GPU Gems 2: Programming Techniques for High-Performance Graphics and General-Purpose Computation First Edition
Use the Amazon App to scan ISBNs and compare prices.
"Devoted" by Dean Koontz
For the first time in paperback, from Dean Koontz, the master of suspense, comes an epic thriller about a terrifying killer and the singular compassion it will take to defeat him. | Learn more
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Frequently bought together
Customers who viewed this item also viewed
From the Back Cover
""GPU Gems 2" isn't meant to simply adorn your bookshelf--it's required reading for anyone trying to keep pace with the rapid evolution of programmable graphics. If you're serious about graphics, this book will take you to the edge of what the GPU can do."
"--Remi Arnaud, Graphics Architect at Sony Computer Entertainment""The topics covered in "GPU Gems 2" are critical to the next generation of game engines."
"--Gary McTaggart, Software Engineer at Valve, Creators of "Half-Life "and" Counter-Strike
This sequel to the best-selling, first volume of "GPU Gems" details the latest programming techniques for today's graphics processing units (GPUs). As GPUs find their way into mobile phones, handheld gaming devices, and consoles, GPU expertise is even more critical in today's competitive environment. Real-time graphics programmers will discover the latest algorithms for creating advanced visual effects, strategies for managing complex scenes, and advanced image processing techniques. Readers will also learn new methods for using the substantial processing power of the GPU in other computationally intensive applications, such as scientific computing and finance. Twenty of the book's forty-eight chapters are devoted to GPGPU programming, from basic concepts to advanced techniques. Written by experts in cutting-edge GPU programming, this book offers readers practical means to harness the enormous capabilities of GPUs.
Major topics covered include: Geometric ComplexityShading, Lighting, and ShadowsHigh-Quality RenderingGeneral-Purpose Computation on GPUs: A PrimerImage-Oriented ComputingSimulation and Numerical Algorithms
Contributors are from the following corporations and universities:
1u Maddox Games
Armstrong State University
GSC Game World
Massachusetts Institute of Technology
Siemens Corporate Research
Siemens Medical Solutions
Sony Pictures Imageworks
Stony Brook University
Technische Universitat Munchen
University of California, Davis
University of North Carolina at Chapel Hill
University of Potsdam
University of Tokyo
University of Toronto
University of Utah
University of Virginia
University of Waterloo
Vienna University of Technology
VRVis Research Center
Section editors include NVIDIA engineers: Kevin Bjorke, Cem Cebenoyan, Simon Green, Mark Harris, Craig Kolb, and Matthias Wloka
The accompanying CD-ROM includes complementary examples and sample programs.
About the Author
Matt Pharr is a software engineer at NVIDIA. Matt is also the coauthor of the book Physically Based Rendering: From Theory to Implementation (Morgan Kaufmann, 2004).
Randima (Randy) Fernando is Manager of Developer Education at NVIDIA.
- Item Weight : 3.4 pounds
- Hardcover : 814 pages
- ISBN-10 : 0321335597
- ISBN-13 : 978-0321335593
- Product Dimensions : 7.62 x 1.6 x 9.5 inches
- Publisher : Addison-Wesley Professional; First Edition (March 13, 2005)
- Language: : English
- Best Sellers Rank: #1,254,148 in Books (See Top 100 in Books)
- Customer Reviews:
Top reviews from the United States
There was a problem filtering reviews right now. Please try again later.
The book is divided into six parts: geometric complexity, shading, high-quality rendering, general purpose computation on the GPU, image oriented computing, and numerical algorithms. A part has anywhere from five to twelve chapters. Each chapter is written by a different author but the format and style is consistent. The chapters have an introduction, discussion of the problem or technique, conclusion, and references. The material is presented with color illustrations and occasionally some pseudo-code or code fragments. Generally, the material is extremely current and very approachable to read.
As a sequel to its well received predecessor, the text focuses on taking advantage of the computational power and features of today's high-powered GPU boards. The first part of the book, geometric complexity, emphasizes this with chapters dedicated to batch rendering, using multi-streaming, hardware occlusion, and displacement pixel-shaders. Each chapter illustrates how operations traditionally performed on the CPU can be moved into the GPU for efficiency and greater effect.
The subsequent two parts on shading and rendering continue along the same theme: improved performance by using hardware functionality found on the GPU. Each topic considers the performance ramifications and GPU capabilities when discussing the problem domain of a rendering technique and factors it into the final solution. For example, chapter 10 considers irradiance environment maps for fast lighting - but with a twist - using the GPU to do the calculations in real-time. In doing so, the book's real value becomes apparent.
The fourth part on general purpose GPU computation is an interesting addition to the text. The chapters illustrate methods of offloading traditional CPU tasks by exploiting the inherent parallel nature of modern GPU hardware. Since the book features Nvidia hardware, the architecture and performance capabilities largely focused on their products.
In the fifth part of the book, hardware assisted image creation and analysis is considered. By using context clues from the spatial, texture, or lighting data - additional refinements can be made to a scene prior to rasterization. The topics presented in this part are further refinements of the text's main theme (using the GPU fully) and are specific solutions to uncommon problems - or approaches to rasterization. None the less as GPUs continue to evolve, the topics presented in this section will undoubtedly become more common.
Finally, the sixth part of the book provides several non-traditional graphics examples to illustrate calculating data on the GPU: solving linear equations, options pricing, and numerical simulation - just to name a few. As using the raw floating point power of modern GPU is a growing trend - these sections were quite interesting and well done.
The included CD-ROM contains examples to 28 of the 48 articles in the book. In most cases, the example material includes source code as well as pre-compiled binaries to help illustrate the topic presented in the text. In order to run the majority of the samples, Cg must be installed on the host computer. In addition, the CD-ROM provides access to Nvidia's software development kit, Cg toolkit, performance tools, and several helpful reference links to on-line sites.
GPU Gems 2 provides a cutting edge view of the capabilities found in today's video cards. The selected articles illustrate that every part of the rendering process can be enhanced in some fashion by fully using the underlying hardware. As such, this book is essential to anyone working with modern GPUs.
I have to admit, though, my interest lies largely in the last 20 chapters. In this section, authors from university and industry research teams describe "GPGPU" - general processing on GPUs. This puts the incredible computing power of the GPU to use on tasks from linear algebra and differential equations to finance, computer vision, fluid flow, and medical imaging, instead of rendering viewable pictures. GPGPU promises huge performance increases over standard CPUs, but imposes huge barriers to realizing that promise. GPUs achieve their high performance by tailoring their physical architecture to a specific class of computations. That class is large, granted, but still covers only a tiny portion of today's compute-intensive tasks - and if your computation doesn't match the GPU model, you're just out of luck. These chapters offer tips'n'tricks for overcoming the architectural barrier, for rethinking applications in terms that GPUs can handle effectively.
GPGPU has been around for only a few years, largely as isolated acts of individual cleverness. No organized body of knowledge and practice exists for explointing this computational resource, and none seems likely to exist for some years to come. Every body of knowledge goes through that stage, "button collecting" of scattered, unrelated facts, the necessary elements from which larger patterns will some day be drawn. This collection, even if hit-or-miss for any one reader's needs, does its part to collect today's techniques and to disseminate that knowledge. Maybe some day, GPGPU will be as common and systematic as C programming is today - until then, anthologies like this are what we have, and this is a good one.
Top reviews from other countries
Combined with nVidia's online content and code samples its a magnificent source of information. This kind of information is otherwise widely disseminated and hard to collate. This book is very well put together and a decent reference for the initiate in removing intensive and massively parallel computation onto the GPU. My own simulations execute 30-60 times faster than on the CPU - well worth it.
The otherwise quite steep learning curve is flattened considerably by this book. Buy this book if your considering GPGPU systems.