14 DAY TRIAL // Just a couple of days away from the rumored release of AMD's next-generation DirectX 11-capable GPUs, the Sunnyvale, California-based chip maker has announced today that it has submitted its OpenCL GPU implementation for certification by the Khronos Working Group. According to the chip maker, this brings its products closer to delivering support for OpenCL, using both the GPU and the central processing unit (CPU), and consequently helping developers optimize their applications to use the resources of both the GPU and CPU, for a more efficient balance of the workload.
“The momentum behind AMD’s OpenCL implementation and ATI Stream technology has been building throughout the year, with an increasing number of software and hardware vendors joining AMD in supporting the open standard,” said Ben Bar-Haim, vice president of AMD software development. “Submitting a GPU conformance candidate to the standards body means we are encouraging software developers to fully utilize the processing capabilities of both our multi-core CPUs and our GPUs to create a more compelling user experience.”
According to AMD, the certification of the new GPU implementation will enable the company to become the only semiconductor provider that will offer support of OpenCL on both its GPU and CPUs. The support will be enabled by the upcoming ATI Stream SDK v2.0 for both the GPU and CPU, expected to be released later this year in its full version. In addition, AMD also announced that it had completed the migration of its Brook+ project to SourceForge, a measure that will provide the developer community with an online location where developers can control and manage open-source software to work with and evolve the Brook+ code.
The chip maker also announced that its much-anticipated Radeon family of DirectX 11-enabled GPUs were expected to be the first to support accelerated processing on the GPU, through the DirectCompute technology.