Intel’s smartphone: “The x86 power myth is finally busted.”


“The x86 power myth is finally busted. While the X900 doesn’t lead in battery life, it’s competitive with the Galaxy S 2 and Galaxy Nexus… …If you’ve been expecting the first x86 smartphone to end up at the bottom of every battery life chart, you’ll be sorely disappointed. “

Intel’s(s intc) first commercially available Android(s goog) smartphone went under the microscope at the always thorough site, Brian Klug’s summary on power consumption for Intel’s Medfield-powered X900 is both surprising and exciting. It means that Intel has finally been able to deliver what it has promised for the better part of three years: A capable x86 system on a chip that can rival those based on the ARM(s armh) architecture.

Going forward then, handset makers have another choice for smartphone and tablet chips, which should, at the very least, be concerning to companies such as Qualcomm (s qcom), Texas Instruments (s txn), Nvidia (s nvda) and others in this space. Klug’s review shows a very capable Android 2.3 device — that can also run Android 4.0 — with solid graphical prowess, decent imaging capabilities and performance that matches most other current Android phones running on ARM chips.

Klug notes that most people wouldn’t know the handset is running on a different type of chip: Nearly all of the apps he tried ran perfectly fine. One lone exception was Netflix(s nflx). That should be expected, however, since Netflix rolled out its app piecemeal to begin with as it initially tested different processors with its DRM system.

Since 2010, I’ve been sour on Intel’s mobile efforts — so has Om –but to be fair, all the company had were promises and not products. But I began to have a change of heart when I saw an Intel-powered Android 4.0 tablet in January. And after reading Klug’s well-written review of the Xolo X900, I think it’s safe to say that Intel won’t allow ARM-based chips to pitch a shutout in the mobile market. There’s room for Intel to improve, for sure, but at least they’re inside the ballpark and now part of the game. Let’s see if any handset makers want to take a swing.


Cameron Mulder

The way Intel might be able to get some type of traction in the phone market would be to include Open Source drivers with it’s phones. This would instantly get the geek/dev market behind X86 phones since they would be able to make much more interesting custom builds of android (and other OS’s) without the current driver issues you hit on just about all current android phones.


Kevin, yes, that’s fine for most people, but that’s not how they market the phones. If that was the case, you and I and lots of other people wouldn’t be buying the latest dual or quad core devices (would you have paid $700 for this phone, like you did for the GN?).

I’m not convinced it will perform as well as similarly equipped devices, the review wasn’t that in-depth. The display is not as big and doesn’t have the same resolution as other high end phones, so it will consume less power, and leave more for the SoC. Intel probably also did a better job of power management, which if implemented on the ARM-based devices, would yield additional power savings. My point is that if all things are equal, the Atom core cannot compete in power consumption with the ARM. In this comparison, all things are not equal.

While most people won’t need the advantage that a multi-core SoC provides, when you start to have multiple apps running, the extra core or 3 will be an advantage. While I think it’s safe to say that Intel can make a chip for phones, I think it’s a stretch to say that they are competitive with the high end ARM solutions. And I also don’t think they will like selling chips into the low end market – Intel does not like to make CPUs that sell for under $10, which is where the low end market will take them. They were born into an industry where they can command $40-$800 per chip, and they don’t have the discipline to compete in a sub-$10 market.


The Intel SoC is only a single core chip – all of the others are dual core. If we’re going to compare power consumption, don’t you think the specs should be similar? Are are you saying we should accept Apple’s premise, that CPU, clock speed, memory, etc. don’t matter, only the benchmarks (which systems can be optimized for) matter?

Kevin C. Tofel

Great point, Ken, but we’re talking about totally different architectures. I don’t care how many cores a phone has, but I do care about overall performance and the related battery consumption. If the Intel phone runs as quickly as most other current ARM chips and gets similar battery life, that’s fine for most people, no?


As a Microsoft employee, I’m excited about what this could mean for the Slate PC in the enterprise. It looks like Microsoft’s strategy to remove enterprise features from Windows RT (Windows 8 on ARM processors) may not matter as much…

Simon Cohen

Intel Medfield: no worse than comparable ARM chips. Just me or is that kind of sad? What reason would a manufacturer have to switch to Intel unless there is a price advantage? Perhaps Intel is willing to sell these chips at a break-even price? Doesn’t sound likely to me, but what else do they have to lure handset makers with?

Kevin C. Tofel

Simon, this is just a starting point for Intel; until now it didn’t have a product that was comparable. No hardware makers may go with the current solution, but they’ll be looking at it as it matures I think.

Comments are closed.