2020 iPhone 5G models may feature 5nm Apple A14 SoCs

Apple's A12 is manufactured by TSMC[1] using a 7nm FinFET process
Apple’s A12 is manufactured by TSMC[1] using a 7nm FinFET process

Monica Chen and Jessie Shen for DigiTimes:

TSMC expects its third-quarter revenues to be driven mainly by growing demand coming from the smartphone and IoT sectors, and believes the fourth quarter will be even stronger thanks mainly to chip demand for smartphones and HPC devices, according to company CFO Lora Ho.

Ho, speaking on the sideline of a June 18 shareholders meeting, reiterated her remark made during the meeting that TSMC has seen a ramp-up in chip demand for 5G devices.

TSMC plans to spend more to expand further its 7nm process production capacity and build up capacity for its newer 5nm node, according to Ho.

TSMC has become “a little bit more aggressive” with regards to its 5nm production ramp-up, said company CEO CC Wei during a Q&A session at the investors conference. The foundry is on track to move the node to volume production in the first half of 2020. An acceleration in the worldwide 5G development will lead to an increase in demand for TSMC’s 5nm and 7nm processes, Wei believes.

MacDailyNews Take: Of course, a 5nm Apple A14 would be a significant leap that will usher in iPhones, iPads, and, quite likely, Macs with markedly greater performance and noticeably reduced power consumption!


  1. Apple will have a six-month window of advantage if they do implement 5nm processors. After that, Qualcomm will flood the market with their own 5nm Snapdragons (865?) and then Apple will have no advantage. Anything Apple can do, Qualcomm will do as well and sell it for less. Currently, Qualcomm is offering a Snapdragon 855 Plus with higher clock speeds to negate Apple’s A13 Fusion processor, but Qualcomm is touting it for gaming and VR use which is a smart marketing strategy to get Android smartphone manufacturers to buy those components.

    A 5nm A14 isn’t going to boost iPhone sales if Apple charges over $1100 for a base model iPhone. I think iPhones are more than powerful enough for most consumers and Apple should simply build a higher-featured and less expensive iPhone because that’s what most consumers would be happy with. As I see it, most consumers just want cheaper smartphones and don’t have a need for greater processing power. Apple could take an A10 SoC, put it into a $300 iPhone and possibly sell millions of them in India and Africa.

    1. Snapdragons aren’t going to match A class chips any time soon ( if ever) 5nm or not the advantages go well beyond that. However as you allude to phones are becoming over powered for what they do. So a lot will depend on the exact impact of AR VR AI on the platform whereupon its latent power may well come into its own.otherwise it will be on other products like the iPad maybe Macs too where the benefits cone I go their own. As for the iPhone few people buy it because it has superior and faster silicon to be honest because only a few even realise it though no doubt if it were slower people would soon notice thanks to the media.

    2. It’s not just foundry node pitch (14 nm, 10 nm, 7 nm, 5 nm, etc.) that is significant nor is the clock rate (XX GHz) all that important. Apple’s Ax series processors are much better at instructions per clock. Apple’s Ax series being specifically tuned to iOS and iPad OS give them a distinct advantage. These and more keep the Ax series well ahead of any other ARM architecture — including Qualcomm’s implementation.

      Hopefully that lead will continue for several more years, and it likely will unless Apple decides it does not need to do so because of some idiotic concept that phone and tablet processors are as fast as they need to be.

      I assume you do know that some parts of Siri, Maps, and other functions are done back at Apple because the processing power and data sets needed are too much for ANY current Ax series processor. What if you had much more powerful Ax series processors that could do 99% (or even 100%) of that processing on your Apple device? What if it were faster when it does? What if when you asked Siri to do something there was not a 1-3 second lag for a response.

      The Ax series processors will have to be 20x to 50x faster and have access to 20x to 100x the data on the device, before we’ll even come close enough to being as fast as they need to be. Hell, by the time we get there, other applications and uses will come along to move those goal posts by another factor of 10 or more.

      Back when I got my first Mac Plus after struggling for a while with a 128K Thin Man Mac, I was blown away. By upgrading it to 4 MB of RAM and hooking it up to a 30 MB SCSI hard drive, I had a moment (well a few days) thinking, “Wow. Why would I ever need anything more powerful than this?” That faded after a few days when I started running up against the boundaries of its capabilities.

      This will ALWAYS be the case. No one will ever build a processor that has more than enough capabilities forever. THAT is reality. Anyone who says otherwise is divorced from reality.

  2. A 5G iPad Pro on iPadOS is going to be a tour-de-force in mobile personal computing, possibly a black swan event that not even Apple sees coming. I hope they don’t dawdle and lollygag with bringing 5G to the iPad and FFS it would be a whole lot more important to the MacBook Pro than the TouchBar.

    An iPad Pro capable of download speeds reaching 100Mb/s and above is a serious work machine, in the office, at home, or on the move.

    1. It would certainly be a serious work machine for the hour or two until AT&T, Verizon, or other cell provider started throttling the user for exceeding his monthly quota!

Reader Feedback

This site uses Akismet to reduce spam. Learn how your comment data is processed.