GPGPU Programming for Games and Science

K20720

ISBN: 978-1-4665-9535-4

9 781466 595354

90000

Computer Game Programming

GPGPU

Programming for Games

and Science

David H. Eberly

Eberly

GPGPU Programming for Games and Science demonstrates how to achieve

robustness, accuracy, speed, and/or easily maintained, reusable, and readable

source code to tackle practical problems in computer science and software

engineering. The book primarily addresses programming on a graphics processing

unit (GPU) while covering some material also relevant to programming on a central

processing unit (CPU). It discusses many concepts of general purpose GPU

(GPGPU) programming and presents several practical examples in game program-

ming and scientific programming.

The author first describes numerical issues that arise when computing with

floating-point arithmetic, including making trade-offs among robustness, accuracy,

and speed. He then shows how single instruction multiple data (SIMD) extensions

work on CPUs. The core of the book focuses on the GPU from the perspective of

Direct3D 11 (D3D11) and the High Level Shading Language (HLSL). The book goes

on to explore practical matters of programming a GPU and discusses vector and

matrix algebra, rotations and quaternions, and coordinate systems. The final

chapter gives several sample GPGPU applications on relatively advanced topics.

Features

• Presents SIMD extensions to the CPU as well as a large collection of

approximations for standard mathematics functions

• Provides examples of how to draw using a combination of vertex, geometry,

and pixel shaders and how to execute a compute shader

• Shows you how to copy data between CPU and GPU and offers guidance

on parallelizing data transfer

• Explains how to work with multiple GPUs regarding D3D11 adapter

enumeration and device creation

• Discusses IEEE floating-point support on a GPU

• Illustrates how to use the Geometric Tools Engine and the resource

types in D3D11, with the source code available on a supporting website

GPGPU

Programming for Games

and Science