Copyright © 2007 by the Association for Computing Machinery, Inc.
Permission to make digital or hard copies of part or all of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed for
commercial advantage and that copies bear this notice and the full citation on the first
page. Copyrights for components of this work owned by others than ACM must be
honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on
servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request
permissions from Permissions Dept, ACM Inc., fax +1 (212) 869-0481 or e-mail
Sandbox Symposium 2007, San Diego, CA, August 04-05, 2007.
© 2007 ACM 978-1-59593-749-0/07/0008 $5.00
Journey of Discovery: The Night Journey Project as “Video/Game Art”
*University of Southern California, School of Cinematic Arts, Interactive Media Division; EA Game Innovation Lab
Figure 1: “Whirlwind” created by using an
additive filter effect in The Night Journey
Figure 2: Interlacing and additive filter
applied to forest environment in The
Figure 3: Video integration into 3D
environment in The Night Journey.
This paper describes the development of a video/game art project
being produced by media artist Bill Viola in collaboration with a
team from the USC Game Innovation Lab, which uses a
combination of both video and game technologies to explore the
universal experience of an individual’s journey towards
enlightenment. Here, we discuss both the creative and technical
approaches to achieving the project’s goals of evoking in the
player the sense of undertaking a spiritual journey.
Keywords: Game design, innovation, iterative process,
prototyping, playtesting, emotion in games, post processing, 3D
environments, game visuals.
The Night Journey project is a game project conceived by Bill
Viola, an internationally acclaimed artist and MacArthur fellow,
which uses both video and video game technologies to explore the
universal story of an individual’s journey towards enlightenment.
The game is being developed in collaboration with a team from
the USC Game Innovation lab led by Tracy Fullerton and
including Todd Furmanski, Kurosh ValaNejad, Kira Perov, Scott
Fisher and Andreas Kratky.
The project team has spent the past year exploring narrative,
visual and procedural themes related to the topic of enlightenment
and the possibilities for the project allowed by the game medium.
The overall objective has been to create a work that stretches the
boundaries of what games may be possible of communicating
with its unique content and mechanics.
Visual inspiration for this project has been drawn from the prior
works of Bill Viola, which afford reference for 3D objects, scenes
and presences in the world; provide textures for the landscape and
objects; and permeate the world itself, creating a bridge between
the “real” and the “imagined,” memory and experience.
Procedural inspiration is based in a set of deign goals that have
arisen from a central question asked early in the design process:
what is the “game mechanic” of enlightenment? How can we
abstract and systemize such an intensely personal, yet archetypal
The development of this game is still in progress; and the overall
success of its design remains to be seen. Early playtesting results
will be discussed below, which show the difficulty of designing a
game that, like all of Bill Viola’s prior work, requires that players
bring to it a high degree of personal interpretation, as well as the
willingness and skills necessary to play a modern video game.
However, as progress has been made, it has also become clear that
the juncture of the game design process and the process of the
individual artist provides an interesting area of research in and of
itself, regardless of the final product. This paper will describe
aspects of the Night Journey design process in order to point out
possible intersections of games and art in terms of intent,
execution, and experience.
2. Project Background and Design Goals
The Night Journey project was conceived by media artist Bill
Viola, whose work has been described as the “pursuit of
enlightenment through attention to transcendent experience.”
[Judson 1995] Viola began working in video art in the early
1970s, when the medium was just emerging. While he was still
attending art school at Syracuse University, he was showing his
work in exhibitions with people like Nam June Paik, Bruce
Naumann, Richard Serra, Peter Campus – all leading early video
artists. Early on, Viola realized that the “other half” of the raw
material of video was “the human perception system … the
viewer, or the viewing experience – the other half of the system.
It’s the dynamic interaction between these two systems, not just
the technology and language of video alone, that is the
fundamental nature of the medium.” [Bellour and Viola 1985]
This recognition of the viewer/participant is critical to Viola’s
work, which includes video, installations and, most recently, a
visualization of the Wagner opera Tristan und Isolde. The Night
Journey project sprung from the realization that computers and
video games were part of a major historical shift in imagery. As
early as 1985, Viola was already thinking ahead towards the level
of today’s computer graphics: “I see the technology moving us
toward building objects from the inside out rather than from the
outside in … soon images will be formed out of a system of logic,
almost like a form of philosophy – a way of describing an object
based on mathematical codes and principles rather than freezing
its light waves in time.” [Bellour and Viola 1985]
Viola began working to define the Night Journey with Kevin
Teixeira of Intel in 2002. The specification written by Teixeira
was a starting point for the USC Game Innovation Lab team,
providing goals for interactivity, visual concepts and scenarios.
One important idea described in this early specification was
“explorable video,” which became a touchstone for the team in
creating the visual aesthetic for the game. [Teixeira and Viola
The current goals for the Night Journey project were developed in
a series of creative meetings in which the team discussed both Bill
Viola’s inspiration for the piece in the illuminated manuscripts of
historical mystics as well as the potentially innovative nature of
the gameplay as it might relate to the Journey of Enlightenment.
Several high-level goals emerged from these meetings which have
guided the design process. These were:
The desire to evoke in the player’s mind a sense of the
archetypal journey of enlightenment through the
“mechanics” of the experience – i.e. what the player is
doing in the game.
The desire to create a world of “explorable video,”
integrating the work and sensibility of Viola’s prior art
into the game world.
The importance of creating an experience that would
appeal to (and be accessible to) both art patrons who
might play it in a gallery setting, and also game players,
who might access it through another from of
The process of setting high-level player experience goals as part
of an innovative game development process is an integral part of a
playcentric design process, which co-author Fullerton has
described elsewhere. [Fullerton et al. 2004] In general, the key
difference in the playcentric approach and a traditional game
design process is in the type of design goals which are set and the
methodology for reaching those goals during the production.
“Play-centric design is design and technology at the service of the
player experience.” [Fullerton et al. 2006] Throughout the design
and development process, prototyping and playtesting are used to
confirm the success, or make changes to, the system as designed,
measuring actual player experience against the high-level goals.
More will be said on this topic in the prototyping and playtesting
3. Game Structure
The Night Journey is a 3D first person game, and like most such
games, is organized around a navigable landscape. Unlike most
games, however, the Night Journey is not broken into a series of
“levels,” but is instead designed as a progressive “layering” of
experiences over the course of play. While experienced as
mysterious and obscure, as will be shown below, there actually is
a simple underlying layout for The Night Journey landscape.
The player begins in the intersection of four key geographical
spaces: forest, mountains, desert and sea, each with their own
sense of “infinity” stretching out to the horizon. In addition to the
four key geographic areas, there is also a vertical pole reaching to
the heavens and falling through the landscape. These “6 degrees
of freedom” (Figure 4) were sketched out by Bill Viola early in
the design process.
The play begins at that singular point of day which is not day or
night. The sun, sinking below the horizon, still lights the sky, as
does the moon, already rising. As the player begins their journey,
they fall down the vertical pole to the “landing spot.” Throughout
this initial fall, they have control of their gaze, but not their
imminent downward motion. At the highest point of the fall, the
player can see out across the immense vista of the game landscape
in all directions: a huge, unknowable space that would take
lifetimes to explore.
Figure 4: “6 degrees of freedom” in the world, design sketch by
Bill Viola (left) and early landscape design (right).
Upon landing, the player is free to traverse the game landscape
toward one of four distant goals, viewable as bright points of light
during the fall. Along their way, they will encounter other points
of interest and exploration. The basic controls consist of
“looking,” “moving” and “reflecting.” These are accessed using a
PlayStation 2 gamepad, with looking assigned to the right thumb-
stick, moving to the left thumb-stick, and reflecting to the “x”
button. Reflecting is a way of interacting with the world that
evokes a layering of imagery, and conceptual space, upon the 3D
world. When players choose to reflect, the world “transforms”
under their gaze. Reflecting also transforms the player, though
they may not realize this immediately. The more time a player
spends reflecting, the faster they will be able to move through the
world and the higher their viewpoint, until finally they begin to
glide over the landscape, barely touching the tips of the trees at
their highest level.
Figure 5: Overhead views of the game landscape at two stages of
design, with the sea at the top, desert at the bottom, mountains to
the left and forest to the right. At center is a series of canyons
that form a mandala when seen from above.
Reflecting will also maintain the fading twilight, keeping darkness
from approaching. By reflecting on specific points of beauty and
interest in the landscape and allowing these points to “fill them”
with light, players will be able to keep darkness at bay and
traverse the landscape a much longer period of time.
Eventually, however, night will fall. At the end of “twilight,”
when the sky has darkened to black, the player will “fall asleep.”
While they sleep, they will be transported to a new area of the
landscape and gently dropped back to resume their journey.
Should they reach the points of light in any of the four areas, they
will find a “St. John’s hut” – a spiritual hermitage in the form of
an old trailer, a cave, the ruins of a building, or an abandoned
cabin. In each of these “huts” is a dream of their journey, a
procedurally created video piece cued by the places and objects
they have visited.
3.1 Expressive geography
Most first person 3D games strive to create a recognizable,
geographically consistent terrain which the player’s and designers
have come to think of as “realistic.” This, combined with
interface features such as “mini-maps,” provides contextualization
and guidance for the player in game worlds, an important feature
when the goals of the level are built around motivating movement
from one point to another, and creating dramatic staging for
moments of gameplay.
For The Night Journey, we imagined the geography as an
expressive element, rather than a practical one, changing with the
player’s perceptions, actions, with time, movement, perspective
and overall offering the potential for a different interpretive
experience each time the player accesses the game. We wanted to
build what Bill has referred to as a “poetic landscape.” “Sense of
place has always been of primary importance in my video work,”
he says. “Sometimes the landscape becomes the subject of a
work, other times it shares the moment in balance with an action
taking place in it, yet, always its energy is present and felt for
what it is – the natural raw material of the human psyche.”
Upon landing, as described above, the player is able to look
around, and to move at the relative speed of a human being.
While most 3D games “empower” the player by allowing them to
move at relative speeds about 40-50 miles per hour across the
virtual terrain, the player of the Night Journey must be satisfied
with pedestrian speeds calculated to enforce our goals of moving
slowly and looking deeply at each moment of the experience.
(This restriction on the player’ movement is slowly and subtly
released as they learn to reflect.)
Figure 6: The “Great Tree” and the mountain peaks.
The underlying terrain itself has fixed landmarks, in order to
allow the player navigate, such as the “Great Tree” in the center
of the world, which is visible from many points in the landscape.
This tree, along with the mountain peaks and the full moon, allow
the player to situate themselves relative to map, no matter how far
they travel. This is important because the player of the Night
Journey will receive no “mini-map” to guide their progress.
As mentioned above, one way for the player to “push back” the
coming darkness is to find and “reflect” on special points in this
expressive landscape. These points are generally based on scenes
from prior works by Bill Viola, as will be described below.
Figure 7: Owl in a tree.
Figure 7 shows an owl who will respond to the presence of the
player if they wait long and patiently enough. Figure 8 shows a
vulture circling lazily over the player. If the player reflects on
these points, they will be rewarded with the layering of visual
imagery as well as a subtle boost in movement.
Figure 8: Circling vulture.
The concept of a game environment that requires interpretive
projection on the part of the player and organizes itself
geographically, visually and aurally around the sense of aporia
and epiphany embodied in the archetypal spiritual journey is very
exciting to us. We feel that this concept of creating an
“expressive geography,” both interactively and visually, is an
important part of achieving our first design goal: to evoke in the
player’s mind a sense of the archetypal journey of enlightenment
through the “mechanics” of the experience – i.e. what the player is
doing in the game. The next section will explore these
“mechanics of enlightenment” in more depth.
3.2 “Mechanic of Enlightenment”
One of the key goals for the game arising from the initial design
meetings is the notion that the procedural mechanics and the
message communicated to the player through their interaction
with these mechanics should express, in some way, the sensibility
of the spiritual journey within the player.
Many games use a risk/reward system for motivating player
interaction. Typical examples of core game mechanics that
involve weighing risk vs. reward include combat, resource
management, puzzle-solving, construction, etc. The goals of the
Night Journey preclude using these existing models as the basis
for a core mechanic; rather, we began to look for mechanics that
might offer an “action/reward” cycle, rather than a “risk/reward”
Examples of such mechanics include prior games by the USC
Game Lab, including Cloud and flOw. In these games, risk is
minimized or absent, but action is still rewarded. Inaction is not
punished, and players are judged under no time constraint or
scoring system. [Fullerton et al. 2006] Other games that we find
work on this precept of “action/reward” vs. “risk/reward” include
such disparate titles as Myst and its sequels, Second Life,
Nintendogs, and Animal Crossing. In thinking about the appeal of
these games, the team recognized that by diminishing the notion
of risk, they invited players to explore, both geographically and
The original design document gave us a place to begin our design
process, in that it described visual puzzles as a way of making
sure the player “changed perspective” before moving forward in
the game. In our early discussions, we decided against
implementing these puzzles directly, instead hoping to create a
more subtle, emergent game environment filled not with literal,
solvable puzzles, but rather with complex moments of visual and
procedural reflection, memory, transformation, and change that
the player must find and experience in order to progress.
An example of such a moment includes the re-creation of a
dilapidated shack in the forest area of the landscape. (Figure 9)
This shack, which could be the retreat of Zen Buddhist poet
Ryokan, is just as easily seen as an old hunting cabin, or a
forgotten storage hut, is hidden within a grove of trees. The light
from the shack shines through the trees as the player approaches,
though there is no “readily explicable light source” as per Bill
Viola’s written description of the scene.
Approaching the shack thins out the trees and reveals more of it
until you arrive at the clearing and the shack itself. This shack is
a 3D model of one seen in a hand-held shot from an earlier piece
by Viola. In the Night Journey, however, it becomes not only a
location and a visual reference, but also a memory space and a
portal through time.
As the player moves closer, their view slips freely between
several moments of the shack in time: fully restored, partially
destroyed, fully destroyed, and a wood structure reclaimed by the
forest. If they reflect on the shack, these moments will be
combined with the visual “reality” of the original video footage,
calling into question the relationship between the “real time” 3D
world we are navigating and the “reality” of the video footage on
which it is based. Upon entering the shack, the player will
transition to a procedurally created “dream” of their journey to
As mentioned above, finding and accessing such moments will
also “transform” the player, by increasing their movement and
perspective. Each time a player “reflects” their speed is slightly
(almost imperceptibly) increased until finally they will begin to
glide, rather than walk. Gliding higher and higher until they are
skimming the tops of the trees in forest they have walked through
earlier in the game.
4. Visual Structure
Early explorations for the visual style included examples from
different periods of artistic practice, each of which showed an
interesting interplay between mental and real spaces that intrigued
us in relation to the idea of the spiritual and physical journey of
the player. Experimentations with light and shadow, abstraction
and illusion led the team to believe a black and white environment
like those pictured above might best suit the game.
While we have remained committed to the choice of black and
white, as the project has developed, we tended to move away from
external references toward the reflexive use of imagery and
captures from Bill Viola’s prior works as both visual reference
and actual material for texturing and building of the 3D world.
As can be seen by Figures 6, 7 and 8, the look and feel of the
game does not in any way resemble a typical 3D game
environment. This is related to the discussion above about
creating an “expressive” landscape, one which had a sense of
“presence” rather than “realism.” Viola’s video work often
includes “low-tech” imagery, using 30-year-old cameras to create
an “enhanced” grain like “salt and pepper.” [Gayford 2006] Our
methods for creating this look and feel included the creation of
several custom technical solutions, visual processing techniques
for our 3D environment that will be discussed below.
4.1 References to Prior Work
An archive of work from Bill Viola was created during the design
research phase of this project. While not exhaustive, this archive
did point to a number of stylistic motifs that could be replicated as
effects in the 3D environment. For example, a number of the
pieces were shot with a “tube” camera, a form of early video
technology, which created distinctive artifacts in the images.
These were caused by the use of a pick up tube rather than a
charge-coupled device of current video cameras, and the blur and
burn they caused were a desired effect in the images we were
attempting to mimic.
Figure 9: Reference images from Bill Viola’s archive used to
design video effects for the 3D environment.
The decision, made early in the research process, to implement
our entire 3D scene in black and white made it possible to mix
various video references from different times and environments
with a greater sense of coherence and so served a primarily
aesthetic purpose. On a technical level, it also allowed for the use
of a set of custom post-processing methods which will be
discussed in the next section. Early work in the process included
the creation of 3D objects based on those found in the archive
images. So, for example, the caravan of people or the hut in the
wilderness seen in Figure 9 were modeled and animated to be
included in the 3D environment.
Our research into this area revealed that most prior work in non-
photorealistic rendering has focused on the mimicry of an artist’s
hand – using recognition of lines, and rendering these as brush
strokes to recreate the look of a painterly or sketched rendering,
for example. [Hertzmann and Perlin 2000] [Lake et al. 2000]
[Lander 2000] [Landsdown and Schofield 1995] [Markosian et al.
1997] [Mohr and Gleicher 2001] In our case, however, the
desired effect was not that of an artist’s hand, but an artist’s eye
combined with a particular imaging technology. Therefore, our
methodology would more appropriately be applied to the scene
after its normal rendering approach, rather than during.
4.2 Integration of Video and 3D Imagery
A basic problem with this approach, both technical and aesthetic,
was one of aesthetic integration: the visual quality of the video
references from Bill Viola’s work was at odds with the distinctive
attributes of a real-time 3D environment. Traditional 3D game
environments, even that using advanced rendering techniques
such as pixel or vertex shaders, have a recognizable “look and
feel” that is quite different from that of video imagery. Like
earlier attempts to make video look more like film, using post-
production effects to simulate grain or scratches, the aesthetic
goals here demanded a new technique for rendering the 3D scene
in such a way as to bring its visual qualities in line with those
expected from video images.
The specifics of this problem, and our solution, were also tied to a
specific artist’s visual style; so, instead of striving to replicated
realistic, highly detailed video imagery, the project has focused on
producing 3D video effects that mimic artifacts such as blur, burn,
glare, and interlacing as a part of the desired look and feel for the
4.3 Custom Post-Processing Methods
Several different methods were used to integrate video effects into
the 3D game environment of The Night Journey. The techniques
described here use artifacts such as those describe above as a
starting point for “degradation” (or, as we like to say
“enhancement”) of the original 3D scene as rendered in order to
achieve a look and feel more visually coherent with the integrated
Figure 10: Pre-processed landscape image.
The first method employed was a video raster effect using
interlacing and additive blending. This post-processing effect
does not use any pixel or vertex shaders for a number of reasons.
While some of these desired effects can be emulated with a pixel
shader (in some cases much faster), the whole effect currently
relies on logic branching, texture and object agnosticism, and
other elements outside of the limitations of typical hardware
shaders. The freedom acquired does come at the expense of
performance, but a careful balance of parameters can keep the
scene to within real-time frame rates (10-40 fps).
The process is achieved in this way:
A low resolution version of the scene is rendered normally.
The screen is copied to a texture.
Pixels are examined, and then a sprite corresponding to
variables such as size and value is drawn on the final screen.
Parameters change how many pixels are sampled, how large
the new "sprite pixels" are, etc.
The previous post-processed image has also been stored in
memory. This image is dimmed slightly, and the new final
screen is added. When interlaced this creates the video-like
effect, as well as increasing performance.
Figures 10 through 13 show the 3D scene with various stages of
the process. Figure 10 depicts a typical game scene rendered
normally. Note the subtlety of the light and shadows on the
objects and contours of the landscape. This is intentionally done
in order to allow for heightened contrast in the image after the
effect has been applied. Depth can be implied using this effect
even on 2 dimensional objects, such as flat texture mapped trees
such as the one on the left side of the image. The lower resolution
of the original scene render, combined with the softening
produced by the interlacing (Figure 11), lends verisimilitude to an
appropriately prepped 3D scene.
Figures 11 and 12: Same image with interlacing (left) and
additive filter (right).
Additive blending (Figure 12) tends to round the edges of the
geometry and reduce many of the side effects of the interlacing.
Also, elements drawn in the previous frame were still visible,
causing trailing on the image if there is rapid motion. Bright areas
in particular "burn in" to the screen using this process much like
the early video artifacts.
In addition, a “grain” effect was created to increase the visual
relationship to Viola’s “grainy cam” footage. The grain was
achieved by randomly adjusting the value of a given "sprite/pixel"
by a ranged amount, positive or negative, from the original pixel
value. This "grain level" could be adjusted dynamically to find a
suitable scale of "error". Low values would be almost
undetectable, while high values would be practically
indistinguishable from white noise or television static. By
carefully controlling the random offsets in color value, this digital
"grain" emulates the similar artifacting of Viola’s “grainy cam”
tube camera. (It is also reminiscent of grainy film stocks or
digital cameras in low-light settings.) Varying the value of a
given sprite/pixel per frame allows a wider range of greys to be
available, decreasing the "blow out" typical of additive blending,
and avoiding the "posterizing" effect that would often leave
distinct banding of specific shades of grey.
Figure 13: Final image, interlacing plus additive filter.
Overall, the visual methods employed in the projected have
tended away from the “realistic” solutions employed by
commercial games and toward a more expressive, poetic visual
solution, and one which (we hope) does justice to the work which
5. Textual Sources
One of our original design goals included integration of several
inspirational of texts including: Rumi, the 13th century Islamic
poet and mystic; Ryokan, the 18th century Zen Buddhist poet,
Shankara; the 8th century Hindu mystic and commentator on the
Upanishads; and St. Anthony, the 3rd century Christian desert
father. As we have moved through the design process, we have
gone through a number of potential ways of including those texts,
from direct visual and aural reference, to a carefully constructed
set of “narrative” moments that could be “found” in the landscape
in any pattern, morphing meaning from play to play.
For example, an early text from Plotinus (204-270 AD) provided
by Bill Viola game inspiration for a particular interactive moment
as the player crosses a creek bed. The entire quote went as
"Withdraw into yourself and look, and if you do not find
yourself beautiful yet, act as does the creator of a statue
that is to be made beautiful: he cuts away here, smooths
there, he makes this line lighter, the other purer, until a
lovely face has grown upon his work. So do you also:
cut away all that is excessive, straighten all that is
crooked, bring light to all that is overcast, labor to make
all one glow of beauty and never cease chiseling your
statue until there shall shine out on you from it the…
splendor of virtue, until you shall see the perfect image
surely established in the stainless shrine.
When you know you have become this perfect work,
when you are self-gathered in the purity of your being,
nothing now remains that can shatter that inner unity,
when you find yourself wholly true to your essential
nature, … -- when you perceive that you have grown to
this, you now become new vision: now call up your
confidence, strike forward yet a step -- you need a guide
no longer -- strain and see!"
In the early version, the player would hear snatches of this quote
as they entered the water of the creek – short, almost
unintelligible pieces, layering on one another like the water
rushing by, until the text itself seemed interwoven into the water
and the landscape.
Later iterations of the textual references included a set of custom
texts, all of which were written to reference historical works, but
took liberties with tone and perspective so as to create a random
set of story pieces which could flow together in any order they
were encountered, forming the journey of a single person out of
the archetypal writings of many mystics.
At this point, we are finding that these direct references may, in
fact, fade away giving preference to an experiential journey
heavily inspired and influenced by their vestigial presence. Bill
Viola has explained to the team, “there is text in all my work,”
though that text may not be directly visible to the viewer, since it
has “fallen away” as the piece becomes more and more visual.
6. Game Design Process
As part of our playcentric process, once high-level design goals
were set, the team’s first step in the game design process was to
create a paper model of our ideas and playtest it. This prototype
was a playable system that helped to focus on several of the
important issues surrounding the merger of our ideas, player
motivation, basic game procedures and consequences of actions.
From our discussions with Bill Viola, this creation of a paper
prototype was intriguing, but somewhat outside of his usual
process, which tended towards reading, writing and drawing
sketches of ideas. As part of the merger of the game design
process and his particular artistic process required by this project,
however, he was more than willing to participate and playtest
these rough initial designs.
6.1 Paper Prototyping
The playtests for the paper prototype were done very casually and
internally, as part of an ongoing discussion regarding the game
structure. Several game design students, the game innovation lab
design team and Bill Viola were the only test subjects. The
learning and discussion that came from this prototype centered on
the notion of how the player accesses and understands the
underlying rules of the game world.
Paper prototyping, as mentioned above, is a central part of the
play-centric method of game design. However, the design of this
prototype assumed that the focus of the player would be on
procedural mechanics, rather than on creative interpretation. It
became clear during this process that we needed to test more than
just system mechanics, but also the player’s interpretive
experience as well.
The paper prototype was created as a single-player game map
with another participant acting as the AI for the game logic. After
lengthy discussion, a set of test procedures were created for the
prototype map to guide the player on their journey. These were
imagined as “ways of being” in the world, such as being “still”
versus striking out quickly. Based on the choices the player
makes, rules were applied to the “landscape,” making orientation
and visual reference more or less difficult.
Figure 14: Bill Viola and Tracy Fullerton playtesting the Night
Just as traditional film and video may assume an underlying
narrative to “carry” the experience of the view along; so does
traditional game design assume a reliance on repetitive,
systematic actions within a goal-based risk/reward system to drive
player interaction with the game. Our discussions with Bill Viola
made us realize that while the overarching structure of the
prototype worked well enough to describe the physical layout of
the game, the model itself could not articulate the creative,
interpretive experience required for our high-level goals.
In the end, it was decided to begin implementing the basic game
map and “day/night” cycle structure in a rough digital
environment, and to concurrently build a set of tools which would
allow us to implement the visually and procedurally emergent
game environment described earlier in this document.
6.2 Digital Prototyping & Playtesting
The digital prototype was next created using an internal game
engine, Bushido. The engine was modified in several ways to
realize the game, including input via PlayStation 2 contoller,
integration of real-time video playback within the 3D space,
dynamically controlled post-processing effects and changes to
these effects based on location and proximity to certain landscape
points. Content for three of the four world areas (forest, desert
and ocean) has been implemented and temporary sound and video
assets integrated as placeholders. Initial playtests are currently in
progress and scheduled to continue through the Summer of 2007.
At this writing, three sets of playtests have taken place, over 6
weeks, involving 13 playtesters, divided into two main groups:
non-game players and game players. All of the playtesting
subjects were drawn from volunteers who responded to email
advertisements sent to USC faculty and students. Two separate
emails were sent, one to find players who were non-gamers but
might be likely to visit this game in a gallery exhibition, and
another to solicit experienced game players interested in new
genres of play. Volunteers from both groups were screened to
make sure that none had been involved in the design of the game
or had seen or played it prior to the testing session.
The following are the overall demographics of the test subjects:
Age 18-34: 46% Over 35: 54%
Gender Male: 62% Female: 38%
Plays games? Yes: 38% No: 62%
Visited gallery or
museum in last 6
Yes: 62% No: 38%
Familiar with Bill
Yes: 54% No: 46%
Over the course of the six weeks of testing, game play and
features were iterated from session to session, in order to respond
to player difficulties and suggestions. Several high-priority
findings, divided by gamers and non-gamers, include:
Figure 15: Playtesting the Night Journey digital prototype.
Hesitancy to begin interacting – prefer to wait until
“told” to act.
Difficulty navigating – do not have a good sense of 3D
environments or use of game controllers.
Interested in the “mysterious” look and feel; liked
exploring the world.
Did not feel a need for externally provdided objectives,
one tester stated that he felt “it’s my job to find the
hidden visual opportunities.”
Wanted to experience visual drama akin to Bill Viola’s
work, as one tester stated, “I’m looking for flames.”
No difficulty navigating 3D space.
Do not like to be restrained in speed; do not like
“automatic” features like having the game take over
camera control. Want to be in control all the time.
Interested in “novel” look and feel; one tester called it
“moody,” another “ghostly.”
Did not mind having to find their own objectives, one
tester suspected there were “hidden” goals he had to
find and was interested in doing so.
Liked moving fast, when that feature was available.
None of the game players were familiar with Bill
Viola’s work, but one wondered about the association
between “dreams and reality” in the game based on his
Overall, the tension between the expectations of the two main
groups of testers has revolved around experience in navigating 3D
worlds, as well as skill and confidence using the game controller.
Both groups have responded very positively to the look and feel
of the game, and have found it interesting and appealing to
explore the game world, finding and setting their own play goals.
Because of this, we feel confident that we will be able to achieve
most, if not all, of our initial design goals over the course of the
remaining playtesting and iteration process. These goals are, as
stated above, to create a sense of the “journey of enlightenment”
through the core game mechanics, to create a world of “explorable
video” and to make a work that appeals (and is accessible) to both
art patrons and game players. We feel that we are very close to
the second and third goals, and are still iterating on the first, most
As we move forward -- playtesting and iterating -- design
possibilities have become clearer to both the team and to Bill
Viola and it often seems that our prototyping process has in many
ways continued and extended the original discussion as to high-
level design goals. Unlike a typical game design process, which
would be focused on feature development and level design, or
even a playcentric process, which would tend to have fairly stable
high-level goals and a flexible, iterative methodology for reaching
those goals, the Night Journey’s design process has been an
ongoing philosophical trek throughout. In many ways, this is the
influence of Viola’s process on the project. In speaking about his
work he has said, “I sometimes think of my work as ‘rational
inspiration.’ I don’t like the way things are done in films: an
inspiration gets written up and set out as a sort of blueprint. The
act of shooting the film becomes a matter of following the
blueprint and reconstructing the original inspiration. Even though
some of my work is precisely predetermined down to the
individual shots, the experience I’m having while recording is still
connected to the work. It’s really the reason why I do it,
ultimately. It is important for me not always to know what I am
doing.” [Bellour 1985]
This seeming tension between the game design process and
Viola’s description of his process, while at first disconcerting, no
longer concerns us – in fact, the entire team has become clearer
and clearer as to the benefits of this extremely iterative process as
we proceed. It is not a journey of milestones and deliverables, but
one of discoveries and insights, much like the one we are trying to
create. In many ways, it is a productive merger of game design
methodologies and one particular artist’s process. Whenever we
voice the concern that a new idea may not be “doable” we receive
a happy confirmation from Viola that this is how we know it is
“worth doing,” because it is untried territory.
The Night Journey is an ongoing project, and, as mentioned in the
introduction to this paper, it remains to be seen how successful its
design elements will be in the end. However, the early playtesting
has given good results, and the team itself has already found its
own success in the sense that we have found a common territory
for collaborative work between the different processes of the
various artists involved.
Overall, we feel that we have found the beginnings of a
fascinating interplay between the process of game design and the
process of one particular visual artist. As opposed to the focus on
“fun” that traditional game design uses as a benchmark, the
learning-based goals set by “serious” games, or even the high-
level experience goals set by our own playcentric method, we
believe we are on a path to discovering how aesthetic goals and
the “voice” of a particular artist may be integrated into a game
design through a collaborative, iterative process.
The current phase of the Night Journey project is funded by the
Annenberg Center for Communication. Many thanks to
Jonathan Aaronson and the individuals at the Annenberg Center
for making this work possible. Additionally, a number of other
groups and individuals have contributed to the support of this
Bill Viola Studios http://www.billviola.com
National Endowment for the Arts, www.nea.gov
USC School of Cinematic Arts, Interactive Media
USC EA Game Innovation Lab,
BELLOUR, R., VIOLA, B. “An Interview with Bill Viola”
October (Vol. 34), Autumn, 1985.
FULLERTON, T., CHEN, J., SANTIAGO, K. ET AL. “That
Cloud Game: Dreaming (and Doing) Innovative Game Design”
Sandbox SIGGRAPH proceedings, 2006.
FULLERTON, T., HOFFMAN, S. and SWAIN, C. Game Design
Workshop: Designing, Prototyping and Playtesting Games. CMP
Books, San Francisco, CA, 2004.
GAYFORD, M. “Expanded Memories: Expanded memories: to
mark his major exhibition opening in Japan this month, Bill Viola
talks about his philosophy of art” Apollo, October, 2006.
HERTZMANN, A., PERLIN, K. “Painterly Rendering for Video
and Interaction” Proceedings, First International Symposium on
Non-Photorealistic Animation and Rendering, June, 2000.
JUDSON, W. “Bill Viola: Allegories in Subjective Perception”
Art Journal (Vol. 54, No. 4), Winter, 1995.
LAKE, A., MARSHALL, C., HARRIS, M. ET AL. “Stylized
Rendering Techniques for Scalable Real-Time 3D Animation”
Proceedings, Non-Photorealistic Animation and Rendering, June,
LANDER, J. “Under the Shade of the Rendering Tree” Game
Developer Magazine, February, 2000.
LANDER, J. “Shades of Disney” Opaquing a 3D World, March,
LANDSDOWN, J., SCHOFIELD, S. “Expressive Rendering: A
Review of Nonphotorealistic Techniques” IEEE Computer
Graphics and Applications, May, 1995.
MARKOSIAN, L., KOWALSKI, M., TRYCHIN, S. ET AL.
“Real-Time Nonphotorealistic Rendering” Proceedings, ACM
MOHR, A., GLEICHER, M. “Non-Invasive, Interactive, Stylized
Rendering” Proceedings, ACM Symposium on Interactive 3D
TEIXEIRA, K., VIOLA, B. “The Night Journey: Design
Overview & Content Assessment, Rev. 4.0 October 9, 2002”
Internal document, 2002.