» Processor and Core Enumeration Using CPUID
This article explains how applications can use OS APIs and CPUID on a system with AMD processors to discover the number of logical and physical processors, the number of cores, and the association between cores and physical processors. Sample code that can be ported across different operating systems is provided.

» Introduction to DirectX's Direct3D 10

• Use of geometry shaders to combat the small-batch performance problem
• Enable render target masks

» OpenMP: More than Just Optimizing Loops
Tips for using OpenMP pragmas to parallelize non-loop code and add barriers

» Radeon® Sample Code
Samples primarily built on the basic Direct3D® sample framework that is provided with the DirectX® 7 SDK.

» Radeon® SDK Samples & Documents
The one-step center for all your source-code needs for ATI GPUs.

» Multi-Core Event Based Profiling
Listing 1 is a typical multithreaded application that starts up one thousand threads in sequence with each thread looping ten million times and then waiting for the rest of the threads to show up. This sample application is provided to illustrate how AMD CodeAnalyst can be used to profile Java applications.

» Using the New AMD Performance Library
This article explains how APL (now known as Framewave) was used to write a program that reads a .bmp file, brightens or darkens it, and then writes it out to a new file, to give a sense of how the library is designed. The sample application is included for download.

» Multithreaded Producer-Consumer: The Easy Way
View a Win32 implementation of a producer-consumer arrangement, using a single thread that queues data blocks for the producer and a thread pool for the consumer.

» Using Threaded Queues to Get Around Slow I/O
The sample application included in this article shows how to use a circular queue in C++ to address that dreaded I/O bottleneck.

» Supersizing Java: Large Pages on the Opteron Processor, Part 2
View the LinkedListTest sample application, which runs through a large linked list three times, and then prints out the elapsed time for the operation.

» Taking Advantage of Concurrent Programming for Windows, Part 1: The Simplest Multi-core Parallelism That Could Possibly Work
» Taking Advantage of Concurrent Programming for Windows, Part 2: Multi-core Programming in .NET
Download sample programs that illustrate the benefits of parallelization using an image processing task as an example.

» Performance Optimization of 64-bit Windows Applications using Visual Studio
The Mandel project is a Visual Studio C/C++ performance demo that illustrates several key concepts in Windows programming and moves progressively through different performance optimizations.