by Anton Shilov
01/14/2014 | 10:08 PM
Advanced Micro Devices on Tuesday finally began to sell its new-generation A-series “Kaveri” accelerated processing units. The new chips boast with all-new Steamroller x86 cores, heterogeneous system architecture (HSA), new-generation graphics engine, improved media processing units as well as a set of AMD’s other latest technologies.
General specifications of AMD A-series “Kaveri” have been known for a long time: the new APUs feature two dual-core x86 Steamroller modules (two integer units, one FPU with two 128-bit FMACs, one MMX unit per module) for a total of four cores; Radeon R7 graphics processing unit with up to 512 stream processors (eight GCN compute units), dual-channel DDR3 memory controller with hUMA technology, various special-purpose accelerators, including AMD TrueAudio DSPs [digital sound processors] second-generation video coding engine (VCE) and fourth-generation unified video decoder (UVD) units that enhance quality of playback and add new video encoding formats.
At present AMD sells A10-7850K and A10-7700K accelerated processing units with unlocked multiplier and 95W thermal design power. Later in the quarter the company will ship A8-7600 models with reduced power consumption. The most advanced AMD APU (model A10-7850K) costs $173 in 1000-unit quantities, whereas its younger brother (A10-7700K) is priced at $152. The A10-7850K and A10-7700K APUs will be bundled with EA’s Battlefield 4 video game.
“AMD maintains our technology leadership with the 2014 AMD A-Series APUs, a revolutionary next generation APU that marks a new era of computing. With world-class graphics and compute technology on a single chip, the AMD A-Series APU is an effective and efficient solution for our customers and enable industry-leading computing experiences,” said Bernd Lienhard, corporate vice president and general manager of client business unit at AMD.
The new APUs support heterogeneous queuing that defines how processors interact equally and enables uniform visibility into the entire memory space by both GPU and CPU. Kaveri’s hUMA [heterogeneous unified memory architecture] shared memory controller fetches data to both x86 and GCN processing units as well as acts like a dispatcher/hub.
Thanks to HUMA and heterogeneous queuing, applications can easily run processes on different types of cores, such as x86 CPU or GCN GPU compute units. As a result, AMD claims that its new Kaveri features twelve compute units in total. While the claim is technically correct, it should be kept in mind that not all applications can offload computing to stream processors of graphics adapters, hence, a number of programs will take advantage of only four x86 processors.
The new series of APUs support all the technologies that AMD introduced along with its latest family of graphics processors – which feature improved second-gen GCN architecture – including TrueAudio as well as DirectX 11.2 and Mantle application programming interfaces.
For previous generations of its high-performance APUs (i.e., Llano, Trinity and Richland), AMD chose 32nm silicon-on-insulator process technology that was developed by the company itself and GlobalFoundries with high-performance microprocessor units in mind. Those process technologies allowed AMD to ensure high clock-speeds for its x86 cores, such as Bulldozer and Piledriver. Unfortunately, that manufacturing technology was not exactly tailored for high transistor density of graphics processing units. As a consequence, AMD had to make certain compromises, which did not allow it to integrate all the hardware it wanted.
With Kaveri, AMD chose to use GlobalFoundries’ 28SHP (28nm super high performance) process technology that was designed with various types of chips in mind. Improved transistor density and thinner elements of 28nm process tech allowed the company to integrate whopping 2.41 billion of transistors into a 245mm2 Kaveri die, which is 85% higher transistor count compared to Trinity/Richland design (246mm2, 1.303 billion transistors, 32nm SOI). Now the company has four Steamroller cores, hUMA [heterogeneous unified memory architecture] memory controller, Radeon graphics processing unit with 512 stream processors (AMD calls them eight GPU compute units) and a lot of various special-purpose hardware inside its flagship APU.
The high-density “universal” process technology has its drawbacks too: it lacks thick metal stacks that make possible to run microprocessors at high frequencies. As a result, the company sacrificed extreme clock-rates for x86 general-purpose processing cores in favour or forward-looking APU design with a strong focus on heterogeneous computing.
While overclockers and performance enthusiasts may not appreciate AMD’s choice, the company claims in its documents that thanks to more advanced Steamroller micro-architecture (with up to 20% higher instruction per clock [IPC] performance), the Kaveri will not be slower than currently-available Richland APUs and will even beat them in many cases.
The AMD A-Series APU processor-in-a-box (PIBs) for the AMD A10-7850K and AMD A10-7700K, which started shipping in Q4 2013, are available starting today. The AMD A8-7600 will be shipping in Q1 2014.
AMD believes that thanks to scalability of its Kaveri design (which is supposed to be available in 15W – 95W thermal design power envelopes), it will be able to successfully address various personal computers, including desktops, notebooks and servers, with its new A-series accelerated processing units in 2014.