Los Angeles Convention Center, Room 403AB
This fifth consecutive Beyond Programmable Shading SIGGRAPH course continues with the core mission of the course: discussing how GPU and CPU parallel computing are used in conjunction with the traditional 3D real-time rendering pipeline to improve image quality and performance or reduce power consumption. The course also explores future directions in real-time rendering innovation. Unlike previous years, this course assumes that the audience is already familiar with modern real-time rendering CPU and GPU hardware architecture and programming models.
The course focuses on applications of GPU and CPU compute to solve real-time illumination problems. It includes an updated discussion of game developers' requests to the rendering research community, a deep dive on power-efficient rendering algorithms, and a panel on how past research in this field has and has not affected the game and graphics hardware industry.
Introduction to Beyond Programmable Shading 2012
Five Major Challenges in Real-Time Rendering
Intersecting Lights With Pixels: Reasoning About Forward and Deferred Rendering
Dynamic Sparse-Voxel Octrees for Next-Gen Real-Time Rendering
Power-Friendly GPU Programming
Panel: From Publication to Product: How Recent Graphics Research Has (And Has Not) Shaped the Industry
Experience with a modern graphics API (OpenGL or Direct3D), including basic experience with shaders, textures, and frame buffers, and/or background in parallel-programming languages. Some experience with parallel programming on CPUs or GPUs is useful but not required.
Researchers and engineers who are interested in advanced graphics techniques that use parallel programming techniques on GPUs and multi-core architectures. Graphics and game developers who are interested in integrating these techniques into their applications.
Advanced Micro Devices, Inc.