GDC is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

View, browse and sort the ever-growing list of sessions by pass type, track, and format. With this Session Viewer, you can view GDC 2023 session details, speakers and share your favorites via social media. You will be able to build your schedule and access it during the show via export or with the Mobile App, once live. Sessions do fill up and seating is first come, first serve, so arrive early to sessions that you would like to attend.

Machine Learning Summit: Parameter Selection for ASTC Mapping Compression Using Machine Learning

Wunong Li  (Senior Programmer, Thunderfire Universe X Studio)

Pass Type: All Access Pass, Summits Pass

Topic: Programming

Format: Lecture

Vault Recording: TBD

Audience Level: All

As game hardware is constantly being updated, artists are producing mapping formats that range from 1k to 2k to 4k and even 8k. High-precision mapping makes the game world more realistic, but it also places a greater performance demand on the storage space, memory and GPU computing ability.

To meet this challenge, new mapping compression formats are being proposed, with the most recent being the ASTC format. The ASTC format is currently being used on a large number of mobile devices due to its high compression ratio and low loss of accuracy.

In this session, Wunong Li, Senior Programmer at Thunderfire Universe X Studio, will combine the algorithmic ideas of reinforcement learning with the real-life industrial scenario of mapping compression to show how this new solution can further improve the efficiency of mapping compression and balance the loss of accuracy. Moreover, he'll show everything you need to know about applying this technique to your project.

Takeaway

Attendees will learn how this new solution can further improve the efficiency of mapping compression and balance the loss of accuracy—by combining the algorithmic ideas of reinforcement learning with the real-life industrial scenario of mapping compression.

Intended Audience

This is mainly for game developers, programmers, and machine learning researchers.