By Srinivas Nori
Today our friends at ARM announced the launch of their newest processor targeted at the rapidly growing market for mid-range mobile devices. The ARM Cortex-A12 processor is expected to offer a significant performance uplift and direct upgrade path from the highly successful Cortex-A9 processor, while matching the energy-efficiency of its predecessor.
Most of the attention these days goes to the latest and greatest high-end superphones and tablets—and of course ARM has processors to serve this important segment—but the market for entry-level and mid-range smartphones is where the real growth is expected to occur in the coming years. For example, a recent report by ABI research projected that shipments of sub-$250 smartphones will grow from 259 million units in 2013 to 788 million units in 2018, at which point they will make up nearly half (46%) of the global smartphone market.
What do consumers want in an entry-level smartphone? They expect similar levels of performance and battery life as enabled by application processors for high-end smartphones, but in a more cost-effective system. Delivering this functionality is no small challenge, and it requires a tight partnership between SoC design and process technology to optimize performance, power, and cost.
We have been collaborating for years to optimize ARM processors for GLOBALFOUNDRIES leading-edge process technology. Today, in conjunction with the launch of the Cortex-A12 processor, we announced new power, performance and cost optimized POP™ IP technology offerings for the ARM Cortex-A12 and ARM Cortex-A7 processors for our 28nm-SLP High-K Metal Gate (HKMG) process technology. You can read more here about this latest milestone in our multi-year collaboration with ARM. The upshot is that this combination of ARM’s processor IP and our leading-edge process technology will enable a new level of system performance and power efficiency with the optimum economics necessary to serve the market for mid-range mobile devices. GLOBALFOUNDRIES’ 28nm-SLP process technology and associated ARM POP IP for the Cortex-A12 processor enables up to 70 percent higher performance (measured single-thread performance) and up to 2x better power efficiency in comparison to a Cortex-A9 processor using 40nm process technology. Designers can achieve even higher performance by trading off for lower power efficiency, depending on their application needs.
But of course the technology industry continues its relentless march forward, and we have no plans to stop there. We are already collaborating to optimize ARM processor IP for our next-generation 14nm-XM technology. Our 14 nm-XM offering is based on a modular technology architecture that uses a 14nm FinFET device combined with elements of GLOBALFOUNDRIES’ proven 20 nm-LPM process, which will give SoC designers the benefits of FinFET technology with reduced risk and time-to-market. The XM stands for “eXtreme Mobility,” and it is truly optimized for mobile SoC designs.
Back in February at the Common Platform Technology Forum, we announced results from the industry’s first implementation of a dual-core ARM® Cortex™-A9 processor using FinFET transistors. We found the following results for a Cortex-A9 processor implemented on 14nm-XM technology. Similar results are expected for Cortex-A12 processor implementations.
Here are the technical details:
- High-performance, energy-efficient ARM processors implemented on 28nm-SLP typically use 12-track libraries. However at 14nm-XM FinFET technology, much higher performance and more energy-efficient ARM processors can be implemented using 9-track libraries resulting in further die-size reductions.
- At constant power, the frequency achieved with 14nm-XM technology based implementation (using 9-track libraries) is expected to be 61% faster than the frequency achieved with 28nm-SLP technology based implementation (using 12-track libraries).
- At constant frequency, the power consumed by 14nm-XM technology based implementation is expected to be 62% lower than the power consumed by 28nm-SLP technology based implementation.
- The performance-power efficiency of 14nm-XM technology based implementation (expressed as DMIPs/milliwatts) is anticipated to be more than twice that of the 28nm-SLP technology based implementation, while using half the silicon area.
The mobile revolution has only just begun. We are excited to see where this dynamic industry will go next, and you can be sure we will continue collaborating with innovative partners like ARM to bring the next generation of connected devices to life.
Srinivas Nori is director, SoC innovation at GLOBALFOUNDRIES. In this role Srinivas owns the GLOBALFOUNDRIES strategy and realization for ARM ecosystem based solutions. Srinivas also oversees the exploration, identification, evaluation and collaborative offering of innovative SoC solutions.