As a shareholder in Intel (INTC 6.53%), I had a lot to be excited about. In fact, I even called it a "veritable blowout" in my analysis of the results. However, the company's mobile and communications group continues to be a drag on the company's earnings, with organic operating expenses in excess of $3 billion per year. Given this level of spending, Intel apparently wants to become a strong competitor to Qualcomm (QCOM +0.15%) in smartphones.
While I do believe that Intel will eventually be successful in phones (and it may be as early as 2015), it's important to take a look at what went wrong with Intel's 2014 smartphone platforms -- Merrifield and Moorefield.
Why did Merrifield and Moorefield fall flat?
While Intel often gets a lot of flak for perceived inefficiencies on the applications processor side of things, the fact is that Intel's Merrifield and Moorefield platforms were actually very competitive on the performance and power side of things. Leadership, I daresay, in the case of Moorefield.
The problem, unfortunately, is that these platforms were late to market. It find it very difficult to believe that Merrifield and Moorefield were "late" because of issues with the actual system-on-a-chip design -- Intel's 22-nanometer production has been healthy for quite some time. Further, Merrifield and Moorefield weren't anything far out relative to what Intel has delivered before.
No, what seems to have kneecapped both Merrifield and Moorefield was very likely the lack of a timely and competitive communications platform to go along with them.
Integration probably wasn't the biggest problem -- features and power were
Intel's XMM 7160 LTE category 4 modem lacks support for TD-SCDMA (important in China), LTE-TDD (again, important for China), and CDMA (important in North America for carriers such as Verizon). The CDMA bit isn't too big of a problem, as most international carriers don't employ the standard, but in general the XMM 7160 is the poor man's Qualcomm MDM9615, which first made its debut in 2012, built on a 28-nanometer manufacturing technology.
Even Qualcomm's lowest-end, bottom-barrel Snapdragon products integrate a more fully featured modem. Not only that, but those chips are also built on a more sophisticated manufacturing technology (Qualcomm's is built on Taiwan Semiconductor's (TSM 6.29%) 28-nanometer LP process; Intel's on TSMC's 40-nanometer process), which gives Qualcomm a power and size advantage.
Can XMM 7260 fix things?
The XMM 7260, which is much more competitive, appears to have been delayed into Q3 or Q4 of this year. While Moorefield paired with the XMM 7260 would be very competitive, the platform's availability timing missed the MWC 2014 launch window.
While Moorefield + XMM 7260 could be ready in time for launch in devices at the IFA 2014 show in September, it'll be tough to compete with the Snapdragon 805 paired with the MDM9x35 from Qualcomm for high-end devices. For low-end devices, it's not clear whether the Moorefield platform has a competitive enough cost structure (bill of materials adjusted) to drive mass adoption.
The silver lining surrounding XMM 7260 is that even though it will probably not find a home in too many smartphones, it should be a good chip for cellular-enabled tablets and PCs during 2015, particularly when paired with the upcoming Cherry Trail tablet platform.
Foolish bottom line
This isn't meant to take away from Intel's stellar earnings results. However, given that Intel is spending so much on mobile, it's important to keep our fingers on the pulse of this division. This year is more or less a bust for anything cellular for Intel, but 2015 should be quite a bit better. Stay tuned for the second part of this piece, which should be titled, "Intel Corporation's 2015 Smartphone Plans: Looking Ahead."
