University of Southern California Dissertations and Theses

duOS

(USC Thesis Other)

duOS

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Lu 1

DUOS

By Thomas Lu
Major: Interactive Media and Game Design
Degree: Master of Fine Arts
University of Southern California
May 12, 2017

Lu 2
Table of Contents
Introduction ……………………………………………………………………………….. 3
Design Simpler …………………………………………………………………………… 7
Unique Affordances First ……………………………………………………………........ 9
Depth vs Complexity …………………………………………………………………….. 13
Conclusion ……………………………………………………………………………….. 15

Lu 3
Thomas Lu
Professor Dennis Wixon
CTIN-594
8 March 2017
duOS PostPartum
Introduction
duOS is my Master’s Thesis project in the Graduate Interactive Media and Games
program at USC. I have been working on it since April 2016 and it has challenged every skill
that I have learned thus far. I am a designer by education, engineer by experience, and creative at
heart. My primary goals for USC’s Interactive program was to learn how to design games, how
to engineer games, and how to manage a team to build a game. Although duOS was a
challenging project for me in all these aspects, I will primarily be discussing the challenges
within design in this paper.
duOS is a single player puzzle platformer game where the player controls two robot
characters separately. The player has an option to “record” their actions and play them back to
allow the other robot to interact with the playback. The resulting actions makes it look like two
players are controlling both robots, but in reality only a single player controls them. This allows
the player to control two different characters with a high degree of precision. The player has to
solve puzzles revolving around precision and timing between the two characters. This is the core
of the gameplay.
My experience working with duOS has been a complex undertaking in design. duOS’
main mechanic is complex, and having a complex mechanic means that there are many issues
that can interfere with gameplay and the game’s original design goal. The team had faced three

Lu 4
main challenges over the course of the year pertaining to the design, which were: The core
gameplay loop, the main mechanic and how to introduce it, and the balance between depth and
complexity.
The original design goal was to push the recording mechanic in two separate but related
directions. In one direction, we wanted players to analyze their surroundings and approach a
situation analytically. In the other direction, we wanted players to be able to improvise when put
under pressure during combat puzzles. By balancing these two elements, the game was to be a
combination of fast-paced hectic puzzles and slow paced analytical puzzles.
Design/Experience Goals:
- We wanted players to record and coordinate with their playback actions
- We wanted the player to analytically approach non-combat puzzles rather than brute
forcing them.
- We wanted players to feel under pressure when solving combat-based puzzles.
- We wanted players to feel at fault when they failed so they can revise their approach
duOS has gone through a large amount of design revision since its original inception. The
original idea stemmed from a platformer prototype, when the character died, the player could
interact with their past dead life which would recreate the movements that the player just played.
Although this was a novel idea, playertesters felt that the main gameplay loop, getting help after
dying, was too punishing. This prompted the project for further prototyping. After a few
iterations, the game now prompted players to record their actions at their own leisure and play
them back. The game also had four individual characters that you can individually control and
record. The original idea was to have the player act as a unit controlling all 4 characters. The four
characters were as follows:

Lu 5
MIA

Description: MIA (Multipurpose Intelligence Android) was originally designed as an advanced combat mech for the military.
Created to mimic human mobility and combat, she is intelligent, agile, and nimble. Learning through her advanced AI, she
became disgusted towards violence. She abandoned her post and modified her parts for higher mobility to elude the military.
However, her tracking chip recalled her to the factory. On the team, her speed and high jumps are invaluable. She can also
perform wall jumps and hang onto walls. However, her own modifications resulted in structural instability in her joints resulting
in lower durability.

PLAT

Description: PLAT (Performance Lifting AutomaTon) is a heavy-lifting bot made for harsh environments. A factory malfunction
has welded a large piece of metal to his head. This mistake turned out to be rather handy as he can serve as a platform for his
allies to stand on. He can stand on spikes and withstand an incredible amount of heat. Although he can take harsh environments
with his above-average health, he has low speed making it difficult for him to navigate in fast paced environments.

Lu 6
CORA

Description: CORA, (Construction Robot Assistant)Originally a welding bot, her power core malfunctioned when it was
damaged during an electrical storm. Because of this, her power core operates at 1500% normal capacity. Her energy inhibitors
cannot contain the energy output of her welding apparatus. As a result, she fires super heated plasma from her arms that can
destroy almost anything. She has an average jump, average move speed, and average health.

HB

Description: HB (Heavy Bot) was originally a military defense project. Due to manufacturing defects, there was an imbalance of
metal density in his construction. Because of this, some parts of his body that are nearly indestructible while others are quite
vulnerable. He is a massive character with low speed and a short jump. Although his mobility is limited, his health and armor
more than makes up for it. He is by far the most massive and tank-like of the group - though he doesn’t do well with spikes or
high heat.

Lu 7
Design Simpler
These four characters were at the core of the gameplay. However, the initial prototype did
not showcase the different characters’ abilities. Instead, the four characters were identical. This is
where we ran into our first design flaw - the game was too cumbersome to play with four
characters.
During the initial tests - players would often have to do the same or similar things up to
four times, either to complete a task or to traverse the level. This system interfered with our
gameplay flow since the repetitive actions did not offer anything more than traversing the level
again. This lead to one of our first major design decisions: cut down the number of characters.
Our first intuition was to cut the number of characters into two, by combining the abilities of of
the characters. The result was MIA (MIA + CORA) and PLAT (PLAT + HB). We decided the
number on 2 since we found that two characters mitigated the complexity of character
management without detracting from the main mechanic.

MIA & PLAT’s redesign

Lu 8
MIA now has the shooting ability of CORA, and PLAT has the shield ability of HB but can also raise his shield to become a
platform.
The change in design direction was met positively - now players were commenting that
the pace of the game was faster and more action-packed. The decreased number of controllable
characters also allowed for easier puzzle design and engineering tasks. We created a vertical
slice of the game thereafter, which was also met with positive reception. After we finalized our
vertical slice prototype, we began work on full-scale production of the game.
During the pre-alpha stages of the game, we ran into a few more problems circulating
around the two characters. One of the issues dealt with the abilities of the two characters. MIA
and PLAT, while they operated well separately, did not have any abilities or affordances that
could help out one another in a puzzle setting. The game’s initial idea had a balance between
combat and puzzle solving, but during the early production phases, the game was slowly moving
towards more of a puzzle-based game. Since some of the initial abilities were more or less
designed for combat - we needed to find a way to make these abilities more puzzle suited. We
came up with the following: PLAT could slam the ground to launch MIA into the air, and MIA
could shoot PLAT to power up his jumps. However, neither ability made it into the final version
of the game, and this will be explained later.

Lu 9
Unique Affordances First
After we finished designing the characters - we began building out different levels to
gauge difficulty. Since there were a lot of things the player had to learn about the game, we had
to introduce each affordance to the player slowly as the game progressed. Our initial idea was to
introduce each individual character’s affordances first and then the recording mechanic. We
wanted to introduce it this way since we believed that the character’s individual abilities were
easier to understand, and once the player understood them, they would better understand how
they worked together with the recording mechanic.
However, after numerous playtest sessions - we realized that this was not the best way to
introduce our main mechanic. In our playtests, we found out that after the player had figured out
each characters’ affordances, players were adamant on solving puzzles with what they were
introduced with first. Players felt that the recording mechanic was more of a side ability rather
than the main mechanic. As a result, players would often try to brute force puzzles without the
recording mechanic which led us to believe that our design was not working.
After a series of convincing playtests, we knew we needed a new approach to the game.
As a result, we decided to start over. From what we gathered from our playtests, we decided that
what we needed to do was to introduce the recording mechanic as soon as possible - making it
the first thing the players see. We also decided to make puzzles that were only possible if players
controlled both characters precisely - which led us to building little nodes that players have to hit
at the same time.

Lu 10

Notice the nodes have a timer and players have to hit them before the timer elapses.
Our new approach yielded positive results, both for the design team and the player
experience. In the new approach, we introduced the recording mechanic first before anything
else, and slowly taught the player other affordances of the character one at a time as the game
progressed. The first thing that the design team noticed was that the puzzles were much easier to
create and much easier to understand from a design perspective. Now that the designers only had
to focus on one element of the puzzle, it allowed the team to focus on that specific element (e.g.
should it teach the player a new concept, or reinforce a concept with practice? Should it reinforce
the new ability or add a twist to it? etc). This way, when we introduced new elements of the
game, such as a new character ability, we were able to isolate the ability, and figure out how it
related to the recording mechanic and build from it.
With this new approach - players used the recording mechanic much more than before.
This reached our design goal of coordinating and cooperating with their own recorded self.
Players also approached puzzles and obstacles with a more analytical approach in comparison to
the previous iteration. This new approach tackled the problem from a different angle - introduce

Lu 11
the mechanic that makes the game unique before anything else, and slowly introduce small
nuances (such as abilities), in order to make the core of the recording mechanic deeper.
This design philosophy was inspired by The Legend of Zelda. The Legend of Zelda series
has exploration and combat at the core of the gameplay. Each new item and unlock in the game
gave Link a new way to explore, but it never detracted from its core. A great example is the
game’s hookshot power up. The hookshot not only allows players to explore places they’ve
never been by latching onto surfaces and pulling the player towards them, but also has an
offensive capability as well. When a new item is added into Link’s arsenal, it opens up a new
perspective to approach situations, but does not detract from its core gameplay. We wanted to
emulate this concept from The Legend of Zelda series - since we believed that it was a more
effective way to introduce a variety of abilities.
The new approach also had a higher focus on the timer switch obstacle. Previously,
players had to solve a number of different puzzles involving a variety of goals. This could be
hitting a switch, reaching a place that is unreachable, or destroying blocks in certain orders.
Though this kept the gameplay interesting for hardened veterans of puzzle platformers, anyone
who had any less experience would be frustrated and confused on how to approach the situation.
In our second iteration, we focused our gameplay down to just hitting these switches with the
recording mechanic. We approached the problem of puzzles to “how do I solve this puzzle”
instead of “what should I do to solve this puzzle”. This new approach allowed for easier puzzle
design due to a much more well-defined goal.

Lu 12

When the hover ability was introduced in this section, we made puzzles that were only possible when PLAT hovered and MIA
uses him as a moving platform.

Lu 13
Depth vs Complexity
In the new iteration however, the design of the abilities started becoming a problem.
PLAT and MIA’s offensive capabilities almost saw no usage. The design team believed that the
abilities did not add much to the game. Because of this, I started looking into some research and
theory on why the game felt this way.
During the course of my research, a topic that I found most interesting was the
relationship of Depth vs Complexity in games. The concept of both is as follows: “Depth is the
number of emergent experientially different possibilities or meaningful choices that come out of
one ruleset” and “Complexity is the mental burden of data from rulesets that the player has to
store” (ExtraCredits). Some examples of games with a lot of complexity are MMORPGs, where
there are a large number of choices and rulesets in the game. These large number of choices and
rulesets may overwhelm the player since there are so many things to remember. Depth is having
a one ruleset that can perform a variety of meaningful choices in the game. This allows simple
design with a lot of playability since the player can perform only a few actions to gain many
meaningful outcomes. According to Extra Credits, “it is the designer’s job to get the maximum
amount of depth with the minimum amount of complexity” (Extra Credits). These concepts gave
me a new perspective on my game, and I began striving for more depth in my game.
An example of a game with great depth is Portal. Portal is a game with a single concept
in which the game expands to over three hours of puzzles. You can shoot a portal onto any
surface, and enter and exit from these portals, but by using the gravity and physics of the world,
these portals can turn into man cannons, or it can help launch pieces of an objective to other parts
of the world. As the game progresses, the player is introduced to new objects and affordances
that play around the main mechanic. Projectiles and boxes are introduced, and all of them obey

Lu 14
the rulesets of the portals. The main takeaway here is that everything new that was introduced in
Portal somehow related back to its main mechanic.
This leads back to the design of MIA and PLAT - their combative abilities (shooting and
ground pounding) were just adding another ruleset to the game without supplementing the main
mechanic. The ground pound ability can launch MIA up into the air as well as destroy blocks
below. Although it could be used in some niche situations where PLAT could not reach MIA, its
applications are limited and are only used in gimmicky puzzles. It was a ruleset that introduced
very little meaningful choice, but added a layer of complexity. MIA’s shooting ability can
destroy blocks in front of MIA at a distance, and can increase PLAT’s jump height if PLAT is
hit. This ability is also limited since it could only fire at obstructions, and none of these
obstructions really needed the recording mechanic to overcome. Furthermore, the ability’s
secondary function to power up PLAT was another complex rule that had to be introduced,
which was not initially intuitive. In the book Game Design Vocabulary, it states that by
“establishing relationships between verbs, we give ourselves more opportunity to design
choices” and that “we want our verbs to be as developed as possible “ (Anthropy 17) (Anthropy
18). In a sense, duOS’ combative verbs, shooting and ground pounding, had little relationship to
the main verb - which is recording and playing back. A combination of this, and its complex and
nuanced rules, the designers felt that they added very little to the game’s overall design.
Although they were enjoyable to use, the team decided to remove the abilities from the game.

Lu 15
Conclusion
In duOS’ current state, the playtest results are generally positive. The changes in design
philosophy did not influence the players’ perception of difficulty. Although we had simplified
the game to better fit for a wider range of audiences, the game’s overall difficulty rating has not
changed with hardened veterans of the genre. The design team believes that the overall changes
to the game’s structure and design philosophy has been generally successful.
The takeaways from duOS is clear - simplicity is better. Every single major design
problem that duOS suffered from can all be traced back to this concept. From the multitude of
characters, the plethora of abilities introduced, and the introduction of the recording mechanic -
all these problems can be solved if the game had a simple outlook from the start. This is not to
say that a complex game is a bad game, but rather a game needs to relate all of its affordances to
its main and core mechanic. If you can describe your game’s core gameplay in a few simple
verbs, make sure that any other verbs that are used in the game relates to back to the core. We
have found that this concept has been the driving force behind our successful design decisions.

Lu 16
Works Cited
Anthropy, Anna. A Game Design Vocabulary. New Jersey: Addison-Wesley, 2014. Print.
ExtraCredits
https://www.youtube.com/watch?v=jVL4st0blGU

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Distance: a cooperative communication game to long-distance players

PDF

Light at the End of the Tunnels: level design and its relationship to a spectrum of fear

PDF

Fall from Grace: an experiment in understanding and challenging player beliefs through games

PDF

Revisions: an exploration of metafiction and metaphors in game design

PDF

The make of The Surveillant: a thesis project ""postpartum""

PDF

Super Opera Squad

PDF

Moloch: creating games with alternative mental state goals to move beyond flow

PDF

The Toymaker's Bequest

PDF

Bottles

PDF

Life On A String: an ink painting narrative game

PDF

Your presence is present enough: a thesis project postpartum

PDF

Bardcore!

PDF

The Death Mask: a study in interactive mystery

PDF

Stepstone Island

PDF

Come with Me: a cooperative game focusing on player emotion

PDF

FRKN WKND and video game mixtapes: developing talent and experience through video game mixtapes

PDF

OCTOBO: the interactive storytelling plush octopus

PDF

Southland

PDF

Wetware: designing for a contemporary dilemma

PDF

The Toymaker’s Bequest: a defense of narrative‐centric game design

Asset Metadata

Creator Lu, Thomas T. (author)

Core Title duOS

School School of Cinematic Arts

Degree Master of Fine Arts

Degree Program Interactive Media

Publication Date 04/27/2017

Defense Date 04/05/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag game,game design,mechanics,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Lemarchand, Richard (committee chair), Fullerton, Tracy (committee member), Wixon, Dennis (committee member)

Creator Email thomastianlu@gmail.com,thomastl@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-368661

Unique identifier UC11255901

Identifier etd-LuThomasT-5278.pdf (filename),usctheses-c40-368661 (legacy record id)

Legacy Identifier etd-LuThomasT-5278.pdf

Dmrecord 368661

Document Type Thesis

Rights Lu, Thomas T.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA