NVIDIA at SIGGRAPH 2004

NVIDIA will be participating in and sponsoring a number of different sessions and other events at SIGGRAPH 2004. These sessions and events are listed below for your convenience.

SPONSORED SESSIONS

The latest GPUs, like the NVIDIA Quadro FX 4000 and others from the GeForce 6 Series, support significant new features, such as reading textures in vertex programs, advanced flow control in fragment programs, floating-point filtering, floating-point blending, and geometry instancing. This lecture dissects practical examples of Shader Model 3.0 in action for use both to enrich visual complexity and improve performance.

The rapid pace of GPU performance advances comes in parallel with the increasing convergence between films, games, and surrounding media -- games are being recast as films, films create matching games, and the technology of imaging is increasingly shared between all facets of the storytelling media. This lecture covers high-performance, practical real-time applications of shading technology to create expressive and realistic shading to compete and complement the best of movie effects. Examples will be given primarily via the NVIDIA FX Composer shader authoring tool, with application for games and DCC users -- whether programmers or artists.

The complexity of modern GPUs necessitates the use of a sophisticated toolset when creating, debugging and tuning your real-time applications. Leveraging our intimate knowledge of the GPU and driver, NVIDIA has created a suite of performance analysis and optimization tools that identify performance bottlenecks and report critical performance metrics. This lecture showcases the toolset for performance optimization while tackling bottlenecks in an actual application, through a step-by-step case study using NVPerfHUD, FX Composer, and other tools.

Programmable GPUs supporting 32-bit floating point calculation are extremely well-suited for many image processing tasks. The lecture examines various methods for harnessing the power of the GPU for pixel processing. Attendees will learn about pixel buffer objects, multiple draw buffers, floating-point filtering and blending, getting data quickly to and from the GPU, high dynamic range (HDR) and GPU-based paint applications.

COURSES

Course #1 -- Real-Time Shading 
Kurt Akeley, NVIDIA Corporation (and others from universities and industry)
Sunday, Full Day, 8:30 am - 5:30 pm
West Hall B

For SIGGRAPH 2004, the course has been redesigned to address today's real-time shading capabilities and provide more practical information for practitioners. The morning sessions cover the more advanced technical aspects of creating a shading system. Afternoon sessions cover practical details of real-time shading use, including an overview of recently developed algorithms that run well on today's shading hardware, followed by presentations on the latest hardware developments from several leading hardware vendors. The course concludes with a question-and-answer session, where attendees can ask questions of any presenter or suggest topics of discussion.

Course #4 -- State of the Art in Monte Carlo Global Illumination 
Matt Pharr, NVIDIA Corporation (and others from leading universities)
Sunday, Full Day, 8:30 am - 5:30 pm
Room 515A

Realistic image synthesis is increasingly important in areas such as entertainment (movies, special effects, and games), design, architecture, and more. A common trend in all these areas is the quest for more realistic images of increasingly complex models. Monte Carlo global illumination algorithms are the only techniques that can handle this complexity. Recent advances in algorithms and computing power have made Monte Carlo algorithms very practical and a natural choice for most image-synthesis problems. The purpose of this course is to provide a thorough understanding of the principles of Monte Carlo path-tracing methods and a detailed overview of the most recently developed methods.

Course #16 -- Performance OpenGL: Platform-Independent Techniques 
Tom True, NVIDIA Corporation (and others from industry)
Monday, Half Day, 8:30 am - 12:15 pm
Room 502A

The purpose of this course is not to compare vendors' hardware implementations of the OpenGL pipeline, but to work at a higher level where a change in data format or method of sending data to OpenGL could affect performance. It draws generalizations about performance by conducting experiments on specific OpenGL implementations (which remain anonymous). For example, in almost all cases, passing signed image data into OpenGL yields an order-of-magnitude performance decrease.

Course #26 -- Real-Time Shadowing Techniques 
Mark Kilgard, NVIDIA Corporation (and others from leading universities)
Tuesday, Full Day, 8:30 am - 5:30 pm
Petree Hall D

Shadows heighten realism and provide important visual cues about the spatial relationships between objects. But integration of robust shadowing techniques in real-time rendering is not an easy task. In this course on how shadows are incorporated in real-time rendering, attendees learn basic shadowing techniques and more advanced techniques that exploit new features of graphics hardware. The course explains the differences among these algorithms and their strengths and weaknesses. Implementation details, often omitted in technical papers, are provided. And throughout the course, the tradeoffs between quality and performance are illustrated for the different techniques.

Course #32 -- GPGPU: General-Purpose Computation on Graphics Hardware 
Mark Harris, Ian Buck, and Tim Purcell, NVIDIA Corporation (and others from several leading universities)
Wednesday, Full Day, 8:30 am - 5:30 pm
Petree Hall D

This course provides a detailed introduction to general-purpose computation on graphics hardware (GPGPU). It emphasizes core computational building blocks, ranging from linear algebra to database queries, and reviews the tools, perils, and tricks of the trade in GPU programming. It also presents some interesting and important case studies on general-purpose applications of graphics hardware. The course presenters are experts on general-purpose GPU computation from academia and industry, and have presented papers and tutorials on the topic at the annual SIGGRAPH conference, Graphics Hardware, Game Developers Conference, and elsewhere.

NVIDIA-Sponsored Workshops

HLSL Shader Workshop: Introductory
Microsoft Corporation, sponsored by NVIDIA and AMD
Monday & Wednesday, 10:30 am and 3:45 pm
Tuesday & Thursday, 1:45 pm
Room 402A

Microsoft invites you to try your hand at writing shaders and other real-time effects! This hands-on workshop introduces real-time shader programming with the Microsoft DirectX 9.0 High-Level Shader Language (HLSL). Attendees learn how to harness the power of the latest GPU technology through a language as familiar as C while using a powerful shader IDE. Developers learn how to integrate shaders with their applications and create new effects that are only possible with HLSL.  All machines used for the labs are the latest AMD64-based computers with NVIDIA’s latest GPUs. The workshop, taught by the experts from Microsoft who actually developed HLSL, also includes tips for writing efficient shaders and strategies for debugging them. The workshop is intended for experienced graphics programmers familiar with fundamental 3D graphics techniques, including simple matrix math. All machines used for the labs are the latest AMD64-based computers with NVIDIA’s latest NV4X-based GPUs.

Register now by e-mailing HLSL@microsoft.com. Make sure to specify your preferred and alternate choice of sessions and times.

HLSL Shader Workshop: Advanced
Microsoft Corporation, sponsored by NVIDIA and AMD
Monday & Wednesday, 1:45 pm
Tuesday & Thursday, 10:30 am and 3:45 pm
Room 402A

In this hands-on Microsoft session, attendees develop a variety of next-generation advanced shader effects using the Microsoft DirectX 9.0 High-Level Shader Language (HLSL). The focus is on practical, high-quality techniques that can easily be integrated with attendees' ongoing and future projects to help set them apart from the pack. Topics include High Dynamic Range (HDR) and Pre-Computed Radiance Transfer (PRT) lighting techniques as well as implementation specifics such as usages of floating-point filtering, and floating-point blending. All machines used for the labs are the latest AMD64-based computers with NVIDIA’s latest GPUs. The workshop, taught by the experts from Microsoft who actually developed HLSL, also includes strategies for writing performant shaders, as well as an overview of potential applications in graphics and general-purpose scenarios. The advanced session is for experienced coders who are already conversant with writing shaders and want to take their knowledge to a higher level. All machines used for the labs are the latest AMD64-based computers with NVIDIA’s latest NV4X-based GPUs.

Register now by e-mailing HLSL@microsoft.com. Make sure to specify your preferred and alternate choice of sessions and times.

NVIDIA-SPONSORED EVENTS

ON THE SHOW FLOOR

NVIDIA technology will be shown in the following booths on the show floor:

• Adobe (#1361)
• Alias (#1312)
• Discreet (#1700)
• EON Reality (#1155)
• Kaydara (#1300)
• HP (#919)
• Newtek (#1147)
• Softimage (#1547)
• and many more...