This isn’t another Antennagate.
A mini firestorm erupted today over a report that Apple was purposely limiting the performance of the modem chip in the iPhone 7.
Bloomberg reported that Apple was intentionally throttling the performance of the Qualcomm modem chip inside some versions of the iPhone 7, in order to make it more similar to the Intel chip it was using in some other versions of the phone.
That’s true, at least according to sources I talked to.
Apple supposedly did this for a simple reason: It wanted the cost savings and flexibility of having two chip sources but also wanted the different versions of the phone to be as similar as possible (to avoid angering AT&T and T-Mobile and their customers in the U.S., for example).
The result is that the modem in the Verizon/Sprint version of the phone, while just as good and perhaps a bit better than the one in the AT&T/T-Mobile version, is technically not fully living up to its potential.
The real, and harder, question is just how big of a deal this is. The short answer is that it probably results in some loss of download speed, but in the context of other variables like network quality, signal strength and specific phone activity, it is hard to isolate in real usage.
A more complete explanation requires getting into the weeds and trying to answer a theoretical-but-difficult question: What would the performance of an iPhone 7 be if it used the full potential of the Qualcomm chip?
We don’t know because one doesn’t exist. Since we can’t answer that, we have to look at the question a couple of other ways.
One can compare an iPhone 7 with a Qualcomm X12 modem to one with an Intel XMM 3360 chip. Bloomberg did that, in conjunction with some analyst firms, and found that the Qualcomm chip still slightly outperformed Intel. But this only tells us that Apple was able to mostly level the playing field, not how uneven it was to begin with.
One can also look at how a phone with an unhobbled Qualcomm chip performs and compare that to the Qualcomm-powered iPhone, but that introduces all sorts of other variables as well.
For their parts, Intel and Qualcomm both declined comment. Apple, repeating the statement it made to Bloomberg, said that “every iPhone 7 and iPhone 7 Plus meets or exceeds all of Apple’s wireless performance standards, quality metrics, and reliability testing.”
“In all of our rigorous lab tests based on wireless industry standards, in thousands of hours of real-world field testing, and in extensive carrier partner testing, the data shows there is no discernible difference in the wireless performance of any of the models,” Apple said.
But, of course, that just means Apple did its intended job of limiting the Qualcomm chip in such a way that it performs similarly to the Intel one. That answer sheds no light on how fast the Qualcomm iPhone 7 would be absent such limitations.
It’s also worth noting that it is not unheard of for companies to have two component sources, creating some performance differences.
With the iPhone 6S, for example, Apple had some of its processors made by Samsung’s manufacturing plants and others by Taiwan Semiconductor Manufacturing Co. The result was not an enormous performance disparity, but some users noticed enough of a difference, especially in battery life, to raise the issue.
Nor is it unheard of for product makers to limit some features in order to make the performance of a device consistent regardless of component differences, as Apple appears to have done here.
That said, Apple is clearly valuing its own cost savings and supply chain flexibility over peak performance. Had it wanted the greatest speed and compatibility, it could have just used Qualcomm’s chip across all the models.
Having made the decision to use Intel modems for some phones and Qualcomm for others, analysts say, Apple was probably wise not to allow one model to have a significant advantage.
“Apple chose to use different modems for leverage and redundancy, and carriers had no real say in that,” Jackdaw Research analyst Jan Dawson told Recode. “So it would be a little unfair if Apple allowed, say, Verizon and Sprint to say that their iPhone performs better than AT&T and T-Mobile’s.”
Dawson added that device makers often choose not to use all the potential features of a component, and do so for a variety of reasons.
“The reality is, of course, that every phone maker makes decisions about how to balance battery life and performance across a variety of components, and so you rarely see a phone that’s maxed out on all possible performance parameters because it would be terrible for battery life,” he said.