According to RBC Capital Markets analyst Doug Freedman, Intel’s current mobile strategy isn’t working, since most OEMs and device customers don’t seem to be too interested in Intel’s x86 chips for mobile. One solution that could see the company owning a significant portion of the mobile chip market quickly would be to buy an ARM chip maker such as Mediatek.
Mediatek has been a rising star in China’s mobile market over the past few years. It saw its market share double from 7.3 percent in 2007, to 14.4 percent in 2013, and it doesn’t show signs of slowing down.
In fact, Mediatek represents an increasing threat to Qualcomm globally, and it's already dominant in China because of its lower-priced processors for low- and mid-range mobile devices. Qualcomm has had to sell stock CPU cores from ARM just so it can match Mediatek's prices; otherwise, Qualcomm’s low-end chips would likely not be very competitive at that level of the market.
If Qualcomm is having a hard time pushing back on Mediatek’s growth, Intel's x86 CPUs -- which are barely present at all in the mobile market -- stand no chance at the low-end. Intel is used to higher profit margins on its chips, and having a low chip volume in the market also means its R&D costs are high compared to the revenue the company gets in return for it.
"Instead of Intel continuing to spend $4 billion-$6 billion a year to enter the market (higher end of spending range as it achieves success), hypothetically, an acquisition of MediaTek may reallocate Intel's best-in-class under-utilized fabs and financial resources to a rising star in the SoC world, solidifying MediaTek's market position," Freedman said.
Doug Freeman thinks that Intel could use the billions of dollars it spends every year to gain small percentage of the mobile chip market to acquire Mediatek instead and thereby could potentially get 30 percent of the market within another seven years. That doesn’t sound like a very exciting goal for Intel, given the expected timeframe, but it’s probably the best Intel could hope for right now.
The only problem is Intel is not particularly fond of the idea that it should be making ARM chips because the company thinks that will seriously undermine its still very profitable x86 business -- which could be a well-founded fear.
As Intel would need to keep raising the performance of its ARM/Mediatek chips to stay competitive with others in the market, it would also slowly make its x86 chips less and less relevant in more product categories (tablets, slim notebooks, low-end PCs, etc) where those ARM chips could be used.
Intel could also intentionally try to keep its ARM chips at the low-end of the market, just as it tried for years with Atom, so that it doesn’t get into low-end notebooks and cannibalize its Core i3 series of chips. However, that would be a mistake, as it would only make Intel's ARM chips less sought after in the market, and the competition would take Intel’s place in those products, anyway.
The idea that Intel should buy an ARM chip maker like Mediatek is certainly one that Intel should seriously consider, and indeed the company may well have already. But Intel also needs to remember that it owned an ARM division before (XScale), and that didn’t end very well. Before it goes ahead and buys yet another ARM company, Intel needs to figure out a strong strategy and how it can avoid repeating the mistakes that led to the failure of the Xscale division.