Skydance Silicon Valley

All 3 Game Engines patents center on removing manual camera control from players and automating that responsibility through various algorithmic approaches. One application describes a system that trains machine learning models on telemetry data to determine camera positioning, angles, zoom levels, and transitions for each interactive environment, effectively placing camera direction in the hands of the engine rather than the player. A second patent takes a different approach, using a real-time scoring algorithm to evaluate pre-placed viewpoints and switch between them dynamically, with a buffer mechanism included to prevent jarring rapid cuts. The third addresses a specific problem that arises when camera perspective shifts suddenly during cinematic sequences: the system detects when a cut occurs and automatically adjusts the player's control scheme relative to the new camera orientation, either immediately or after tracking how the player adapts, reducing the disorientation that typically follows an abrupt change in viewpoint.

The 2 UI/UX patents both reimagine how players move through time in narrative games, though each focuses on a distinct aspect of that navigation. One describes a timeline interface that allows scrubbing both forward and backward through any game moment, with the density of available scrubbing points adjusting automatically depending on context, offering finer control during cutscenes and limiting jumps during gameplay to predetermined narrative moments. The other focuses specifically on forward movement through unplayed content, allowing players to skip ahead on a visual timeline to sections they have not yet reached, while a machine learning component detects frustration patterns and surfaces skip suggestions, and keyframe previews are designed to avoid revealing story details before players encounter them organically.