There's no denying that Intel's (INTC -0.38%) yield issues with its 14-nanometer manufacturing technology had impacted just about all of the company's product segments. For example, volume manufacturing of Broadwell processors, which was supposed to begin in late 2013, didn't actually start until the second quarter of 2014.
This delay not only affected the Broadwell family of processors, but it seemed to push out the 14-nanometer Atom processors for tablets, phones, low-cost PCs, and even servers. Intel's executives stressed that moving to next-generation manufacturing technologies is hard and getting harder, but at the same time reiterated their commitment to "Moore's Law."
In light of the 14-nanometer delays and the apparent difficulties in moving to next-generation manufacturing technologies, investors might want to know what is happening with Intel's 10-nanometer process and products.
What we know so far
During last year's Intel investor meeting, CEO Brian Krzanich displayed a slide that said 10-nanometer products would go into production by the fourth quarter of 2015. However, at that time, Intel management also expected its 14-nanometer to go into production by the first quarter of 2014.
Intel refused to comment on its 10-nanometer plans at the 2014 investor meeting, with management claiming that talking about its 14-nanometer plans too early had given the competition room "to adjust."
That said, given the 14-nanometer production timeline, as well as the product road maps that the company presented at the investor meeting, it's not hard to guess when Intel is likely to bring 10-nanometers into volume production.
Cannonlake in 2016, 10-nanometer Atom in 2017
Given that Intel said it plans to launch its second-generation 14-nanometer PC parts based on the Skylake architecture during the "second half of 2015," with the ramp-up extending into 2016, I expect the first 10-nanometer PC parts, based on the Cannonlake architecture, to go into production in early 2016 for launch in mid-to-late 2016.
On the Atom side of things, the schedule is a little trickier. Intel has claimed that Broxton, the high-end 14-nanometer Atom, will come in "2016," as will the midrange and low-end variants known as SoFIA MID and SoFIA LTE 2. I expect Broxton will be a first-quarter 2016 part, while SoFIA LTE 2 and SoFIA MID will be mid-2016 parts, given that Intel indicated the first SoFIA LTE will launch at some point in the middle of 2015.
Given this potential launch time frame, if Intel executes really well, I could see the successor to Broxton launching in the first quarter of 2017, with the lower end later transitioning to 10-nanometers once the process has matured and costs have come down.
Has Intel's manufacturing lead shrunken?
I'm sure many investors' concern here is whether Intel's manufacturing lead has shrunk.
It's difficult to answer this question yet, as it depends on a number of key variables that remain unknown. Taiwan Semiconductor claims that its 16 FinFET Plus technology will go into production around July 2015, while Samsung claims that its 14-nanometer production will begin by the end of this year.
However, the "names" of the manufacturing technologies aren't all that meaningful; the density and performance characteristics of the processes are also critical. Intel claims that it has a meaningful logic density lead over TSMC's 16-nanometer and Samsung's 14-nanometer processes based on the published gate pitch and metal pitch metrics. However, we don't yet know how much of a lead, if any, Intel has in transistor performance.
The answers to these questions should become clearer at the International Electron Devices Meeting in December, where both Taiwan Semiconductor and Intel will present detailed papers on their respective processes.
