Digital technology, media and intellectual property
Random header image at GB Media

The iPhone 7 product cycle: here is the problem when you use 2 chips

October 25th, 2016 |  Published in Apple



25 October 2016 – For years, Apple has had its modem chips supplied solely by Qualcomm. This isn’t a unique method of procuring components for Apple, which has generally preferred to stick with one supplier per component type when it can. Business reasons for such a decision are obvious: with one supplier, Apple can better ensure it is putting out a quality product that performs similarly across the entire global user base, and the company only has one supplier to keep an eye on to ensure proper quality control. On the other side, single suppliers have strong bargaining power and it exposes Apple to supplier risk (supplier not meeting volume requirements of components needed to meet device demand).

This changed with the iPhone 7 product cycle, with Intel finally getting a major win within its mobile division on what is, by most accounts, the most important smartphone platform globally. No surprise that this was viewed as a major win for Intel as it catapulted the company onto the mobile stage after years of struggles. Intel shareholders rightly celebrated that victory, and it removed some of the bad taste that has existed after Intel botched the acquisition of Infineon Technologies’ Wireless Solutions business.

For clarity there, Infineon was the original supplier in 2007 on the first iPhone, and their 2010 acquisition by Intel was hoped to be a means of getting Intel a foothold within the key mobile market. Unfortunately, after the acquisition, Apple dropped that Infineon unit as a supplier in favor of Qualcomm. Ouch.

Unsettling News

However, two different chips developed on different technologies (20nm node for Qualcomm, 14mm for Intel) that are produced by two different suppliers often means different performance. For those investors who follow Apple closely, they are likely already familiar with the Cellular Insights report discussed below. For those who are not, Cellular Insights took the time to benchmark Qualcomm versus Intel performance on these two chips, and the results were interesting to say the least. It appears Qualcomm’s MDM9645M modem outclasses Intel’s XMM7360 modem in situations where signal is less than ideal.



See the results (left) that were performed on Band 4, which is the most common LTE spectrum band you’ll find in use here in North America. It was found that as reference signal receive power (RSRP) is degraded, the Intel chip begins to significantly underperform versus the Qualcomm alternative. For more detail on how the test is performed, see Cellular Insights’ explanation:

“Consistent with our previous reviews, our focus has been on measuring the highest achievable LTE throughput in clean channel state, starting at RSRP value of -85dBm, and incrementally reducing radiated power level while maintaining Block Error Rate (BLER) under 2%. This allows us to measure RF sensitivity of the device under test while eliminating inter-cell interference and fully controlling the radiated environment. This also allows for high reproducibility and consistency of our tests, and takes into account the performance of the entire RFFE subsystem.”

Now, there are caveats here. Note that throughput remained high, even at the far end of the test range (>10Mbps). Beyond that, just because a phone could support 150Mbps of data, that doesn’t mean that the carrier will allow or support that much data most of the time, even via carrier aggregation. Further, most consumers are never going to notice the difference between the two chips, even if they could see those types of data speeds. For typical phone usage (browsing webpages, e-mail, Youtube video streaming), it simply isn’t going to result in a noticeable difference in performance. Nearly all consumers would never know the difference.

However, the problem for Apple here is the news story. Just because the problem doesn’t make much difference to the consumer when it comes to their data consumption habits, that doesn’t mean that those same consumers won’t pitch a fit. “Why is yours better than mine? Not fair!” exists on school grounds across the United States, and those feelings don’t go away with age. So just because it shouldn’t be a big deal, that doesn’t mean that the issue won’t potentially negatively impact either iPhone sales or Apple’s reputation. News outlets have been running with the story for several days now and it is starting to get more and more press. Anything Apple related is going to get clicks, especially anti-Apple dialogue, so expect this story to hang around.

Timing could not be worse as the last thing Apple needs is a negative news story heading into the most important selling quarter of the year. While I do think the end impact will be minimal, I think it could be a data point on Apple returning to sourcing components from single suppliers whenever possible, even if that strategy impacts gross margin. Remember that Apple likely went the multiple supplier route to hopefully reduce overall costs of building the phone. Forcing Qualcomm to compete with Intel likely would have yielded some lower component costs in the future, and Apple might have to reconsider that strategy going forward. Since a success story here could have meant a little bit of a tailwind on gross margins in the future. It’s a shame as a shareholder to see such a rough start to Apple trying to begin to better manage supplier risk.

About the author

Email | All posts by

"The mind that lies fallow but a single day sprouts up follies that are only to be killed by a constant and assiduous culture."
Latest Videos

Un aperçu de la FIC 2017 / A quick look at FIC 2017 (Lille, France)

Cybersecurity: a chat with John Frank, Vice President EU Government Affairs for Microsoft

From Legaltech NYC 2017: a chat with Andy Wilson of Logikcull

5G is coming ... and it's going to blow you away. Yes. Really.

The Internet of Things ... or the cybernetic consortia? (Part 1)

From the Mobile World Congress 2016: an introduction