SIGGRAPH 2011, the world's premier conference on computer graphics and interactive techniques, offered a unique opportunity for NVIDIA to present to developers from around the world on several new technologies. From exploring the impact of GPUs on state-of-the-art CG and interactive design, advanced ray tracing, rendering. and tessellation. Below are the key presentations from NVIDIA engineers and partners.

Real-time Ptex (Per-Face Texture Mapping)

John McDonald, NVIDIA

Brent Burley, Walt Disney Animation Studios

Originally invented for offline rendering, Ptex (http://ptex.us/) is a texturing technique invented for film rendering that makes several signficant improvements over traditional texture mapping. This technique for rendering per-face texture-mapping datasets in real-time on commodity Direct3D 11 hardware eliminates the texture-unwrap step of the art creation process while producing an image that avoids texture discontinuities frequently found in models.

Stereoscopy from A to Z

This comprehensive course on stereoscopy for graphics computer is intended for technical artists and graphics developers working on animations, cartoons, games or real-time projects looking to acquire a solid understanding on the fundamentals and advanced techniques of stereo, and the art of making it look elegant and comfortable for the audience.

Real-Time Eulerian Water Simulation Using a Restricted Tall Cell Grid

Nuttapong Chentanez, NVIDIA PhysX Research

Matthias Müller, NVIDIA PhysX Research

This paper proposes a new real-time Eulerian fluid simulation using a hybrid grid of regular cubic cells on top of a layer of tall cells above an arbitrary terrain. Novel multigrid method and several fluid solver modifications are proposed. Real-time frame rate of 30fps on real-world scenarios is achieved on a modern GPU.

Solid Simulation with Oriented Particles

Matthias Müller, NVIDIA PhysX Research

Nuttapong Chentanez, NVIDIA PhysX Research

This paper proposes to use oriented particles instead of traditional tetrahedral meshes to simulate rigid, plastic and soft bodies as well as one, two and three dimensional structures such as ropes, cloth and volumetric objects. Using oriented instead of simple particles has several advantages: first, particles can be represented by ellipsoids which approximate surfaces more accurately than spheres. Second, the simulation is stable even in sparse regions of the mesh. Third, the full transformation stored in the particles can be used for robust skinning of graphical meshes.

Generating Displacement From Normal Map for Use in 3D Games

Kirill Dmitriev, NVIDIA

Evgeny Makarov, NVIDIA

This algorithm can automatically convert normal map to displacement map to reduce the cost of creating tessellated 3D game assets.

Advanced Rendering Solutions

Phil Miller, Director of Software Product Management, Professional Solutions Group, NVIDIA

The full range of advanced rendering solutions and frameworks from NVIDIA will be jointly explained in this insightful product and technology discussion and demonstration. Come learn about the latest possibilities for incorporating advanced rendering - from leveraging mental ray, adding iray, creating with OptiX, combining multiple solutions, or building custom pipelines.

OpenGL and CUDA®-Based Tessellation

Philippe Rollin, Applied Engineer, NVIDIA

Brent Oster, Senior Applied Engineer, NVIDIA

Miguel Ortega, Modeling & Texture Lead, Digital Domain

Tessellation in OpenGL has some limitations that can be overcome by writing your own tessellation in CUDA--allowing for higher-order surfaces, custom UV coordinate determination, and advanced features such as trimming curves. This part of the session will cover how to set up tessellation of simple cubic Bezier patches in CUDA, including how to generate UV coordinates and compute tessellated vertices and triangles in parallel. It will also explore advanced tessellation using other basis functions, higher-order curves, and trimming curves.

Parallel Nsight™ 2.0 and CUDA 4.0 for the Win!

Jeff Kiel, Manager of Graphics Tools, NVIDIA

The CUDA parallel computing architecture enables dramatic increases in computing and graphics performance by harnessing the power of the GPU. Parallel Nsight is the first development environment for CUDA and graphics application development integrated into Microsoft Visual Studio.

Together, CUDA 4.0 and Parallel Nsight 2.0 make it easier than ever for developers to upgrade existing CPU-only applications to massively parallel applications that make best use of both the CPU and the GPU. Session attendees will learn:

  • How CUDA 4.0 makes it easier than ever to improve application performance.
  • How to utilize the features of Parallel Nsight 2.0 that allow you to easily integrate compute and graphics.
  • How to visualize an application's API calls and workloads across the CPU and GPU.
  • How to inspect textures and GPU pipeline state at any time during your application's run.

Parallel Nsight supports many common GPU Computing/graphics APIs, including CUDA C/C++, OpenCL, DirectCompute, Direct3D, and OpenGL.

Using the GPU to Create a Seamless Display from Multiple Projectors

Steve Nash, Applied Engineer, NVIDIA

Rajeev Surati, Co-Founder, Scalable Display Technology

This talk will explore a new method of using the GPU to adjust the pixel intensity and screen coordinates to create a visually seamless desktop from multiple projectors.

Tools for Mobile Photography and Vision

Joe Stam, Manager, Mobile Computer Vision

Kari Pulli, Senior Director of Research, NVIDIA

NVIDIA views computational photography and computer vision as important driving areas for the future of mobile, desktop, and server computing. This talk gives an overview of current research and development in these areas, as well as how this relates to our current hardware and software roadmaps.

GPU Ray Tracing

Phil Miller, Director of Software Product Management, Professional Solutions Group, NVIDIA

Learn the latest approaches in levering GPUs for the fastest possible ray tracing results from experts developing and leveraging the NVIDIA OptiX ray tracing engine, the team behind NVIDIA iray, and those making custom renderers. Multiple rendering techniques, GPU programming languages, and optimal hardware configurations will be covered in this cutting-edge discussion.