These Companies Will Profit from Intel's Demise

On Tuesday, a monumental piece of news leaked out from the technology world, and in response, people merely sighed. The news amounts to nothing less than the final piece of the puzzle needed to shatter "Wintel," the combination of Microsoft (Nasdaq: MSFT  ) Windows operating systems and Intel (Nasdaq: INTC  ) processors that came to define the PC era of computers.

The seeds of Wintel destruction
Weeks ago, a colleague had e-mailed me asking my opinions on Intel. Much like other tech heavies, the stock looks downright cheap. Intel is a cash machine, spinning off more than $10 billion a year in free cash flow. The company also maintains nearly $21 billion in cash sitting ready on its balance sheet.

Intel's value is largely impaired thanks to investor fears over tablets and smartphones, an area in which the company isn't currently competitive. But investors who believe the near-term threat from these devices is exaggerated are largely correct. Smartphones and tablets combine to make a pretty big wave, but they're mainly hitting netbooks. However, Intel wasn't getting stellar revenues from its Atom line that's found in netbooks to begin with, so this loss is overblown. Add on strength in businesses refreshing PCs and servers, and Intel should be fine in the short run.

However, things do look a bit more ominous in the long term. I told my colleague to be on the lookout for the shattering of Wintel, which could happen if Microsoft announced a version of Windows that's compatible with the ARM architecture championed by ARM Holdings (Nasdaq: ARMH  ) .

Why Wintel matters
People might be familiar with the term "Wintel" but unfamiliar with the implications of the term and the way it has enabled Intel's near-monopoly dominance. Essentially, consumer versions of Windows, the dominant operating system of this generation, have worked only on x86 processors, the architecture championed by Intel. The term initially referred to systems bundling Intel processors and the Microsoft operating system, but today it also includes processors from Advanced Micro Devices (NYSE: AMD  ) , which are also based on Intel's x86 architecture.

This "Wintel" combination has largely held up through the last generation of computing. It has left Intel as the titan of semiconductors while smaller firms engaged in pricing bloodbaths and brutal cyclical booms and busts that left their value in the gutters. However, Wintel now faces its greatest threat.

Meet ARM, bringer of death
On Tuesday, reports from Bloomberg leaked out that Microsoft was developing a version of Windows for ARM-based processors. That means that instead of just working on smartphones and tablets, ARM processors can now be put in everyday laptops running Windows. That opens up a whole segment of consumer spending for ARM-based processors.

The news was met with minimal fanfare; it's long been known that Intel and ARM were gunning for one another. However, investors failed to realize the broader implication of this move. Since ARM took a dominant market share in the mobile space, the question has been whether Intel could encroach on ARM's space. Till now, we've seen Intel on the offensive with ARM preparing to defend itself -- but now, thanks to Microsoft, ARM has a beachhead on the PCs we use in our everyday life. ARM is officially on the offensive.

ARM: more than meets the eye
This news is apt to be shrugged off. People will point out that the power of Intel's chips are vastly superior to ARM processors and that ARM's simply too underpowered for the needs of modern computer users. However, that line of thinking should be proved wrong for a couple reasons.

  1. Users no longer place the same premium on horsepower: Some of the hottest computing trends of the past year feature systems that were initially panned as "underpowered." Given the trade-off between having a slightly underpowered laptop and vastly superior battery life, today's computer user may be more inclined to take the option with better battery life.  Also, although ARM might be behind today, it's quickly bridging the performance gap to a level that would prove acceptable in today's laptops.
  2. ARM features a powerful alliance backing it: ARM doesn't actually produce the processors seen in your phone; rather, it licenses the technology behind them. Instead, companies such as Qualcomm (Nasdaq: QCOM  ) , Apple (Nasdaq: AAPL  ) , Marvell, Texas Instruments, Samsung, and NVIDIA (Nasdaq: NVDA  ) have placed large amounts of investment behind the architecture. Thanks to its inability to crack the mobile world, Intel has allowed a powerful grouping of companies allied behind its rival technology to emerge.

This shift should come sooner than users expect. The example of Intel's relationship with Apple proves especially illustrative in this respect.

Thanks in part to a dispute where Intel has been bullying NVIDIA around, Apple has continued to use ancient Core 2 Duo processors on its Macbook Air line. That's an old processor that should have long ago been replaced, yet in spite of this limitation, the Macbook Air is wildly popular.

What does that foretell? As we shift to a cloud-based world where most of our computer usage is browser-based, people care less and less about the processor! With Apple pouring hundreds of millions into developing its ARM-based A4 processor, what's to stop that company from programming a new version of OS X to run on ARM-based processors and later releasing a Macbook Air using its own A4 processor? Within a few Macbook refreshes, Intel could find itself in serious trouble.

The bottom line
Intel is facing what I believe to be its greatest challenge yet. With Wintel shattering, Microsoft comes out comparatively better. Users will still want the operating system they're used to, but they have less incentive for loyalty toward the unseen processor powering their computer. If competition is open to a vast array of companies producing cheap processors for Windows thanks to its new ARM-enabled version, that hurts Intel's pricing power at the very least. Intel's long-standing competitive advantage of being the leader in the only architecture for Windows is destroyed. Make no mistake: Prices will fall.

When in three years you look at your next laptop and you see a sticker that reads "Qualcomm Inside," rest assured that the seeds for that shift were planted this week.

A huge opportunity awaits the companies producing ARM-based chips, and a dangerous future that's not fully appreciated by many investors awaits Intel. My recommendation on how to invest in this news shouldn't come as a surprise.

If you're looking for other companies at the forefront of growth trends, The Motley Fool has created a brand new free report called "The Motley Fool's Top Stock for 2011." In it, we reveal the little company set to profit from the broadband Internet expansion. Get instant access by clicking here -- it's free.

Eric Bleeker owns shares of no companies listed above. Intel and Microsoft are Motley Fool Inside Value choices. Apple and NVIDIA are Motley Fool Stock Advisor recommendations. The Fool owns shares of and has bought calls on Intel. Motley Fool Options has recommended buying calls on Intel. Motley Fool Options has recommended a diagonal call position on Microsoft. The Fool owns shares of Apple, Microsoft, Texas Instruments, and Qualcomm. Try any of our Foolish newsletter services free for 30 days. We Fools don't all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. The Motley Fool has a disclosure policy.


Read/Post Comments (27) | Recommend This Article (47)

Comments from our Foolish Readers

Help us keep this a respectfully Foolish area! This is a place for our readers to discuss, debate, and learn more about the Foolish investing topic you read about above. Help us keep it clean and safe. If you believe a comment is abusive or otherwise violates our Fool's Rules, please report it via the Report this Comment Report this Comment icon found on every comment.

  • Report this Comment On December 24, 2010, at 4:46 PM, techy46 wrote:

    Blah, Blah, Blah. Intel's building out SoC (System on Chip) to match Apple's A4 and scaling down from 45nm to 22nm while proliferating Atom chps across mobile devices. Probably room for competitors as there always has been in niches but Intel's demise? You gguys publish some many misleading crisis headlines I'm glad I've got enought knowledge (42 years system development) to know or find out what's the truth. Unfortunately, a lot of idiots will rush to trade on your misinformation.

  • Report this Comment On December 24, 2010, at 4:47 PM, ConstableOdo wrote:

    I doubt if Intel will be in any trouble. Microsoft will continue to crank out Windows for Intel-powered machines until the end of time. I think most of the Intel machines will be ones owned by corporations. I don't think consumers will have any problem at all leaving the Wintel platform. I doubt the average consumer even cares what platform they're using as long as they can get what they need done.

    There should continue be a fair demand for powerful Intel multi-core processors as long as Windows dominates the computer industry. They'll continue to need powerful processors in order to run that bloated Windows and Microsoft Office of yesteryear at a decent clip. Besides, corporations are chock full of certified Windows-based IT managers that won't want to lose their livelihood and will try to hinder the changeover at any cost. And there's still those gaming nerds who require nearly the power of a supercomputer just to get high framerates for life-like graphics.

    I like those fast and powerful Intel multi-core processors. They're just like the supercars of the past. Full of horsepower and guzzling gas like it was water. However, the industry needs to shift to more efficient hardware. Small ARM-powered desktop boxes replacing those humongous Intel floor models. Everything you need for computing in the size of a Mac Mini or just built into a thin display. Think of the electricity savings over a couple of years.

    Anyway, I'm sure Intel will be able to build power-efficient chips if they want to but undoubtedly, they're going to have to make some serious changes in their current processor roadmap. I think it's time for consumers to move past Windows and power-hungry processors.

  • Report this Comment On December 24, 2010, at 5:56 PM, TMFRhino wrote:

    Techy46,

    No worries. I'm sure your well constructed critique of how Intel will be able to maintain its premium pricing in the face of added competition will surely sway readers away from "idiots" like me.

    ConstableOdo,

    Its true they could very well withstand this threat like others before, however, even if we see Windows for Intel chips until the end of time (very well could), without the "Wintel" advantage their selling prices could fall taking their stock price down with it. That's really the crux of this situation, in my opinion.

    Fool on!

    -Eric

  • Report this Comment On December 24, 2010, at 8:53 PM, ems52082 wrote:

    Let me get this straight, people are still doubting the strength of ARMH which has appreciated like 400% in the past few years. Are you insane to doubt this company will continue to rise. The extremely high PE ratio was placed on the stock before the public still even realizes what ARMH does and before Microsoft entered the picture. ARMH will double and Intel will give you the same dividend it always has with zero growth. I've placed my bets. ARMH is the future if you believe it isn't stop buying so many tablets I'm sick of people making my stock go up so much :) boooyahhh!!!

  • Report this Comment On December 24, 2010, at 9:02 PM, mytk7 wrote:

    i think there will be room for everybody since the chip type of market is just keep on growing with more creative

    way to use electronics.

    That being said the 'threats' that the author see is too far , 5-6 years, and not established as Intel is not being idle,

    secondly i would prefer more computer power that i can run my applications well any day of the week than another some more battery life,

    i can always find an outlet to work and charge my battery.

    Knowing Microsoft it will take several years until they come with a decent o.s. for tablets.

    The hype behinde the iPad is overblown, i went to Best buy today, people look at it but no body takes it seriously.

  • Report this Comment On December 24, 2010, at 9:09 PM, techy46 wrote:

    ERic - The correct title for your article should be "These companies WOULD profit from Intel's Demise". You should then offer both sides of the equation instead of a prejudiced rant of anti-Intel FUD. All of the ARM licensees pay about $0.11 per core to use ARM's IP. Intel's was licensee but opted to use their IP and investment in 22nm fab rather than add a 2nd line of processors. Wp7 (phones) already runs on Snapdragon, Wt7 (tablets) will shortly and Intel can produce plenty of cores at 45-22nm to compete with lower cost designs and that's why they've licensed TSMC to produce those low power chips. Lastly, being constantly connected and dependent on the "cloud" of Apple or Google servers doesn't make sense to a lot of serious personal computer users like it does browsers and pedestrians.

  • Report this Comment On December 24, 2010, at 9:49 PM, predfern wrote:

    If ARMH and QCOM are such great stocks, then why are they not Rule Breakers picks?

  • Report this Comment On December 25, 2010, at 7:28 AM, WunMoshen wrote:

    Merry Christmas fools!

    I usually try to be open minded about these things and not respond because **** happens.

    I am going to keep this basic. Lets be real; At one point people felt like the net book was going to steal the show from the laptop which proved to be wrong.

    Microsoft decides to develop Windows around Arm holdings processor; I don't see any threat towards Intel UNLESS 75% of the software developers rewrite & optimize their applications in a way where it renders multiple cores irrelevant or obsolete.

    (You know what I'm talking about ?) I'll tell you what Im talking about.

    Performing task(s) on a meager 1 Ghz processor that would usually require a minimum of 2.5 Ghz on dual core architecture, which I doubt very seriously will happen.

    But hey I could be wrong...

  • Report this Comment On December 25, 2010, at 11:45 AM, rav55 wrote:

    x86-64 instruction set allows software and pc's to design the world. ARM which is 32 bit and is basically a cheap low performance cpu has NO SUCH capability.

    Yes netbooks are the next big thing but just what are they? They are Internet Connection appliances and that is all.

    Netbooks (and down) can not run Autocad, Catia, Matcad, or any of the multithreaded capable 32 and 64 bit capable design packages.

    Netbooks and other mobile appliances are great for email, showing off pictures but they CAN NOT PROCESS.

    And yes they do not perform any where near as well as x86-64 cpu's and now APU's.

    The net book is best described as either using the "low end" of the performance niche with x86-64 ala AMD Fusion or barely the high end performance of ARM. But since ARM will never run x86-64 that will be the limit for it's capability. Faster means more energy gets used. So if it runs faster and burns up more energy then what is gained against x86-64?

    In fact Windows 8 will support 128 bit computing which places the ARM Cortex well at the bottom of the heap. AND AMD Bulldog just might as well be called the first 128 bit cpu.

    Do we need 128 bit computing? Did we need 64? 32?

    It is an x86-64 bit world out there. Low performing appliances powered by ARM need not apply.

  • Report this Comment On December 25, 2010, at 11:58 AM, motsig wrote:

    This is really nothing new. The argument of RISC vs. CISC based processors for windows has been going on for many years. Now you have this new spin of it being ARM.

    Intel (x86) architecture is a complex instruction set architecture. That means it can do many things efficiently in one clock cycle that the ARM will take many cycles to accomplish.

    Take a look at how effective your cell phone or iPAD are at even the most basic web browsing when you have integrated media. Now, try to do some complex tasks on these devices.

    Intel/AMD are not at risc today (pun intented). As ARM scales up their capabilities AMD/Intel will scale up their integration as their doing with their new APU architectures.

    Forget about it.

  • Report this Comment On December 25, 2010, at 3:46 PM, NOTvuffett wrote:

    ARM microprocessors not in the same league as Intel's. Apples and oranges. Rav55 has it right. AMD only serious competitor for high power cpus.

    When IBM selected the Intel x86 chips I was disappointed. The Motorola 64000 was much better. I think computer technology would be further along if they had.

  • Report this Comment On December 25, 2010, at 7:01 PM, techy46 wrote:

    Title for next article "These Companies Will Profit from Steve Jobs Demise"

  • Report this Comment On December 26, 2010, at 11:39 AM, rav55 wrote:

    I do not understand WHY anyone believes that the ARM Cortex cpu can threaten Intel, AMD or Microsoft.

    The ARM cpu and risc instruction set is at best a lawn mower engine. The latest Intel and AMD offerings are best compared to Detroit Iron. Lawn mow engines cut your grass. Detroit iron gets you where you want to go quickly and in style! Okayl so you burn some gas doin it.

    ARM just is not in the same league as x86 and never will be. But Microsoft sees a threat to it's O/S (monopoly?) and for some reason seems to think that users would prefer an Android netbook that can't run MS Office.

    Now you can putt-putt anywhere you want on a gallon of gas using ARM. But if you want to get there faster then you need more performance. The RISC instruction set just is not up to it. Especially if it is run on an anemic little cpu.

    The same people who believe that ARM means the downfall of Intel would also have you believe the Tegra II will solve Nvidia's problems.

    Tegra II is not even Nvidia's technology it is a licensed ARM cortex cpu!! Tegra II is ARM wearing a new dress.

    If you do not understand history then you will be doomed to repeat it. Apple was an independent and proprietary OEM. Apple tried to succeed against x86 and x86-32 and they failed. So Steve Jobs rewrote Mac O/S to run using x86-32 and x86-64 instructions and now Apple is a huge competitor to Microsoft. And the world is responding. AutoDesk just released AutoCAD to run on Mac O/S!!

    That will NEVER happen with ARM or Tegra.

  • Report this Comment On December 26, 2010, at 7:39 PM, TheBlindCat wrote:

    @buyaluminum - "It is like trying to take on the U.S. Army and Navy! Just forget it, beat it!"

    I believe that is the same thing they said about the Spanish Armada. How ironic that those crafty Brits are behind this "giant killer"?

  • Report this Comment On December 27, 2010, at 9:31 AM, marceloavai wrote:

    Intel/AMD are not at risc today (pun intented). As ARM scales up their capabilities AMD/Intel will scale up their integration as their doing with their new APU architectures.

    Forget about it.

    http://www.avilas.com.br

  • Report this Comment On December 27, 2010, at 9:57 AM, rabinnh wrote:

    Wow, where to begin. It's just not that simple.

    Question 1; is Windows a good brand now? I don't think so. In fact, most college kids I know have had it. Windows has caused them to lose work and miss deadlines. All of them want Macs or (gasp) Linux.

    Also, I know that the predictions for Linux on the desktop have been happening for 10 years, but distros like Ubuntu are really easy to use now, and just like in 2005 when I first started seeing executives carrying Macbooks, I'm starting to see regular people using Ubuntu. 90% of corporate Windows users use Office and the Internet. How long before corporations realize that Linux and OpenOffice will suffice for those users and that they can save $millions (I should know. For last 15 years I have worked for some of the leading Systems Management vendors in world capitalizing on the difficulties of managing corporate Windows desktops. Thanks, Microsoft).

    And let's say Microsoft releases Windows for the ARM tomorrow (BTW, Windows CE has been on the ARM for years, so it's not exactly a huge leap), it's rather useless until all the applications vendors decide that it's profitable to release ARM versions of their software. Until then you have Windows running . . . Windows.

    And lastly, in spite of the hype, the trend really is to move user applications to the Internet (the much hyped "Cloud"). And what do you need in the data center? Big honking servers with powerful multicore processors capable of virtualizing a lot of machines in a small space that are reasonably energy efficient. And who leads by a mile in that market? None other than Intel.

    There is an entire ecosystem out there, and saying that Intel is doomed because Microsoft will release Windows on the ARM is simpleminded at best. It assumes that Windows desktops are the be all/end all of technology forever, and we can already see that that is not the case.

  • Report this Comment On December 27, 2010, at 3:22 PM, jav1231 wrote:

    The primary problem with this article is that the author knows very little about the IT industry. I recall listening to a stocks program about 4-5 years ago where the pundit said to a caller, "No don't invest in AMD. They're a new-comer to the chip market and it's going to be hard for them to compete with Intel." I believe that was the Motley Fool show. AMD had/has been making chips for decades. Go crack open an old Compaq clone and you'll likely find AMD chips on network adapters and other circuitry.

    Now they're hailing the age of ARM. The only problem is that ARM is not sufficiently powerful enough to reap the entire PC market. It's just not going to happen. Someone has to program all those nifty games and apps that are squeezed into the ARM's architecture. ARM doesn't have that power. Could it in the future? Perhaps. Anything is possible. But Intel could just as easily take off with yet another processor. It doesn't matter. ARM will not tip Intel. The only people who believe such things are people outside looking in or people inside hoping to profit from such hype. Both are wrong.

    That's not to say ARM won't take off. That's not to say there's a market for them. But to say they're Intel killers is insanity at most and ignorance at least. I'll remind you that tablet PC's were going to rule the roost back in the early 2000's. It took a new processor and the public's acceptance as an augment to their PC to make it finally happen. (Actually, it took Apple!) Netbooks were going to make peace and love fall from the sky like pure snowflakes. Uh-huh....now ARM is going to kill Intel. Careful, Motley, your "Fool" is showing.

  • Report this Comment On December 27, 2010, at 3:49 PM, TheBlindCat wrote:

    @jav1231

    "The primary problem with this article is that the author knows very little about the IT industry."

    Unfortunately, IT guys in general know very little about what makes a company a good investment.

    I too was once naive enough to think because a company made great products and was a wonderful place to work (thinking Sun Microsystems) that that translated into a good investment.

    A quick look at AMD's stock performance will show you that, long term, it has proved to be a very poor investment.

    What Eric has correctly pointed out, is the fact that ARM will eat into Intel/AMD's margins, it is inevitable.

    ARM is already ubiquitous everywhere outside of mainstream PCs, laptops and servers and even this is changing as ARM is aggressively encroaching into all those traditional x86 environments.

    If you are only considering traditional PCs, I can see how you might think Intel and AMD are the big fish, but spend a little time looking at sales trends and you will reach the conclusion that PC's are becoming a smaller and smaller piece of the computing pie.

    The vast majority of the worlds ~7Billion people simply don't need anything more robust than a modern phone or tablet. Frankly speaking your typical office worker can get by with Email, Word Processing and a spreadsheet.

    ARM has reached the threshold where basic computing can be provided more cheaply from both an initial investment cost and subsequent energy consumption.

    The energy consumption/heat dissipation issues are also why companies are looking at ARM for basic server infrastructure.

  • Report this Comment On December 30, 2010, at 8:22 AM, rw1270 wrote:

    I think what people totally miss in this conversation is SERVERS!!!! All those milions of gadgets that are powered by whatever need cloud to operate. Guess who will make chips for those servers.

    Also, those "PC is dead" people simply forget that tablets have great potential to bite into low/mid end laptops, but there is still sizable desktop and high end lapto market for enterprise market and that one is not going away any time soon (and investments were quite frozen). If anything, that market will start growing again as a simple replacement cycle.

    I think all those prediction for PC demise are simply looking at one segment of the market and proclaim the rest doesn't matter. So far it looks the market agrees (Intel's valuation is downright stupid if you don't think it will experience crash in sales). I think Mr. Market will change his mind in next year, but what do I know...

  • Report this Comment On December 30, 2010, at 6:24 PM, rav55 wrote:

    "AMD readies next-generation netbook/tablet PC platform"

    http://www.digitimes.com/news/a20101229PD214.html

    This IS HUGE. And unexpected. It is also an Atom killer and of course Tegra II is just a little underpowered. Krishna was not expected until 2012.

    Oh yeah and foul weather was responsible for sinking maybe 24 of 50 or so ships lost of the original 151 ships of the Spanish fleet and weather forced the withdrawal of Spanish forces.

  • Report this Comment On December 31, 2010, at 1:44 PM, ffbj wrote:

    To paraphrase the Death of Intel has been greatly exaggerated. Having been a laggard this year it might actually be a good time to buy it.

  • Report this Comment On December 31, 2010, at 7:54 PM, TheBlindCat wrote:

    @rav55 - It doesn't really count if you lose ships to the weather while retreating ... you still got your arse kicked.

    "The Armada achieved its first goal and anchored outside Gravelines, at the coastal border area between France and the Spanish Netherlands. While awaiting communications from Parma's army, it was driven from its anchorage by an English fire ship attack, and in the ensuing naval battle at Gravelines the Spanish were forced to abandon their rendezvous with Parma's army.

    The Armada managed to regroup and withdraw north, with the English fleet harrying it for some distance up the east coast of England. A return voyage to Spain was plotted, and the fleet sailed north of Scotland, into the Atlantic and past Ireland, but severe storms disrupted the fleet's course. More than 24 vessels were wrecked on the north and western coasts of Ireland, with the survivors having to seek refuge in Scotland. Of the fleet's initial complement, about 50 vessels failed to make it back to Spain. "

  • Report this Comment On December 31, 2010, at 9:36 PM, TheBlindCat wrote:

    "I think what people totally miss in this conversation is SERVERS!!!!"

    I don't think this was missed at all. The suggestion is that Server chips are under the same pressure from lower cost alternatives that will reduce margins (if not volume).

    That said, the number of servers is dwarfed by the number of client devices.

    To put things in perspective:

    Stack Overflow uses 4 servers to handle the following traffic:

    -16 million page views a month

    -3 million unique visitors a month (Facebook reaches 77 million unique visitors a month)

    - 6 million visits a month

    Web Tier

    - 2 x Lenovo ThinkServer RS110 1U

    - 4 cores, 2.83 Ghz, 12 MB L2 cache

    - 500 GB datacenter hard drives, mirrored

    - 8 GB RAM

    - 500 GB RAID 1 mirror array

    Database Tier

    - 1 x Lenovo ThinkServer RD120 2U

    - 8 cores, 2.5 Ghz, 24 MB L2 cache

    - 48 GB RAM

    A fourth server was added to run superuser.com. All together the servers also run Stack Overflow, Server Fault, and Super User.

  • Report this Comment On January 01, 2011, at 7:50 AM, TheBlindCat wrote:

    "And lastly, in spite of the hype, the trend really is to move user applications to the Internet (the much hyped "Cloud"). And what do you need in the data center? Big honking servers with powerful multicore processors capable of virtualizing a lot of machines in a small space that are reasonably energy efficient."

    Two assertions above:

    1) Applications moving to the cloud.

    2) powerful multicore processors capable of virtualizing a lot of machines.

    While I agree with both, the end result is debatable.

    It could just as easily play against intel/AMD overerall:

    1) When applications move to the cloud, smaller more efficient processors like those found in phones, tablets and netbooks are now able to support the vast majority of use cases, reducing the number of traditional PCs sold (remember that the number of clients must, for economic reasons, significantly outnumber servers).

    2) While virtualization does have it's limits for certain low-latency/high performance apps, it is used extensively to REDUCE the number of physical machines in a data center, so this is something of a double-edged sword.

  • Report this Comment On January 01, 2011, at 4:10 PM, crawlfish wrote:

    This sounds like a good argument but one problem. Microsoft is a very bloated operating system and it will run very slowly on a ARMS platform. That is why cell phones rarely run Microsoft. The descendants of Unix like Apple and Linux mainly run on the ARMS platform. I think that Microsoft is just as vulnerable if not more so than Intel. If you have a net book, put Linux on it and dual boot it with Windows being the other operating system and you will see what I am talking about. As smart phones take up more market share Microsoft will loose share to different variations of Unix. Android is a Linux operating system. And Apple's I Phone doesn't run on Windows either but uses a form of BSD.

  • Report this Comment On January 03, 2011, at 12:03 AM, jorgie3237 wrote:

    For over 10 years microsoft has been looking for an excuse to port their OS to another processor. Recent ARMs are just starting to have enough performance to run Windows. Intel has been held back by Windows overhead on mobile devices and thus working deligently to enable other OSes on Atom. In 2011 Atom will likely have equivalent power/performance to ARM then will leap ahead when they scale to 22nm.

  • Report this Comment On January 08, 2011, at 1:00 PM, baldheadeddork wrote:

    In late, but I think you're misreading this, Eric.

    Microsoft needs an ARM version of Windows to compete in tablets, but that doesn't mean death to the Wintel x86 platform. There is still a massive performance capability difference between mobile and desktop OS, and the hardware that runs each.

    I don't think most users appreciate how much work they're doing on a desktop or a laptop. It's expected now that even a bottom-end $400 system will be able to run four or five applications at the same time. You can't do that with an SoC. When they offer multitasking its a virtual form like Windows had in the early 90's, and forget about running basic multi-threaded desktop tasks like consumer video editing in any form on a SoC.

    How far are we from SoC's having that basic capability? Generations, and that's if they want to go in that direction. Stepping up to desktop performance will raise the cost of SoC's for mobile devices that won't use it.

    This is when someone pipes up with "Yeah, but user applications are moving to the cloud so you won't need that much power." No, they are not because it's exceedingly cost prohibitive.

    Thin clients work well when you have a pool of users running one program and nothing else. Call centers were a great example long before Salesforce came along. Moving CSM from thin clients on a local network to the internet was a logical and even an obvious move.

    But get beyond that and things get hairy in a big hurry. Let's look at a very small example. Have an office of, say, a dozen CPA's and two or three clerical staff. They're going to run the same eight to ten applications, but they're going to be using and moving between them as needed. Providing that capability means your backend hardware requirements just exploded. Quickbooks server for a dozen clients that use it extensively will want a dual proc/quad core Xeon setup. Now you need to replicate that for the Office suite, database clients, all of their other applications, and then tie it together with a monster client OS server that can bring all of the pieces together.

    You've already exceeded the cost of individual desktop systems several times over, but you still need to significantly upgrade your network. All of the work the OS and applications did locally now needs to go across the network and that's going to eat a ton of bandwidth. You'll need gigabit to the workstations, more than that between the severs, and forget about more than one or two users on thin client wifi at any time.

    And you're still buying workstations. You still need a box with a processor, a motherboard with video, audio and LAN chipsets, a power supply and some RAM at each workstation. You don't need a hard drive and the processor and memory requirements are lower, but the costs savings per workstation will not be that great.

    Last, central computing means a single point of failure for all of the clients. If the client OS server or the network go down, everything comes to a stop. You can add failover servers and build redundancy into your network, but this adds thousands of dollars into the cost.

    Remember, this is just talking about one office of fifteen users, running relatively mild, common desktop applications. No video editing, no cad work, and thin clients are still hugely more expensive than desktop systems. Yet people have got it into their heads that cloud computing means moving everything to a Google or Amazon server and piping out work from hundreds of applications to millions of clients simultaneously. It's a fairy tale.

    Intel's problem isn't ARM or SoC. They missed that market, but remember it's a small margin market without a lot of upside to their other products. That usually doesn't work out well for tech companies. I think you can make a case that Intel was smart to give it a pass.

    I think Intel's core problem is their development cycle. It's still running at about the same pace it was in the mid 90's when processor speed was the real choke point in computer performance. Back then it made sense but now the performance gains are much more incremental and Intel is under much greater cost pressures. They need to stop trying to prove Moore's Law is still valid and think about the lifecycle of their products as it relates to the market.

Add your comment.

Sponsored Links

Leaked: Apple's Next Smart Device
(Warning, it may shock you)
The secret is out... experts are predicting 458 million of these types of devices will be sold per year. 1 hyper-growth company stands to rake in maximum profit - and it's NOT Apple. Show me Apple's new smart gizmo!

DocumentId: 1410293, ~/Articles/ArticleHandler.aspx, 10/1/2014 6:40:34 AM

Report This Comment

Use this area to report a comment that you believe is in violation of the community guidelines. Our team will review the entry and take any appropriate action.

Sending report...


Advertisement