Tencent filed 5 patents across 4 categories: UI/UX (1), Networking (1), Game Mechanics (2), and AI & Machine Learning (1).
The Game Mechanics filings describe a release-to-cast capture control for pet RPGs and a system that carries species interactions from the open world into turn-based combat. The UI/UX patent shows hero attribute compatibility during MOBA draft phases, while the Networking application addresses scalable event synchronization for cloud-based virtual worlds. The AI & Machine Learning filing covers natural language command processing for NPC teammates, using environmental context to interpret player speech.
The Game Mechanics filings both streamline pet interactions in RPG settings. One patent replaces manual aiming and physics-based throwing with a single release action that triggers a capture spell and immediately shows the result, removing the need to account for wind, gravity, or collision. The other maintains the relationship between two pets when they transition from exploration to combat, so if they showed a particular species interaction in the open world, they enter turn-based battle in a matching combat mode, and the player can fight either or both of them.
A Networking patent tackles the scalability ceiling that appears when virtual worlds span multiple regional servers. Traditional cross-region play required every server to synchronize state with every other server, creating an all-pairs communication load that eventually maxed out a single server's capacity. This filing decouples computation cycles across servers and treats inter-region events as one-directional signals that each receiving server processes during its own next cycle, turning continuous state mirroring into scheduled event delivery.
The UI/UX application focuses on MOBA draft efficiency. When a player hovers over a hero candidate, the avatar frame displays whether that hero's team-function attribute matches or conflicts with a teammate's already chosen character, offering a comparison cue at the moment of selection instead of forcing players to pick independently and then adjust their lineup over multiple rounds.
An AI & Machine Learning patent uses virtual environment perception to improve voice command accuracy for NPC teammates. Instead of relying solely on a generic speech-to-text model that analyzes words in isolation, the system feeds environmental information gathered by the player character and the NPC into the recognition step, helping disambiguate similar-sounding phrases and ensuring the recognized action aligns with what is actually present in the game world.
All data sourced from USPTO patent filings. Google Patents may take several weeks to index recent publications. If a link is unavailable, search for the patent number at USPTO Patent Public Search.