This dissertation examines how the algorithmic processes and logics of 3D videogame engines shape our experiences of embodiment game play. Videogame engines are software tools used to author and develop videogames, enabling development across platforms. Commercially licensed game engines, like the Unity engine, are used by studios of varying sizes (from small independent to large AAA studios) to develop hundreds of thousands of individual game titles. Game engines define many of the core functionalities of videogames. These often-ignored processes are essential to how players experience embodiment through virtual, on-screen bodies and their own embodied performances of play off screen. This project examines 3D videogames developed after 1993, ranging from AAA titles to independent art games, tracking the gradual shift in videogame development away from modeling object appearance and towards modeling and simulating object behavior. This shift is marked by the rise of physical simulation, a process governed and parameterized by game engines. Each chapter considers how embodiment has been reshaped by the turn to physical simulation in video games by placing a particular simulation technique in a broader social and historical context. This work conceptualizes bodies as having both representational and computational elements, addressing physical simulations that make representations of human bodies visible through bouncing light, computational regimes for how objects like bodies collide and interact, and procedural animations that control how on-screen bodies, our avatars, move when we play with them. In the first chapter I discuss the process of subsurface scattering, a form of rendering that dictates how skin (the visible surface of any object) interacts with light. Using Detroit: Become Human (Quantic Dream, 2018) as a case study, I demonstrate that this physically based approach to rendering imposes a prototypical whiteness on representations of “realistic human skin” at a programmatic level. I contrast this approach with indie developer Tale of Tales’ refusal to render the body of their protagonist in Sunset (2015) in anything other than real-time reflections. This non-physically based rendering technique challenges the logics of “realistic” skin shaders and the imperatives of software efficiency and performance. The second chapter focuses on the processes of collision detection and response, which seek to model the interactions between solid objects in videogame worlds. This chapter outlines how interpenetration came to be a prohibited relationship between two video game bodies, culminating in a discussion of the relationship between video games and queer sexual orientations in the virtual reality game Summer Lesson (Bandai Namco, 2016). The third chapter explores video game embodiment through ragdoll physics. In videogames, a ragdoll is a procedurally animated body that is limp and floppy, much like its analog namesake, but also interacts with the 3D simulated environment in which it is situated. In this chapter I theorize an inverse relationship between physical performances of mastery and visual displays of mastery. Looking to games like Half Life 2 (Valve, 2004), I trace the development of ragdoll physics from its origins in the first-person shooter genre, where its use emphasizes player control over the videogame world through spectacular performances of death, to later games, like Octodad: The Dadliest Catch (Young Horses, 2014), that use ragdoll physics as the quality of the player avatar, disorienting players with prior embodied literacy. I go on to claim that this haptic disorientation enacts a form of somatic entrainment, such that the successful player of fumblecore games become virtuosic human performers, forced to attend to the intricacies of simulated physics. On the whole, this dissertation considers how the software ontologies that inform the representation of bodies within video games become relational—between bodies within the video games and the world outside the game.