CUDA Workshop

All CUDA Workshop presentations are available for download here.

Full Conference        Full Conference One-Day        Courses        Exhibits Only

Thursday, 16, December, 9:00 am-6:00 pm
Room 317B/C, Coex Convention & Exhibition Center
Presented in English, translated simultaneously to Korean / 영어로 발표 됨 (한국어 동시 통역)

Languages, Tools, and APIs for GPU Computing
9:00-9:50 am
This talk covers the key features and differences between the major programming languages, APIs, and development tools available today. It also explains several high-level design patterns for consumer, professional, and HPC applications, with practical programming considerations for each.

Scott Fitzpatrick
NVIDIA Corporation

Introduction to Programming for CUDA C
10:00-10:50 am
This presentation teaches the basics of programming GPUS . For anyone with a background in C or C++, it explains everything necessary to start programming in CUDA C. Explore the various APIs available to CUDA applications and learn the best (and worst) ways to employ them in applications. The concepts are illustrated with step-by-step walkthroughs of code samples.

Samuel Gateau
NVIDIA Corporation

Computational Photography, Real-Time Plenoptic Rendering
11:00-11:50 am
Learn how to change the focus of an image after it has been captured. Get the latest information on GPU-based plenoptic rendering, including a demonstration of refocusing, novel view generation, polarization, high-dynamic-range imaging, and stereo 3D. Learn how GPU hardware enables interactive plenoptic rendering with high-resolution imagery, and how this process opens up entirely new possibilities for modern photography.

Todor Georgiev
Adobe Systems Incorporated

Andrew Lumsdaine
Indiana University

Using 3D Stereoscopic Solutions in Broadcast and Film Production
12:00-12:50 pm
Film, television, and the entire digital-content industry are undergoing a stereoscopic 3D revolution. 3D stereoscopic techniques provide an entirely new way to tell a story to your audience. Learn how software tools and hardware work together to make stereoscopic 3D production and distribution possible.

Andrew Page
NVIDIA Corporation

Developer Tools for the GPU: Parallel Nsight & AXE
1:00-1:50 pm
Learn about NVIDIA's latest software-development tools. The first half of this session explains how NVIDIA's range of professional software makes it easy for application developers to maximize the power of professional GPUs to deliver ultimate experiences and capabilities within their products and for their customers. Many of these offerings benefit end users directly, as they push the boundaries of what's possible in graphics and visual computing

The second half of the session covers NVIDIA’s Parallel Nsight development environment for Windows. Parallel Nsight, fully integrated into Visual Studio 2010, introduces native GPU debugging and platform-wide performance analysis tools for both computing and graphics developers.

Phillip Miller
Samuel Gateau
NVIDIA Corporation

DX11 Tessellation
2:00-2:50 pm
This talk provides a comprehensive overview of Direct3D11 tessellation technology. It not only demonstrates how to leverage Direct3D11 tessellation to take your games to the next level, but also addresses crucial practical issues. It also presents a few breathtaking practical examples and demonstrates how these advanced techniques have been adopted into gaming source engines.

Tianyun Ni
NVIDIA Corporation

Using the GPU for Visual Effects for Film & Video
3:00-4:30 pm
Learn from NVIDIA partners who are working with the GPU to improve visual effects for film and video.

"Avatar" Case Study: Accelerating Out-Of-Core Ray Tracing of Sparsely Sampled Occlusion with Pantaray
3:00-3:30 pm
Modern VFX rendering pipelines confront major complexity challenges: a film like "Avatar" requires rendering hundreds of thousands of frames, each containing hundreds of millions or billions of polygons. Furthermore, the process of lighting requires many rendering iterations across all shots. This talk presents the architecture of an efficient out-of-core ray-tracing system designed to make rendering precomputations of gigantic assets practical on GPUs. The system, dubbed PantaRay, leverages development of modern ray-tracing algorithms for massively parallel GPU architectures and combines them with new out-of-core streaming and level-of-detail rendering techniques.

Luca Fascione
Sebastian Sylwan
Weta Digital

Additional speakers and topics coming soon.

CUDA and Fermi Optimization Techniques
4:45-6:00 pm
This talk presents a detailed technical overview of implementing CUDA on Fermi for various algorithms: Monte Carlo simulation, finite-difference method, and some useful image processing algorithms. Learn how to implement these algorithms on Fermi-based CUDA architecture and how to use new Fermi features and optimize to achieve optimum performance on these applications.

Hyungon Ryu
NVIDIA Corporation

Presentations will be available for download on the SIGGRAPH Asia 2010 and NVIDIA web sites soon after the workshop.