Skip navigation links
Tools
SDKs
Libraries
Samples & Demos
Docs
Zones
Community
Support
AMD Developer Inside Track
Skip Navigation LinksHome > Docs > Videos > AMD Developer Inside Track
The AMD Developer Inside Track is a monthly video series with the sole purpose of giving software developers and inside, behind-the-scenes, look at new technologies coming out of AMD and AMD’s software partners.
Aparapi Open Sourced
Gary Frost, AMD Runtimes Team


Learn how Aparapi works and how you can contribute to the open source project. (10:05)

Share |
 Embed 
SimNow Live Demo
Juan Flores, AMD SimNow Team


AMD SimNow team member Juan Flores gives an in-depth demonstration of SimNow, a fast and configurable x86 and x86-64 functional simulator that demonstrates next generation hardware before the silicon becomes available. Juan walks developers through the environment and demonstrates a piece of the SPEC CPU benchmark using AMD's next generation instruction set architectures. (10:05)

Share |
 Embed 
Introduction to Aparapi
Gary Frost and Tom Deneau, AMD Runtimes Team


The AMD Runtimes team discusses the recently released Java-to-OpenCL™ API, Aparapi, and how it can help Java developers parallelize code without necessarily having to learn a new programming language. Aparapi is a Java API for arranging data parallel solutions to be executed on the GPU, which can convert Java to OpenCL on the fly. This video discusses the features and applications of Aparapi, as well a preview of what’s to come in the next release. (10:25)

Share |
 Embed 
“Evergreen” Graphics Architecture
Justin Hensley, Design Engineer, Office of the CTO


This video gives you an in-depth description of GPU architecture with specifics about our “Evergreen” graphics processors. A deep understanding of GPU architecture is key to fully unlocking the power of Heterogeneous computing and optimizing those graphic intensive applications to their greatest potential. (18:51)

Share |
 Embed 
“Bulldozer” Processor Topology
Elsie Wahlig, AMD Fellow


Processor Topology has changed dramatically over the years and will continue to evolve to be more complex. Elsie Wahlig, AMD Fellow in the Software Division, steps through a processor block diagram for the “Bulldozer” Module and highlights the areas that have changed over what is currently available today with the “Magny-Cours” Processor. She also talks about what kind of developer really needs to know this level of processor detail and what tools they can use to get all the topology information they need. (9:26)

Share |
 Embed 
Introduction to SimNow™
Juan Flores, Member of Technical Staff


Did you know that AMD offers a free application called SimNow that actually simulates future AMD platforms? This video features Juan Flores, MTS talking all about what SimNow is, who would could benefit from using it, and gives examples on what type of instructions, like AVX and XOP, are available in SimNow that aren’t available in current silicon. (12:43)

Related Links:

Download SimNow

Share |
 Embed 
Introduction to the AMD Core Math Library (ACML)
Chip Freitag, Senior Member of Technical Staff, Software Development Engineer


Get a first hand introduction of the AMD Core Math Library from Chip Freitag, a 17 year AMD veteran, currently working on the ACML Team. He talks about what ACML is, how it differs from ACML-GPU and who could really take advantage of it.  He also gives some insight into what the team is working on now to improve ACML and when we will have access to those new features. (7:36)

Related Links:

ACML Download
ACML Data Sheet
ACML-GPU
AMD Performance Libraries

Share |
 Embed 
CodeAnalyst Supports Visual Studio 2010
Frank Swehosky


CodeAnalyst integrates into Visual Studio 2010. Get a first hand introduction from Frank Swehosky, a 10 year AMD veteran on the CodeAnalyst Team. He talks through the integration process of Visual Studio 2010 and the analysis and possible optimizations that can be done on C/C++ and .NET managed code. Check CodeAnalyst often for regularly scheduled end of quarter refreshes.
Share |
 Embed 
Xen Summit
Tom Woller, Principal MTS Engineer


This episode features Tom Woller, Principal MTS Engineer speaking why AMD is involved with virtualization. Also, Ian Pratt, chairman of Xen along with Rodney Schmidt, AMD Software Eningeer, speak openly about what is coming up for Xen and AMD in the virtualization space. (09:40)
Share |
 Embed 
How to Optimize Image Convolution
Udeepta Bordoloi, MTS Software Engineer in the Stream Computing Group


Udeepta Bordoloi walks though several different ways to optimize an image convolution algorithm. He explains the thinking behind tweaking the workgroup size, type of memory used, and data types that result in incredible speedups. If you use convolution you will definitely not want to miss this step-by-step optimization video. (18:15)
Share |
 Embed 

Saif Ali, AMD Software Engineer, Advanced Technology Group
Christopher Jess, Studio|GPU™, Technical Director

This episode of the AMD Inside Track Video series brings the AMD GDC demos to you. Software Engineer, Saif Ali, walks through three examples of how OpenCL can be used to create realistic cloth, destruction and fluid particle simulation using the updated Pixelux and Bullet Physics offerings. And our partners from Studio|GPU™ (Mach Studio Pro is bundled with our FirePro Graphics cards) show us how to use tessellation to change the texture of a character in real time! (10:47)

Share |
 Embed 
Using CPUID

Randy VanderHeyden, MST Software Engineer, ISV Team

This video features Randy VanderHeyden, a 17 year AMD veteran, talking about the best practices on how to use, and how not to use the CPUID instruction. He demonstrates how to find the processor core count with CPUID using the code provided in the Processor and Core Enumeration Using CPUID article (Thanks for the code Tracy Carver!). Also, for those of you using Visual Studio, there is very detailed information available on MSDN about the CPUID Intrinsic. (7:22)

Share |
 Embed 

AMD x86 Open64 Compiler Suite Team Insights

Mike Vermeulen, Manager of the x86 Open64 Team
Roy Ju, Architect and Software Engineer
Michael Lai, Software Engineer

This video features AMD’s Open64 Compiler Team talking about Open64 origins, unique features of this compiler, such as multi-core scalability optimizations and other important optimization flags you should not miss such as loop nest optimizations (LNO), and interprocedural analysis (IPA) to name a few. Many thanks to the team for giving us this insight directly! (14:23)

Share |
 Embed 

AMD & Adobe: A Software Optimization Story

Justin Boggs, Sofware Engineer CPU
Thomas Fortier, Software Engineer Graphics

Justin Boggs and Thomas Fortier are AMD software engineers that work with Adobe to make sure their software is optimized for AMD. This video footage was taken earlier this year when they were part of an Adobe Users group multithreading panel. Hear what they had to say about AMD CPU and GPU optimizations with Adobe. (9:37)

Share |
 Embed 

Vision Launch Recap: A Developer’s Perspective

Check out the AMD VISION launch event from a software developer perspective. Get the details about DirectX 11 development straight from Codemasters, Rebellion, EA and Turbine. You’ll see six monitors being driven from one graphics card, get the info on the new DX11 games coming up and more importantly, learn about the new levels of graphical detail that DX11 hardware can bring to the table.

Many thanks to Gareth Thomas, Senior Programmer from Codemasters, Chris Kingsley, CTO of Rebellion, Kevin O’Leary, Product Manager of EA and Nate Jones, VP of Corporate Business Development at Turbine. (5:55)

Share |
 Embed 
Introduction to OpenCL
Michael Houston, GPG System Architect

Learn about OpenCL, what the transition to OpenCL will be like, what applications are ideal for OpenCL and what impact OpenCL will have on future software. Read this blog post  for a summary of this video and some example applications and demos (10:07).

Share |
 Embed 

CommunityOne 2009 West Panel

Experts from AMD, Allinea, Pervasive, and Rogue Wave give their advice and insights for what they recommend when beginning a multithreaded project, whether you are starting from scratch or working with existing single threaded code.  Parallel programming issues are not a one-size-fits-all.  Depending on the algorithm, the data dependencies, and the problem domain there are different approaches to achieving parallelism.  These expert interviews followed an AMD-sponsored  multithreading discussion at CommunityOne 2009 West (12:06).

Share |
 Embed