My choice for this week’s example of interactive art is not exactly art, per se, but adjacent. Detroit: Become Human is a high-budget, triple-A, branching-narrative adventure game centered on a futuristic Detroit and its impending android revolution. Detroit is the latest entry in a collection of graphically sophisticated but lukewarm “interactive movies”, such as Until Dawn, Heavy Rain and Beyond: Two Souls, but unlike its predecessors, Detroit‘s attempts to tear down ludonarrative dissonance while incorporating meaningful gameplay are innovative and fascinating, if not fully successful.
Where most video games attempt to depict a narrative in spite of the form’s limitations, Detroit attempts to make its tropes a conscious move. For example, UI elements such as missions, objective markers and invisible boundaries are presented as part of the in-world HUD through which the three android protagonists perceive the world. This HUD distorts and even breaks as the protagonists gain sentience, and so surpasses being a disconnected layer of information to become narratively essential. In terms of narrative, Detroit surpasses its predecessors by offering a much stronger illusion of meaningful and irreversible decisions, as evidenced by the sophisticated flowcharts that display at the end of each level. On the other hand, Detroit is still obviously an “interactive movie”; levels are still movies interspersed with button pressing or thematically appropriate minigames, alternate scenarios are often just the same lines delivered by different actors, and the need to display the flowchart at the end of each level belies a deliberate appeal to completionist video game players. Perhaps these moves seem so innovative to me because I’ve become accustomed to the standards of gaming and accepting of ludonarrative dissonance, as we discussed in class. Nonetheless, seeing these small “tricks” in a highly produced, highly commercial video game that seems to earnestly try to push the boundaries of its form excites me.
Perhaps more interesting is the game’s laborious production and the implications of what that effort attempts to achieve. Detroit: Become Human is the result of €30 million and six years of development—over two years to research and write 3,000 pages, 250 3D-scanned and mocapped actors for 513 roles, and even a new game engine to support graphical advancements. Detroit betrays the video game industry’s desperation to imitate life. In addition to actors who portray the characters, there are actors who take every possible step in every scene. There are stunt doubles. All these actors are scanned and mocapped. Then the modelers and animators adjust the resulting 3D models and animations side by side with reference videos of the original performance. Eye and eyelid animation are manually added because they are excluded from the mocap data. Facial animations are also animated from scratch in action scenes. All this extended effort to reproduce a live performance is reduced to an awkward moment when a few hundred android figures appear untextured and unanimated in what is supposed to be a triumphant crowd due to hardware limitations.
It’s fascinating to examine how Detroit struggles with its conflicting desires to be interactive, but also narrative, but also commercial.