8 ms·

Intel cuts Atom chips, basically giving up on the smartphone and tablet markets

122 points by Ivoah 10y ago

PaulHoule 10y ago
I think that's a misunderstanding. Atom chips have always been crippled to keep them from cannibalizing more expensive chips. Skylake is a fine tablet chip, in fact, that's really what Skylake is good for. They are probably producing them in high enough numbers now that they can give up on Atom.
- cfallin 10y ago
  Atom isn't crippled so much as it's developed from a very simple (small, low-power, in-order execution) baseline and so hasn't had the time to reach the level of performance maturity and sophistication that the "big cores" (Haswell, Skylake, etc) have. If Intel had the ability to get Skylake performance in an Atom power envelope, they would.
mappu 10y ago
Anandtech coverage: http://www.anandtech.com/show/10288/intel-broxton-sofia-smartphone-socs-cancelled http://www.anandtech.com/show/10288/intel-broxton-sofia-smar...
joezydeco 10y ago
So is Edison still on?
- hga 10y ago
  That or an article linked in a side bar said they were still chasing the IoT.
  - melling 10y ago
    Edison is based on Atom? https://en.m.wikipedia.org/wiki/Intel_Edison https://en.m.wikipedia.org/wiki/Intel_Edison
    - kps 10y ago
      No, the Quark SoCs are 486ish cores, not Atom.
      - Narishma 10y ago
        Edison has both Atom and Quark cores.
    - elijahparker 10y ago
      The Edison has both an Atom running Linux and a Quark running an RTOS
- elijahparker 10y ago
  I have my doubts on the future of the Edison. I was going to use it in a project but ended up giving up and switching to ARM after so many bugs went unresolved across multiple platform releases. The power consumption vs performance is incredible, but it just feels like they're not taking it seriously beyond single quantity orders for individual makers
grizzles 10y ago
I might be biased but I think this is a maybe looks good on paper decision but a terrible long term strategy decision. Just because they can't compete on price doesn't mean they can't compete on features. It was only two months ago that they were showcasing their android platform and practically begging for customers. [1] I've wanted to do a phone for awhile. I'm sure I'm not the only one. In the last few months I contacted people at Intel but I had no response whatsoever to my multiple inquiries (to you know, become a customer). And I have some of the company's most senior management in my LinkedIn. Now I know why. This is kind of pathological behavior on Intel's part imo. The big players were never going to come to their platform. Huawei, HTC, Samsung, LG etc have no reason to change from ARM. They are content to keep doing their cookie cutter thing while fighting a brutal price war in a staid market. On the other hand, the startups out there, those of us who want to do something really out of the box and grow new markets can't get a word in edgewise. It's disappointing. As a technologist, I really hope someone manages to ship an open source handheld one day. That will be the catalyst for the next iteration of smartphone evolution. [1] http://www.theregister.co.uk/2016/02/23/move_over_continuum_intel_shows_android_smartphone_powering_bigscreen_linux/ http://www.theregister.co.uk/2016/02/23/move_over_continuum_...)
- petra 10y ago
  >> open source handheld one day. I would guess that this is beyond Intel's pay grade and it is a political/security decision. Also except for selling openness, what else will sell that you couldn't do today ?
  - grizzles 10y ago
    You can get a raspberry pi with an atheros chipset wifi dongle and you have pretty much a completely open source stack. It just doesn't look very nice to users. >> Also except for selling openness, what else will sell that you couldn't do today ? Anything that doesn't fit into Google's technical design / strategy that sells and customers like it. Android has heaps of preconceived notions about how a handheld should work that are baked into the system and unchangeable. If you are looking for exact examples of things that are not currently technically possible on android, how about a phone with a gps button widget that doesn't require you to leave the maps app to enable/disable gps or (not that I would personally pursue this) the ability to play video in a distributable app widget.
fiatmoney 10y ago
Wow, this is really surprising. Atom chips were steadily increasing their value from generation to generation and were within spitting distance of ARM in terms of mojo / watt and power consumption. Intel's process advantage should really let them dominate that market. I wonder if this implies we'll see ARM chips built in Intel fabs?
- StringyBob 10y ago
  Probably not the type of chip you meant, but as an example here's an ARM CPU on the Intel 14nm process: https://www.altera.com/products/soc/portfolio/stratix-10-soc/overview.smartphone.highResolutionDisplay.html https://www.altera.com/products/soc/portfolio/stratix-10-soc...
rayiner 10y ago
Gotta love an article that leads with sunk cost fallacy. Anyway, axing Atom doesn't mean the end of Intel targeting tablets. Core M is fitting into fanless tablets thinner than an iPad.
- nkurz 10y ago
  Separately, the title is misleading when it says "Intel cuts Atom chips". While "cuts" may be technically correct, Intel isn't giving up on Atom, just canceling the smartphone SoC's that were planned to use it. Silvermont (the basis of the soon to come out new Knights Landing Xeon Phi) and (so far as I know) it's successor Goldmont are still going forward as scheduled. From the (significantly better) AnandTech article titled "Intel's Changing Future: Smartphone SoCs Broxton & SoFIA Officially Cancelled": Also not discussed in greater detail is Intel's future plans for their overall Atom lineup. With Apollo Lake announced just earlier this month, it's clear that Intel's Atom efforts have not been cancelled entirely. We will still see the new 14nm Goldmont cores appear in low-cost PCs under Apollo Lake, most likely in several 11-to-13 inch high volume devices. http://www.anandtech.com/show/10288/intel-broxton-sofia-smartphone-socs-cancelled http://www.anandtech.com/show/10288/intel-broxton-sofia-smar...
educar 10y ago
This is a little puzzling given their 5G investment at mwc - https://newsroom.intel.com/news-releases/intel-accelerates-path-to-5g/ https://newsroom.intel.com/news-releases/intel-accelerates-p...
igravious 10y ago
fta: > but Intel failed to unseat market leader ARM. Intel should, if anything, license ARM designs and make the best slices of silicon out there for ARM. What I'm saying is, talking about unseating is the wrong way to think about ARM.
- petra 10y ago
  It wouldn't be smart of ARM to help Intel and harm it's own ecosystem and especially the chip fabs,because once they are weakend enough, Intel could return to using it's own cores.
- kps 10y ago
  Intel is an ARM architecture licensee.
  - rbanffy 10y ago
    I believe they sold that license about the time they went for Itanic.
  - igravious 10y ago
    Ah, I should have checked. I did read through the ARM architecture page† on Wikipedia and it does not mention that fact. It does say, "In 1994, Acorn used the ARM610 as the main central processing unit (CPU) in their RiscPC computers. DEC licensed the ARM6 architecture and produced the StrongARM. At 233 MHz, this CPU drew only one watt (newer versions draw far less). This work was later passed to Intel as a part of a lawsuit settlement, and Intel took the opportunity to supplement their i960 line with the StrongARM. Intel later developed its own high performance implementation named XScale, which it has since sold to Marvell." I'm not sure how much sense it makes for Intel to be both an ARM licensee and develop Atom. † https://en.wikipedia.org/wiki/ARM_architecture#Core_licence https://en.wikipedia.org/wiki/ARM_architecture#Core_licence
iamgopal 10y ago
Could not they just use their superior manufacturing capability to make e.g. 24 core arm that use less power then competition ? If not for them, it will atleast bite the market share from competitor. This is quite analogous to "Nokia should have embraced android much ago."
- andrepd 10y ago
  Shamelessly stolen from the comments at http://anandtech.com/show/10288/intel-broxton-sofia-smartphone-socs-cancelled http://anandtech.com/show/10288/intel-broxton-sofia-smartpho... "The short answer: margins. The long answer: Intel could absolutely build a killer smartphone SoC on the 14nm process. However it would need to be Core based; the Atom CPU line isn't performant enough. They may need to go back to Ivy Bridge or strip down Skylake a bit to get there, but they could do it. The issue is that smartphone SoCs have very low margins. MediaTek, Rockchip, etc are crushing the market. Atom is designed for margins, not performance. If Intel built a winning Core smartphone SoC, then they'd have to sell it for significantly lower margins than what their other Core parts sell for. And that in turn would put pressure on the rest of their chip stack; why should an OEM pay $250 for Core-M when you could get a similar smartphone SoC for $30? For what it's worth Apple faces a similar challenge. But being vertically integrated means they don't have to share with others, and the total profit off of iPhones/iPads can offset the higher costs involved in developing and fabbing the A-series SoCs. However Intel doesn't have that luxury, and Asus would never let Intel have all of the profit. Smartphone SoCs are a race to the bottom. And Intel as a business has nothing to gain from taking part in that race. The winner is the guy who got badly hurt but didn't die; a Pyrrhic victory."
  - yuhong 10y ago
    PCs were also a race to the bottom. I wonder what would happen if Intel or AMD bought Compaq or another PC manufacturer maybe in the 1990s and dominated the PC market by the mid-2000s.
    - hga 10y ago
      Intel probably wouldn't have been able to sustain the DNA required to also be successful in that market, and they would have had severe antitrust problems. AMD would have had the same first problem (but probably worse since they've historically had much poorer management), and would have likely lost a lot of their business selling to other PC manufacturers they'd now be directly competing with.
pier25 10y ago
Isn't Core M a tablet chip?
cageface 10y ago
It seems like with Intel's superior process and production scale they could easily enter into the ARM market. Is it just not profitable enough for them to do this? Or are their fabs too committed to producing x86 chips? Missing out on mobile entirely seems like a pretty big strategic mistake.
- alrs 10y ago
  Intel bought DEC's ARM business in 1997, tried doing ARM, and then gave up. https://en.wikipedia.org/wiki/StrongARM https://en.wikipedia.org/wiki/StrongARM
  - toyg 10y ago
    To be fair, it was a different world (almost 20 years ago... I feel old). It doesn't matter anyway -- after several years concentrating their immense muscle on battery savings and integrated graphics, they have caught up a lot. CoreM / Skylake can fit most use-cases where ARM stuff would be called for, at least in the high-margin market.
turnip1979 10y ago
Dumb question ... Broxton = Atom? So Atom will be no more?
- Sanddancer 10y ago
  No. The Atom core's going to be used pretty heavily, just not as a tablet/phone core. It's just been moved to pretty much solely the entry level laptop/desktop market.
rbanffy 10y ago
This is a path that must be followed with great attention. Giving up on low-end, entry-level parts often looks great in short term results but may doom the company in the end. One of the reasons we deploy to x86 servers now rather then SPARC, POWER or MIPS is that we develop on x86. When Sun and IBM chose to exit the low-end space and pursue only large clients, they doomed their platforms to the legacy of those clients who already had their systems running on their platforms. How many fresh applications are designed with SPARC or POWER in mind? How many are designed for zSeries mainframes?
stesch 10y ago
Isn't the HoloLens using an Intel Atom processor?
api 10y ago
I don't necessarily think this is a bad decision. Intel currently dominates high performance computing and data center chips, and if they lose that edge they will really lose. Maybe it makes sense for them to instead double down on that and try to maintain their lead in those areas. Mobile markets are high volume but lower margin.

122 points by Ivoah 10y ago

I think that's a misunderstanding. Atom chips have always been crippled to keep them from cannibalizing more expensive chips. Skylake is a fine tablet chip, in fact, that's really what Skylake is good for. They are probably producing them in high enough numbers now that they can give up on Atom.

cfallin 10y ago
Atom isn't crippled so much as it's developed from a very simple (small, low-power, in-order execution) baseline and so hasn't had the time to reach the level of performance maturity and sophistication that the "big cores" (Haswell, Skylake, etc) have. If Intel had the ability to get Skylake performance in an Atom power envelope, they would.

mappu 10y ago

Anandtech coverage: http://www.anandtech.com/show/10288/intel-broxton-sofia-smartphone-socs-cancelled http://www.anandtech.com/show/10288/intel-broxton-sofia-smar...

joezydeco 10y ago

So is Edison still on?

hga 10y ago
That or an article linked in a side bar said they were still chasing the IoT.
- melling 10y ago
  Edison is based on Atom? https://en.m.wikipedia.org/wiki/Intel_Edison https://en.m.wikipedia.org/wiki/Intel_Edison
  - kps 10y ago
    No, the Quark SoCs are 486ish cores, not Atom.
    - Narishma 10y ago
      Edison has both Atom and Quark cores.
  - elijahparker 10y ago
    The Edison has both an Atom running Linux and a Quark running an RTOS
elijahparker 10y ago
I have my doubts on the future of the Edison. I was going to use it in a project but ended up giving up and switching to ARM after so many bugs went unresolved across multiple platform releases. The power consumption vs performance is incredible, but it just feels like they're not taking it seriously beyond single quantity orders for individual makers

grizzles 10y ago

I might be biased but I think this is a maybe looks good on paper decision but a terrible long term strategy decision. Just because they can't compete on price doesn't mean they can't compete on features. It was only two months ago that they were showcasing their android platform and practically begging for customers. [1] I've wanted to do a phone for awhile. I'm sure I'm not the only one. In the last few months I contacted people at Intel but I had no response whatsoever to my multiple inquiries (to you know, become a customer). And I have some of the company's most senior management in my LinkedIn. Now I know why. This is kind of pathological behavior on Intel's part imo. The big players were never going to come to their platform. Huawei, HTC, Samsung, LG etc have no reason to change from ARM. They are content to keep doing their cookie cutter thing while fighting a brutal price war in a staid market. On the other hand, the startups out there, those of us who want to do something really out of the box and grow new markets can't get a word in edgewise. It's disappointing. As a technologist, I really hope someone manages to ship an open source handheld one day. That will be the catalyst for the next iteration of smartphone evolution. [1] http://www.theregister.co.uk/2016/02/23/move_over_continuum_intel_shows_android_smartphone_powering_bigscreen_linux/ http://www.theregister.co.uk/2016/02/23/move_over_continuum_...)

petra 10y ago
>> open source handheld one day. I would guess that this is beyond Intel's pay grade and it is a political/security decision. Also except for selling openness, what else will sell that you couldn't do today ?
- grizzles 10y ago
  You can get a raspberry pi with an atheros chipset wifi dongle and you have pretty much a completely open source stack. It just doesn't look very nice to users. >> Also except for selling openness, what else will sell that you couldn't do today ? Anything that doesn't fit into Google's technical design / strategy that sells and customers like it. Android has heaps of preconceived notions about how a handheld should work that are baked into the system and unchangeable. If you are looking for exact examples of things that are not currently technically possible on android, how about a phone with a gps button widget that doesn't require you to leave the maps app to enable/disable gps or (not that I would personally pursue this) the ability to play video in a distributable app widget.

fiatmoney 10y ago

Wow, this is really surprising. Atom chips were steadily increasing their value from generation to generation and were within spitting distance of ARM in terms of mojo / watt and power consumption. Intel's process advantage should really let them dominate that market. I wonder if this implies we'll see ARM chips built in Intel fabs?

StringyBob 10y ago
Probably not the type of chip you meant, but as an example here's an ARM CPU on the Intel 14nm process: https://www.altera.com/products/soc/portfolio/stratix-10-soc/overview.smartphone.highResolutionDisplay.html https://www.altera.com/products/soc/portfolio/stratix-10-soc...

rayiner 10y ago

Gotta love an article that leads with sunk cost fallacy. Anyway, axing Atom doesn't mean the end of Intel targeting tablets. Core M is fitting into fanless tablets thinner than an iPad.

nkurz 10y ago
Separately, the title is misleading when it says "Intel cuts Atom chips". While "cuts" may be technically correct, Intel isn't giving up on Atom, just canceling the smartphone SoC's that were planned to use it. Silvermont (the basis of the soon to come out new Knights Landing Xeon Phi) and (so far as I know) it's successor Goldmont are still going forward as scheduled. From the (significantly better) AnandTech article titled "Intel's Changing Future: Smartphone SoCs Broxton & SoFIA Officially Cancelled": Also not discussed in greater detail is Intel's future plans for their overall Atom lineup. With Apollo Lake announced just earlier this month, it's clear that Intel's Atom efforts have not been cancelled entirely. We will still see the new 14nm Goldmont cores appear in low-cost PCs under Apollo Lake, most likely in several 11-to-13 inch high volume devices. http://www.anandtech.com/show/10288/intel-broxton-sofia-smartphone-socs-cancelled http://www.anandtech.com/show/10288/intel-broxton-sofia-smar...

educar 10y ago

This is a little puzzling given their 5G investment at mwc - https://newsroom.intel.com/news-releases/intel-accelerates-path-to-5g/ https://newsroom.intel.com/news-releases/intel-accelerates-p...

igravious 10y ago

fta: > but Intel failed to unseat market leader ARM. Intel should, if anything, license ARM designs and make the best slices of silicon out there for ARM. What I'm saying is, talking about unseating is the wrong way to think about ARM.

petra 10y ago
It wouldn't be smart of ARM to help Intel and harm it's own ecosystem and especially the chip fabs,because once they are weakend enough, Intel could return to using it's own cores.
kps 10y ago
Intel is an ARM architecture licensee.
- rbanffy 10y ago
  I believe they sold that license about the time they went for Itanic.
- igravious 10y ago
  Ah, I should have checked. I did read through the ARM architecture page† on Wikipedia and it does not mention that fact. It does say, "In 1994, Acorn used the ARM610 as the main central processing unit (CPU) in their RiscPC computers. DEC licensed the ARM6 architecture and produced the StrongARM. At 233 MHz, this CPU drew only one watt (newer versions draw far less). This work was later passed to Intel as a part of a lawsuit settlement, and Intel took the opportunity to supplement their i960 line with the StrongARM. Intel later developed its own high performance implementation named XScale, which it has since sold to Marvell." I'm not sure how much sense it makes for Intel to be both an ARM licensee and develop Atom. † https://en.wikipedia.org/wiki/ARM_architecture#Core_licence https://en.wikipedia.org/wiki/ARM_architecture#Core_licence

iamgopal 10y ago

Could not they just use their superior manufacturing capability to make e.g. 24 core arm that use less power then competition ? If not for them, it will atleast bite the market share from competitor. This is quite analogous to "Nokia should have embraced android much ago."

andrepd 10y ago
Shamelessly stolen from the comments at http://anandtech.com/show/10288/intel-broxton-sofia-smartphone-socs-cancelled http://anandtech.com/show/10288/intel-broxton-sofia-smartpho... "The short answer: margins. The long answer: Intel could absolutely build a killer smartphone SoC on the 14nm process. However it would need to be Core based; the Atom CPU line isn't performant enough. They may need to go back to Ivy Bridge or strip down Skylake a bit to get there, but they could do it. The issue is that smartphone SoCs have very low margins. MediaTek, Rockchip, etc are crushing the market. Atom is designed for margins, not performance. If Intel built a winning Core smartphone SoC, then they'd have to sell it for significantly lower margins than what their other Core parts sell for. And that in turn would put pressure on the rest of their chip stack; why should an OEM pay $250 for Core-M when you could get a similar smartphone SoC for $30? For what it's worth Apple faces a similar challenge. But being vertically integrated means they don't have to share with others, and the total profit off of iPhones/iPads can offset the higher costs involved in developing and fabbing the A-series SoCs. However Intel doesn't have that luxury, and Asus would never let Intel have all of the profit. Smartphone SoCs are a race to the bottom. And Intel as a business has nothing to gain from taking part in that race. The winner is the guy who got badly hurt but didn't die; a Pyrrhic victory."
- yuhong 10y ago
  PCs were also a race to the bottom. I wonder what would happen if Intel or AMD bought Compaq or another PC manufacturer maybe in the 1990s and dominated the PC market by the mid-2000s.
  - hga 10y ago
    Intel probably wouldn't have been able to sustain the DNA required to also be successful in that market, and they would have had severe antitrust problems. AMD would have had the same first problem (but probably worse since they've historically had much poorer management), and would have likely lost a lot of their business selling to other PC manufacturers they'd now be directly competing with.

pier25 10y ago

Isn't Core M a tablet chip?

cageface 10y ago

It seems like with Intel's superior process and production scale they could easily enter into the ARM market. Is it just not profitable enough for them to do this? Or are their fabs too committed to producing x86 chips? Missing out on mobile entirely seems like a pretty big strategic mistake.

alrs 10y ago
Intel bought DEC's ARM business in 1997, tried doing ARM, and then gave up. https://en.wikipedia.org/wiki/StrongARM https://en.wikipedia.org/wiki/StrongARM
- toyg 10y ago
  To be fair, it was a different world (almost 20 years ago... I feel old). It doesn't matter anyway -- after several years concentrating their immense muscle on battery savings and integrated graphics, they have caught up a lot. CoreM / Skylake can fit most use-cases where ARM stuff would be called for, at least in the high-margin market.

turnip1979 10y ago

Dumb question ... Broxton = Atom? So Atom will be no more?

Sanddancer 10y ago
No. The Atom core's going to be used pretty heavily, just not as a tablet/phone core. It's just been moved to pretty much solely the entry level laptop/desktop market.

rbanffy 10y ago

This is a path that must be followed with great attention. Giving up on low-end, entry-level parts often looks great in short term results but may doom the company in the end. One of the reasons we deploy to x86 servers now rather then SPARC, POWER or MIPS is that we develop on x86. When Sun and IBM chose to exit the low-end space and pursue only large clients, they doomed their platforms to the legacy of those clients who already had their systems running on their platforms. How many fresh applications are designed with SPARC or POWER in mind? How many are designed for zSeries mainframes?

stesch 10y ago

Isn't the HoloLens using an Intel Atom processor?

api 10y ago

I don't necessarily think this is a bad decision. Intel currently dominates high performance computing and data center chips, and if they lose that edge they will really lose. Maybe it makes sense for them to instead double down on that and try to maintain their lead in those areas. Mobile markets are high volume but lower margin.