hcNewsFlash - Learn about the latest hc news, tools, events, training opportunities, and job opportunities in the heterogeneous computing arena.
Accelerated Apps

Marc screenshot

 

Advanced Nonlinear Simulation Up to 9x Faster using OpenCL™
Marc – is a powerful, general-purpose, nonlinear finite element analysis solution which can be used to accurately simulate the response of your products under static, dynamic and multi-physics loading scenarios.  Marc has been optimized for parallel execution using OpenCL™, after which, a 7.47M DOF trimmed car body frequency response analysis was up to nine times (9x) faster. Explore other applications that are leveraging AMD’s heterogeneous technologies here.

Update

APU13 Collage

 

AMD Developer Summit (APU13) Registration now Open!
APU13 is three intensive days that put you at the epicenter of the latest developments in heterogeneous computing, OpenCL™, C++ AMP and related technologies. Join us in sunny San Jose, California Nov. 11-14 for exclusive demonstrations, OpenCL™ and C++ AMP tips and much more, including prestigious keynote speakers. Register today.

HSA Foundation

 

New heterogeneous compute event during Siggraph 2013!
The HSA Foundation will be hosting a Birds of a Feather event July 24 from 1-2 p.m. at the Anaheim Convention Center in Room 202B.
Click here for more information.

OpenCL Logo

 

New Release – APP SDK v2.8.1 for OpenCL™ Development
AMD Accelerated Parallel Processing (APP) SDK v2.8.1 is now available. The SDK also includes samples for AMD’s C++ template library, codename Bolt, as well as samples for the latest OpenCL™ accelerated OpenCV (Open Computer Vision) library. Learn more.

Bolt Logo

 

BOLT v1.0 Release
Bolt v1.0 source code has been released on GitHub.  Bolt provides an STL compatible library of high level constructs for creating accelerated data parallel applications. Release v1.0 includes numerous features, including several performance optimizations, TBB multi-threaded code paths for several routines, serial code paths for all routines and more.  Learn more about Bolt.

CodeXL Logo

 

New Release – AMD CodeXL v1.2
A new version 1.2 of CodeXL is available for download. AMD CodeXL is a comprehensive tool suite that enables developers to harness the benefits of AMD CPUs, GPUs and APUs. It includes powerful GPU debugging, comprehensive GPU and CPU profiling and static OpenCL™ kernel analysis capabilities. Read about feature improvements in the blog by Marty Johnson, director of Product Management or download the v1.2 release here.

OpenCL Logo

 

Computer Vision with OpenCL™ Acceleration
Exciting developments are unfolding in the world of computer vision this year! With the latest releases of OpenCV (the world’s most popular computer vision library), your applications can now tap the full power of modern heterogeneous computing. Read more in this blog post by Dr. Harris Gasparakis, the OpenCV-CL project manager, technical lead and evangelist. More...

OpenJDK Logo

 

HSAIL-based GPU offload: the Quest for increasing Java Performance Begins
Sumatra is the first open source project that aims to integrate GPU offload capabilities directly into the JVM. This blog provides insights into how Java programmers can leverage HSA to increase application performance.  More...

Education and Events
 Training  Date  Location
OpenCL™ Programming Course by AccelerEyes September 4-5, 2013 Boston, MA
C++ AMP Programming Course by Acceleware September 10-13, 2013 Boston, MA
OpenCL™ Programming Course by AccelerEyes October 9-10, 2013 Houston, TX
OpenCL™ Programming Course by Acceleware October 22-25, 2013 Houston, TX

 
Tricks and Tips

Rahulgarg writes on the AMD Developer Forums: “I finally have a Richland (A10-5750) based system and can clarify a few things for anyone looking for this info. . . The GPU supports cl_amd_fp64 but not cl_khr_fp64 . . . Thus, for Radeon 8650G (which as 384 cores) for example, the throughput for various fp64 instructions is: 1. ADD: 384/8 = 48 adds/cycle;  .  . . ”

Read the full post here and visit the AMD Developer Forums for other useful tips and tricks.

Submit your Tips&Tricks to hcNewsFlash@amd.com

Jobs
AMD is Hiring!

Featured Jobs:

Submit Heterogeneous Compute job listings to hcNewsFlash@amd.com

Networks

Connect with the heterogeneous community and be a part of this new era of computing!

     LinkedIn FaceBook  Twitter YouTube
©2013 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, Radeon and combinations thereof are trademarks of Advanced Micro Devices, Inc.

Advanced Micro Devices, Inc., One AMD Place, P.O. Box 3453, Sunnyvale, California 94088-3453 USA
AMD Developer Central | Copyright | Privacy | Cookie Policy | Trademark Info