The only way MS can succeed in the mobile business is with x86

[OneOfTheseDays]

They will never, I repeat never, gain enough market share to adequately challenge Apple/Google and create the kind of development community they need to gain equal footing unless they can bring x86 to the phone world.

They are already doing this with Windows 8 and the Surface, and I predict the x86 Win8 tablets will absolutely dominate the RT versions and soon obsolete them. Having Intel at your side is a humongous boon for them, and Intel has the R&D chops to bring x86 down to the mobile space with the performance that not even ARM can match.

Once Haswell and co. arrives we will see a major shift in the tablet space, and Win8/x86 will dominate Android alternatives. Apple will always have it's place in the world, but I fully believe the opening is there to wipe out Google.

 

[wirednuts]

wipe out google? not likely

 

[dagamer34]

Do you Intel would have started putting it's serious A game forward if Windows RT didn't exist though? Do you think Atom would be getting multiple die shrinks in 2013 and 2014 as well as much improved graphics if it didn't have to compete with nVidia, Samsung, TI, and Qualcomm ARM chips? Do you think Haswell would be pushing into lower and lower power envelopes if it didn't sense ARM wanting to take some of its business in the low power server space?

Even if Windows RT completely fails because Intel builds a competent, low power, low cost x86 CPU, it will only because Microsoft and all of the ARM competitors pushed them there, because lord knows AMD doesn't fill that role anymore.

x86 on phones is a joke though, because Apple and Microsoft show no signs of adopting Intel chips, and Intel isn't yet price competitive enough for Android OEMs to take a serious look at what they are offering. Maybe next year.

 

[Puddle Jumper]

I agree that haswell will likely end the use of ARM in high end tablets, even quad core Cortex A15 doesn't have any chance of competing with it. That could be an issue for apple though as I would expect them to continue with their own SoC which would not be remotely competitive with Windows 8 tablets.

 

[bearxor]

I disagree. With the advancement of windows 8 as a platform, and the push Microsoft is giving Surface, I think we'll see a lot of apps cross compiled and since UI an navigation are similar between Windows 8 and Windows Phone 8, I think it will make an impact with people.

Maybe not a HUGE inpact, but I wouldn't be surprised if Microsoft is sitting at 8-12% WP8 market share this time next year, which is roughly 2-3x what they're at now I believe.

 

[lothar]

I disagree. With the advancement of windows 8 as a platform, and the push Microsoft is giving Surface, I think we'll see a lot of apps cross compiled and since UI an navigation are similar between Windows 8 and Windows Phone 8, I think it will make an impact with people.

Maybe not a HUGE inpact, but I wouldn't be surprised if Microsoft is sitting at 8-12% WP8 market share this time next year, which is roughly 2-3x what they're at now I believe.

Who is Microsoft going to steal market share from?

 

[dagamer34]

Who is Microsoft going to steal market share from?

Android? Do you think most people who buy Android phones do so because they like Android or because they are cheap? My bet is on the latter. Symbian, Blackberry, and Windows Mobile are also likely to disappear to be subsumed by the remaining 3.

We also need to remember that this market is still definitely growing and over 40-50% of people in the US do not own a smartphone. I don't think we'll see dumbphones continue to exist in 5 years at least in the developed world, so there's still a good number of people to attract to the Windows Phone platform.

 

[wirednuts]

i do agree that android is popular because its cheap.

windows might have a problem competing that way though. maybe not

 

[Dominato3r]

i do agree that android is popular because its cheap.

windows might have a problem competing that way though. maybe not

Is it?

With Android, there's a high chance you're paying Microsoft or someone else, along with dealing with other patent issues and other legal stuff.

With Windows, you pay the license fee, but you're also protected by Microsoft from any legal issue that may arise from the software end (ie: something like the bounce back and double tap deal). There's also the part where Microsoft deals with OS/software updates, and not you as an OEM.

 

[tfinch2]

Once the Intel machine gets rolling, I believe ARM is in serious trouble.

 

[jiffylube1024]

I think Windows RT is dead on arrival. It's a chicken and egg scenario. It's a nice platform but it has no Apps; why would you be a guinea pig for RT when you get get an Android or iOS tablet with way more Apps today?

Windows 8 Pro tablets, I agree should become very popular, especially after haswell comes out and allows for new, smaller hardware.

 

[Red Storm]

The advantage to the Win 8 store is that apps will appear across phones, tablets, laptops, and computers. That's a lot of exposure.

The downside is that it seems like a lot of app makers don't realize that not supporting something like keyboard input in your app is pretty darn stupid and aggravating when you browse for apps on a laptop or desktop (or with a tablet too).

 

[OneOfTheseDays]

x86 tablets will hands down be the most functional on the market, even besting the iPad. Once Intel gets Haswell out the door they will have the power savings and performance that is unmatched on the market.

 

[BenSkywalker]

Once the Intel machine gets rolling, I believe ARM is in serious trouble.

Absolutely, remember when Intel decided to take over non x86 computing with Itanium? Threw their billions at it and......oh wait.... bad point

Remember then Intel decided they were going to take over the ARM market with XSCale? Threw a billion at it and......oh wait.......bad point

Remember the time Intel decided to put ATi and nVidia out of the GPU business with Larrabee, threw several billions at it and.......oh wait.....bad point

The reality is when Intel steps out of the mainstream x86 market they have a history of failure. Intel is the best fabrication company on the planet, that happens to make x86 chips. Outside of their brilliant fabrication engineers, it isn't like they have proven themselves to be very formidable.

 

[OneOfTheseDays]

Absolutely, remember when Intel decided to take over non x86 computing with Itanium? Threw their billions at it and......oh wait.... bad point

Remember then Intel decided they were going to take over the ARM market with XSCale? Threw a billion at it and......oh wait.......bad point

Remember the time Intel decided to put ATi and nVidia out of the GPU business with Larrabee, threw several billions at it and.......oh wait.....bad point

The reality is when Intel steps out of the mainstream x86 market they have a history of failure. Intel is the best fabrication company on the planet, that happens to make x86 chips. Outside of their brilliant fabrication engineers, it isn't like they have proven themselves to be very formidable.

So how does any of that apply to x86 Windows 8? I have full confidence that they will be able to deliver a complete x86 platform that is as good, if not better, as anything ARM can muster in the coming years in the mobile world.

 

[tfinch2]

The reality is when Intel steps out of the mainstream x86 market they have a history of failure. Intel is the best fabrication company on the planet, that happens to make x86 chips. Outside of their brilliant fabrication engineers, it isn't like they have proven themselves to be very formidable.

That's the thing though. They are going to (or have already started with Medfield) attack mobile computing with x86. It will take some time to be successful, but it will happen.

I disagree that Android will be dethroned on the phone side. It will just be x86 Android.

 

[dagamer34]

So how does any of that apply to x86 Windows 8? I have full confidence that they will be able to deliver a complete x86 platform that is as good, if not better, as anything ARM can muster in the coming years in the mobile world.

Their true skill is semiconductor fabrication, not CPU or GPU design. Their 1-node fabrication advantage often hides their failures which were clearly visible during the Athlon 64 days. Even now where the real competition occurs, they only have 2 actual design wins in a Lenovo phone sold in China and the RAZR i, and no one talks about them as serious competitors in the market. And Intel's been pushing the Moorsetown/Medfield promise for the past four years. So forgive us if we are a bit skeptical.

Here's Anand's 5 year old article on Atom and what it was supposed to be: http://www.anandtech.com/show/2493/21

 

[podspi]

Absolutely, remember when Intel decided to take over non x86 computing with Itanium? Threw their billions at it and......oh wait.... bad point

Remember then Intel decided they were going to take over the ARM market with XSCale? Threw a billion at it and......oh wait.......bad point

Remember the time Intel decided to put ATi and nVidia out of the GPU business with Larrabee, threw several billions at it and.......oh wait.....bad point

The reality is when Intel steps out of the mainstream x86 market they have a history of failure. Intel is the best fabrication company on the planet, that happens to make x86 chips. Outside of their brilliant fabrication engineers, it isn't like they have proven themselves to be very formidable.

That is the beauty of their plan. Instead of inventing something new or creating a new market, they're just lowering the power envelope on high-speed x86 designs. I won't claim to know the future about whether Intel is going to wipe out ARM, MIPs, etc, but now that Atom is actually shipping in a smartphone (RAZR i), and doesn't suck, I think we need to acknowledge that the tired story of Intel only being able to do high-powered high-performance CPUs is not true.

That's the thing though. They are going to (or have already started with Medfield) attack mobile computing with x86. It will take some time to be successful, but it will happen.

I disagree that Android will be dethroned on the phone side. It will just be x86 Android.

Yea, Android was designed to be more-or-less hardware agnostic, and this will only get better with time. At the moment, the hardware really isn't to the point where you can abstract from native code if you require high-performance, but we're quickly getting to the point where you'll be able to for most applications.

Anyway, it looks like Metro apps are similarly designed so that they can be cross-compiled trivially, so I disagree that they need x86. As long as they can get the app ecosystem up and running, they should be more or less hardware agnostic, provided developers care enough to tick a few check boxes, and the dev tools are good enough.

 

[Geekbabe]

Once the Intel machine gets rolling, I believe ARM is in serious trouble.

QFT

Big changes are coming

 

[Genx87]

Dominate in what ways? Volume? I highly doubt it unless Microsoft prices one at sub 200 bucks.

 

[lothar]

Android? Do you think most people who buy Android phones do so because they like Android or because they are cheap? My bet is on the latter. Symbian, Blackberry, and Windows Mobile are also likely to disappear to be subsumed by the remaining 3.

We also need to remember that this market is still definitely growing and over 40-50% of people in the US do not own a smartphone. I don't think we'll see dumbphones continue to exist in 5 years at least in the developed world, so there's still a good number of people to attract to the Windows Phone platform.

WP8 can't compete with free.
Unlike Windows, Google doesn't charge licenses for the OS. The only company that can is Apple, and that's because they have their own niche customers.
How much does Microsoft charge for a WP8 license?
I heard sometime ago that they charge $95 for Windows RT tablet license, not sure about WP8 though.

That's true, but it won't make much of a difference.
Based on that chart, iOS shipments grew by a whopping 57%, but yet it only gained a 1% market share from that 57% increase in shipments.
For WP7/8 to go from 3.6% to 12% market share in a year, that's going to be a hell of a lot.

 

[Zink]

Intel may bring full laptop performance down to the tablet level in the future but ARM has managed to progress very quickly. This is the first time that we have had a 2x performance speedup from the iPhone two years in a row with 4 -> 4S -> 5. Hands on, an iPhone 5 doesn't feel that different from the using a 4S but it is a noticeably faster. By the time Haswell/Broadwell is competitive in high end tablets in 2014 we will have two more ARM generations and a die shrink hopefully bringing ARM 4x the performance we have today at very low prices that Intel might not be able to compete with. I don't think a product that feels a bit faster but costs $100 more is going to steal major market share and that may be all that Intel + MS brings.

It doesn't look like Intel is able to aggressively move into phone or tablet markets as we've seen with Atom phones so they will have to slowly gain market share. Windows RT looks like one more time they are coming behind because there seems to be a lack of Win8 Atom tablets to compete with ARM. Next gen Atom looks interesting but it doesn't seem that different from the Atom we have today because ARM will also have improved. Additionally, Windows RT is building momentum for the Windows platform away from x86. At first this will only be for consumer applications but in 10 years professional apps may have followed.

Intel already has Atom that is competitive with ARM so why can't I buy it? Maybe Intel is slowly building up to a chip that will be 2x faster than ARM and they will maintain that lead but I'm not sure if it would suddenly be scooped up into a ton of devices and steal market share.

 

[OneOfTheseDays]

Intel may bring full laptop performance down to the tablet level in the future but ARM has managed to progress very quickly. This is the first time that we have had a 2x performance speedup from the iPhone two years in a row with 4 -> 4S -> 5. Hands on, an iPhone 5 doesn't feel that different from the using a 4S but it is a noticeably faster. By the time Haswell/Broadwell is competitive in high end tablets in 2014 we will have two more ARM generations and a die shrink hopefully bringing ARM 4x the performance we have today at very low prices that Intel might not be able to compete with. I don't think a product that feels a bit faster but costs $100 more is going to steal major market share and that may be all that Intel + MS brings.

It doesn't look like Intel is able to aggressively move into phone or tablet markets as we've seen with Atom phones so they will have to slowly gain market share. Windows RT looks like on more time they are coming behind because there seems to be a lack of Win8 Atom tablets to compete with ARM. Next gen Atom looks interesting but it doesn't seem that different from the Atom we have today because ARM will also have improved. Additionally, Windows RT is building momentum for the Windows platform away from x86. At first this will only be for consumer applications but in 10 years professional apps may have followed.

Intel already has Atom that is competitive with ARM so why can't I buy it? Maybe Intel is slowly building up to a chip that will be 2x faster than ARM and they will maintain that lead but I'm not sure if it would suddenly be scooped up into a ton of devices and steal market share.

With the downfall of AMD, I don't believe the pressures are there for Intel to have to spend vast R&D resources on the desktop platform side. All of the innovation and growth is coming from the mobile space.

I fully believe once Intel, and I think they already have, dedicates the brunt of their R&D resources towards mobile x86 solutions they will obliterate anything ARM can put out on the market. They have the software partner they need in Microsoft to really revolutionize the tablet world IMHO.

 

[dagamer34]

With the downfall of AMD, I don't believe the pressures are there for Intel to have to spend vast R&D resources on the desktop platform side. All of the innovation and growth is coming from the mobile space.

I fully believe once Intel, and I think they already have, dedicates the brunt of their R&D resources towards mobile x86 solutions they will obliterate anything ARM can put out on the market. They have the software partner they need in Microsoft to really revolutionize the tablet world IMHO.

Intel can make the chips, but OEMs have to be willing to use them. Apple won't be using them, and Microsoft has shown no sign of supporting x86 in Windows Phone 8. Samsung already designs and fabs its own mobile chips so I see little reason for them to go with Intel. With a huge chunk of the market gone, who's left? Motorola? HTC? Sony? LG? I don't see Intel just walking all over Qualcomm, especially when they build the mobile basebands and radio transcievers which go into a HUGE chunk of all cell phones sold today.

 

[tommo123]

With the downfall of AMD, I don't believe the pressures are there for Intel to have to spend vast R&D resources on the desktop platform side. All of the innovation and growth is coming from the mobile space.

I fully believe once Intel, and I think they already have, dedicates the brunt of their R&D resources towards mobile x86 solutions they will obliterate anything ARM can put out on the market. They have the software partner they need in Microsoft to really revolutionize the tablet world IMHO.

not just microsoft - google too with android. weren't intel working on x86 optimizations for android?

i don't think intel is going to just back microsoft. i mean the wintel days are gone and microsoft has clearly hedged their bets with ARM. if intel has another conroe moment in the next 2-3 years, then every smartphone/tablet bar the very low end ones might run on intel chips - even iphones/ipads. apple would have to move back to intel wouldn't they? i mean assuming the mobile atoms by then obliterate the ARM cpus.

interesting times and not bad for the desktop either. we'll get improvements just geared towards low power which isn't a bad thing. if only GPU makers had this kind of competition from the low power side of things then maybe we could have the same kind of improvements in efficiency on that end of things too