You've been logged out of GDC Vault since the maximum users allowed for this account has been reached. To access Members Only content on GDC Vault, please log out of GDC Vault from the computer which last accessed this account.

Click here to find out about GDC Vault Membership options for more users.


Session Name:

Machine Learning Summit: Multi-Modal Based Frame Rate Prediction


Frame rate is a key indicator which measure the fluency of a game. Assuming that we can predict frame rate by a model, we can take measures (such as reducing the rendering quality in advance) to ensure the smooth running of a game when the frame rate is predicted to drop. This lecture will take "Conqueror's Blade", a recently released multiplayer action competitive game, as the experimental platform. Here, we will introduce how Booming Tech collects and stores relative data at the scene of game latency and how to implement a frame rate prediction model to mine those key features that affect frame rate. In this experiment, the data we collected are all general game data. In other words, you can also collect similar data in your own game and easily reproduce our model, thereby helping you iterate the strategy of game optimization.

Did you know free users get access to 30% of content from the last 2 years?

Get your team full access to the most up to date GDC content

  • Game Developers Conference 2022
  • Ruidong Feng
  • Booming Tech & Netease
  • free content
  • Machine Learning Summit
  • Programming