At this point, investor expectations are quite low for Apple's (AAPL 0.38%) fiscal year 2016, as the company's iPhone 6s/6s Plus have seemingly failed to generate enough customer interest to keep iPhone sales flat, let alone grow, for the year.
I believe part of the problem is the very difficult year-over-year comparisons Apple will have to face during the fiscal year, but another part of it is that Apple didn't quite deliver a big enough improvement relative to the prior generation phones in ways that mattered to consumers.
One area where Apple has routinely impressed, particularly vis-a-vis the competition, is with its chip designs. Over the last few generations, Apple has delivered clearly leadership processor architectures and has been the first to migrate its applications processors to new chip manufacturing technologies, giving it a nice edge over the competition.
However, there is one potential competitive issue that Apple's iPhone 7 could face mid-cycle that investors should keep a close eye on.
Top Android vendors could transition to 10-nanometer chips while Apple stays on 16FF+
Samsung (NASDAQOTH: SSNLF) was the first company to ship smartphones with 14-nanometer silicon with the Galaxy S6. Apple followed about five months later, being the second major user of Samsung's 14-nanometer chip manufacturing technology and the first major user of TSMC's (TSM 2.79%) 16-nanometer FinFET Plus technology with both the A9 and the A9X in the iPad Pro.
Apple is expected to, once again, utilize TSMC's 16-nanometer FinFET Plus technology to build the A10 chip that will power the iPhone 7/7 Plus (and it is widely reported that TSMC will be the sole source, here). I believe that thanks to significant architectural enhancements, Apple should be able to deliver a nice performance boost over the A9 and ultimately performance leadership.
However, it would seem that there is a reasonable chance that Apple's iPhone 7/7 Plus could fall behind in applications processor power/performance mid-cycle.
Both TSMC and Samsung (NASDAQOTH: SSNLF) claim they will enter mass production on their respective 10-nanometer chip manufacturing processes by the end of 2016. Although I do think Apple could be one of the first customers of this process, there is a reasonable chance that Qualcomm (QCOM 1.05%) -- hungry to try to enable its customers to grab share from Apple at the high end of the smartphone market -- could launch a 10-nanometer Snapdragon processor during the first half of the year.
In this case, a number of Android vendors could roll out products using 10-nanometer-based processors, affording them a potentially significant performance/power advantage over Apple's iPhone 7/7 Plus phones for a period of several months.
That said, it takes some time to go from production start to products on the shelves
If we look back at statements made by TSMC on its Jan. 2014 earnings call, management said its 20-nanometer process had begun mass production. The first product based on this technology, the Qualcomm MDM9x35 stand-alone modem, didn't arrive in commercially available devices until mid-2015.
More importantly, though, it wasn't until the third quarter of 2014 that TSMC reported that a material portion of its revenue came from sales of 20-nanometer wafers, which was almost certainly due to the iPhone 6/6 Plus ramp (Apple's A8 chip was the first major chip built on this technology).
This makes sense if you think about the time it takes to go from wafer production start to wafers being shipped to customers (in this case, Apple).
The typical time it takes for a wafer to go from "blank" wafer to fully processed is around three months, with this time increasing as process technologies become more complex; TSMC says that at 16-nanometer, it's more along the lines of 1.5 quarters, or four-and-a-half months.
If we assume that wafers on 10-nanometer go in on Dec. 31, and if we assume an even longer cycle time (the formal term for the time it takes from wafer in to processed wafer out) of around two quarters, then we shouldn't expect shipments of 10-nanometer product to customers until around mid-2016.
Which, unsurprisingly, is exactly when Apple is expected to be well into the builds of its iPhone 7s/7s Plus to support a launch in the fall of 2017.
It really depends on when 10-nanometer goes into production
If Samsung and/or TSMC begin 10-nanometer production in, say, October of 2016, then I could see Android flagships getting the proverbial "jump" on Apple vis-a-vis Qualcomm aggressively pushing to a 10-nanometer Snapdragon.
However, if it's more along the lines of a production start for major customers in very late 2016, then Apple should be right there at the bleeding edge of the transition from 16-nanometer to 10-nanometer.
