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

DEM Solutions

  EDEM® by DEM Solutions®
EDEM® is a particle flow simulation technology from DEM Solutions®. This tool predicts the performance of bulk particle handling and processing equipment, providing engineers with key design and process information. DEM Solutions CEO, Dr. John Favier, says “In recent tests, we saw that the new OpenCL™ implementation of our EDEM simulation engine employing FirePro™ professional graphics runs up to 25 times faster than on a single CPU.” Read more about EDEM, or explore other apps using heterogeneous computing technologies.

AMD Developer Summit

  AMD Developer Summit 2013
The AMD Fusion Developer Summit (AFDS) has a new name: AMD Developer Summit 2013 or APU13 for short. APU13 will be in San Jose, CA on November 11-14. Registration is expected to open in May. If you missed last year’s summit, you should watch this highlight video. If you are interested in presenting at APU13, you can find more information on how to present on the CFP page or review more event information on the APU13 site.

Bolt Logo

  Bolt Beta Released to Open Source
Bolt Beta v1.0 is now available as open source! Bolt provides an STL compatible library of high level constructs for creating accelerated data parallel applications. For this release, we now support C++ AMP in addition to OpenCL™ as underlying supported compute technologies. Learn more about Bolt Beta 1.0

AMD at GDC image

  AMD @ GDC 2013
At this year’s Game Developers Conference (GDC), AMD demoed game development & digital content creation tools as well as the latest Gaming Evolved™ games in glorious Eyefinity resolutions. The Tomb Raider demo with TressFX hair drew in the largest crowds. AMD also announced the Sky series of Radeon products for cloud gaming. Battlefield™ 4 was unveiled, & this amazing demo of the DICE Frostbite™ 3 engine was powered by Radeon 7990 hardware. Don’t miss AMD’s GDC 2013 technical presentations.


  APPML v1.10 Now Available
Accelerated Parallel Processing Math Libraries (APPML) v1.10 are now available! APPML are software libraries containing FFT and BLAS functions written in OpenCL™ and designed to run on AMD GPUs. This new version adds the complete set of L1 routines for BLAS. Learn more about APPML v1.10.

Documentation Library

  Documentation Library
We’re putting together a library of the best articles and research on topics such as heterogeneous computing, OpenCL™ and performance & optimization. Additional resources will be added to the library throughout the year. Email us with a link to your favorite article on these or other relevant topics. We’ll review your suggestion, and if it’s a good fit, add it to the library.


  From the Twitterverse…
@ExtremeTech tweets “Why is AMD so much better than NVIDIA at Bitcoin mining? http://www.extremetech.com/computing/153467-amd-destroys-nvidia-bitcoin-mining


  clMAGMA v0.1 Beta Now Available
If you’re looking for an OpenCL™ accelerated LAPACK inspired library, then you are in luck. A beta of clMAGMA is now available. This release provides OpenCL™ implementations for MAGMA's one-sided dense matrix factorizations (LU, QR, and Cholesky)—extending MAGMA's support to include AMD GPUs. In support of the clMAGMA beta release, AMD is making available a beta release of our linux BLAS library v1.7.257, which you can read about and download here.
Education and Events
 Training  Date  Location
OpenCL Programming Course by AccelerEyes July 10-11 2013 San Jose, CA
OpenCL Programming Course by AccelerEyes September 4-5 2013 Boston, MA
OpenCL Programming Course by AccelerEyes September 25-26 2013 Washington D.C.
OpenCL Programming Course by Acceleware October 22-25 2013 Houston, TX
OpenCL Software Development Downloads

OpenCL Training Online

Recommended Reading

University hcResources

Regional hcResources

Send comments and suggestions for the newsletter to hcNewsFlash@amd.com and follow us on Twitter @hcNewsFlash

Tricks and Tips

realhet writes on the AMD Developer Forums: “In OpenCL…there is no such thing as goto/call/return. There are only loops because VLIW hardware doesn't supports any kind of goto. It supports only if blocks and loop blocks….However on the new GCN there are several ways to alter the instruction pointer: S_BRANCH, S_CBRANCH_xxx – conditional/unconditional branches [and] S_SETPC, S_SWAPPC, S_GETPC : directly manipulate IP. And some more, you can check in the Southern Islands ISA manual.”

Learn more on the AMD Developer Forums.

Submit your Tips&Tricks to hcNewsFlash@amd.com

AMD is Hiring!

Featured Job:

Submit Heterogeneous Compute job listings to hcNewsFlash@amd.com


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