INTC - Intel

[Hielko]

It wouldn't surprise me if Intel chips are the future for smartphones. The reason is simple: developing new chips is extremely capital intensive, you need to spend massive amounts on R&D. The biggest player can spend the most on R&D while investing a smaller percentage on R&D per chip. This is exactly why Intel got a dominant position in the desktop/server cpu-business.

 

If that how it's going to play out is ofcourse uncertain. Currently they don't have a big market share in this sector, and while I expect that a decent amount of the money they spend on desktop cpu R&D can also be used for mobile phone cpu's there are also R&D costs specific for smartphone CPU's. And maybe players like Samsung or Apple could achieve sufficient economies of scale w.r.t. R&D development costs / the market is big enough to support multiple players.

[Ross812]

It wouldn't surprise me if Intel chips are the future for smartphones. The reason is simple: developing new chips is extremely capital intensive, you need to spend massive amounts on R&D. The biggest player can spend the most on R&D while investing a smaller percentage on R&D per chip. This is exactly why Intel got a dominant position in the desktop/server cpu-business.

 

If that how it's going to play out is ofcourse uncertain. Currently they don't have a big market share in this sector, and while I expect that a decent amount of the money they spend on desktop cpu R&D can also be used for mobile phone cpu's there are also R&D costs specific for smartphone CPU's. And maybe players like Samsung or Apple could achieve sufficient economies of scale w.r.t. R&D development costs / the market is big enough to support multiple players.

 

Intel has been shrinking their architecture for years now and is within 5 years of getting a desktop/laptop chip with a low enough power demand to be used in a phone. Intel hasn't been sitting still either. They developed the atom processor to bridge the gap in the meantime while their i3,5,7 mobile processors become relevant for super mobile computing. Haswell is due for the first quarter of 2013 with a TDP of 10W. A Haswell prototype demonstration it can run at just under 8 watts at full load. Haswell will put a mainstream iCore chip in tablets with all day (8+) battery life and gets the form factor of these new tablets with desktop style chips down to 12mm (the iPad is at 10mm). The best part is this is due in 1Q 2013. What is coming in late 2013 (shrink to 14nm)-2014-2015? The Haswell i3 i5 and i7 will make any ARM chip on the market used in tablets look silly. The Atom is starting to be incorporated into phones (and some pre-haswell tablets) as a place holder until their main line of chips become viable.   

[Guest rimm_never_sleeps]

onslaught of Intel? Actually waiting for a competitive mobile chip from them has been akin to waiting for Godot. They Were very good when they were bullying AMD for years and had the PC market all to itself, and riding the coattails of Microsoft. Amd has always been a puny, ineffectual competitor. This time, after having totally missed the mobile market, they are going to try to play catch up.  But instead of AMD, they are going up against ARM, Samsung, Apple, and Qualcomm and an entire ecosystem built around the ARM architecture. Intel knows a thing or two about establishing moats around ecosystems. As far as having the capital to win, that may have been true years ago. But not now when there are massive contract fabs like TMC out there who have the capability of investing as much in cap equipment as Intel does.

[bathtime]

onslaught of Intel? Actually waiting for a competitive mobile chip from them has been akin to waiting for Godot.

 

But based on this review they already have a competitive chip:

 

http://semiaccurate.com/2012/04/26/how-well-does-intels-new-phone-work-as-a-phone/

 

"Overall, Intel has done what they promised and exceeded our expectations. If you buy an x86 based phone, be it the Xolo X900 or any of the successors, it should just work the way you expect an Android phone to."

 

A year or two on should allow for significant improvement, no?

[bargainman]

I don't think that is the future,

I think this is the future. Processing power is becoming a commodity and the trend has been for years to shift away from a focus on efficient programming to ease of development. One of the strong points of Android (and I believe one of the reasons of their success) is that it is relatively easy to write platform-agnostic applications, i.e. it can run on a phone, tablet, pc or television without the developer having to worry about device limitations or interface quirks.

 

From the official android website:

Android also gives you tools for creating apps that look great and take advantage of the hardware capabilities available on each device. It automatically adapts your UI to look it's best on each device, while giving you as much control as you want over your UI on different device types. (..) For example, you can create a single app binary that's optimized for both phone and tablet form factors.

Writing and compiling software for specific CPU's is, for most developers, a thing of the past. Servers are virtualized, websites run on any platform, and the same thing is happening for phones and tablets.

 

The moat of Intel has, as far as my analysis goes, nothing to do with their instruction set. Everybody can copy it. Their moat is that it takes immense amounts of capital and knowledge to fit as many transistors as possible on a very small chip and that they are incredibly good at that.

 

Historically their focus has always been on performance for desktops / servers, where power was not really an issue. But due to the rise of mobile computing performance is perhaps not that relevant anymore. Power consumption is increasingly important and it will take Intel some time to adjust to that (at least that's what I tell myself ;) ).

 

 

The funny thing is that on the one hand you are talking about power consumption being more important, and on the other hand you are saying that compiling software for particular platforms is no longer an issue.  These are actually somewhat opposing statements.  A program not compiled for a particular platform will be lower performant, and in particular it will be lower performant per power consumption unit.  This was actually one of the main points behind a recent C++ conference where they claimed there was a resurgence in C++.  Power has become a big issue.  For servers it's an issue because the cost of a server farm is something like 70%+ power consumption!  As such programmer time is not the highest cost that needs to be optimized anymore.  On the other hand for tablets and phones, battery life is not increasing with moore's law, it's a linear curve, not an exponential one.  As such '#of computer instructions per power unit' has also taken on more importance since battery life is limited and not growing fast (plus the addition of features to tablets and phones like bigger higher resolution screens cause more power draw).  That is one reason why Apple's iPhone seemed much snappier than android for a while even though the hardware specs would indicate parity.  (Apple uses Obj-C/C++, and Android was using Java *exclusively* to start).  That's also the reason that Google opened up native (C++) coding for Android more recently since they realized a pure Java environment would not work given the constraints.  That said, the question is, how long will it take for the hardware to get to the point where programmers can use higher level interpreted languages without too much fall out.  The C++ conference I mentioned was probably at least a year ago.  By next year it's possible it won't matter anymore.

 

The interesting thing is that the computing industry goes through these cycles over and over, usually everytime a form factor change happens.  So there were mainframes, then minis, micro computers, then laptops.  Then for a while things got stagnant, MSFT tried to go to a fully managed environment in C#, but realized they needed to leave the door open for highly performant apps.  Then we got phones and tablets and everyone realized that speed and performance were important again. It's a stair step that happens again and again.  Computing power gets so powerful, that programmer time is more important, so we move to higher level languages.  Then computing power get so powerful that it gets moved to a smaller form factor, and then power becomes important again..  and on and on..

[Guest rimm_never_sleeps]

onslaught of Intel? Actually waiting for a competitive mobile chip from them has been akin to waiting for Godot.

 

But based on this review they already have a competitive chip:

 

http://semiaccurate.com/2012/04/26/how-well-does-intels-new-phone-work-as-a-phone/

 

"Overall, Intel has done what they promised and exceeded our expectations. If you buy an x86 based phone, be it the Xolo X900 or any of the successors, it should just work the way you expect an Android phone to."

 

A year or two on should allow for significant improvement, no?

 

Yes they finally have one that's close, although the graphic performance is not up to ARM standards. But they promised a competitive one in 2010. Intel well knows that it takes more than technical parity when you are on the outside looking in. Intel vanquished the RISC architecture in the 90s. By all accounts these chips were way better designs than the x86. But there was a moat around X86. So it's more than just getting to technical parity. It's also about going up against a powerful entrenched ecosystem. And that is going to be much tougher for Intel than getting to technical parity. To break through they may have to give margin crushing sweetheart deals to customers simply to gain share.

[Guest valueInv]

It wouldn't surprise me if Intel chips are the future for smartphones. The reason is simple: developing new chips is extremely capital intensive, you need to spend massive amounts on R&D. The biggest player can spend the most on R&D while investing a smaller percentage on R&D per chip. This is exactly why Intel got a dominant position in the desktop/server cpu-business.

 

If that how it's going to play out is ofcourse uncertain. Currently they don't have a big market share in this sector, and while I expect that a decent amount of the money they spend on desktop cpu R&D can also be used for mobile phone cpu's there are also R&D costs specific for smartphone CPU's. And maybe players like Samsung or Apple could achieve sufficient economies of scale w.r.t. R&D development costs / the market is big enough to support multiple players.

 

Intel has been shrinking their architecture for years now and is within 5 years of getting a desktop/laptop chip with a low enough power demand to be used in a phone. Intel hasn't been sitting still either. They developed the atom processor to bridge the gap in the meantime while their i3,5,7 mobile processors become relevant for super mobile computing. Haswell is due for the first quarter of 2013 with a TDP of 10W. A Haswell prototype demonstration it can run at just under 8 watts at full load. Haswell will put a mainstream iCore chip in tablets with all day (8+) battery life and gets the form factor of these new tablets with desktop style chips down to 12mm (the iPad is at 10mm). The best part is this is due in 1Q 2013. What is coming in late 2013 (shrink to 14nm)-2014-2015? The Haswell i3 i5 and i7 will make any ARM chip on the market used in tablets look silly. The Atom is starting to be incorporated into phones (and some pre-haswell tablets) as a place holder until their main line of chips become viable. 

 

You left out the most critical point - Intel is mostly locked out of the two largest smartphone vendors:

 

Apple: Designs it owns chips and greatly prefers proprietary technology.

Samsung: Has is own chip design and manufacturing and is unlikely to switch to Intel.

 

Both of these vendors have huge marketshare and account for majority the industry's profits. Intel is left fighting for scraps at unprofitable vendors with declining marketshare. Not a very promising business.

 

Even their own traditional partner MSFT has introduced their ARM-based devices first.

[FrankArabia]

So if INTC is left completely out of the mobile phone space, are they screwed? given the current level of earnings (which don't include mobile revenues) I think they're doing ok.

[eclecticvalue]

So if INTC is left completely out of the mobile phone space, are they screwed? given the current level of earnings (which don't include mobile revenues) I think they're doing ok.

 

Couldn't Intel get into that space. Because with a lot of cash on the Balance sheet. I am sure they can make inroads in that space.

[Palantir]

So if INTC is left completely out of the mobile phone space, are they screwed? given the current level of earnings (which don't include mobile revenues) I think they're doing ok.

 

Couldn't Intel get into that space. Because with a lot of cash on the Balance sheet. I am sure they can make inroads in that space.

 

They can get into the space, but what are their profit margins going to be? Will they have an edge? It seems to me that this stuff is going to be commoditized in the future...

[Palantir]

So if INTC is left completely out of the mobile phone space, are they screwed? given the current level of earnings (which don't include mobile revenues) I think they're doing ok.

 

They're not screwed, but just in a declining market. You have to also keep in mind, their main partner - MSFT, has started using ARM chips on the Surface, and as such, they may not be able to count on MSFT in the future.

[Ross812]

It wouldn't surprise me if Intel chips are the future for smartphones. The reason is simple: developing new chips is extremely capital intensive, you need to spend massive amounts on R&D. The biggest player can spend the most on R&D while investing a smaller percentage on R&D per chip. This is exactly why Intel got a dominant position in the desktop/server cpu-business.

 

If that how it's going to play out is ofcourse uncertain. Currently they don't have a big market share in this sector, and while I expect that a decent amount of the money they spend on desktop cpu R&D can also be used for mobile phone cpu's there are also R&D costs specific for smartphone CPU's. And maybe players like Samsung or Apple could achieve sufficient economies of scale w.r.t. R&D development costs / the market is big enough to support multiple players.

 

Intel has been shrinking their architecture for years now and is within 5 years of getting a desktop/laptop chip with a low enough power demand to be used in a phone. Intel hasn't been sitting still either. They developed the atom processor to bridge the gap in the meantime while their i3,5,7 mobile processors become relevant for super mobile computing. Haswell is due for the first quarter of 2013 with a TDP of 10W. A Haswell prototype demonstration it can run at just under 8 watts at full load. Haswell will put a mainstream iCore chip in tablets with all day (8+) battery life and gets the form factor of these new tablets with desktop style chips down to 12mm (the iPad is at 10mm). The best part is this is due in 1Q 2013. What is coming in late 2013 (shrink to 14nm)-2014-2015? The Haswell i3 i5 and i7 will make any ARM chip on the market used in tablets look silly. The Atom is starting to be incorporated into phones (and some pre-haswell tablets) as a place holder until their main line of chips become viable. 

 

You left out the most critical point - Intel is mostly locked out of the two largest smartphone vendors:

 

Apple: Designs it owns chips and greatly prefers proprietary technology.

Samsung: Has is own chip design and manufacturing and is unlikely to switch to Intel.

 

Both of these vendors have huge marketshare and account for majority the industry's profits. Intel is left fighting for scraps at unprofitable vendors with declining marketshare. Not a very promising business.

 

Even their own traditional partner MSFT has introduced their ARM-based devices first.

 

Samsung and Apple both use Intel CPUs in their ultra books. My whole thesis with Intel is the coming parity between Intel's flagship Core-I processors and super mobile devices. This will start with tablets in late Q1 and early Q2 next year. At this time next year the processor in your laptop can be built into your iPad. Intel has the Atom processor for low end tablets and phones until their real processor series has a low enough power demand for those devices. Do you honestly believe Apple will continue to use ARM chips if Intel can build chips capable of dual booting the latest OSX operating system on their iPad? 

[FrankArabia]

The surface is their hardware device right? don't think that will go anywhere....MSFT's key strength is their Windows and Office....everything else is semi to non important......why would you buy a surface if you could buy a google nexus?

 

that said I am still trying to wrap my head why INTC can't one day make chips equivalent to ARM? I still haven't heard an answer I can understand.

[Guest valueInv]

Samsung and Apple both use Intel CPUs in their ultra books. My whole thesis with Intel is the coming parity between Intel's flagship Core-I processors and super mobile devices. This will start with tablets in late Q1 and early Q2 next year. At this time next year the processor in your laptop can be built into your iPad. Intel has the Atom processor for low end tablets and phones until their real processor series has a low enough power demand for those devices. Do you honestly believe Apple will continue to use ARM chips if Intel can build chips capable of dual booting the latest OSX operating system on their iPad?

 

Yes, because:

1, OSX is not built for touch and hence, not suited for tablets

2, Apple always prefers its own proprietary technology. Otherwise they wouldn't have invested 4 years in building their own chips.

[Guest valueInv]

The surface is their hardware device right? don't think that will go anywhere....MSFT's key strength is their Windows and Office....everything else is semi to non important......why would you buy a surface if you could buy a google nexus?

 

http://www.intomobile.com/2012/10/17/preorders-32gb-surface-rt-without-black-touch-cover-sell-out/

[Palantir]

The surface is their hardware device right? don't think that will go anywhere....MSFT's key strength is their Windows and Office....everything else is semi to non important......why would you buy a surface if you could buy a google nexus?

 

that said I am still trying to wrap my head why INTC can't one day make chips equivalent to ARM? I still haven't heard an answer I can understand.

 

They can, and they are...but you haven't answered the follow on question - what's their edge/competitive position/profit margins in a scenario where cool cores are the dominant processor?

 

You have a lot more research to do. There are very good reasons for the Surface (both to make and to buy) over the Nexus....again, like Intel, MSFT also faces declining markets for its Windows products...the same headwinds Intel has, MSFT has as well.

[ItsAValueTrap]

They can get into the space, but what are their profit margins going to be?

You could look at the margins of Qualcomm, Broadcom, Marvell, etc.  Their margins are reasonably high.  Even the smaller players are making money even though they don't get to scale their R&D design costs over huge volumes.

Of course in the future things can change.

 

I don't think that scale is that big an advantage in R&D.  If you look at Intel versus AMD, there have been times where AMD was putting out designs just as good as Intel's.  Here's my analogy... the Yankees spend the most money on their baseball team but they don't always win the World Series.  Intel outspends AMD... it has the best design most of the time but not all of the time.

 

Will they have an edge?

For some reason, in semiconductor MANUFACTURING, scale is a huge advantage due to economies of scale.  The other factor is government subsidies.  Some manufacturers are kept alive by gov't subsidies and all semiconductor manufacturers receive some form of government subsidy.

 

Intel, Samsung, and TSMC have the greatest scale.  There are a bunch of smaller players like AMD's Globalfoundries, Freescale, etc. etc.

Samsung makes smartphone SoCs for Samsung and Apple.  Apple just sued Samsung so it is possible that Apple is going to go to somebody else.

TSMC is the world's largest contract fab and is making smartphone SoCs for many of the ARM-based SoCs.

 

I believe that Intel has the manufacturing edge as they have the most advanced fabs and the largest scale.  Its competitors won't move to finFET for a few years.  It is ahead in terms of process size.

[ItsAValueTrap]

But not now when there are massive contract fabs like TMC out there who have the capability of investing as much in cap equipment as Intel does.

TSMC looks like it spent $7.3B USD on capex for its last fiscal year versus $10.7B for Intel

 

TSMC

As of 2011-12-31

213962.5 New Taiwan dollars = 7343.5781 US dollars

 

INTC

53 weeks ending 2011-12-31

-10,764.00

[Palantir]

You could look at the margins of Qualcomm, Broadcom, Marvell, etc.  Their margins are reasonably high.  Even the smaller players are making money even though they don't get to scale their R&D design costs over huge volumes.

Of course in the future things can change.

 

Thing is though, this open marketplace will only lead to margin compression down the line, especially as in the future as hardware manufacturers can start making their processors in house the way Apple is doing. This makes Intel's value proposition more tenuous.

 

It seems to me ARM holds the cards here, as they just own the IP, and collect royalties, and aren't involved in the production or manufacture. It's like ARM is Coke, and the chip firms are bottlers.

 

 

 

Intel, Samsung, and TSMC have the greatest scale.  There are a bunch of smaller players like AMD's Globalfoundries, Freescale, etc. etc.

Samsung makes smartphone SoCs for Samsung and Apple.  Apple just sued Samsung so it is possible that Apple is going to go to somebody else.

TSMC is the world's largest contract fab and is making smartphone SoCs for many of the ARM-based SoCs.

 

I believe that Intel has the manufacturing edge as they have the most advanced fabs and the largest scale.  Its competitors won't move to finFET for a few years.  It is ahead in terms of process size.

 

Intel certainly has an advantage in manufacturing, but I'm skeptical it is enough to offset the aforementioned weaknesses. In Intel, everything is integrated - architecture, design, manu etc. With ARM, each of these components has been separated. It is possible Intel could claw back market share, but we will have to wait.

 

The other dynamic you have to keep in mind is that OEMs will not want Intel to become too dominant again, so I think they will try to keep spreading work between different companies and keeping some in house.

[stylized_fact]

It seems to me ARM holds the cards here, as they just own the IP, and collect royalties, and aren't involved in the production or manufacture. It's like ARM is Coke, and the chip firms are bottlers.

 

Is there a possibility that MIPS can become the equivalent of Pepsi?

[Guest rimm_never_sleeps]

let's think this through. how has the supposed intel manufacturing prowess been working for them in Mobile? you could make the case that their manufacturing prowess has compensated for their noncompetitive mobile designs and has actually allowed them to be marginally competitive in 2012. But is that something to be bullish about? good manufacturing, marginal designs?

 

I agree that the manufacturing angle for Intel is way overrated. what's more important is the market dynamics (outside looking in), the ecosystem you have to challenge (ARM), your competition (arm, apple samsung and qualcomm), your mobile designs vs theirs (atom/arm).

 

it was also mentioned that Apple could move the entire product line to Intel, which I believe is extremely remote. But then there is this...

http://semiaccurate.com/2012/09/24/an-update-on-apple-moving-away-from-intel/

[FrankArabia]

They can get into the space, but what are their profit margins going to be?

You could look at the margins of Qualcomm, Broadcom, Marvell, etc.  Their margins are reasonably high.  Even the smaller players are making money even though they don't get to scale their R&D design costs over huge volumes.

Of course in the future things can change.

 

I don't think that scale is that big an advantage in R&D.  If you look at Intel versus AMD, there have been times where AMD was putting out designs just as good as Intel's.  Here's my analogy... the Yankees spend the most money on their baseball team but they don't always win the World Series.  Intel outspends AMD... it has the best design most of the time but not all of the time.

 

Will they have an edge?

For some reason, in semiconductor MANUFACTURING, scale is a huge advantage due to economies of scale.  The other factor is government subsidies.  Some manufacturers are kept alive by gov't subsidies and all semiconductor manufacturers receive some form of government subsidy.

 

Intel, Samsung, and TSMC have the greatest scale.  There are a bunch of smaller players like AMD's Globalfoundries, Freescale, etc. etc.

Samsung makes smartphone SoCs for Samsung and Apple.  Apple just sued Samsung so it is possible that Apple is going to go to somebody else.

TSMC is the world's largest contract fab and is making smartphone SoCs for many of the ARM-based SoCs.

 

I believe that Intel has the manufacturing edge as they have the most advanced fabs and the largest scale.  Its competitors won't move to finFET for a few years.  It is ahead in terms of process size.

 

both are excellent companies with certain advantages and disadvantages. I think you know what they're. MSFT is a bit more expensive at this point but its a good buy at anywhere around 25. even at these levels, i don't think you lose money buying MSFT.

 

we'll see how the surface performs. I just don't bother with forecasting stuff like this as it is ultimately in my view a short term type of thing where i have zero advantage (I don't even own a smartphone, i have an old BB). if it does well fine, but i don't think it'll take off like the iPad. i will say Excel is crucial to my work and to many others as well. i bet on MSFT on their key products that have high barriers and customer loyalty. everything else is hopeful.

[Palantir]

It seems to me ARM holds the cards here, as they just own the IP, and collect royalties, and aren't involved in the production or manufacture. It's like ARM is Coke, and the chip firms are bottlers.

 

Is there a possibility that MIPS can become the equivalent of Pepsi?

 

I'm not knowledgeable about this, could you go into this point a little bit? Is MIPS the architecture that Texas Instruments also been producing for?

[ItsAValueTrap]

There is a writeup on VIC for Intel:

 

http://www.valueinvestorsclub.com/value2/Idea/ViewIdea/78064

 

It's a different perspective... some of the things I have a different opinion on.

[stylized_fact]

It seems to me ARM holds the cards here, as they just own the IP, and collect royalties, and aren't involved in the production or manufacture. It's like ARM is Coke, and the chip firms are bottlers.

 

Is there a possibility that MIPS can become the equivalent of Pepsi?

 

I'm not knowledgeable about this, could you go into this point a little bit? Is MIPS the architecture that Texas Instruments also been producing for?

 

MIPS was a pioneering RISC design and has long been available as 64 bit.  The company operates like ARM Ltd, licensing their designs to interested parties.  They are used in networking applications primarily, I believe.  The company has been subject to takeover speculation recently: 

 

http://www.ft.com/intl/cms/s/2/b019b99a-0198-11e2-83bb-00144feabdc0.html