The Intel Atom Thread

[Sweepr]

Microsoft's assault on cheap Android tablets starts this week

Microsoft promised cheaper Windows tablets with its "Windows 8.1 with Bing" option for manufacturers, and they’re now starting to arrive. Toshiba is one of the first to announce a 7-inch Windows tablet this week at the IFA tradeshow in Berlin, signalling a new price point for mainstream manufacturers to battle the cheap Android alternatives. The new Encore Mini is a 7-inch (1024 x 600) Windows 8.1 tablet running a quad-core Intel Atom processor with 16GB of internal storage and an SD slot if you need more space. It’s a pretty bare-bones machine, but it’s priced accordingly: it retails for just $119.99 and is available now.

Either way, get used to these cheap Windows tablets as they’re about to get a lot more common. Microsoft is arming its closest partners with free versions of Windows in order to battle with the increasing amount of cheap Android tablets that are flooding the market. Toshiba might be one of the first to hit such low price points, but HP is also planning to release 7- and 8-inch versions of its new "Stream" PC line for just $99 this holiday season, both running Windows. Others are bound to follow, all while Microsoft prepares to merge its Windows RT and Windows Phone operating system to better cater for small tablets, albeit ARM-based ones, that will arrive next year.

Acer's also announcing an 8-inch (1280 x 800 IPS) Iconia Tab 8 W with Windows 8.1 and one year of complimentary Microsoft Office 365 Personal. It features up to eight hours of battery in a 9.75-mm thick chassis weighing 0.8 pounds. It'll begin shipping in October starting at 149 euro, and to "Pan America" in November for $149.99.

www.theverge.com/2014/9/3/6100741/toshiba-encore-mini-windows-tablet

Acer announces $150 Bay Trail-based Iconia One 8 and Iconia Tab 8 W tablets

Acer's taken more than a passing fancy to Chrome OS of late, but at this year's IFA, the company's showing a rekindled love for its affordable tablet range. Its first new slates since the beginning of the year come in two sizes, with the more portable 8-inch form factor also offering a choice of platform. The Iconia One Tab 8 runs Android 4.4 KitKat on a quad-core Intel Atom processor (the Z3735G, if you're interested), with an 8-inch, 1,280 x 800 IPS display up front. Otherwise, it's got most of the standard features you'd expect on a tablet, like a pair of cameras and microSD slot for storage expansion. Now picture essentially the same hardware configuration, but instead running Windows 8.1 with Bing, and you've got the Iconia Tab 8 W. Successor to the Iconia W4, the Tab 8 W also boasts up to eight hours of battery life and one free year of Office 365. Whether you prefer Google's or Microsoft's OS, both 8-inch slates will launch next month in Europe for €150, and in the US in November for $150. [...]

www.engadget.com/2014/09/03/acer-ifa-windows-android-iconia-tablets/

Acer introduces Switch 11 2-in-1 tablet, updates the Switch 10

The Aspire Switch 11 will come in two versions. The entry-level SW5-111 features an 11.6 inch, 1366 x 768 pixel display, an Intel Atom Z3745 quad-core processor, 2GB of RAM, 64GB of eMMC storage and 2 USB 2.0 ports.

Acer will also offer a more powerful SW5-171 model with support for up to an Intel Core i5 Haswell processor, 4GB of RAM, and up to 128GB of storage with a SATA hard drive. This model has USB 3.0 and USB 2.0 ports.

Both models feature 802.11n WiFi, Bluetooth 4.0, micro HDMI ports, and SD cards. The Switch 11 should be available in October for $399 and up.

Acer’s new Switch 10 SW5-012 will be available with either a 1280 x 800 pixel screen or a 1920 x 1200 pixel display, an Intel Atom Bay Trail processor, 2GB of RAM, 32GB to 64GB of storage, and up to 8 hours of battery life… or 7 hours of run time if you have the full HD model. It launches this month for $330 and up and comes with Microsoft Office 365 Personal pre-loaded.

http://liliputing.com/2014/09/acer-introduces-switch-11-2-1-tablet-updates-switch-10.html

HP Stream 7, 8 Windows tablets pop up in Europe

The HP Stream tablets are small Windows devices Intel Bay Trail chips, and a starting price of €179 in Europe. The 8 inch model will sell for €249 and up. Both models will likely be cheaper when they launch in the United States.

The HP Stream 7 features a 7 inch, 1280 x 800 pixel display, an Intel Atom Z3735G processor, 1GB of RAM, and 16GB of storage. It runs Windows 8.1 with Bing software.

The Stream 8 has similar hardware, but a larger screen.

http://liliputing.com/2014/09/hp-stream-7-tablet-pops-europe.html

Toshiba's Radius 11 puts a Chromebook's price on a convertible Windows 8 laptop

Looking at it sitting among Toshiba's range of laptops, the new Satellite Radius 11 doesn't immediately stand out. But push back against its staid appearance and you'll find its screen flexing a full 360 degrees. This unassuming 11.6-inch Windows 8.1 machine includes a touchscreen and a hinge that converts it into an impromptu tablet when the need arises. That's still not all that impressive, but then Toshiba told us the price for the Radius, which is an extremely affordable £329 in the UK. Lenovo's Yoga PCs have led the way with this variety of flexible computer, and they are still more refined in their design than Toshiba's Radius, but they've never been quite this cheap. Of course, that necessitates an otherwise barebones spec sheet for the Radius, with 4GB of RAM and 500GB of HDD storage allied to an Intel Celeron processor and 1366 x 768 resolution.

The Radius 11, meanwhile, will arrive in late October or early November. Again, no word on price just yet. In the meantime, we've got hands-on shots aplenty. Check 'em out.

www.theverge.com/2014/9/3/6100747/t...books-price-on-a-convertible-windows-8-laptop
www.engadget.com/2014/09/03/toshiba-chromebook-2-satellite-radius-11/?ncid=rss_truncated

Toshiba's Chromebook 2 adds a 1080p screen, cuts down on size and weight

Toshiba isn't showing off a whole lot here at IFA, Europe's biggest trade show. That said, the stuff it has announced at least manages to run the gamut. Today, for instance, the company unveiled both a slimmed-down Chromebook 2 and the Satellite Radius 11, a small-screened Windows convertible. Starting with the former, the Chromebook 2 has the same 13-inch screen size as its predecessor, along with the same textured plastic design. This time, though, it's thinner and lighter, at 2.95 pounds, with the sort of compact dimensions you'd expect to find on a 12-inch system. Toshiba bumped up the resolution too, so that you can now get it with a 340-nit, 1,920 x 1,080 IPS panel (the base model still has a 1,366 x 768 display). Depending on the resolution, you can expect either 11.5 hours of battery life on the lower-res version, or nine hours with full HD.

Toshiba won't be making people wait too long to try the new machines, either: the Chromebook 2 will be available as of October 5th in two different SKUs. The 1080p model with 4GB of RAM will retail for $329 — quite a bit less than Samsung's comparable Chromebook 2, which is priced at $399. There's also a lower-end SKU that keeps the original 1366 x 768 screen and only comes with 2GB of RAM; it'll be priced at $249.

www.theverge.com/2014/9/3/6097461/toshiba-chromebook-2-1080p-screen-cuts-down-on-size-and-weight
www.engadget.com/2014/09/03/toshiba-chromebook-2-satellite-radius-11/?ncid=rss_truncated

 

[liahos1]

amazing values for full windows and i think a year of free office 365.

 

[witeken]

Lots of low-res devices, unfortunately. At least the CPU is good, although 14nm would have been nicer. I also wonder how those devices will affect Windows' tablet market share.

 

[Shivansps]

Personally im thinking about buying this one, its a 3735D with 2GB and 64-bit ram
http://www.pandawill.com/voyo-winpa...40d-80-ips-screen-2gb-32gb-silver-p90401.html

but in kinda worried about that "1.46GB usable".

 

[witeken]

1.5GB RAM does't seem like a lot, indeed. Not sure how it will affect performance, though.

 

[witeken]

I always thought that Brian Krzanich had a great strategy with his "if it computes it does it best with Intel" goal, and now I just saw another example of that.

I just read a hands-on impression of Samsung's Gear VR. If using a smartphone as screen is the future of VR, then Intel will also miss out on VR if it doesn't have a presence in smartphones. Too bad Craig Barrett and Paul Otellini didn't realize the importance of that.

 

[liahos1]

Recent Q&A session with BK at the Citi tech conference. Thought it was interesting enough to listen to fwiw

http://www.veracast.com/webcasts/citigroup/technology2014/86202820698.cfm

that was an interesting comment on the pervasiveness of x86 from that anandtech link witeken. I think Intel's goal needs to be ultimately to make x86 pervasive as possible. BK was talking about 50c chips - thats what they are driving towards. at the end of the day they need that pervasiveness otherwise arm will take it all and make themselves the most relevant ISA.

I think they really need to figure out why they havent gotten any traction in phones since that is such a large volume market. If they can't win Apple/Samsung business they need to win Xiaomi and OnePlus One type business. It would also do a lot for the stock since nobody really cares (at least right now) about luxury jewelry/smartwatches sold at Barneys.

 

[witeken]

You have to register?

 

[liahos1]

You have to register?

ya just use your gmail address I guess. I am not a Citi client but accessed the webcast anyway with my work information.

 

[Khato]

I think they really need to figure out why they havent gotten any traction in phones since that is such a large volume market. If they can't win Apple/Samsung business they need to win Xiaomi and OnePlus One type business. It would also do a lot for the stock since nobody really cares (at least right now) about luxury jewelry/smartwatches sold at Barneys.

My current impression is that Intel's primary problem in smartphones is that they've been targeting the premium while not having the features necessary for such in the relevant markets. aka, who's going to buy a premium smartphone without LTE? Due to that their offerings are relegated to the mid-range where pricing doesn't necessarily support their current BoM costs. Now Merrifield and Moorefield could be acceptable in the premium space, but they don't offer much in the way of marketable features over the alternatives.

I'm not certain how much more luck Intel will have trying to get into the smartphone market from the bottom up with SoFIA though. There's definitely more opportunity in that area, but there's also more diverse competition.

 

[ams23]

Too bad Craig Barrett and Paul Otellini didn't realize the importance of [smartphones].

Well it is easy to criticize after the fact, but the reality is that with the original iPhone, Apple was completely unwilling to pay Intel enough money to even cover the forecasted cost of manufacturing the CPU! So Intel would have continued to lose more and more money on the deal as more and more units were shipped out. Now, unfortunately Intel overestimated the cost of manufacturing the CPU and underestimated the market impact of the iOS ecosystem and ARM ecosystem, but in the mid-2000's when these decisions were made, no one could have known for sure how things would play out.

 

[Arachnotronic]

Well it is easy to criticize after the fact, but the reality is that with the original iPhone, Apple was completely unwilling to pay Intel enough money to even cover the forecasted cost of manufacturing the CPU! So Intel would have continued to lose more and more money on the deal as more and more units were shipped out. Now, unfortunately Intel overestimated the cost of manufacturing the CPU and underestimated the market impact of the iOS ecosystem and ARM ecosystem, but in the mid-2000's when these decisions were made, no one could have known for sure how things would play out.

"Only the paranoid survive"

 

[ninaholic37]

On a related note the EeePC is back: http://www.engadget.com/2014/09/03/asus-eeebook-x205/

Hopefully the rebirth if the netbook lures people back from walled garden mobile OS.

Interesting. I never jumped on the Asus T100 bandwagon because it seemed too expensive ($349) for what it is, but this might work.

 

[ams23]

"Only the paranoid survive"

Heh, well sure, but I just don't see how Otellini would have been justified at the time to make that kind of deal. What was he going to tell shareholders: that he expected ARM processors and iOS software to be dominant in mobile devices five years down the road, and as a result, he decided to make a deal with Apple where he had essentially zero negotiating power on pricing and where he was forced to sell Intel x86 CPU's to Apple for less than the forecasted manufacturing cost, even though he had no idea what the actual volumes of this one premium developmental product would be nor did he have any idea on exactly how much impact this one premium developmental product would make on the market?

The reality is that Intel's product lineup, business model, and cost structure were not the right fit at that time for the iPhone developmental product. Apple was looking for a commoditized, low power, price-conscious application processor that was merely "good enough" at that time, not for something particularly special.

Do note that in the beginning when the iPhone first came out, not many people predicted that it would be a huge sales success. The reason the iPhone turned into a huge sales success is due in large part to wireless carrier subsidies. Apple is able to sell extremely high volumes of very premium iPhones in the USA because wireless carriers are willing to subsidize $400+ USD per phone.

 

[Sweepr]

Asustek readies new Bay Trail & Braswell products

In addition to an upgraded version of the T100, which is set to launch at the end of September or early October, Asustek Computer is ready to unveil its new Transformer Book T300 Chi, a larger version of the T100 in January 2015, according to sources from the upstream supply chain.

Asustek will also announce several inexpensive 14- to 15-inch K- and X-series notebooks in the fourth quarter, priced around US$249, using Intel bay Trail-M series processors. The company will also release notebooks using Intel's Braswell platform, which will become available in the second quarter of 2015.

Asustek will also release some Chromebooks in the fourth quarter looking to land orders from North America's education market.

http://www.digitimes.com/news/a20140902PD206.html&usg=ALkJrhiD4jqSem_gPcAHzC0qVqCFhX9Kiw

ASUS Outs $200 Windows 8.1 Notebook

Another entrant in the economical lappie race is ASUS, which debuted perhaps the most intriguing device of the group: a Windows 8.1 notebook that costs a piddly $200.

The 11.6-inch (1366 x 768) EeeBook X205 sports a quad-core Intel Atom chip (1.33 / 1.83 GHz) with 2 GB of RAM, and for connectivity there's 802.11a/b/g/n dual-band WiFi, Bluetooth 4.0, two USB 2.0 ports, micro HDMI out, a headphone/mic combo jack and a webcam. ASUS promises up to 12 hours of battery life under modest use.

ASUS EeeBook X205ASUS EeeBook X205The storage situation is multifaceted; in addition to 32 or 64 GB of onboard eMMC flash storage, you can get up to 115 GB of OneDrive storage free for a year, a whopping 500 GB of ASUS WebStorage free for two years, and you can add another 64 GB of storage via the MicroSD card slot. ASUS managed to subsidize the cost of this incredibly low-price device by loading it with the Bing version of Windows 8.1.

www.tomshardware.com/news/asus-acer-toshiba-windows-chromebook,27595.html

Quad-core Bay Trail, 2GB RAM, 12 hours battery life - that's a lot of value for a $199 Windows notebook.

HP launches new Chromebook 11 and Chromebook 14 with Intel and NVIDIA chips

HP is refreshing its line of Chrome OS laptops. As expected, the company’s new HP Chromebook 11 features an Intel Bay Trail processor instead of the Samsung Exynos 5 ARM-based chip used in HP’s last few 11.6 inch Chromebooks.

The smaller model features an 11.6 inch, 1366 x 768 pixel display, an Intel Celeron N2830 processor, 2GB or 4GB of RAM, 16GB of storage, and up to 8 hours and 15 minutes of run time from a 36Whr battery.

[...] HP says the new Chromebok 11 should be available in the US starting October 5th for $280 and up (although the product page currently shows a list price of $200).

http://liliputing.com/2014/09/hp-launches-new-chromebook-11-chromebook-14-intel-nvidia-chips.html

HP introduces 10 inch Pavilion x2 Windows tablet

HP launched a low-cost 2-in-1 tablet with an 11.6 inch display and an Intel Pentium Bay trail processor in late 2013. Now the company is adding a new model with a 10 inch screen and an Intel Atom processor. The new HP Pavilion x2 is a Windows convertible that gets up to 11.75 hours of battery life.

You can use the Pavilion x2 as a standalone table or use a it like a notebook thanks to a keyboard cover which also acts as a screen protector and a multi-position stand. The tablet is expected to go on sale October 29th for $330 and up.

http://liliputing.com/2014/09/hp-introduces-10-inch-pavilion-x2-windows-tablet.html

ECS Reveals LIVA Product Roadmap

The PC market is warming up to the ECS LIVA 'nettop' that we reviewed back in July. The fanless nature and low cost, as well as 'good-enough' performance serve as positives for the unit. Recently, ECS conducted a media event, where they gave a sneak peek into the future of the LIVA lineup.

One of the first things that needs to be cleared up is the configuration of the currently shipping 32 GB LIVA kit. Our review unit came with the Celeron N2806. Apparently, only the first batch of 600 units sported that SoC. Subsequent production batches have the Celeron N2807. The update addresses one of the improvement aspects we raised in our review - the use of a Bay Trail part enabled with Intel Quick Sync for transcoding. This opens up additional use-cases for the LIVA kit.

Coming back to the product roadmap, ECS is going to bring the 64 GB version into mass production soon. Next month, the company is also planning to release the BAT-MINI motherboard standalone into the market (users can bring their own chassis). In late November, we will see the 2nd generation LIVA. This is also based on Bay Trail, and comes in a solid metal chassis. ECS was mum on the internal configuration, but we can see that the chassis addresses concerns regarding the positioning of the USB ports, amongst other things. Next year the Bay Trail units will get upgraded to Braswell too.

http://anandtech.com/show/8484/ecs-reveals-liva-product-roadmap

 

[Arachnotronic]

Heh, well sure, but I just don't see how Otellini would have been justified at the time to make that kind of deal. What was he going to tell shareholders: that he expected ARM processors and iOS software to be dominant in mobile devices five years down the road, and as a result, he decided to make a deal with Apple where he had essentially zero negotiating power on pricing and where he was forced to sell Intel x86 CPU's to Apple for less than the forecasted manufacturing cost, even though he had no idea what the actual volumes of this one premium developmental product would be nor did he have any idea on exactly how much impact this one premium developmental product would make on the market?

The reality is that Intel's product lineup, business model, and cost structure were not the right fit at that time for the iPhone developmental product. Apple was looking for a commoditized, low power, price-conscious application processor that was merely "good enough" at that time, not for something particularly special.

Do note that in the beginning when the iPhone first came out, not many people predicted that it would be a huge sales success. The reason the iPhone turned into a huge sales success is due in large part to wireless carrier subsidies. Apple is able to sell extremely high volumes of very premium iPhones in the USA because wireless carriers are willing to subsidize $400+ USD per phone.

OK, I understand, but why are MediaTek, Qualcomm, Marvell, Spreadtrum, and a whole host of others able to offer competitive mobile chips while Intel continues to fail extremely hard?

What did these other, less powerful semiconductors see that Intel didn't see?

At any rate, it doesn't matter now. All that matters is that Intel is now spending $3B+ on mobile chip R&D and it still isn't putting out competitive chips in a timely fashion. This is a real problem and why you have seen folks like liahos1 and witeken so frustrated.

 

[Vesku]

OK, I understand, but why are MediaTek, Qualcomm, Marvell, Spreadtrum, and a whole host of others able to offer competitive mobile chips while Intel continues to fail extremely hard?

What did these other, less powerful semiconductors see that Intel didn't see?

At any rate, it doesn't matter now. All that matters is that Intel is now spending $3B+ on mobile chip R&D and it still isn't putting out competitive chips in a timely fashion. This is a real problem and why you have seen folks like liahos1 and witeken so frustrated.

Otellini gave a bit of an explanation when he left. Handset makers, and Apple who was looking to become one, want a SoC at a specific price usually single or low double digits and he along with his board weren't comfortable with spending the billions needed to get into a market with such low unit prices. Note how Intel is having to 'contra-revenue' with Bay Trail because the market is so cost sensitive.

 

[Nothingness]

OK, I understand, but why are MediaTek, Qualcomm, Marvell, Spreadtrum, and a whole host of others able to offer competitive mobile chips while Intel continues to fail extremely hard?

What did these other, less powerful semiconductors see that Intel didn't see?

At any rate, it doesn't matter now. All that matters is that Intel is now spending $3B+ on mobile chip R&D and it still isn't putting out competitive chips in a timely fashion. This is a real problem and why you have seen folks like liahos1 and witeken so frustrated.

IMHO Intel lacks agility, everything is spec'ed long before the ship is ready for customers. Their design cycle seems long probably due to much heavier validation phase (which is good) and much heavier design processes. That works very well for desktop/laptop chips but not for a market that changes very quickly.

 

[ams23]

OK, I understand, but why are MediaTek, Qualcomm, Marvell, Spreadtrum, and a whole host of others able to offer competitive mobile chips while Intel continues to fail extremely hard?

Qualcomm has achieved a dominant position largely due to their modem technology that they can bundle with their application processors. There are some very big wireless carriers in the USA such as Verizon and Sprint that have legacy networks that require Qualcomm's WCDMA modem tech. So until VoLTE becomes more widespread, this will be the status quo.

MediaTek has achieved a strong position largely due to being a very low cost supplier of application processors. They rely almost exclusively on low cost third party IP to develop their application processors, and do not develop the IP in house.

Apple develops their own application processors (through a combination of third party licensing of graphics cores, and ISA licensing of in-house CPU cores). Samsung either develops their own application processors using third party IP, or they use Qualcomm application processors.

Apple and Samsung also continue to receive the most web/TV/print promotion and marketing from wireless carriers, in addition to the largest wireless carrier subsidies to end users too.

So that doesn't leave too much room for other players to gain a foothold in the smartphone space.

 

[witeken]

Heh, well sure, but I just don't see how Otellini would have been justified at the time to make that kind of deal. What was he going to tell shareholders: that he expected ARM processors and iOS software to be dominant in mobile devices five years down the road, and as a result, he decided to make a deal with Apple where he had essentially zero negotiating power on pricing and where he was forced to sell Intel x86 CPU's to Apple for less than the forecasted manufacturing cost, even though he had no idea what the actual volumes of this one premium developmental product would be nor did he have any idea on exactly how much impact this one premium developmental product would make on the market?

The reality is that Intel's product lineup, business model, and cost structure were not the right fit at that time for the iPhone developmental product. Apple was looking for a commoditized, low power, price-conscious application processor that was merely "good enough" at that time, not for something particularly special.

Do note that in the beginning when the iPhone first came out, not many people predicted that it would be a huge sales success. The reason the iPhone turned into a huge sales success is due in large part to wireless carrier subsidies. Apple is able to sell extremely high volumes of very premium iPhones in the USA because wireless carriers are willing to subsidize $400+ USD per phone.

You make some good points, but if you're Intel, there's no way you would not have been able to see this coming. This is exactly what Moore's Law predicted and it happened. It isn't an easy/logical decision if you're a 60%+ margin company with high ASPs, but it would have been the right one.

 

[Arachnotronic]

You make some good points, but if you're Intel, there's no way you would not have been able to see this coming. This is exactly what Moore's Law predicted and it happened. It isn't an easy/logical decision if you're a 60%+ margin company with high ASPs, but it would have been the right one.

"But there's no sense crying over every mistake. You just keep on trying 'till you run out of [cash]" -- GLaDOS.

 

[ams23]

You make some good points, but if you're Intel, there's no way you would not have been able to see this coming. This is exactly what Moore's Law predicted and it happened. It isn't an easy/logical decision if you're a 60%+ margin company with high ASPs, but it would have been the right one.

Intel did predict the miniaturization of computing because they started to invest in and develop low power processors close to ten years ago, but there are outside market forces that they couldn't control such as the use and acceptance of very low cost commoditized ARM processors running on Android rather than Windows, the use of large legacy wireless networks that specifically require one competitor's modem technology, the use of absolutely massive wireless carrier subsidies for high end Apple and Samsung phones which makes the price accessible to countless millions of people who would otherwise not be able to easily afford it, etc.

 

[Arachnotronic]

Intel did predict the miniaturization of computing because they started to invest in and develop low power processors more than ten years ago, but there are outside market forces that they couldn't control such as the use and acceptance of very low cost commoditized ARM processors running on Android rather than Windows, the use of large legacy wireless networks that specifically require one competitor's modem technology, the use of absolutely massive subsidies for high end Apple and Samsung phones which makes the price accessible to countless millions of people who would otherwise not be able to easily afford it.

The real problem is that Intel under-invested in its low-power micro-architectures/IP, failed to develop a robust SoC methodology internally that could lead to many variants of chips produced quickly, and heavily under-invested in communications/wireless technology which is integral to this market.

Intel is correcting course; we'll see where they land in a couple of years.

 

[ams23]

The real problem is that Intel under-invested in its low-power micro-architectures/IP, failed to develop a robust SoC methodology internally that could lead to many variants of chips produced quickly, and heavily under-invested in communications/wireless technology which is integral to this market.

Investing more heavily would not have changed any of the market forces that I mentioned above.

 

[Shivansps]

mmm this looks like a good deal

http://www.aliexpress.com/item/Onda...1-Intel-Z3735-Quad-Core-8-IPS/2009609749.html

it fixes all the issues i hate of the Toshiba tablet...