AMD Ships Trinity-Based Embedded Chips

AMD’s Embedded R-Series accelerated processing units use the Trinity chipset introduced earlier this month

AMD on 21 May announced the Embedded R-Series accelerated processing units (APUs), based on the Trinity chips introduced earlier this month.

The new Trinity APUs, which offer integrated x86 CPUs and discrete-level graphics capabilities, are the second generation of the APU push that AMD first introduced in January 2011.

Embedded R-Series

Up to eight APUs will be offered in the Embedded R-Series lineup, with the chips offering as many as four chip cores and 384 graphics cores. The chip cores will be based on AMD’s latest Piledriver core architecture.

The APUs also will carry AMD’s Radeon HD 7000 Series graphics chips, which support DirectX 11 for improved multimedia capabilities. Some versions of the chip come with PCI Express Gen 2 support.

“AMD pioneered the embedded APU to offer our customers a high-performance, power-efficient, small-form-factor embedded processor,” Buddy Broeker, director of AMD’s Embedded Solutions unit, said in a statement. “By leveraging its seamlessly integrated heterogeneous system architecture, developers can tap into a high-performance and efficient parallel-processing engine to accelerate their graphics- and compute-intensive applications, all while using industry-standard libraries such as OpenCL and DirectCompute.”

AMD officials have outlined a host of areas that can leverage the new embedded chips, from smart cameras and medical systems, to video conferencing, digital signage and point-of-sale machines.

In a 21 May post on AMD’s blog site, Cameron Swen, manager of embedded marketing at AMD, said the new Series-R chips help system designers solve the issues of power consumption when trying to ramp up performance.

Heterogeneous processing

“New processing solutions based on heterogeneous architectures are emerging that enable low-power designers to significantly increase their system performance without adding significant cost or power to the system,” Swen wrote. “Heterogeneous processing itself isn’t new, but what makes these solutions different are the open and royalty-free programming standards for general-purpose computations on heterogeneous systems that are developing around them, such as OpenCL. Using standards like OpenCL helps programmers preserve their expensive source-code investment and easily target and port code between multi-core CPUs, GPUs and new APUs, which combine both x86 processing cores and graphics-processing units on a single die.”

AMD last year launched the Embedded G-Series platform, which offers power envelopes of 5.5 to 18 watts, addresses the needs of very low-power applications, he said. In comparison, the new Embedded R-Series consume 17 to 35 watts, a good complement to the G-Series, Swen wrote.

“So now there are more options available to help you find that ideal balance between performance and power consumption,” he wrote.

AMD announced that a number of system makers – including Advantech-Innocore, Axiomtek, Congatec, iBase, Quixant and iBase – have said they will use the new R-Series in their systems.

AMD officials are aggressively ramping up their second-generation APUs, having unveiled on 15 May the new Trinity A-Series chips aimed at notebooks, ultrathin laptops and desktops.

How much do you know about smartphones? Take our quiz.