View, browse and sort the ever-growing list of GDC sessions by time, pass type, track, and format. With GDC Session Scheduler, you can build your schedule in advance and access it during the show via export or with the GDC Mobile App. Sessions do fill up, so please arrive early to sessions that you would like to attend. Adding a session to your schedule does not guarantee you a seat.
In order to best utilize Session Scheduler, please use the login credentials you've set up on this page as your login credentials on our GDC Mobile App. Your login credentials are the same as your registration login. You can download the app on iTunes or Google Play.
Michael Stallone (Lead Software Engineer, Engine, Netherrealm Games)
Location: Room 25, North Hall
Date: Wednesday, March 21
Time: 2:00pm - 3:00pm
Pass Type: All Access, GDC Conference + Summits, GDC Conference - Get your pass now!
Vault Recording: Video
Audience Level: All
Network latency is an ever present hurdle to overcome when trying to create a compelling multiplayer experience in any game, but can be especially difficult for fighting games. Rollback/re-simulate networking models remove the (often infuriating) local input delay typically associated with a traditional lockstep model. This is achieved by allowing the game simulation to take place without having received input from the remote player(s). This presentation will describe the journey of implementing the rollback/re-simulate model used in 'Mortal Kombat X' and 'Injustice 2'. This lecture will describe the drastic engine changes, optimizations, and tools that are involved in rolling back and simulating up to 8 game frames in 16ms.
Attendees will have an understanding of how the Netherrealm team tackled the challenge of hugely increased simulation costs associated with the rollback networking model for 'Mortal Kombat X' and 'Injustice 2' using aggressive caching, parallelization, CPU optimizations and essential tools.
This talk would be most interesting to those who are looking to reduce the impact of network latency on the user's experience of input lag. A general familiarity with deterministic lockstep network simulation would be helpful, though not required as a brief overview will be provided.