Back in late 2013, Intel's (INTC -0.38%) top brass told investors that a processor code-named Broxton, aimed at both high-end smartphones and tablets, would be production-ready by "mid-2015." By the end of 2014, though, Intel was telling investors that the chip would launch at some point in 2016, but no further specifics were given.
In this article, I'd like to share my thoughts on when I believe Intel will formally launch the chip.
Apollo Lake timeline gives us a clue
Information pertaining to Intel's upcoming low-cost PC platform, codenamed Apollo Lake, recently hit the Web (you can see my write-up on the chip here). Per the initial report from website BenchLife, Apollo Lake will go into mass production in June 2016 and will be formally unveiled in August/September 2016.
Note that Apollo Lake will be based on the same fundamental architecture as Broxton, although it will likely contain some additional PC-specific functionality. At any rate, the Apollo Lake production and launch timeframes should clue us in to when Intel may launch Broxton for tablets and phones.
What does history tell us?
If we look at what Intel did with Bay Trail, we see that Bay Trail-T for tablets and Bay Trail-M/D for low-cost PCs both launched in the third quarter of 2013. However, if we look at what Intel did with Cherry Trail/Braswell, we see that Intel announced first shipments to customers of Cherry Trail in very early 2015, while it held off on such an announcement for Braswell until early April 2015.
At any rate, it seems that a delta of between zero to four months is typical between the launches of the tablet-oriented Atom chips and the low-cost, PC-oriented ones. This tells me to expect Broxton for phones to go into mass production at some point between February 2016 and June 2016, for launches between May and September 2016.
That's a pretty big slip
If Broxton for phones/tablets is "production ready" between February 2016 and June 2016, then -- assuming that "mid-2015" meant August 2015 -- Broxton is anywhere from six to 10 months late. In a market where product cycles are approximately a year, this is quite a significant slip that could materially impact the competitiveness of the platform.
Is there hope for Broxton?
If Intel hasn't changed the product definition of Broxton from what it originally was back in late 2013, then my guess is that Intel has very little chance of delivering a competitive product. However, If Intel was able to bolster the specifications of the chip early enough in the chip's development cycle, there may be hope for the chip yet.
In fact, I can't help but remember the following comment Intel's Brian Krzanich made about a year ago with respect to being able to make product changes relatively shortly before shipping:
[We] have a process coming where we want to be able to adjust products, because we believe these markets haven't been changing up until the last minute before you shipped something, so three or four months before product is going to shipped, we want to get to a point where we could make changes in the characteristics of those products and we are always away from that maybe a year, year-and-a-half, but we believe we can get to a place where we have to kind of flexibility and IP configuration that we can do that.
Maybe Intel will be able to beef up the Broxton system-on-chip to be able to compete in the mid- to late-2016 timeframe, but it's very unclear at this point.
