Intel Foundry bows to the inevitable by embracing ARM physical IP

Chip giant Intel has put an end to decades of trying to beat UK chip designer ARM at its own game by announcing it will be optimizing its 10 nm manufacturing process for ARM-based chips.

Scott Bicheno

August 17, 2016

4 Min Read
Telecoms logo in a gray background | Telecoms

Chip giant Intel has put an end to decades of trying to beat UK chip designer ARM at its own game by announcing it will be optimizing its 10 nm manufacturing process for ARM-based chips.

Historically Intel has made good margin from manufacturing its own chips – primarily those used in PCs and servers – in very expensive specialised plants referred to as fabs. But with the PC market in slow decline it seems Intel no longer needs so much fab capacity for its own products and is looking to take on the likes of TSMC and Samsung in the third-party foundry game, which encompasses pretty much the whole mobile market including Qualcomm and Apple.

One of the main ways we have been able to extract ever more performance from ever smaller pieces of silicon is the continual shrinking of the manufacturing process, which refers to the size of each individual transistor in the chip. The next major process shrink is to 10 nm and if Intel is serious about becoming a major foundry player it needs to be competitive at this level.

Intel has poured billions into becoming a mobile chip player but has never been able to overcome the advantages ARM’s architecture has over its own when it comes to power efficiency – a critical issue in the mobile space. So strong is ARM’s position that even parity would probably never have been enough for Intel and it’s hard to see how any foundry could be competitive without being optimised for ARM physical IP, which is a library of information ARM offers to help with the manufacturing of its chip designs.

“Our 10 nm design platform for foundry customers will now offer access to ARM Artisan physical IP, including POP IP, based on the most advanced ARM cores and Cortex series processors,” said co-GM of Intel Custom Foundry Zane Ball. “Optimizing this technology for Intel’s 10 nm process means that foundry customers can take advantage of the IP to achieve best-in-class PPA (power, performance, area) for power-efficient, high-performance implementations of their designs for mobile, IoT and other consumer applications.”

“Today’s announcement represents what we expect to be a long-term, mutually beneficial partnership with Intel Custom Foundry,” said GM of ARM’s Physical Design Group Will Abbey. “One of the strengths and differentiators of the Artisan platform is the availability of ARM core-optimized IP – what we call ARM PO technology. The value of POP technology for an ARM core on the Intel 10nm process is tremendous, as it will allow for quicker knowledge transfer, enabling customers to lower their risk in implementing the most advanced ARM cores on Intel’s leading-edge process technology.

“Additionally, POP technology enables silicon partners to accelerate the implementation and tape-outs of their ARM-based designs. The initial POP IP will be for two future advanced ARM Cortex-A processor cores designed for mobile computing applications in either ARM big.LITTLE or stand-alone configurations.”

Telecoms.com spoke to Sravan Kundojjala, Analyst at Strategy Analytics, to get his take on the news. “This is a smart move by Intel to take on TSMC, Samsung LS and Global Foundries, and the foundry business has good margins, with TSMC operating margin at around 35%,” he said. “The LG and Spreadtrum mobile foundry deals will help Intel fill its fabs in the face of declining PC sales and if successful the company could even land orders from Qualcomm, Apple and MediaTek.

“Between 2011 and 2015 Intel spent approximately $15 billion on mobile R&D and approximately $3.5 billion on mobile-related acquisitions without much to show for it, so the latest move could provide a much needed boost to Intel’s mobile ambitions. TSMC and Samsung LSI are likely to beat Intel to market in 10 nm by 5-6 months but Intel could offer a credible foundry alternative.”

As Kundojjala said Intel has already announced some foundry deals on the back of this move, including LG, which will start making its own mobile SoCs in partnership with Intel, following the trend established by Apple, Samsung and Huawei, who license ARM designs to make their own chips. Six months seems like a substantial lag in the fast-moving mobile market, however, so Intel will need to compete strongly on price and service too.

The announcement came on the opening day of Intel’s annual developer forum – IDF – at which there was several other juicy morsels well covered by Light Reading’s Brian Santo at the event. Among the big announcements were a couple of Virtual Reality initiatives and new NFV collaborations with AT&T and NEC.

To some extent this move is also an admission by Intel that ARM is likely to dominate IoT devices in much the same way it has mobile. If anything ARM should be even stronger in IoT as it designs chips all the way down to the simplest microcontrollers and Softbank is betting on exactly that. The IoT boom should yield a lot of foundry business and, regardless of its own chip ambitions, Intel needs to support ARM if it wants some of it.

About the Author

Scott Bicheno

As the Editorial Director of Telecoms.com, Scott oversees all editorial activity on the site and also manages the Telecoms.com Intelligence arm, which focuses on analysis and bespoke content.
Scott has been covering the mobile phone and broader technology industries for over ten years. Prior to Telecoms.com Scott was the primary smartphone specialist at industry analyst Strategy Analytics’. Before that Scott was a technology journalist, covering the PC and telecoms sectors from a business perspective.
Follow him @scottbicheno

Get the latest news straight to your inbox.
Register for the Telecoms.com newsletter here.

You May Also Like