The Lost Level: Retro-Terminology in Early Arcade Games
“I think the real indicator will be when somebody confesses that they cried at Level 17.”
-Steven Spielberg on videogames as an artform (in Grossman n.p.)
The formal structure of segmentation exists across media objects. Books have chapters, films have scenes and videogames have levels. Except that isn’t exactly true. Not all books have chapters, not all films have scenes and not all videogames have levels. In this essay I argue that the term “level” is a contingent historical construction that belies both the heterogeneity of terminology used in early videogames and the nuanced multiplicity of the meaning of the word level. In fact, it is precisely the term’s unspecificity that I suspect is why it paradoxically came to be standardized and concretized.
Here I ask two questions: When did the level emerge as an identifiable structure, and when did “level” emerge as a term for that structure? The linguistic echo between these two occurrences is the reason I deem level to be an example of “retro-terminology”, not only in the sense that it is used in “retro” videogames but more pointedly that it can be retroactively applied to levels which exist avant la lettre.[1] While not entirely absent, the term level is conspicuously missing from many classic games. For example, until the New Super Mario Bros. subseries was launched in 2006, the term level was only ever used once in-game (in the title of Super Mario Bros.: The Lost Levels) to describe levels in the Super Mario series of platformers which is synonymous with level design.
To account for the terminology of level and its contemporaries, I employ a mixed-methods grounded theory approach in three phases. Nathan Hook describes grounded theory as a “process [which] starts with data collection, gradually building up categories and forming a theory, before linking that theory to previous literature at the end.” (309) For this study quantitative data on early arcade games was gathered from the YouTube channel Old Classic Retro Gaming, specifically its playlists “70s Arcade & Console Games (1970–1979)”, “Early 80s Video Games (1980–1983)”, “Early 80s Video Games (1980 -1983) Pt 2”, “Early 80s Video Games (1980–1983) Pt 3”, “Mid-1980s Video Games (1984–1985)” and “1983–1985 Arcade Laserdiscs” (“Old Classic Retro Gaming”).
As with all archives this database is partial, consisting of games available on noncommercial emulators. I restricted the inclusion criteria to released arcade games, partly for manageability and partly because console games are often paired with manuals that contain additional information not gleaned from the game itself. Additionally, only arcade games that refer to their level progression either textually or symbolically were included. This provided me with a corpus of 304 games released between 1979 and 1984.
Secondly, for further insight I looked at the paratexts (mainly manuals) of various versions of a paradigmatic case study: Donkey Kong. These paratexts reveal that even after the term level had entered the videogame lexicon, its meaning was multifaceted and rarely static. Lastly, reviewing some of the existing level design and game studies literature I offer a theory of where level design comes from.
Findings
Five major themes can be derived from the quantitative analysis of the collected data. 1) A majority of the selected games did not use the term level, nor was level the most frequently used term 2) There was a lack of consistency of terminology even among games of the same developer 3) There are significant regional differences in the terminology used 4) Complex notation systems for game progression were commonplace in early arcade games and 5) It was not until 1979 that arcade games textually or symbolically indicated progression.
The most commonly used term in the corpus was “round”, being present in 65 games, 57 of which used the term exclusively.[2] “Level” by comparison was the second most used term with 52 games, 40 of which used the term exclusively. The next most frequently used terms were “stage” (n = 23), “mission” (n = 16), “pattern” (n = 16), “wave” (n = 15), “scene” (n = 14), “course” (n = 9), “room” (n = 7), “phase” (n = 7), “sector” (n = 6), “act” (n = 5), “base” (n = 4), “maze” (n = 4), “game” (n = 4), “part” (n =3), “track” (n = 3), “area” (n = 3), “screen” (n = 2), “floor” (n = 2), “rank” (n = 2) “checkpoint” (n = 2) “zone” (n = 2) and “section” (n = 2). An additional 15 games used a unique term exclusively, while 7 used a unique term in conjunction with a previously mentioned term and 2 used multiple unique terms.
While some patterns did emerge, the dataset was characterized by an overall lack of consistency among games by the same developer. However, in some cases a term may be associated with a particular developer. For example, Data East and its subsidiary Data East USA accounted for 11 of 16 games (68.75%) that used the term “pattern”, while Williams accounted for 6 of 15 games (40%) that used the term “wave”. These two examples are also illustrative of the regional differences in terminology used between Japanese and Western developers. The terms “level”, “mission”, “wave”, “phase” and “room” were used by a majority of Western-developed games while the terms “round”, “stage”, “pattern”, “scene” and “course” were used by a majority of Japanese-developed games. It is difficult to ascertain any cultural reason for these regional differences, as the terms appear almost arbitrarily divided between East and West, although the preference for “round” over “level” may be partially explained by the perception of the English /l/ sound as an /r/ by some Japanese speakers.
As alluded to above several games used multiple terms to describe progression. In some instances this may have simply been a discrepancy between terms used in the attract mode, graphical user interface, bonuses, level select or interstitial screens. For example, Space Dungeon uses the terms “depth” and “level” interchangeably in this way. In other cases two different terms may denote a nested progression logic. For example, in Q*bert each “level” consists of four “rounds”. This sort of nested progression was later made famous by the “dash” notation of Super Mario Bros.
Other symbolic forms of representing progression exist in many early arcade games. Namco’s Galaxian, Pac-Man and Galaga are particularly notable in this regard. Galaxian uses flag notation in which progression is marked with a series of red triangular flags in the bottom right corner of the screen, going up to eight flags. In total, 9 games in the corpus used a variation of this flag notation. In Pac-Man a variation on flag notation which I call object notation is used, where progression is marked by a series of fruits and other objects. In total 16 games in the corpus used object notation, though often with less variety than Pac-Man. Lastly, Galaga used a unique medal notation, where progression through stages is marked by a series of medals denoting different increments (1, 5, 10, 20, 30 and 50).
Another symbolic means of representing progression is the progress bar. Although Konami’s Scramble contains the ur-example of a progress bar, the first game to use one was Data East’s Astro Fighter from 1979. In Astro Fighter the progress bar is segmented into five sections which each contain a sprite that represents an enemy type, which flashes when it is currently being fought. Interestingly, the progress bar in Astro Fighter progresses from right to left rather than left to right as in Scramble. After the fifth boss fight the game repeats back at the right-most enemy pattern. A total of 9 games in the corpus used a variation on the progress bar.
Lastly, a small number of games used terminology for the moments in-between levels rather than the levels themselves. The oldest example is SNK’s Ozma Wars from 1979, a space-themed shoot ’em up game. Ozma Wars begins with a “docking time” where the player’s energy is refueled by a larger ship, and these docking times recur throughout the game incrementing in number each time. Nichibutsu’s Moon Cresta and Universal’s Zero Hour, released the following year, turn these docking times into a mini-game in the vein of Atari’s Lunar Lander. SEGA’s vector-based Space Fury uses the concept of docking as both upgrade and level selection.
In addition to Galaxian, Astro Fighter and Ozma Wars, three other games from 1979 indicated progression through terminology. These are Irem’s Andromeda and Space Beam and Atari’s Monte Carlo. This is not to say that earlier games did not contain level progression, as will be explored in the “Discussion” section below.
“Structure Changes Shape”: Donkey Kong as Nexus of Meaning
With the data from the corpus of early arcade games we can answer the second of my two research questions: When did “level” emerge as a term for identifiable structures of progression in videogames? The answer is 1980, as evidenced by four games released by Taito that year: Crazy Balloon, Polaris, Steel Worker and the racist shooting game Indian Battle. Of these four, Crazy Balloon is the most notable for featuring levels as we understand them today, as discrete designed spaces. However, even though the term level was used in arcade games by 1980 does not mean that what level meant was universal. Thus, I argue that a statement like “the game Donkey Kong has four levels”, while true by contemporary standards, is still an application of retro-terminology.
Donkey Kong is a historically significant videogame for many reasons, however one of these reasons that has been largely unremarked upon is how it serves as a nexus of meaning for the term level. That is to say that at least five different meanings of level are accounted for within Donkey Kong. These are: level as tier, level as rank, level as repetition, level as difficulty and level as design.
Firstly, Donkey Kong contains levels as in the tiers or floors that make up each screen, although no official media refers to these as levels. Secondly, Donkey Kong contains levels as in the rankings on the high score table, which are referred to in the game as “rank”. Thirdly, Donkey Kong contains a direct reference to the term level in the top right corner of the screen, where blue text reads “L=” followed by the current level number. However, this L does not refer to the level as in the current screen, but to the level as in the current repetition.[3] The original Donkey Kong consists of several screens each 25 “meters” apart, four of which are unique (these screens have been likened to the four-part kishōtenketsu story structure of yonkoma manga) (Komanome n.p.; Paumgarten n.p.). Once the “rivet” screen is completed the game returns to the first screen and the level increases by one,[4] with the highest possible level being 22 due to technical constraints. Each repetition up to the 5th level also increases the difficulty by increasing the speed of obstacles. Thus, in Donkey Kong level as in repetition and level as in difficulty are intermeshed. The manual for the 1982 ColecoVision version of Donkey Kong makes explicit reference to difficulty level, while using the term “round” for repetition: “One round of play consists of five screens in this order: Ramps, Rivets, Elevators, Rivets, and Elevators. Complete one round and you move on to the next, starting with Ramps at a higher level of difficulty.” (8)[5]
The instruction card of the original arcade Donkey Kong makes reference to levels as designed spaces only obliquely, stating: “If Jumpman reaches top, Donkey Kong takes the lady higher up, and structure changes shape.” The manual for the 1983 Nintendo Entertainment System (NES) version refers to its three screens as “rounds” followed by a roman numeral (5). The manual for the 1982 Intellivision version uses the term “skill level” to refer to difficulty while also making reference to game progression: “When Mario removes the last rivet, Donkey Kong™ takes the girlfriend to a still higher level.” (7) Interestingly, this seems to imply not a temporal reset to the beginning of the game as in other versions, but that the game takes place on an infinitely tall series of structures. In 1987, Nintendo released The Official Nintendo Player’s Guide which explicitly referred to the NES version of Donkey Kong’s screens as levels (Nintendo 154). A recently translated 1997 article by Donkey Kong programmer Hirohisa Komanome uses both the terms “stage” and “level”, as well as an analogous term for level design in “screen composition” (n.p.).
Donkey Kong demonstrates that the term level condenses several distinct yet related meanings. However, it does not tell us where levels, as a term and as a structure, originated from. Below I review some of the level design and game studies literature to bring clarity to this subject.
Discussion: The Origins of Level Design
Luditecture or Lucidography?
I had a strange experience in an undergraduate course about game engines. The instructor was comparing the terminology used within two engines, GameMaker and Unity. He was puzzled by GameMaker’s nomenclature of “room”, and wondered why its developers hadn’t opted to use what to him was the more straightforward term “scene”, as per Unity’s naming convention. As a long-time GameMaker user this had never occurred to me–if anything I found it counterintuitive. I believe there are a few reasons for this difference of opinion. The first is that GameMaker is a largely 2D engine, and thus new rooms have a finite and preset amount of horizontal and vertical space; they are defined by their limits. Unity on the other hand was designed as a 3D engine first, and scenes are defined by the presence of a virtual camera, stretching boundlessly in xyz coordinate space.
In this anecdote we can see two interpretations of the same thing, one employing an architectural metaphor (room) and the other employing a cinematic metaphor (scene). We might transfer this dichotomy from the realm of game engines and onto videogames themselves. Where does videogame level design come from, architecture or cinema? In the game studies and level design literature, architectural approaches abound (Iacovoni; Walz; Totten; Götz and Gerber; Whistance-Smith). Despite the anxieties of early game theorists about the influence of film studies, straightforwardly cinematic approaches are far more sparse (King and Krzywinska; Walther; Moh and Zaidi).
Behind this dichotomy are two broader concerns: architecture as a spatial medium and cinema as temporal medium. Spatiality has long been a key concern of canonical game studies texts (J. H. Murray; Newman; Nitsche, Video Game Spaces; Calleja) and there are several edited volumes on the subject including Space Time Play, Ludotopia, Game | World | Architectonics and Video Games and Spatiality in American Studies. Recent monographs have approached spatiality from established fields such as cultural studies (S. Murray) or by engaging with emerging research on embodied cognition (Whistance-Smith).
Temporality by comparison has been somewhat neglected in game studies, for example despite second billing in the title of Space Time Play only two chapters mention the word time in their title. The first decade-or-so of game studies production did see individual chapters on temporality by Espen Aarseth and Patrick Crogan, as well as research by Michael Nitsche (“Mapping Time”) and José P. Zagal and Michael Mateas (“Temporal Frames”, “Time in Video Games”). In 2016 a German-language volume entitled Time to Play was published, but it was not until Christopher Hanson’s Game Time published two years later that the field saw a monograph dedicated to the subject. Also worth noting are texts that address both spatiality and temporality either separately (Wolf; Juul) or together (Alvarez Igarzábal).
Of course, it is insufficient to say that architecture is purely a spatial medium and cinema purely temporal, as with videogames they are both a combination of both. Following Martin Picard, who describes videogame levels as “unit[s] of place (and time) in the progression of a game”, my approach is both spatial and temporal (or rather spatiotemporal) and neither architectural nor cinematic but attending to the uniqueness of videogames as a medium (170).
In the corpus of early arcade games we can see this dichotomy between the spatial and temporal in the terminology used. Most of the terms are spatial in nature, such as “course”, “room”, “base”, “track”, “area”, “screen”, “floor”, “maze”, “checkpoint”, “section”, “zone” and “sector”, while a few are temporal; “round”, “phase”, “act” and “docking time”. Others can be read as both spatial and temporal such as “stage”, “wave”, “scene”, “pattern”, “part” and, indeed, “game”. The term “level” reflects this overall tendency towards the spatial but could also potentially be read temporally.
Zagal et al. are somewhat insistent that the term level be understood primarily spatially, defining it as “a recognizable subspace of the gameworld.” (182) They argue that games like Pac-Man “do not have spatial levels because the maze is always the same” (183). The error Zagal et al. make here is one of retro-terminology; who says that a level must be inherently spatial while rounds are inherently temporal? As shown by the corpus this distinction was not made by early arcade game developers. Furthermore, in the case of Pac-Man the assertion that each maze in the game is the same location is an interpretation that is inextricable from the expectation of differentiation in levels garnered by later games, such as Donkey Kong.
Dungeons & Drivers
It serves to consider the etymology of the term level outside the context of games. The word level was first used as a noun in the mid-14th century meaning a “tool to indicate a horizontal line” derived from the Old French livel and ultimately the Latin libella, a diminutive of libra (“Level”). In the early 15th century the word became used as an adjective and then as a verb in the mid 15th century (Ibid.). Circa 1600 the word level took on a new “[f]igurative meaning in reference to social, moral, or intellectual condition” (Ibid.). This early modern hierarchical emphasis of level translates to its contemporary usage within videogames, as well as prior architectural uses i.e. “levels” of a building.
A key flashpoint in the history of levels and their relation to games was the publishing of the first edition of the tabletop role-playing game (TTRPG) Dungeons & Dragons (D&D) by Gary Gygax and Dave Arneson in 1974. As noted by Zagal et al., D&D used the term level in two distinct senses (182). The most prominent usage was the levels of character progression. Under the subheading “Levels” Gygax and Arneson write: “There is no theoretical limit to how high a character may progress, i.e. 20th level Lord, 20th level Wizard, etc. Distinct names have only been included for the base levels, but this does not influence progression.” (8) Though sometimes conflated, a table on the same page illustrates the amount of experience points (XP) that correspond to each level. This concept of “measured progression” was previously featured in Arneson’s Blackmoor TTRPG campaigns (Peterson 95).
However, a second usage of the term level is present in D&D’s first edition. Under the heading “Preparation for the Campaign” Gygax and Arneson write: “First, the referee must draw out a minimum of half a dozen maps of the levels of his ‘underworld’, people them with monsters of various horrid aspect, distribute treasures accordingly, and note the location of the latter two on keys, each corresponding to the appropriate level.” While one may read the first instance of “level” in this quotation as being metaphorical–“levels of underworld” being akin to Dante’s circles of hell–the second instance makes clear that the term level is being used to refer to the design of discrete spaces, in this case taking the form of two-dimensional paper maps. When combined, the two meanings of progression and design present in D&D set the stage for the adoption of the term level by videogames in the 1980’s.
In his industry-standard textbook Level Design: Concept, Theory, and Practice, Rudolf Kremers briefly considers the prehistory of videogame level design, looking at sports, board-game layouts, pinball machines and D&D (13–15). Ed Byrne’s earlier Game Level Design makes similar observations about D&D and pinball, stating that “[g]ameplay needs a vessel in which to exist.” (6–7) Kremers concludes that “level design is not exclusive to video games [and] never exists purely on its own terms.” (15) While I am inclined to agree with Kremers and Byrne, they also overlook a particularly relevant example from the sporting world: golf.
I am not the first to make this connection between golf and videogame level design. In their Donkey Kong retrospective Art Maybury describes both domains as “[t]he organization of multiple courses to challenge the player’s skills in a specific order that is understood and scored as parts of a whole” (n.p.). One plays a “round” of golf holes akin to the leveled repetitions of screens of Donkey Kong. We might also consider how the iconography of golf is present in videogames, from the previously mentioned flag notation of early arcade games to the flagpoles that mark the endings of most Super Mario Bros. levels. It is no coincidence then that Mario has appeared in many golf videogames between 1984 and 2021.
If golf is an ancestor of level design, then it follows that the earliest example of a golf videogame may provide us with some insight into the historical development of levels. Indeed, this appears to be the case with Computer Golf! for the Magnavox Odyssey 2, released in December of 1978. Computer Golf! features nine holes each with their own distinct layout. The screen even changes to a “zoomed-in” perspective/sub-level once the player successfully lands the ball on the green. The game also formalizes a videogame convention that the player’s avatar does not necessarily have to travel between levels; “a good walk spoiled” becomes a good warp spoiled.
Interestingly, a black-and-white mini-golf game from Atari aptly titled Atari Mini Golf was also developed in 1978 but never released. Atari went on to release Miniature Golf in 1979 and Golf in 1980, both for the Atari 2600. The fact that these miniature versions of golf so closely resemble the non-miniature versions speaks both to the toyetic nature of early videogames and the recursive mise en abyme of adaptations of adaptations that nonetheless bear the name of “golf”. It is also fascinating that developers of the 1970’s had the technological capacity to produce videogames with designed levels, yet it did not occur to them to do so until tasked with translating a real-world activity. Consider also the technological context of Computer Golf! and how the modularity of home console gaming with its swappable cartridges mirrors the modularity of level design with its discrete segments.
Conclusion
Does Computer Golf! then fulfill an answer to the question “when did the level emerge as an identifiable structure?” Perhaps. We might also consider early videogames which leveraged procedural generation, such as Side Trak, Blasto and The Amazing Maze Game, as well as shooting games like Gun Fight (also known as Western Gun) and Boot Hill in which objects such as cacti appear more numerous each round of play, suggesting at least some tenuous link between time and space. Early role-playing and adventure games, whether graphical like Gary Whisenhunt and Ray Wood’s dnd or text-based like William Crowther and Don Woods’ Colossal Cave Adventure are also potential candidates. It all depends on how one defines the term “level”, for while many would likely agree that they are spatiotemporal units, these units are not uniform. That is, there is a large degree of variance between levels of the same videogame, let alone how levels compare across myriad examples. Some may say that this variance is the very point of levels to begin with. Therefore, because of the ontologically ambiguous status of the videogame level, the most accurate answer to the above question I can provide is sometime in the mid-to-late 1970’s.
[1] In his study of the etymology of the term “first-person shooter” Carl Therrien draws similar conclusions when he says that “usage of the expressions ‘first-person’ / ‘first-person shooter’ occurs retrospectively on many objects that were not referred to in the exact same terms.” (n.p.)
[2] This does not include Jack the Giantkiller, which uses the text “R=” that presumably stands for “round”.
[3] In fact, the L may not stand for level at all but instead “loop” as evidenced by the manual for the 1983 NES version of Donkey Kong (8).
[4] See Zagal et al. 184 for a detailed table of Donkey Kong’s screen progression.
[5] Note that these screens differ from the arcade version.
Works Cited
Aarseth, Espen. “Aporia and Epiphany in Doom and The Speaking Clock: The Temporality of Ergodic Art.” Cyberspace Textuality: Computer Technology and Literary Theory, edited by Marie-Laure Ryan. Indiana University Press, 1999. 31–41.
Aarseth, Espen, and Stephan Günzel, eds. Ludotopia: Spaces, Places and Territories in Computer Games. transcript Verlag, 2019.
Alvarez Igarzábal, Federico. Time and Space in Video Games. A Cognitive-Formalist Approach. transcript, 2019.
Bonner, Marc, ed. Game | World | Architectonics: Transdisciplinary Approaches on Structures and Mechanics, Levels and Spaces, Aesthetics and Perception. Heidelberg University Publishing (heiUP), 2021.
Byrne, Ed. Game Level Design. Charles River Media, 2005.
Calleja, Gordon. In-Game: From Immersion to Incorporation. MIT Press, 2011.
Crogan, Patrick. “Gametime: History, Narrative, and Temporality in Combat Flight Simulator 2.” The Video Game Theory Reader, edited by Mark J. P. Wolf. Routledge, 2013. 275–301.
Donkey Kong. ColecoVision version game manual, Coleco, 1982, www.archive.org/details/donkey-kong-game-manual-coleco-vision-1982_202105/page/8/mode/2up.
Donkey Kong. Intellivision version game manual, Coleco, 1982, www.gamesdatabase.org/Media/SYSTEM/Mattel_Intellivision//Manual/formated/Donkey_Kong_-_1982_-_Coleco.pdf.
Donkey Kong. Nintendo Entertainment System version game manual, Nintendo, 1983, www.nintendo.co.jp/clv/manuals/en/pdf/CLV-P-NAAEE.pdf.
Götz, Ulrich, and Andri Gerber, eds. Architectonics of Game Spaces: The Spatial Logic of the Virtual and Its Meaning for the Real. transcript Verlag, 2019.
Grossman, Lev. “The Art of the Virtual.” TIME, 8 Nov. 2004, content.time.com/time/subscriber/article/0,33009,995582,00.html.
Gygax, Gary, and Dave Arneson. Dungeons & Dragons: Single Volume Edition. Tactical Study Rules, 1974.
Hanson, Christopher. Game Time: Understanding Temporality in Video Games. Indiana University Press, 2018.
Höltgen, Stefan, and Jan Claas van Treeck, eds. Time to Play: Zeit und Computerspiel. Glückstadt: Verlag Werner Hülsbusch, 2016.
Hook, Nathan. “Grounded Theory.” Game Research Methods: An Overview, edited by Petri Lankoski and Staffan Björk. ETC Press, 2015. 309–320.
Iacovoni, Alberto. Game Zone: Playgrounds Between Virtual Scenarios and Reality. Springer Science & Business Media, 2003.
Juul, Jesper. Half-Real: Video Games Between Real Rules and Fictional Worlds. MIT Press, 2005.
King, Geoff, and Tanya Krzywinska, eds. Screenplay: Cinema/Videogames/Interfaces. Wallflower Press, 2002.
Komanome, Hirohisa. “Donkey Kong: A Record of Struggle.” Translated by Alex, 2023, www.shmuplations.com/donkeykong/.
Kremers, Rudolf. Level Design: Concept, Theory, and Practice. CRC Press, 2009.
“Level.” Online Etymology Dictionary, www.etymonline.com/word/level.
Maybury, Art. “Donkey Kong [1981].” Arcade Idea, 17 Aug. 2020, www.arcadeidea.wordpress.com/2020/08/17/donkey-kong-1981/.
Meinel, Dietmar, ed. Video Games and Spatiality in American Studies. Walter de Gruyter GmbH & Co KG, 2022.
Moh, Ian Soon Lee, and Syed Fawad Mustafa Zaidi. “Game Cinematography: Towards Understanding Relationship between Spatial Distortion and Game-play.” Proceedings of the 2019 3rd International Conference on Virtual and Augmented Reality Simulations. 2019.
Murray, Janet H. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. MIT Press, 1997.
Murray, Soraya. On Video Games: The Visual Politics of Race, Gender and Space. Bloomsbury Publishing, 2017.
Newman, James. Videogames. Routledge, 2004.
Nintendo. The Official Nintendo Player’s Guide. Tokuma Shoten, 1987.
Nitsche, Michael. “Mapping Time in Video Games.” DiGRA Conference. 2007.
Nitsche, Michael. Video Game Spaces: Image, Play, and Structure in 3D Worlds. MIT Press, 2008.
“Old Classic Retro Gaming.” YouTube, 8 Apr. 2010, www.youtube.com/@oldclassicgame.
Paumgarten, Nick. “Master of Play.” The New Yorker, 12 Dec. 2010, www.newyorker.com/magazine/2010/12/20/master-of-play.
Peterson, Jon. The Elusive Shift: How Role-Playing Games Forged Their Identity. MIT Press, 2020.
Picard, Martin. “Levels.” The Routledge Companion to Video Game Studies, edited by Mark J. P. Wolf and Bernard Perron. Routledge, 2023. 170–177.
Therrien, Carl. “Inspecting Video Game Historiography through Critical Lens: Etymology of the First-Person Shooter Genre.” Game Studies 15.2 (2015).
Totten, Christopher W. An Architectural Approach to Level Design. CRC Press, 2019.
Von Borries, Friedrich, Steffen P. Walz, and Matthias Böttger, eds. Space Time Play: Computer Games, Architecture and Urbanism: The Next Level. Basel: Birkhauser, 2007.
Walther, Bo Kampmann. “Cinematography and Ludology: In Search of a Lucidography.” Dichtung Digital. Journal für Kunst und Kultur digitaler Medien 6.1 (2004): 1–21.
Walz, Steffen P. Toward a Ludic Architecture: The Space of Play and Games. ETC Press, 2010.
Whistance-Smith, Gregory. Expressive Space: Embodying Meaning in Video Game Environments. Walter de Gruyter GmbH & Co KG, 2022.
Wolf, Mark J. P., ed. The Medium of the Video Game. University of Texas Press, 2002.
Zagal, José P., and Michael Mateas. “Temporal Frames: A Unifying Framework for the Analysis of Game Temporality.” DiGRA Conference. 2007.
Zagal, José P., and Michael Mateas. “Time in Video Games: A Survey and Analysis.” Simulation & Gaming 41.6 (2010): 844–868.
Zagal, José P., Clara Fernández-Vara, and Michael Mateas. “Rounds, Levels, and Waves: The Early Evolution of Gameplay Segmentation.” Games and Culture 3.2 (2008): 175–198.
