“I have been writing about Moorestown since Intel started talking publicly about it in 2007, so the official unveiling of Intel’s first x86-based SoC aimed at the smartphone market marks the end of a long journey. Moorestown’s appearance also marks the beginning of another journey, as Intel prepares to face down ARM in its quest to win handset and tablet makers over to the x86 camp. In many ways, this is the biggest and most important Intel product launch since the original Atom introduction.”
… Global warming is an obvious fact. 😉
Irony apart, Intel does not communicate on the power consumption. And as an embedded software engineer, let me say, in this kind of market… It is quite hype (Media ! Media !)
So, as usual with new Intel chip, let see in one year…
Why the atom chips haven’t moved to Intel’s 32mn process is a bit confusing. Their goal is lower power consumption and to allow for increased battery life of the device they are inside. So why are they still using the 45mn process when Intel’s 32mn process is rather mature?
Well, they’re able to manage leakage better on 45 nm, it looks like, from Ars Technica’s article.
I have heard of ARM9 designs having less power consumption on 180 nm than 65 nm, simply due to the leakage on the smaller processes.
from tfa:
“””
You won’t just fire up Ubuntu—or even Chrome OS or Android or an x86 port of webOS—on your Moorestown-based tablet and expect to get any kind of battery life from it. No, making a Moorestown-based product sip power like a smartphone will take a lot of custom, Intel-led effort, and this will constrain the platform’s potential applications a bit.
“””
Well, MeeGo is free and you can base an arbitrary linux distro on it, just like many distros use debian as base currently. I don’t see this as a problem – pm has always been about hw and sw playing together. You can’t get huge battery life by just picking a low power cpu and letting it poll freely.
BTW – fast cpu can rush to sleep earlier.
They don’t have the designs nor talent to do it.
A more sensible statement might point out that ARM has had 20-25 year head start, and has continued to improve the product design over that time. The same reasons why it would be very difficult to any cpu architecture to take space away from x86 (well, that and the installed base of software).
It really doesn’t make any sense to deride a company trying to overcome nearly insurmountable odds. Its like picking on a championship boxer for his poor calculus skills. Sure its true, but he’s got other things going for him. He might decide to test your calculus skills against his fists.
No designs, maybe, but I’m pretty sure they have the talent. The question is whether they want to sink a lot of development money into a low-margin, maybe-high-volume market.
They’ll get some wins due to x86 familiarity, influence, and alliances (I’d think Microsoft would find it easier to develop WinPhone 7 on x86 than ARM). But as BillShooter said, ARM has a hefty head start.
I don’t x86 familiarity being much of a selling point as once the kernel has been built for the chip architecture, the rest of the OS and applications don’t care too much whether it’s x86 or ARM as the applications would have to be cross-compiled anyway and higher level languages are chip-set independent.
Few developers really need to worry about the architecture of a CPU. They let the compiler worry about that.
They’re not afraid to throw billions at R&D if they are insulted. Look at how they came back after AMD’s X2.
They’ll throw a few billion on mobile research without even blinking.