What's Hot Now at AMD

If you want faster apps, you’ll want to use heterogeneous computing technologies. We’re here to make that happen with plenty of tools and resources. And if you’re thinking about architecture, HSA is going to rock your world.

OpenCL™ Training

Check out these upcoming OpenCL™ training opportunities. Register for a scheduled class or request a custom session on site for you and other developers on your team. Acceleware has two upcoming OpenCL™ training classes in May and June, and Applied Parallel Computing LLC offers OpenCL™ training in nearly 60 locations across Europe.

EVS '15: Understanding Adaptive Machine Learning Vision Algorithms

Don't miss this session at the Embedded Vision Summit where Harris Gasparakis, Ph.D., discusses machine learning algorithms applied to visual computing tasks such as detection, recognition, and tracking, with an emphasis on fully exploiting the OpenCL 2.0/HSA capabilities of modern hardware architectures. Learn more and register for EVS '15 here.

AMD GPU Services Library 2.2 Released

The AGS 2.2 update includes a new function to query the amount of video memory exposed on AMD GPUs and APUs. It also includes a console mode sample and an Eyefinity® sample.Learn more and download AGS 2.2 here.

Khronos Reveals Vulkan API for High-efficiency Graphics and Compute on GPUs

Khronos has announced the availability of technical previews of the new Vulkan™ open standard API for high-efficiency access to graphics and compute on modern GPUs! See what all the excitement is about in this "Next Generation OpenGL® Initiative" here.


Videos about New Tools & Technologies


OpenCL 2.0 Demystified

AMD provides full support for the OpenCL™ 2.0 specification, as well as support for some optional features like Fine Grain Shared Virtual Memory. There are a lot of new and powerful capabilities in this version of the standard. We know that programming for a new API and new capabilities can be daunting. If you’d like to learn what you can do, and how to do it, check out the OpenCL 2.0 Demystified blog series. AMD engineers walk you through the code and show you how features like Shared Virtual Memory, Pipes, Device Enqueue and many more actually work. Read the blogs, and learn what you can do. Happy Coding!

Conference, Webinar and Event Presentations

Want to take another look at an interesting conference presentation? Missed an event you hoped to see? Browse our collection of technical presentations from developer conferences, industry events and Webinars around the world covering a wide variety of heterogeneous computing topics such as Mantle, C++, rendering, shaders, game development, Java, graphics, libraries, OpenCL™ and much more.

Everything You Always Wanted to Know About HSA*

What is Heterogeneous System Architecture? Why does it matter? In this whitepaper, Insight 64 Research Fellow Nathan Brookwood explains why, even if HSA doesn’t entirely change your life, it will change the way you use your desktop, laptop, tablet, smartphone and the cloud.

The Heterogeneous System Architecture: It's (Not) All About The GPU

The use of GPUs in computation intensive tasks has had an ever increasing impact across all platforms. At the same time, GPUs have several properties that make it a challenge to unlock their full benefit in many real-world applications. The HSA was created to address many of these current shortcomings while providing a great update for already established software models. Learn more in this GPGPU7 keynote by Paul Blinzer, chairperson of the System Architecture Workgroup of the HSA Foundation.



Blog Posts

  • AMD Releases clRNG, an OpenCL™ Random Number Library
    May 5, 2015
    clRNG 1.0 is an OpenCL library that generates uniform random numbers. This library is now available as open-source on GitHub under the clMathLibraries organization, with other popu
  • AMD CodeXL 1.7 is here!
    April 28, 2015
    Spring time – the flowers are blooming, the bees are buzzing, and we have a new AMD CodeXL release to show off what we’ve been working on through the winter. ...

Forum Posts

  • error in BLAS - ACML
    May 21, 2015
    it seems old (bug), using ACML 5.3.1 > ## PR#4582 %*% with NAs> stopifnot(is.na(NA %*% 0), is.na(0 %*% NA))> ## depended on the BLAS in use.> > > ## found from fa
  • Video wall questions
    May 21, 2015
    Hi, I'm working in a company developing large interactive video wall using FirePro solutions.I need to ask some technical questions concerning framelocking and sync between mu

Follow AMDDevCentral on Twitter