Updates
Latest Tweet
What's New?
Check out for latest innovation, a computer based training video collection
Like this Page
GPU Gems 2: Programming Techniques for High-Performance Graphics and General-Purpose Computation Review by calvinnme
A comprehensive resource - but not for the novice
Truly Matt Pharr and NVIDIA are at the top of their game, and his "GPU Gems" series is certainly the only one of its kind for graphics professionals that are familiar with GPU's and shading languages already, and want to exploit them for the maximum speed and effect achievable using today's technology. It is not a "how-to" book on shading or GPUs or even advanced graphics. There are equations and code interspersed throughout the book, with bits of wisdom that are very instructive for the initiated. However, if you are a home-grown graphics programmer who knows C, or C++, or Java, and even some assembly language, plus you are familiar with image processing and computer graphics techniques, that will probably not be enough to get you through this book. I would say that this is a five star book for high-level professional graphics programmers who work with GPU specifics and shaders daily, and I would say it is a two or three star book for everyone else.
The one section of the book that is pretty accessible to anyone with knowledge of computer architecture and computer graphics would be section 4 of the book, which is about general purpose computation on GPU's themselves. That section has a series of articles that comprise an excellent tutorial on GPU's, what they are, and what they can do. It is the best material in print I have seen on the subject.
If you want a good introduction into the concept of writing shaders plus practice with an actual shading language, try "OpenGL Shading Language, 2nd Edition" by Rost, assuming you already know OpenGL. For a great on-line resource for modeling and graphics that will get you going in the right direction of knowing what the authors of these articles know, type "Elias Hugo" into Google and check out the first address shown. There is a wealth of on-line articles, complete with math and pseudocode, available there. Then, maybe, you will be ready to sift some knowledge from this "GPU Gems" series.
I notice that the table of contents is not shown by Amazon, so I list the articles here:
PART 1:GEOMETRIC COMPLEXITY
Towards Photorealism in Virtual Botany
Terrain Rendering using GPU-Based Geometry Clipmaps
Inside Geometry Instancing
Segment Buffering
Optimizing Resource Management with Multi-Streaming
Hardware Occlusion Queries Made Useful
Adaptive Tessellation of Subdivision Surfaces with Displacement Mapping
Per-Pixel Displacement Mapping with Distance Functions
PART 2:SHADING, LIGHTING, AND SHADOWS
Deferred Shading in STALKER
Real-Time Computation of Dynamic Irradiance Environment Maps
Approximate Bidirectional Texture Functions
Tile-Based Texture Mapping
Implementing the Mental Images Phenomena Renderer on the GPU
Dynamic Ambient Occlusion and Indirect Lighting
Blueprint Rendering and "Sketchy Drawings"
Accurate Atmospheric Scattering
Efficient Soft-Edged Shadows Using Pixel Shader Branching
Using Vertex Texture Displacement for Realistic Water Rendering
Generic Refraction Simulation
PART 3:HIGH-QUALITY RENDERING
Fast Third-Order Texture Filtering
High Quality Antialiased Rasterization
Fast Prefiltered Lines
Hair Animation and Rendering in the Nalu Demo
Using Lookup Tables to Accelerate Color Transformations
GPU Image Processing in Apple's Motion
Implementing Improved Perlin Noise
Advanced High-Quality Filtering
Mipmap Level Measurement
PART 4:GENERAL PURPOSE COMPUTATION ON GPUS: A PRIMER
Streaming Architectures and Technology Trends
The GeForce 6 Series GPU Architecture
Mapping Computational Concepts to GPUs
GPU Computation Strategies and Tips
Implementing Efficient Parallel Data Structures on GPUs
Flow Control Idioms
GPU Program Optimization
Stream Reduction Operations for GPGPU Applications
PART 5:IMAGE-ORIENTED COMPUTING
Octree Textures on the GPU
High-Quality Global Illumination Rendering Using Rasterization
Global Illumination using Progressive Refinement Radiosity
Computer Vision on the GPU
Deferred Filtering: Rendering from Difficult Data Formats
Conservative Rasterization
PART 6:SIMULATION AND NUMERICAL ALGORITHMS
GPU Computing for Protein Structure Prediction
A GPU Framework for Solving Systems of Linear Equations
Options Pricing on the GPU
Improved GPU Sorting
Flow Simulation with Complex Boundaries
Medical Image Reconstruction with the FFT