December 04, 2012 by Tony DeYoung
The new APP SDK 2.8 includes dozens of new and improved samples for OpenCL, Aparapi and C++ AMP that deliver significantly faster performance than APP SDK 2.7 – up to 2.3x faster on average in nine key benchmarks.
The APP SDK 2.8 also includes a preview version of AMD’s new open source C++ template library, codename “Bolt.”
Bolt is an STL compatible template library of data parallel primitives and provides a standard way to develop an application that can execute on either a regular CPU, or use any available OpenC™ capable accelerated compute unit, with a single code path.
V2.8 also SDK also improves and extends OpenCL capabilities by including support for the Direct3D 11 sharing Khronos extension in addition to including 64-bit atomics.
July 31, 2012 by Tony DeYoung
HSAIL is a a virtual byte code and virtual machine designed for parallel compute on heterogeneous devices that was introduced as part of the HSA rollout at AFDS. It is an intermediate representation that dynamically compiles at run-time for the device.
HSAIL was designed from the beginning to be agnostic to CPU and or GPU and whatever other heterogeneous hardware is available. While OpenCL will remain a primary choice for developers, HSAIL opens the door to other modern high level programming languages such as Java and C++ and thus mainstream coders.
This Slideshare presentation discusses HSAIL including code examples.
Also signup for the August 28 Webinar Overview of HSAIL - the basis for implementing an HSA platform agnostic open-source OpenCL runtime.
July 11, 2012 by Tony DeYoung
Code Devine gives first impressions of using C++ AMP as opposed to Cuda or OpenCL.
- The good: Easy to get started, good integration with Visual Studio and the code is not tied to a specific hardware vendor like Cuda.
- The bad: Profiling tools need work and compile times are long.
June 27, 2012 by Tony DeYoung
Beyond3D has teamed up with Microsoft and AMD to sponsor a coding challenge focused on Heterogeneous Computing using Microsoft’s C++ AMP.
“We wanted a somewhat equivocal task, that forced thought to be invested in the design and potentially allowed for the optimal solution to be a heterogeneous one, as opposed to a purely CPU or GPU one. Which brought us, to the somewhat famous, somewhat old, definitely challenging Travelling Salesman Problem.
The contest began June 12 and ends August 31, 2012. Microsoft and AMD will act as referees. AMD is providing the quad-core CPU, Radeon GPU prizes.
February 23, 2012 by Tony DeYoung
C++ AMP (Accelerated Massive Parallelism) is a small set of open specification language extensions (two of them) and a single library (amp.h) that makes general purpose GPU programming (aka GPGPU) a first class, seamless experience in modern C++.
These videos are short clips from the C++ AMP development team.
- What C++ AMP is and why it matters
- The design of C++ AMP, how it works and why.
- Meet the C++ AMP development team: Don McCrady, Amit Agarwal, Charles Fu, Lingli Zhang, Simon Wybranski, Steve Deitz and Weirong Zhu