Programming with OpenGL: Advanced Rendering

Organizer:
Tom McReynolds
Silicon Graphics

Copyright ©1997 by Tom McReynolds and David Blythe.
All rights reserved

SIGGRAPH `97 Course

Abstract:

This course moves beyond the straightforward images generated by the novice, demonstrating the more sophisticated and novel techniques possible using the OpenGL library.

By explaining the concepts and demonstrating the techniques required to generate images of greater realism and utility, the course helps students achieve two goals: they gain a deeper insight into OpenGL functionality and computer graphics concepts, while expanding their ``toolbox'' of useful OpenGL techniques.

Speakers

David Blythe David Blythe is a Principal Engineer with the Advanced Systems Division at Silicon Graphics. David joined SGI in 1991 and has contributed to the development of RealityEngine and InfiniteReality graphics. He has contributed extensively to implementations of the OpenGL graphics library and OpenGL extension specifications.

Prior to joining SGI, David was a visualization scientist at the Ontario Centre for Large Scale Computation. David received both a B.S. and M.S. degree in computer science from the University of Toronto.

Email: blythe@asd.sgi.com

Celeste Fowler Celeste Fowler is a software engineer in the Advanced Systems Division at Silicon Graphics. She worked on the OpenGL imaging pipeline for the InfiniteReality graphics system and on the OpenGL display list implementation for InfiniteReality and RealityEngine.

Before coming to SGI, Celeste attended Princeton University where she did research on radiosity techniques and TA'd courses in computer graphics and programming systems.

Email: celeste@asd.sgi.com

Brad Grantham Brad Grantham currently contributes to the design and implementation of Silicon Graphics' high-level graphics toolkits, including OpenGL++, a scene graph toolkit for OpenGL. Brad previously worked on the Windows 95 port and Java bindings for Cosmo 3D and, before that, worked in the IRIS Performer group.

Before joining SGI, Brad wrote UNIX kernel code and imaging codecs. He received a B.S. from Virginia Tech in 1992, and his previous claim to fame is MacBSD, BSD UNIX for the Macintosh.

Email: grantham@sgi.com

Simon Hui Simon Hui is a software engineer at 3Dfx Interactive, Inc. He currently works on OpenGL and other graphics libraries for PC and consumer platforms.

Prior to joining 3Dfx, Simon worked on IRIS Performer, a realtime graphics toolkit, in the Advanced Systems Division at Silicon Graphics. He has also worked on OpenGL implementations for the RealityEngine and InfiniteReality. Simon received a B.A. in Computer Science from the University of California at Berkeley.

Email: simon@3dfx.com

Tom McReynolds

Tom McReynolds is a software engineer in the Core Rendering group at Silicon Graphics. He's implemented OpenGL extensions and done OpenGL performance work. He currently works on IRIS Performer, a real-time visualization library that uses OpenGL.

Prior to SGI, he worked at Sun Microsystems, where he developed graphics hardware support software and graphics libraries, including XGL.

Tom is also an adjunct professor at Santa Clara University, where he teaches courses in computer graphics using the OpenGL library. He has also presented at the X Technical Conference, SIGGRAPH '96, and SGI's 1996 Developer Forum.

Email: tomcat@asd.sgi.com

Paula Womack Paula Womack manages the OpenGL group at Silicon Graphics. She is also a member of the OpenGL Architectural Review Board (the OpenGL ARB) which is responsible for defining and enhancing OpenGL.

Prior to joining Silicon Graphics, Paula worked on OpenGL at Kubota and Digital Equipment. She has a B.S. in Computer Engineering from the University of California at San Diego.

Email: womack@asd.sgi.com

 
• Contents
• List of Figures
• 12.3.1 Introduction
  • 1.1 Acknowledgments
  • 1.2 Course Notes Web Site
• 2 About OpenGL
• 3 Modelling
  • 3.1 Modelling Considerations
  • 3.2 Decomposition and Tessellation
  • 3.3 Capping Clipped Solids with the Stencil Buffer
  • 3.4 Constructive Solid Geometry with the Stencil Buffer
• 4 Geometry and Transformations
  • 4.1 Stereo Viewing
    • 4.1.1 Fusion Distance
    • 4.1.2 Computing the Transforms
    • 4.1.3 Rotate vs. Shear
  • 9.2 Depth of Field
  • 4.3 The Z Coordinate and Perspective Projection
    • 4.3.1 Depth Buffering
  • 4.4 Image Tiling
  • 4.5 Moving the Current Raster Position
• 5 Texture Mapping
  • 5.1 Review
    • 5.1.1 Filtering
    • 5.1.2 Texture Environment
  • 5.2 MIPmap Generation
  • 5.3 View Dependent Filtering
  • 5.4 Fine Tuning
  • 5.5 Paging Textures
  • 5.6 Transparency Mapping and Trimming with Alpha
  • 5.7 Billboards
  • 5.8 Rendering Text
  • 5.9 Projective Textures
  • 5.10 Environment Mapping
  • 5.11 Image Warping and Dewarping
  • 5.12 3D Textures
    • 5.12.1 Using 3D Textures
      • 3D Textures vs. MIPmaps
    • 5.12.2 3D Texture Portability
    • 5.12.3 3D Textures to Render Solid Materials
    • 5.12.4 3D Textures as Multidimensional Functions
  • 5.13 Procedural Texture Generation
    • 5.13.1 Filtered Noise Functions
    • 5.13.2 Generating Noise Functions
    • 5.13.3 High Resolution Filtering
    • 5.13.4 Spectral Synthesis
    • 5.13.5 Other Noise Functions
    • 5.13.6 Turbulence
    • 5.13.7 Example: Image Warping
    • 5.13.8 Generating 3D Noise
    • 5.13.9 Generating 2D Noise to Simulate 3D Noise
    • Trade-offs Between 3D and 2D Techniques
• 8.4 Blending
  • 6.1 Compositing
  • 6.2 Advanced Blending
  • 6.3 Painting
  • 6.4 Blending with the Accumulation Buffer
• 7 Antialiasing
  • 7.1 Antialiasing Points and Lines
  • 7.2 Polygon Antialiasing
  • 7.3 Multisampling
  • 7.4 Antialiasing With Textures
  • 7.5 Antialiasing with Accumulation Buffer
• 8 Lighting
  • 8.1 Phong Shading
    • 8.1.1 Phong Highlights with Texture
    • 8.1.2 Spotlight Effects using Projective Textures
    • 8.1.3 Phong shading by Adaptive Tessellation
  • 8.2 Light Maps
    • 8.2.1 2D Texture Light Maps
    • 8.2.2 3D Texture Light Maps
  • 8.3 Bump Mapping with Textures
    • 8.3.1 Tangent Space
    • 8.3.2 Going for higher quality
  • Blending
    • 8.4.1 Why does this work?
    • 8.4.2 Limitations
  • 8.5 Choosing Material Properties
    • 8.5.1 Modeling Material Type
      • Dielectrics
      • Metals
      • Composite Materials
      • Other Materials
    • 8.5.2 Modeling Material Smoothness
• 9 Scene Realism
  • 9.1 Motion Blur
  • Depth of Field
  • 9.3 Reflections and Refractions
    • 9.3.1 Planar Reflectors
      • Planar Reflections and Refractions using the Stencil Buffer
      • Planar Reflections using Texture Mapping
      • Interreflections
    • 9.3.2 Sphere Mapping
      • Using a Sphere Map
      • Generating a Sphere Map for Specular Reflection
      • Multipass Techniques and Interreflections
      • Other Sphere Mapping Techniques
      • Limitations of Sphere Mapping
  • 9.4 Creating Shadows
    • 9.4.1 Projection Shadows
      • Projection Shadow Trade-offs
    • 9.4.2 Shadow Volumes
      • Multiple Light Sources
      • Shadow Volume Trade-offs
    • 9.4.3 Shadow Maps
      • Shadow Map Trade-offs
    • 9.4.4 Soft Shadows by Jittering Lights
    • 9.4.5 Soft Shadows Using Textures
• 10 Transparency
  • Screen-Door Transparency
  • 10.2 Alpha Blending
  • 10.3 Sorting
  • 10.4 Using the Alpha Function
  • 10.5 Using Multisampling
• 11 Natural Phenomena
  • 11.1 Smoke
  • 11.2 Vapor Trails
  • 11.3 Fire
  • 11.4 Clouds
  • 11.5 Water
  • 11.6 Light Points
  • 11.7 Other Atmospheric Effects
• 12 Image Processing
  • Introduction
    • 12.1.1 The Pixel Transfer Pipeline
    • 12.1.2 Geometric Drawing and Texturing
    • The Frame Buffer and Per-Fragment Operations
  • 12.2 Colors and Color Spaces
    • 12.2.1 The Accumulation Buffer: Interpolation and Extrapolation
    • 12.2.2 Pixel Scale and Bias Operations
    • Look-Up Tables
      • Pixel Maps
      • The Color Table Extension
      • The Texture Color Table Extension
      • The Pixel Texture Extension
    • 12.2.4 The Color Matrix Extension
      • Equivalent Functionality Without SGI_color_matrix
      • Scale and Bias
      • Conversion to Luminance
      • Modifying Saturation
      • Hue Rotation
      • CMY Conversion
      • YIQ Conversion
  • 12.3 Convolutions
    • Introduction
    • 12.3.2 The Convolution Operation
      • Two-Dimensional Convolutions
      • Separable Filters
    • 12.3.3 Convolutions Using the Accumulation Buffer
    • 12.3.4 The Convolution Extension
    • 12.3.5 Useful Convolution Filters
      • Line detection
      • Gradient Detection (Embossing)
      • Smoothing and Blurring
      • High-pass Filters
      • Laplacian Filter
      • Sobel Filter
  • 12.4 Image Warping
    • 12.4.1 The Pixel Zoom Operation
    • 12.4.2 Warps Using Texture Mapping
      • Line Integral Convolution
• 13 Volume Visualization with Texture
  • 13.1 Overview of the Technique
  • 13.2 3D Texture Volume Rendering
  • 13.3 2D Texture Volume Rendering
  • 13.4 Blending Operators
    • 13.4.1 Over
    • 13.4.2 Attenuate
    • 13.4.3 MIP
    • 13.4.4 Under
  • 13.5 Sampling Frequency
  • 13.6 Shrinking the Volume Image
  • 13.7 Virtualizing Texture Memory
  • 13.8 Mixing Volumetric and Geometric Objects
  • 13.9 Transfer Functions
  • 13.10 Volume Cutting Planes
  • 13.11 Shading the Volume
  • 13.12 Warped Volumes
• 14 Using the Stencil Buffer
  • 14.1 Dissolves with Stencil
  • 14.2 Decaling with Stencil
  • 14.3 Finding Depth Complexity with the Stencil Buffer
  • 14.4 Compositing Images with Depth
• 15 Line Rendering Techniques
  • 15.1 Hidden Lines
  • 15.2 Haloed Lines
  • 15.3 Silhouette Edges
• 16 Tuning Your OpenGL Application
  • 16.1 What Is Pipeline Tuning?
    • Three-Stage Model of the Graphics Pipeline
    • 16.1.2 Finding Bottlenecks in Your Application
      • Application bottlenecks.
      • Geometry bottlenecks.
      • Rasterization bottlenecks.
    • 16.1.3 Factors Influencing Performance
  • 16.2 Optimizing Your Application Code
    • 16.2.1 Optimize Cache and Memory Usage
      • Minimizing Cache Misses.
    • 16.2.2 Store Data in a Format That is Efficient for Rendering
      • Minimizing State Changes.
    • Per-Platform Tuning
  • 16.3 Tuning the Geometry Subsystem
    • 16.3.1 Use Expensive Modes Efficiently
    • 16.3.2 Optimizing Transformations
    • 16.3.3 Optimizing Lighting Performance
    • Advanced Geometry-Limited Tuning Techniques
  • 16.4 Tuning the Raster Subsystem
    • 16.4.1 Using Backface/Frontface Removal
    • Minimizing Per-Pixel Calculations
      • Avoid Unnecessary Per-Fragment Operations.
      • Use Simple Fill Algorithms for Large Polygons.
      • Use the Depth Buffer Efficiently.
    • 16.4.3 Optimizing Texture Mapping
    • 16.4.4 Clearing the Color and Depth Buffers Simultaneously
  • 16.5 Rendering Geometry Efficiently
    • Using Peak-Performance Primitives
    • 16.5.2 Using Vertex Arrays
    • 16.5.3 Using Display Lists
    • 16.5.4 Balancing Polygon Size and Pixel Operations
  • 16.6 Rendering Images Efficiently
  • 16.7 Tuning Animation
    • 16.7.1 Factors Contributing to Animation Speed
    • 16.7.2 Optimizing Frame Rate Performance
  • 16.8 Taking Timing Measurements
    • 16.8.1 Benchmarking Basics
    • 16.8.2 Achieving Accurate Timing Measurements
    • 16.8.3 Achieving Accurate Benchmarking Results
• 17 List of Demo Programs
• 18 Equation Appendix
  • 18.1 Projection Matrices
    • 18.1.1 Perspective Projection
    • 18.1.2 Orthographic Projection
  • 18.2 Lighting Equations
    • 18.2.1 Attenuation Factor
    • 18.2.2 Spotlight Effect
    • 18.2.3 Ambient Term
    • 18.2.4 Diffuse Term
    • 18.2.5 Specular Term
    • 18.2.6 Putting It All Together
• 19 References
• References
• About this document ...