’s character. It helped that Alone in the Dark was a Lovecraftian horror tale, set in an abomination-filled mansion that echoed the tradition of gothic horror.
Alone in the Dark‘s limitation-enabled visual style ended up being a good fit for its horror genre gameplay. It was such a good fit, in fact, that four years later, Capcom, developing its own horror title for the Sony PlayStation home console, stole it. Resident Evil (Capcom, 1996), with its themes of bioterror and viral outbreak, seems, on paper, to be far removed from the Lovecraftian leanings of Alone in the Dark. But, in setting their game in a large mansion, Capcom opened the door for a similarly gothic-inspired visual style. Here’s the opening of Resident Evil in motion:
The gothic has a long tradition in literature, but, being a visual medium, Alone in the Dark and Resident Evil‘s borrowings are largely from gothic horror in the cinema. Here, I refer to a tradition of horror films, with roots in German Expressionism and James Wales’ 1930s-era horror films, but for whatever reason becoming especially popular in the 1960s, that typically dwells upon ghosts, vampires, or Satanists inhabiting an Old Dark House.
Let’s look at a few gothic horror films from the 1960s, from the US, UK, and Italy, paying special attention to the ways they frame architecture and people. One technique we see is the use of canted angles, achieved either through positioning the camera such that the architecture itself looks askew …
… or by using expressionistic low-key lighting to stretch shadows like claw-marks across the screen:
Also on display: deep staging, especially when combined with deep shadows, which create pools of blackness in the image, swallowing up some of our information about the architecture’s spatial layout. (Sometimes, the deep staging will also present an opportunity to stick some grotesque statuary in the foreground, pulling our attention away momentarily from any moving figures, towards the edge of the frame.)
Another hallmark: unusually high or low angles:
Here, we see some of the same techniques on display in Alone in the Dark and Resident Evil:
That addresses the historical background of the technologies shaping these design decisions. Next, we move onto a play session, and then a day in which students discuss the effects of these decisions.
Lesson Two — Pushing, Pulling, and Screen Direction
Shifting between multiple static camera angles can help give your game an interesting visual edge, but it does arrive with drawbacks. For one, controlling the on-screen character, through repeated shifts in the scene’s framing, can be cumbersome.
Many critics have claimed that Resident Evil‘s cumbersome controls are actually an asset, making the gameplay more anxiety-inducing, and therefore more horrific, than it would otherwise be.
Laurie N. Taylor claims that the disorienting camera “led to a gameplay style where players cannot master the controls,” and that this “prevention of mastery subverts the typical gaming apparatus because most gaming interfaces are presented as mere extensions of the gamer.” [i] This feeds into her concept of the ludic-gothic, in which the usual defining feature of the Gothic, the transgression of boundaries, becomes “tied to gameplay” [ii]
Tom Bissell cites a moment in his play experience that is probably familiar to many who have played the game:
“you stagger back into the hallway to give yourself more room to maneuver, but the camera switches in such a way as to leave you unaware of the zombie’s exact location … You fire blindly down the hall, toward the moaning, with no guarantee that your shots are hitting the zombie or coming anywhere close to it.” [iii]
Let’s dig deeper into the sense of spatial confusion that controlling Resident Evil so often engenders, using the vocabulary of filmmaking.
First, a basic film studies question: What is the line, or the axis of action? (Here, I count of students who have been through Intro to Film to explain this concept to the rest of the class.)
The axis of action, or simply “the line,” is an imaginary line drawn either between two characters (in conversation scenes) or drawn by movement direction (in action scenes). The line establishes screen direction. It establishes the arc that defines the “180º rule.” If filmmakers follow this rule, they preserve screen direction.
The 180º rule is simple: If you are filming a conversation scene with two people talking, you should set up your camera placements so that one character will consistently be on the right side of the screen, and the the other will consistently be on the left. We can define your range of possible camera placements by a 180º arc:Meanwhile, in a scene in which the axis of action is defined by motion, to preserve screen direction, you’ll want to set up your camera placements so a person or vehicle moving in a continuous direction will consistently be presented as moving left to right, or right to left:Let’s take a look at the 180º rule in action, with a clip from John Huston’s The African Queen (1951). Notice how the path of the boat down the river forms our axis of action, and Huston’s adherence to the 180º rule means that screen direction is always preserved: the boat always moves from right to left in the frame.
There’s a fair amount of camera setups in that clip: a high-angle long shot, a standard-angle long shot, and some medium close-ups. But through all of the changes, screen direction is preserved.
Sometimes you can get really close to the line without crossing it. Notice how, in the medium close-ups of the following scene in which Humphrey Bogart’s character pulls the boat, Bogart is almost heading straight into the camera—but not quite. He still drifts ever-so-slightly to the left of the frame as he trudges forward, which allows for the preservation of screen direction as we cut to the differently-framed medium shots of Katherine Hepburn:
Transitioning now from pulling a boat to pushing a step-stool, let’s see how well the original Resident Evil preserves screen direction:
Resident Evil is at a disadvantage to the African Queen, as it doesn’t have an editor making careful choices about each cut. Instead, its editing is procedural: the game decides what angle to cut on based on where the player is currently positioned in the room’s flooplan. Basically, this room is broken into thirds, from west to east—easily demarcated by the marble tiles on the floor.
If our character is positioned in the western-most third of the room, we are greeted with a a high-angle shot from the east, looking at the double-doors on the west side of the room (that is, right where we entered as we entered the room). Let’s call this angle 1.
As we move to the middle of the room, the camera’s position reverses, and we are greeted with a low-angle shot, from the west, looking up at the statue in the room’s center. Let’s call this angle 2.
Finally, as we move to the eastern-most third of the room, we cut to another high angle shot, this time with the camera positioned in the center of the room, right above the statue, pointing east toward the two doors on the east wall of the room. This shot is very deliberately constructed: by placing the statue’s water jug in the extreme foreground, it guides players’ eyes to the rolled-up map item hidden within it. Let’s call this angle 3.
This breakdown of space and angles is mostly fine for general navigation. It’s only when the player starts maneuvering the step-stool to get at the map that Resident Evil reveals that, if it were to be judged by these standards of classical continuity editing, it would be deemed shoddy filmmaking. The movement of the step-stool presents us with an axis of action, and it is not one that Resident Evil can accommodate very well.
When players first start pushing the step-stool, they’ll be framed by by angle 3. The cut from angle 3 to angle 2 is a smooth one: The screen direction of the step-stool’s movement being preserved. It moves from left to right, in both shots.The cameras are placed in such a way that players are encouraged to stop pushing the step-stool before they transition from angle 2 back to angle 1, so that transition is generally without problems. The transition between angle 1 back to angle 2, however, is a tricky one, in ways that can be maddening.The problem is this: if the player has the step-stool lined up where it needs to be, angle 1 is very nearly “on the line“—that is, the virtual camera is positioned almost exactly perpendicular to the axis of action (similar to what we had in the second African Queen clip). Angle 2, meanwhile, is absolutely on the line, from the reverse angle.
Now, shooting on the line is technically legal in classical continuity editing. You’ll sometimes see it used as a way to gradually segue from one screen direction to another, as it is considered a means of “resetting” the 180º rule. But Resident Evil‘s use of two nearly- or exactly-on-the-line shots, from reverse angles, is particularly ill-suited to the activity players are attempting to accomplish in this moment. The place where they must fine-tune the position of the step-stool so that it properly allow them to retrieve the map is precisely the spot on which the game makes the determination of which angle to cut to. The result is a succession of rapid cuts between reverse angles, with the player-character moving toward and then away from the camera and back again, just as the player most needs to concentrate on what’s visually going on in the scene.
From here, I have students play through a few short scenes of Resident Evil, noting aloud when they’re especially confused. To what degree does the games much-noted awkwardness come down to a failure to adhere to the rules of continuity editing? Can students spot any shot transitions in which the rules of screen direction are outright broken?
When I first taught this lesson in April 2013, I had students also play through a short section of Capcom’s 2002 remake of Resident Evil for the Nintendo Gamecube console. With this, I wanted to demonstrate that styles born out of the creative navigation of technological constraints could outlive those constraints. The Gamecube certainly had no problem with detailed real-time 3D rendering (Resident Evil 4, released three years later, makes that clear). But in the 2002 version of Resident Evil, Capcom still sticks to static camera angles with characters moving across pre-rendered backgrounds. The message is clear: this is now an aesthetic choice, one that Capcom wanted to maintain as a visual hallmark of their series.
In 2015, there were a couple of developments that I will likely mention in this lesson, should I teach it again. First, Capcom remade the original Resident Evil again—this time, starting with the Gamecube version as a template, and upping the visual quality from there. There are points to be made about the game industry cannibalizing its past here, to be sure, but it’s also interesting to note that even by 2015 the notion of “old-school” static-framed survival horror held enough draw among fans to make this a sensible economic choice for Capcom.
More evidence that static framing in horror games still has its admirers: 2015 also saw the release of White Night by developers OSome Studios. With dramatic, pre-set camera angles accentuating its eye-popping high-contras black and white visual style, the game demonstrated that developers outside of Capcom are also interested in returning to cameras untethered to the player’s moving character:
We’ll have to wait and see if this constitutes a genuine return to form. Until then, though, these games at least provide a solid grounds for discussing the push and pull of technological constraints and aesthetic choice.
[i]. Taylor, Laurie N. “Gothic Bloodlines in Survival Horror Gaming.” In Horror Video Games: Essays on the Fusion of Fear and Play. Edited by Bernard Perron. Jefferson, NC: McFarland & co., 2009. Pg 52. (I assign this chapter as reading before this lesson.)
[ii]. Taylor, “Gothic Bloodlines in Survival Horror Gaming,” pg 50.
[iii]. Bissell, Tom. “Headshots.” In Extra Lives: Why Video Games Matter. New York: Vintage Books, 2011. Pg 24. (This is my second assigned reading for this lesson.)