Why Intel’s foundry tech couldn’t save its mobile business
Intel’s process technology leadership couldn’t save the company’s mobile division because it wasn’t designed to do so. Smartphone and tablet OEMs wanted devices with integrated LTE radios; Intel didn’t have them. Even the SoFIA partnership with TSMC never came to market, apparently because Intel couldn’t secure enough volume to kickstart production.
Intel’s 14nm problems delayed its next-generation tablet processors from 2014 to 2015. Its 10nm node, once expected to secure enormous economies of scale over TSMC, has been pushed to 2017 as well. I don’t think these delays played a huge role in Intel’s decision to leave the mobile market, but they may have influenced it. In 2012, Intel still expected to be on 10nm by 2016 with EUV (extreme ultraviolet lithography) ramping towards full production. The now-canceled 450mm wafers weren’t expected in-market quite this soon, but Chipzilla expected to recognize significant cost savings from moving to the larger wafers in the 2018 to 2020 timeframe — cost savings that could’ve further improved its standing against Samsung, TSMC, and GlobalFoundries.
If Intel had begun reorienting towards Atom when it launched the chip in 2008, it might’ve weathered these delays and cancellations without much trouble. Failing to do so left its beleaguered mobile business facing higher-than-expected costs and minimal revenue.
Intel didn’t lose the mobile market because Atom’s performance and power consumption didn’t compete with ARM; research and evaluation showed that Atom was capable of matching ARM performance in multiple market segments. It lost the mobile market because it didn’t make the changes that would have allowed it to compete on cost with products manufactured at TSMC and Samsung. The exacting rules and unique layouts that drove Intel to the top of one market could not be easily adapted to others, and Intel was unwilling to risk its position at the top of the conventional x86 market for a risky payoff in mobile. There’s no evidence that keeping XScale or developing ARM products would have changed that — if anything, the ARM division would’ve been under even more pressure to ensure it never became a threat to the x86 business.
Accent on: ...the now-canceled 450mm wafers weren’t expected in-market quite this soon... Canceled???