According to a fairly old article on technology news site Digitimes, Intel's (INTC -0.30%) smartphone chip launch schedule looked like this:
- Fourth quarter of 2013 – Merrifield, 22nm/2 core Silvermont/PowerVR G6400
- First half of 2014 – Moorefield, 22nm/4 core Silvermont/PowerVR G6430
- First quarter of 2015 – Morganfield, 14nm/4 core Airmont?/???
According to Intel's latest roadmap, the just-announced Merrifield is a first-half 2014 affair, with Moorefield coming in during second half of the year, likely very early, though. There is no Morganfield scheduled for the first quarter of 2015, but there is a chip known as Broxton scheduled for "mid-2015." So, what's going on here?
Intel needed to move quickly to Broxton
To not get too lost in code names, it's important to define them upfront. Silvermont is a brand-new CPU design built on the 22-nanometer manufacturing process. Airmont is a slight enhancement of this design but built on Intel's upcoming 14-nanometer process. The CPU following this is known as Goldmont, which is a brand-new design on the 14-nanometer process. This is the CPU at the heart of the next-generation Broxton system-on-chip, and it is likely to be a significant leap from Silvermont/Airmont.
For tablets, Intel will launch Cherry Trail, which is based on Airmont, at the end of 2014. However, Intel has not indicated that a smartphone variant of this processor is coming. This leads one to believe one of the following two possibilities is true:
- Intel pulled in the Goldmont-based Broxton in order to be more competitive, scrapping any potential Airmont-based smartphone platform in the process.
- A smartphone-based Airmont product never existed, and the plan was to move to the Goldmont-based Broxton as quickly as possible.
A search on LinkedIn revealed that there was not a single reference to a Morganfield platform and, indeed, the only three system-on-chip products associated with smartphones were to be Tangier, the SoC found in Merrifield, Anniedale, the SoC in Moorefield, and Broxton, as shown from this LinkedIn profile:
It is likely that Intel never had any intention of putting the Airmont CPU core in a smartphone-targeted system-on-chip and wanted to skip straight to Goldmont. This was probably for competitive reasons against the various ARM-based players that have been advancing their micro-architectures rather rapidly, including ARM's own Cortex A series.
Foolish bottom line
This is the right strategic move on Intel's part, particularly as the updated Silvermont core, known as Airmont, is unlikely to drive the kind of performance boost really necessary to be competitive in the mid-2015 time frame against ARM Cortex A57 class processors. The good news, however, is that Silvermont was so power-efficient on 22-nanometers and left so much thermal headroom on the table relative to its peers, that Intel can really flex its muscles with Goldmont. This boost will be achieved both by taking advantage of the next-generation 14-nanometer process, which brings substantial density and performance/power improvements, as well as an expanded thermal envelope.