Intel's Medfield Gingerbread Smartphone Reference Platformby Anand Lal Shimpi on September 13, 2011 5:01 PM EST
- Posted in
- IDF 2011
- Trade Shows
Earlier this morning Paul Otellini showed off Intel's Android smartphone reference platform running a version of Gingerbread. We just got up close and personal with the device as you can see in the gallery below. The phone is 9.5mm thick and features Intel's 32nm Medfield SoC although there's no word on the specifics of the architecture. Intel also told us that Medfield includes a newer generation GPU, but it's unclear whether Intel is using something as powerful as the PowerVR SGX 543MP2 or if it's a bit more tame in the graphics department.
Intel did give us a demonstration of the new ISP (Image Signal Processor) in Medfield that's capable of capturing up to 20 fps at 1600 x 1200. The reference platform could only deliver 10 fps from its 8MP rear sensor, but there's apparently headroom in the architecture.
Intel is claiming competitive battery life/power for Medfield but it's unclear how it will perform compared to the next-generation of ARM based SoCs (Krait, Kal-El, A15). We won't see the first Medfield based solutions until the first half of 2012. Apparently most device manufacturers weren't too pleased with Intel's close partnership with Nokia and Google was likely unhappy with Intel's work on MeeGo. With the Nokia partnership dissolved, Android prioritized and a unified design infrastructure, we may finally see some real momentum behind Atom in smartphones starting next year.
Post Your CommentPlease log in or sign up to comment.
View All Comments
ltcommanderdata - Tuesday, September 13, 2011 - linkGiven the need to support Windows, Intel tends to go with PowerVR GPUs with DirectX support. Other than the SGX535 they've used, their options are SGX545 or the SGX544MP.
Hector2 - Tuesday, September 13, 2011 - linkA 32nm processor is nice, but don't expect Intel to make a splash into hand-helds until 22nm. 22nm ought to really start getting them traction. In the meantime, they can fine-tune everything else and get it ready for prime time. I think 14nm will be really amazing.
name99 - Wednesday, September 14, 2011 - linkAnd ARM will just stand still until then?
This is what is so dumb about tech comparisons.
You compare what's available today, with the competitor today. Or you compare what will be available in two years with the competition in two years.
But you don't say "hah, our 2014 product will totally crush ARM's 2011 product --- we're winning!!!"
readers - Thursday, September 15, 2011 - linkSorry but you don't have a clue about the industry.
And yes, ARM will just stand still, guess why? Intel has THE BEST FAB in the world, ARM has NONE.
ARM will be wishing company like TI, Samsung and TSMC etc to get better than Intel. HOWEVER, None of them has managed to do that in past 7-8 years, in fact, none of them were even on par with intel, and with each of their road-map, that will not happen for at least another 4-5 years.
So you can count on Intel's SOC to have better process than anyone else's, unless they decide to not use their latest tech, it's not a prediction, it's a fact.
sjael - Tuesday, September 13, 2011 - linkA15 is a SoC now? ^.<
I'll go crawl back under my rock..
ioannis - Tuesday, September 13, 2011 - linkSo MeeGo was holding Intel back?
microlithx - Wednesday, September 14, 2011 - linkGoogle was likely unhappy with Intel's work on MeeGo
Of course they were. Google has worked very, very hard to undermine non-Android open source mobile projects, and here goes Intel pushing one not owned by Google. Too bad Google was so successful in suckering so many people into thinking they actually had respect for the existing Linux community. Now actual FOSS alternatives have to struggle to get hardware support.
I wonder if this means that actual Linux distros will see lesser support, now that Intel will be investing time and effort into working on Google's divergent kernel and wholly incompatible userspace.
asv2026 - Wednesday, September 14, 2011 - linkpart 2 video @0.20
name99 - Wednesday, September 14, 2011 - link"we may finally see some real momentum behind Atom in smartphones starting next year."
Why? Intel's two strengths are
- x86 compatability
- kickass process
The x86 compatibility appears to be ever less importance.
The kickass process only starts to matter when it actually delivers a product. Intel's history has been that when a choice needs to be made, marketing not technology will determine that choice. We saw that with iTanic --- not dead yet, but who are we kidding. And we saw that choice with Atom, where every boneheaded piece of nonsense that has been added to x86 over thirty years was dragged into the mobile space, even though it has no relevance to this space.
Intel can continue to go on about how wonderful each new Atom is compared to the previous Atom, or to the mainline x86; but those are not the comparisons of relevance. The comparison of relevance is that ARM is zipping along just fine, able to revise the architecture way faster than Intel, and quite happy to add or remove things from each chip as best suits customer needs. It is awesome just how good Intel's process is --- but they're still carrying a self-imposed 50 pound weight on their backs. Unless, and until, they're willing to give that up, well, good luck with the Windows8 Tablet market. I'm sure the world is dying to buy devices that will cost $200 more than an iPad and offer a quarter of the battery life.
curverider - Thursday, September 15, 2011 - linkPeople who think they get the industry always try to boil things down to process size and clock speeds.
It makes absolutely no difference if TSMC or Global Foundry is 6 mo to a year behind Intel in fab process. It will give Intel no advantage.
Intel is losing the mobile processor space b/c they don't have a good business model for mobile. It's not Intel versus ARM. It's x86: made by one, and only one company versus little-endian ARM: made by dozens of companies, with dozens of options.
Furthermore, the biggest players manufacturer their own chips: Apple, Samsung, LG, and others. SO WHY? WHY? WHY? would they use an Intel chip and give up margin with higher COGS? They would only use an Intel chip if it gave them a massive advantage in performance on both a wattage basis and performance basis. Intel can't deliver that. Not with x86 - not any time soon.
As a designer of mobile devices, I can ensure you, that nothing Intel brings to the table is interesting for us. Sure, if you are making a x86 netbook (or whatever we are terming them now) - basically a little laptop, its great. You need x86, you need high performance, you don't care about the rest. But for everything else mobile, its a real weak value proposition.
Then there is pricing. Intel is simply *not* competitive. Go to findchips.com and run pricing on Atom versus OMAP whatever, or Samsung whatever, 10k unit size. It's not close guys. And guess what - if they get close - by lowering price, it will cripple their margins. But clipping margins won't hurt ARM as much. It doesn't even have fabs. ARM is like a software company. Intel is hardware company. apples and oranges.
no chance. media hype. Furthermore, as some one who understands the core code in Android, and has developed kernel-level code, there is almost nothing of significance that's going to help Intel in their "partnership" with Google. It's a press release by Intel marketing people, trying to create buzz. Google played along b/c they want to keep Android in the fold, and keep ARM from partnering so heavily with Microsoft. But in terms of change in the industry... it simply nothing.
When you see MMI actually put Intel in their mobile devices, then, maybe then, can you give Intel/x86 a position as a serious contender in mobile.
If Intel were smart, they would buy MIPS at its rock bottom price today (which already has a good Android port, and is used by Motorola in set-top), bundle in some SSE2/3 blah blah, and then license it in a deal with Imagination (for an SGX GPU), to other vendors. That would be a high margin business - the could compete directly with ARM. But they won't.