Games, Montage and the First Person Point of View

Games, Montage and the First Person Point of View By Michael Nitsche michael.nitsche@lcc.gatech.edu LCC – Georgia Tech Executive Abstract Montage of moving images is an effect present in games. To investigate this effect we look at the use of the first person point of view (POV) in montage elements controlled by the player. The focus of the analysis lies on montage that combines gameplay and visualization. The interactive control that generates a reference between player and player-character emerges as a condition for meaningful montage. A view to the value of classic film theory to this interactive condition concludes the paper. Montage and Games The visual fragmentation of the interactive playground through larger levels and 3D worlds introduced cinematic elements to games. Once games outgrew a single screen and perspective some form of selection of visible elements became necessary and framing, mise-en-scene and montage entered the video game world. Not surprisingly, a wide range of work addresses the question of camera control in real-time 3 dimensional virtual worlds [Drucker 1994; He/ Cohen/ Salesin 1996; Tomlinson/ Blumberg/ Nain 2000; Courty et.al. 2003], cross-referencing between games and movies [Manovich 2000; King/ Krzywinska 2002] and the use of cut-scenes [Klevjer 2002]. But apart from the very well covered work on interactive cinema (see e.g. Davenport’s group at MIT, Manovich in San Diego, the European SAGAs program) few have looked into the element of montage in real-time 3D worlds. Montage of a film image is understood as the technique and result of selecting, editing, and piecing together separate film clips into a linear sequence. In contrast to film video games pick from a limited pool of pre-defined and rule-driven viewpoints instead of a limited pool of prefabricated shots. As game action depends on the player’s interaction the resulting image is highly variable. This seems to conflict with any relevant theory of montage. Unpredictable combinations complicate the traditional category of montage as a result of the editing and as a distinct process of meaning generation. Following this logic, Manovich identifies an ‘anti-montage tendency in GUI’ [Manovich 2000, 143]. But, in practice, montage as a modus operandi stays present in games. The pictures are still edited and the resulting montage is not only functional – as Poole suggests [Poole 2000, 93] – but also representational. It not only serves the gameplay but also expresses meaning and style. To shape this expression, a game like Gretzky NHL 2005 (989 Sports for EA, TBP) can include 700 different camera positions. Notably the editing here is heavily geared towards the replay mode during which the action is already defined. The cameras show the event in a non-interactive but highly dramatized way that quotes live TV sport broadcasts. Interactive Montage and the First Person Point of View In contrast to the replay mode, interactive montage is tied to the variable gaming situation. Any cut during an interactive montage is triggered by the player in combination with a game feature – not as mere visual effect. The resulting montage cannot be traced any traditional style. Interactive montage operates only within the gaming situation and is a defining element of the play as the cut becomes part of the interaction that shapes the event. While the literature on montage in games is still very thin – the work on element of montage as triggered by the player during the gameplay has not been developed at all (e.g. missing in [Salen/ Zimmerman 2003]). Yet it is in this form of interactive montage where the strengths of the game and the expressive feature of cinema depend on each other. Instead of the much-discussed conflict between the two media the interactive montage offers a shared field where gameplay and cinematic language are closely intertwined and interdependent. Thus, any ‘annihilation’ [Eskelinen 2001] of a discussion of these cinematic features would be misleading – any approach based too closely on film theory [Wolf 2001] would lack the game specifics at work. To avoid such asymmetric concentration on either medium two restrictions apply in this paper’s discussion: 1) The editing has to be integral part of the functional gameplay and connected to an interactive option. Any cut has to be clearly motivated by the underlying game system and initiated by the player. 2) In order to allow for comparison between different games and the development of a general montage form the paper concentrates on cuts either to or from a first person POV – a camera perspective once proclaimed the only feasible one in VR from [Laurel 1986] to [Manovich 2000]. Such a restriction narrows the range of games but provides the necessary frame for in-depth analysis. The reference between the player and the main player-character is at the center of the investigation. Established through interactive access to the virtual character’s behavior and cinematic means that present the character to the player while rooting it inside the virtual world, this reference becomes the joint around which the interactive montage of the first person POV develops. How applicable are classic film theories to this condition? Instead of the assembly of different visual attractors (Eisenstein) the player-character position is the reference for the cut’s efficiency and is re-instated. Such a character-centered approach still allows for combination of seemingly opposing attractors. Instead of a guidance of the audience through the camera (Pudovkin) the camera has to be guided by the player. What we can identify in the interactive montage of first person POVs are elements of Bazin’s realist cinema [Bazin 1967]. Especially his demand for long takes, eye-level camera perspectives and unobstrusive cutting are part of many cuts to and fro first person POVs. Such a reference makes sense in the light of the new "reality" that the game generates in its event generation. Thus, players realizing the event and its visualization might retrace Bazin’s ideals not only in the gaming situation but also in the cinematic presentation and montage of it. While this approach might provide one source for the underlying theory the practical analysis of various montage elements has to be an ongoing process of which this paper is part. References - Bazin, André, What is Cinema? (Berkeley/ Los Angeles: University of California Press, 1967) - Courty, Nicolas/ Lmarache, Fabrice/ Donikian, Stéphane/ Marchand, Éric, ‘A Cinematography System for Virtual Storytelling’, in: Virtual Storytelling. ICVS ’03 (ed. by Olivier Balet/ Gérard Subsol/ Patrice Torguet) (Berlin et al: Springer, 2003), 30-37 - Drucker, Steven M., Intelligent Camera Control for Graphical Environments (Ph.D., Massachusetts Institute of Technology, 1994) - Eskelinen, Markku, ‘The Gaming Situation’, in: Games Studies, vol. 1, 1 (July 2001) at: http://www.gamestudies.org/0101/eskelinen/ - He, Li –Wei/ Cohen, Michael F. and Salesin, David H., "The Virtual Cinematographer: A Paradigm for Automatic Real-Time Camera Control and Directing"