Intel CEO: 'We're porting Android 3.0 for tablets this year' Intel's president and chief executive Paul Otellini says his company is hard at work porting Google's tablet-specific Android 3.0, aka Honeycomb, to the x86 architecture. "We've received the Android code – the Honeycomb version of Android source code – from Google, and we're actively doing the port on that," Otellini told …
Unfortunately for Intel if the tablet can run Windows, no PC OEM will ship a credible Android tablet with it. They will once again promise the world and not deliver. Or worse, they'll ship a few lame tablets to try and poison the market. So the big pc guys are right out. They will ship it with Windows, with the results we have seen for fifteen years: poor sales. They will be "persuaded" to do this.
Then there are Samsung, moto, and the other phone guys. They don't need Intel's chip.
Who's left? The small fry who can't get the cost down to compete.
I wish Intel luck, but I don't see this succeeding as anything exept an extrapeneurial venture. Maybe there's somebody in the consumer space who could pull it off but I don't know who.
Suggestions anyone?
Why would a manufacturer not build a tablet with Android, if the market wants it? I think your faith in MSTs power is a little out of date, unless you are Nokia of course!
They might build any because the ARM is a better solution for Android on a tablet, but if the market exists "they will come".
The answer is simple.
OEM price for Windows exclusivity on hardware is X.
The price for non-exclusivity is Y.
Where Y >>> X.
Now, question, why do OEMs want (need?) to ship Windows? Well, the answer is easy. That's what non-techie Joe McPunter knows. Sven Chubbleson and his designer pals know about Macs, true. But Joe McPunter lives in a Windows world. He uses IE6, Word etc at work, plays MMOs with his friends etc. All these (like it or not) demand Windows and as far as he knows, Windows is it.
So if HP (or someone) sell him a PC that cannot run IE, Word, MegaKill 4000 Ultra-Violence Edition; then he will return it as "broken" rather than try to learn a new system.
Even if that system is superior in many ways.
You can argue until you are blue in the face about how "wrong" all this is and I will agree with many of your points, but that is the way the world is. Linux is at least a decade away from being on the desktop, there's still too much Windows-only entrenchment and momentum.
This is close to right. They're not allowed to do it this way any more, of course.
So they work around that by paying "Co-marketing" dollars. OEMs get money to put the sticker on, and more money to push Windows to the fore in the ads, and more money to put "X recommends Windows 7" on every page of their website, including the one where you select your Linux version. Naturally because margins are razor thin this can be the difference between large points of market share on desktop and laptop Windows boxen that are the bulk of the market.
And coincidentally if you launch a product that will run Windows with anything but Windows on it, difficulties ensue. But it's not _directly_ related. It's just a remarkable coincindence. That all your customers get calls from their Microsoft support reps about every service mishap is just them being mindful of their customer. Ominous warnings about your future products and prospects during their site visit elevator chats is just good account management. Derision and deprecation when Redmond reps visit your mutual integrator partners is just personal opinion. No court could find fault in a properly toned "Oh?", the curl of a lip, the diversion of conversation from your future products as if you were doomed to a fate best not spoken of.
Of course if you get real stiff with the Redmond gang they'll unleash their New Media group on you. God help you then. You'll be getting hate-tweeted by the bloggerati from the arctic to the antarctic, maximizing the downside of every communication and issue - and inventing issues when there are none. The negative articles by well-named authors will be posted, syndicated, rewritten and auto-rewritten in every language around the globe - and then be paraphrased and sent 'round again to keep the negative impressions going for months. But there will be no proof. Best not to risk it if you have a lot to lose.
considering all the emulators are basically already x86 android versions, the linux kernel runs fine on x86 and already has device drivers, and the important bits of android are in java then 'this year' is a rather pessimistic estimate. 'this month' maybe, unless their programmers are contractors paid by the hour, in which case they'll spin it as long as they can.
But didnt intel sell their Arm division a few years back? Sure that ranks as one of their most spectacular errors? Maybe even worse than getting complacent around the P4 vs opteron a few years back.
Why are the completely incapable of accepting non-x86 CPUs? No itanic doesn't count.
There's a lot of cruft in tablets, I don't mean crappy tablets (which there are plenty of) I mean in feature cruft.
For example, I don't need 3G, don't need GPS, don't need a camera (front facing or back facing) or vast amounts of storage. I simply want a sub £200 wifi capable honeycomb tablet that I can choose to expand the storage capacity with an sd card.
Look at the motorola xoom as a prime example of overpriced and stuffed with pointless hardware features (which increase its cost).
Make a sub £200 honeycomb tablet that is 1. 9 or 10 inch capacitive touch screen 2. 1GHz or better ARMv7 processor 3. Around 1GB of RAM 4. If necessary for running apps 1GB of storage capacity 5. An SD card slot for storage expansion 6. Wifi capable. 7. Half decent graphics chip, doesn't have to be a powerhouse tegra or anything. There, push that for £200 or less and it's a sale.
I think Nokia is no more trustable for anything that can compete with Microsoft and they are still claiming to support Meego while not even shipping latest Qt to it.
I think Intel and Linux foundation should get rid of Nokia in that real Linux for smart appliances either in polite way or legal way.
They have a company who left Android porting for a such important framework (qt) to 16 year old, single person. It is not a co-incidence and Intel should fix the real OS and the real problem first.
Android should have a real competitor suitable for OEMs, it will serve to Android too.
I think, instead of buying that Av vendor with horrible image, they should at least add qt to shopping list too and stay behind its Android port as Intel giant with trustable roadmaps and sync releases.
Nokia should be kicked out from anything non Windows. We don't know what else that $1B covers.
This is a futile and pointless exercise by Intel, now they have realised that tablets are popular and that they tend to ship mostly with no Intel components in them.
Looks like a desperate attempt to stop the inevitable slide as people ditch their x86-architecture museum relics for inceasingly-powerful smartphones, tablets and netbooks.
It wouldn't surprise me if PCs and laptops died out in the next few years, replaced by tablets and hybrid thingies like the Motorola Atrix. People will carry a powerful computer in their pockets and simply dock it with a larger screen, keyboard and mouse when the need arises.
'Bye Intel and Windows - it was fun (?) while it lasted......
Just as I think you'd be insane to use an x86 in a phone, I also don't see why you wouldn't use an Atom-class processor in tablet devices: they've got room for big batteries, and they've got big screens. Plus, many of these "tablet" designs will end up used in point-of-sale or information systems, where battery performance is not as much of an issue.
ARM can only go so far before it runs out of steam - it's not a question of technical excellence, but more of what the market wants: phones and tablets are at the high end of ARMs applications, and until OEMs start making ARM desktops, Intel will have better products for the "tablets to workstations" end of the market even with a horrible instruction architecture.
If you add in the better code generation from gcc for x86 than ARM, and the much better Java performance on x86 than any other architecture, Android/x86 makes sense.
Having MeeGo available on the same hardware may not make Android look too sharp, but it'd be nice to be able to compare on a like-for-like basis.
Big Batteries = More expensive, heavier, longer charging.
For what? So you can get the full "Windows Experience"?
What is it about Windows that would enhance the use of a tablet exactly?
Why do you think that a cut down x86 would perform better in a tablet than a made for purpose ARM cpu?
I never mentioned Windows - that would be stupid (although someone would do it).
ARM devices top out around about where Intel gets going. If you want more of a laptop-featured device, and can live with the higher power consumption, it becomes viable to consider Intel's parts. I do think ARM is a nicer design than x86 (okay, ANYTHING is nicer than x86, but ARM is especially nice), but that's not the point.
I'd argue that tablets, being less portable, are not under the sever power constraint that mobiles are, but that's not all: the big advantage of Intel is in driver support. Every peripheral under the sun probably has an x86 driver written for it, many are Open Source. Not so for ARM.
If you think of how many of these "tablet" designs will end up stuck in kiosks and other mains-powered fixtures, the speed of bring-up becomes more important than power budget.
Look, I'm not an iFanboi and haven't been since the early '80s. There's only one Apple product in my house: an iPod touch I got my son a couple Christmases ago that I've only seen twice. I have to use their stuff at work once in a while, so I know a bit about it, but it's not a big part of my gig at all. For myself I prefer to get my BSD closer to the root of the tree. I do Windows a lot more at work and despise it. My personal gear runs Linux because I expect to be the only master of my gear and own utterly the data I put in it. iFans need not feel disparaged here: I wouldn't say their way is bad; just that it's not my preference.
Having said that though, I have to give Apple a lot of credit. Apple ships products. When Apple tells the public about a product, they do it by inviting the press to a product release party. There they describe the product, then they show the product, they they tell you how and when you can get the product - and it's not some mockup product under glass you can't touch that might be available over a year from that day. The press gets to fondle it immediately after. It's produced in millions quantity and available to the common folk inside of 90 days, if you're willing to queue up for it.
Details of future Apple products don't often leak out. You don't often get pictures of complete Apple products months in advance. You never see CAD mockups. They certainly don't give announcements in the press half a year in advance of individual components that might or might not be a useful part of a complete system someday.
When Apple tells us about stuff, it's real stuff. So we put more credibility in what they say. Although I don't personally care for the complete ecosystem vision they have for their products - their "cathedral" if you will - I have to say I have great admiration and respect for how they handle their communications.
How this is relevant to the article is left as an exercise.
Intel has yet to deliver its first significant win in this arena. ARM owns the space.
Or rather, it doesn't. And that's its great strength.
Adopting a licensing business model has given ARM two dimensions of innovation - its own relentless drive to deliver processing throughput and power efficiency, and its licensors' packaging of that processor technology directly into single chip systems and package-on-package combinations.
That produces cost and time-to-market advantages, and fosters competition that benefits OEMs, who have confidence in the entire ecosystem, in which ARM enables the shots rather than calling them. The diseconomies facing OEMs thinking about adopting Intel are substantial.
That's not to say that ARM's position is impregnable. Intel's business model gives it many dollars per chip sold, where ARM collects just a few cents. It has breathtaking cash generation, deriving from the markets it dominates and the gross margin position it harvests. That war chest enables all kinds of attacks on the market it must succeed in if it's to maintain anything like its present position as the world goes mobile.
They must succeed. But that doesn't mean they will. And they are in the unenviable position of seeing both their star weapon - the wealth of software written first for x86 - degrading so unexpectedly rapidly, and of finding themselves facing a new wave of competition in Windows and server markets.
Half a league, half a league, half a league onward. It's a magnificent warhorse, and you have to admire the cavalry charge, but just look at the heavy artillery on every side.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
