Past Issues August 2011 
AMD Developer Central - Code Faster. Faster Code.
 
 
In this issue:
» AMD APP SDK 2.5 Provides Enhanced Performance and Major New Capabilities
» New Webinar: “Advanced OpenCL™ Debugging using gDEBugger”
» Now Available: “Heterogeneous Computing with OpenCL” Book
» CPU-to-GPU data transfers exceed 15GB/s using APU zero copy path
» Now Available: OpenCL Programming Guide
» Experience the Power of More Cores with AMD at VMworld 2011
» AMD Releases Catalyst Beta Drivers for OpenGL 4.2
» AMD gDEBugger’s single-GPU debug capability impresses at SIGGRAPH
» SIGGRAPH attendees show high level of interest in OpenCL™ technology
» AMD FUSION11 7th Annual Technical Forum & Exhibition
» AMD and Acceleware Collaborate to Offer OpenCL™ Education
» AMD APP Math Libraries Absorb Functions From ACML-GPU
» Recently Added to developer.amd.com
» Upcoming Sponsored Events
 developer.amd.com
    Tools
SDKs
Libraries
Technology Zones
Documentation & Articles
Sample Code & Apps
Community
Member Benefits
Events
 
 
 
AMD APP SDK 2.5 Provides Enhanced Performance and Major New Capabilities

AMD APP SDK 2.5 Provides Enhanced Performance and Major New Capabilities
 Links
    RSS
Login
 

HotItem
The latest release of the AMD Accelerated Parallel Processing (APP) SDK adds key performance enhancements for APUs that free applications from the CPU-to-GPU bandwidth limitation imposed by the PCIe bus, achieving effective data transfer rates as high as 15GB/s. Find out what else is new in version 2.5. Also released: AMD APP Profiler v2.3, AMD APP KernelAnalyzer v1.9, and AMD APP Math Libraries v1.4.

Learn More
 
 
New Webinar: “Advanced OpenCL™ Debugging using gDEBugger” New Webinar: “Advanced OpenCL™ Debugging using gDEBugger”

Developing robust parallel computing applications is difficult. In this talk we will introduce the audience to gDEBugger, an OpenCL kernel source code debugger, integrated into Visual Studio™. We will display advanced debugging techniques that help locate hard-to-find OpenCL related bugs.  Join Yaki Tebeka, an AMD Fellow, responsible for AMD’s developer tools, for this live webinar. Yaki brings over 13 years of experience in software, focusing on 3D graphics, heterogeneous computing and developer tools.

September 6th 9:00am PST

OpenCL and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos.

Register
 
 
Now Available:  “Heterogeneous Computing with OpenCL</span>” Book Now Available: “Heterogeneous Computing with OpenCL” Book

Now shipping is the “Heterogeneous Computing with OpenCL” book, intended to support a parallel programming course. Read the blog by co-author Lee Howes to learn why the authors decided to approach this topic the way they did and what you can expect from the book.

Learn More
 
 
CPU-to-GPU data transfers exceed 15GB/s using APU zero copy path CPU-to-GPU data transfers exceed 15GB/s using APU zero copy path

At the AMD Fusion Developer Summit (AFDS), AMD introduced the AMD A-Series processors, formerly known as the “Sabine” platform and “Llano” APU. As a GPU compute device, the A8-3850 with Radeon™ HD 6550D graphics is capable of up to 480 GFLOPS of performance.  In order to fully benefit from this performance, it is essential to be able to transfer data efficiently between the host CPU and the GPU.   Mark Ireton, Product Manager for Compute Solutions at AMD, discusses the zero copy data transfer path that has been introduced to support the tighter coupling between CPU and GPU memory afforded by the APU architecture.

Learn More
 
 
Now Available:  OpenCL Programming Guide Now Available: OpenCL Programming Guide

Introducing the “OpenCL Programming Guide,” a book covering the whole of OpenCL 1.1, written by the people who developed the specification, including Aaftab Munshi, the specification editor and its original designer. OpenCL 1.1 is supported on all major platforms, including Linux®, Windows®, and OS X. This new book covers the OpenCL specification in detail, including a wide selection of detailed examples.

Learn More
 
 
Experience the Power of More Cores with AMD at VMworld 2011 Experience the Power of More Cores with AMD at VMworld 2011

AMD is a platinum sponsor of VMworld 2011 in Las Vegas, August 29th through September 1st.  Join us at Booth #311 to see demos of Live Migration, Memecached, and a sneak preview of our upcoming “Bulldozer” technology.  Read on to learn about our technical sessions and giveaways.

Learn More
 
 
AMD Releases Catalyst Beta Drivers for OpenGL 4.2 AMD Releases Catalyst Beta Drivers for OpenGL 4.2

At SIGGRAPH earlier this month, AMD released a beta driver for the OpenGL® 4.2 API specification announced by the Khronos Group. The new AMD Catalyst™ Beta package for the OpenGL 4.2 API is available for download for use with the Windows 7, Windows Vista, Windows XP and Linux operating systems.  The new specification contains features that both extend functionality, and enable increased application performance – while maintaining full backwards compatibility.

Learn More
 
 
AMD gDEBugger’s single-GPU debug capability impresses at SIGGRAPH AMD gDEBugger’s single-GPU debug capability impresses at SIGGRAPH

At SIGGRAPH this year AMD showcased the latest generation of our OpenCL tools.  On display were the AMD APP Profiler, a comprehensive profiler for OpenCL based applications, and AMD gDEBugger, a sophisticated tool that enables detailed source level debugging of OpenCL kernels in addition to API level debugging for OpenCL and OpenGL applications.  Mark Ireton, Product Manager for Compute Solutions at AMD, recounts what impressed developers most in the demo.

Learn More
 
 
SIGGRAPH attendees show high level of interest in OpenCL™ technology SIGGRAPH attendees show high level of interest in OpenCL™ technology

Mark Ireton reports on the OpenCL momentum he experienced at SIGGRAPH.
AMD isn’t the only company on board – other commercial entities as well as academics are seeing the value in OpenCL.

Learn More
 
 
AMD FUSION<sup>11</sup> 7<sup>th</sup>  Annual Technical Forum & Exhibition AMD FUSION11 7th Annual Technical Forum & Exhibition

AMD FUSION11 Technical Forum & Exhibition is the premier annual Taiwan event that brings academia and industry together with AMD partners to drive platform innovation that achieves greater energy efficiency, higher performance, and new capabilities in tablets, embedded systems, servers, and software experiences.  Join AMD in Taipei, Taiwan on October 5, 2011 for shared insights and demos that span manufacturing and packaging, electrical, embedded, USB, thermal, hardware and software.

Learn More
 
 
AMD and Acceleware Collaborate to Offer OpenCL™ Education AMD and Acceleware Collaborate to Offer OpenCL™ Education

AMD is partnering with Acceleware to offer training courses on techniques for parallel programming in OpenCL specifically tailored for AMD GPUs and APUs. These courses include classroom lectures, hands-on exercises, and a certificate of completion. Find out if training is happening near you, or request training to come to your location.

Next class: Houston, Texas on September 13-14

Learn More
 
 
AMD APP Math Libraries Absorb Functions From ACML-GPU AMD APP Math Libraries Absorb Functions From ACML-GPU

Now that the latest version of the AMD Accelerated Parallel Processing Math Libraries (APPML) is released, availability and support for  ACML-GPU will be discontinued  on October 1, 2011.  APPML contains FFT and BLAS functions written in OpenCL and designed to run on AMD GPUs. The libraries also support running on CPU devices to facilitate debugging and multicore programming.

Take the Survey
 
 
Code Faster Recently Added to developer.amd.com

 AMD Fusion Developer Summit session videos
 SCCM DASH 1.1 Plug-in
 AMD Quad-Buffer SDK



 
 
Upcoming Sponsored Events Upcoming Sponsored Events
  • OpenCL Training with Acceleware
    September 13-14:  Houston, Texas
  • AMD Fusion11 Technical Forum & Exhibition
    October 5:  Taipei, Taiwan
  • Supercomputing
    November 12-18:  Seattle, Washington
 See all Events