Intel eyes China's Internet of Things market

Intel Corp will step up marketing investments in China, as the United States chip giant aims to reshape its image in one of the world's largest chip markets and better tap into opportunities brought by the Internet of Things.
Penny Baldwin, vice-president in the Global Marketing and Communications Group at Intel, said: "The only thing about China is that we need to spend more to expand in this market which is so big and has so many people. We will focus on the millennial generation, which are not only our core consumers but influential business decision makers."
Baldwin made the comment in an interview with China Daily in Beijing. According to her, the US PC chip giant, which has failed to catch up with rival Qualcomm Inc in the mobile Internet era, will spare no efforts to exploit the opportunities brought by the Internet of Things.
"By 2020, over 50 billion devices will be connected and this is exactly what Intel is preparing for," she added, identifying drones, augmented reality, and robots as the company's priorities.
The company is making a major shift in its branding strategy, shifting from a chip supplier behind hardcore manufacturers to a firm that wants to be more close to consumers and be associated with fashion and cultural symbols.
"We want to develop preferences and brand loyalty among consumers, based on which we can boost our brand premium and have more power in hands to negotiate with our manufacturing clients,"
In 2015, Intel has partnered with popular movie makers and TV program producers to expand their marketing campaign. Instead of simply imputing its logos in these programs, Intel invited content producers to shoot their laboratories, giving audience a peak at how employees at the technology giant work in daily lives.

http://www.chinadaily.com.cn/business/tech/2016-08/31/content_26655716.htm

Japan Touts Adjustable Crypto for IoT, Cloud

Japanese engineers claimed to have developed the first secure "cryptoprocessor" able to adjust security levels on the fly without first upgrading the device itself. Previously, the researchers noted, encryption hardware required an upgrade as secure key lengths using approaches like RSA public key cryptography increased beyond 512 bits.
Japan's Science and Technology Agency supported secure cryptoprocessor development conducted by researchers at Okayama University and industrial partner Tokyo Electron Device Ltd. (TYO: 2760). Along with adjustable security levels, the researchers said the new device supports a variety of emerging security approaches, including elliptic curve and pairing-based cryptography.
According to the cryptoprocesser developers, "It is not easy to seamlessly adjust the security level of devices because public key cryptographies are…based on some difficult and complicated mathematic problems." For example, secure key lengths of RSA cryptography increase from 512, 1024, 2048, and then 3072 bits. "Accordingly, cryptoprocessors need to be upgraded together with their arithmetic architectures," they explained.
The new device implemented on a field-programmable gate array (FPGA) board on which the secure processor is integrated is touted as capable of handling a range of security strengths between 512 and 5,120 bits based on elliptic curve and pairing-based cryptography. The developers noted that security levels could be adjusted without hardware modifications.
The ability to scale the security chip is based on the Okayama University researcher's proprietary scheme called the "cyclic vector multiplication algorithm," which is said to provide scalability for security applications.
With a device maker in hand and an FPGA implementation available now, the researchers said they expect the cryptoprocessor to be embedded in a range of "ubiquitous terminals" and Internet of Things devices. The public-key security systems based on RSA and elliptic curve cryptography is widely used for digital authentication of users and devices.
The university researchers said encryption strength provided by the flexible crypto platform could be adjusted depending on the sensitivity of data being transmitted, normal for an email and high for classified data, for example.
Power consumption also could be adjusted according to performance requirements, they added.
Meanwhile, as concerns about IoT and cloud security grow, greater emphasis is being given to designing security into hardware, applications, sensors and instruments. Whether those security features include adjustable encryption schemes like the Japanese FPGA implementation remains to be seen.
Earlier attempts to secure vulnerable RF links between, say, RFID tags and mobile payment systems using encryption algorithms have achieved mixed results. According to one report, researchers recently defeated a security algorithm designed to secure the air interface between devices that would be connected to the IoT.
Hence, observers argue that device security may have to be raised to a level equivalent to critical IT infrastructure. That could create an opening for flexible, FPGA-based approaches such as the Okayama University partnership.

http://www.enterprisetech.com/2016/08/29/japan-touts-adjustable-crypto-iot-cloud/

Intel aims to challenge TSMC over Apple chip orders by 2018

TAIPEI -- Intel's recent pledge to expand its business making chips for others highlights its ambition to snatch chip orders for Apple's popular iPhones from Taiwan Semiconductor Manufacturing Co. as early as 2018, industry experts said.
Intel, the world's largest chipmaker by revenue, announced earlier this month that it will license technology from British mobile chip designer ARM with the aim of securing more business from smartphone companies. LG Electronics will become the first smartphone company to adopt Intel chips following the ARM deal.
While Intel still dominates in the faltering personal computer market, it has been struggling in the mobile era with ARM virtually monopolizing smartphone chip technology.
TSMC, the world's largest contract chipmaker, has thrived partly from its close partnership with ARM during the rapid rise of smartphones over the past decade, but Intel's new alliance with the British innovator could be a game changer.
"TSMC could face tough competition as soon as 2018 or 2019 as Intel is likely to gain orders from Apple by then," Samuel Wang, a veteran semiconductor analyst at research company Gartner, told the Nikkei Asian Review. "Intel has begun to engage with Apple and it aims to grab one or two top-tier customers from TSMC."
Intel fumbled in its previous attempts to grow its foundry business, where it makes chips for others, as the U.S. company prioritized production of PC and server chips for its own use, according to Wang.
But that could change this time around, he said, as Intel now sees its foundry business as a key source of new revenue, which can be used to invest in future technologies.
A senior Taiwanese chip industry executive shared Wang's views. "Intel is definitely the most formidable challenger for TSMC," the executive told the NAR.
"There is no rivalry between Apple and Intel so it's really likely that Apple could shift some orders there. The move is also in line with Washington's policy to encourage U.S. companies to make more products at home," the person said.
Intel declined to comment. Apple did not immediately return a emailed request seeking comment.

Intel Vice President Zane Ball, who oversees the company's foundry business, told developers in mid-August that the U.S. chipmaker can offer a better service than its rivals because it is not just a chip producer.
"We can do design, and we can do assembly and test. We have remarkable sources and we will bring these capabilities to our customers as well." Ball added that with ARM's licensing support, Intel is able to do more in the mobile chip sector.

The move by Intel to promote its foundry business is an attempt to compensate for an ongoing decline in the PC chip market. For the three months ended in June, Intel's net profits halved to $1.3 billion from a year ago though its revenue rose 3% to $13.5 billion.
Intel is currently making only $100 million annually from its foundry business, with a limited number of customers including Chinese mobile chip designer Spreadtrum Communications, a key rival to Qualcomm of the U.S. and MediaTek of Taiwan. Intel has a 20% stake in Spreadtrum.
Up until now, the Santa Clara, California-based company has mostly stuck to designing, producing and marketing chips under its own brand.
Yet with its colossal size, it could still pose serious threats to rivals in the foundry market. Founder and Chairman of TSMC, Morris Chang, said in June that Intel is a potential rival for his company in the longer term.
Intel, TSMC, and Samsung Electronics are the world's three leading chipmakers, and the only players capable of spending big sums to develop cutting-edge semiconductor technology and build advanced production facilities.
They will allocate $9.5 billion, $10 billion, and $11 billion respectively for capital expenditure in 2016, according to research firm IC Insights.
TSMC currently controls 55% of the global foundry market, and also accounts for 55% of global smartphone chip production.
For 2016 and 2017, TSMC is reportedly monopolizing orders for iPhone core processors, while it still had to share Apple business with Samsung in 2015.
Last year, TSMC booked revenue of 843.5 billion New Taiwan dollars ($26.61 billion) from making chips for 470 customers worldwide including Apple, Qualcomm, Nvidia of the U.S., MediaTek, and Huawei of China's chip arm Hisilicon Technology.
Apple contributed to 16% of TSMC's revenue in 2015 and that is expected to increase to around 20% this year.
TSMC spokesperson Elisabeth Sun said competition was always out there and the Taiwanese chipmaker would continue to provide the best services to customers.
"TSMC's mission is to provide the world's No.1 foundry services to customers and we will keep doing that," Sun told the NAR.

http://asia.nikkei.com/Business/AC/Intel-aims-to-challenge-TSMC-over-Apple-chip-orders-by-2018

Intel pushes 3D NAND to inexpensive SSDs

Intel is expanding its lineup of SSDs with its 3D NAND chips with more affordable consumer and enterprise drives.
The 3D NAND chips have a structure that makes SSDs durable and fast but was only available in a handful of drives. The new SSDs support the NVMe protocol, which offers faster throughput than the SATA controller originally designed for hard drives.

The new SSDs include entry-level consumer drives starting under US$100. Gamers may want to wait, however, until Intel releases its crazy-fast Optane SSDs, which the chip maker claims will be up to 10 times faster than regular SSDs.
The consumer SSD 600p series for PCs starts at $69 for a 128GB drive and ranges to $359 for a 1TB drive.

The SSD Pro 6000p for business laptops is slightly more expensive at $79 for a 128GB drive and up to $364 for the 1TB drive.
The PC SSDs plug into m.2 slots and have similar speeds but differ on features. The SSD Pro 6000p has enterprise features like remote erase, where the disk can be wiped remotely. It also has fine-grained power management features not available with the 600p. Those features work with Intel's Core vPro processors, allowing system administrators to take control of remote PCs.
The drives have sequential read speeds of 1,800 megabytes (MBps) per second and write speeds of 560MBps. They have random read speeds of 155,000 IOPS (input-output per second) and write speeds of 128,000 IOPS. On all counts, the 600p series is significantly faster than Intel's SSD 540s, which came out earlier this year.
Intel's new drives also include the enterprise SSD DC P3520 series, which has storage capacities from 450GB ($294) to 2TB ($984). The SSDs plug into 2.5-inch and standard PCI-Express 3.0 slots.

The 2TB drive is the fastest in the series, with random sequential read speeds of 1,700MBps and write speeds of 1,350MBps. The random read speeds are 375,000 IOPs and write speeds are 26,000 IOPS.
Those speeds beat Samsung's 1.92TB PM953 enterprise SSD based on NVMe, which has a sequential read speed of 1,000MBps, write speed of 870MBps, and a random read speed of 240,000 IOPS and write speed of 19,000 IOPS.
Another Intel SSD for data centers called the SSD DC S3520 Series "is ideal for those making the initial transition to SATA SSDs from hard drives," an Intel spokesperson said in an email. They come in capacities from 150GB ($89) to 1.6TB ($739). The drive comes in m.2 and 2.5-inch slot formats.
The SSD DC S3520 is a kind of cache where data is stored temporarily while being processed by computers. It is used to boot systems, serve up commonly used applications or data over the internet, or to store database or analytics data being processed. It's not meant for long-term storage, and has low power-consumption and read and write speeds, making it more appropriate for storage arrays.

http://www.computerworld.com/article/3112635/data-storage/intel-pushes-3d-nand-to-inexpensive-ssds.html

Samsung aims to conquer the memory market with HBM3

rue advances in technology are rare. The expense and difficulty of launching brand-new initiatives means that companies tend to prefer iterative improvements. Every now and then, however, we get the best of both worlds — an iterative improvement that could deliver enormous gains to a wide slice of the consumer market. At Hot Chips, Samsung unveiled a pair of initiatives that could revolutionize computer memory by pushing High Bandwidth Memory further on the one hand, while cutting costs and introducing the technology to all-new markets on the other.
Let’s take them one at a time.

Low-cost HBM clears the path for less expensive devices

As we’ve discussed previously, HBM stacks memory chips on top of each other around a central core. The stacks are all connected by wires that run through each memory die (these are called through silicon vias, or TSVs) and the entire chip structure sits on an interposer layer. The resulting configuration is sometimes referred to as a 2.5D architecture. The advantage is vastly increased memory bandwidth and much lower power consumption compared with GDDR5. The disadvantage is cost. While HBM proved competitive with GDDR5 at high frequencies and loadouts, the technology is currently limited to the top of the graphics market. AMD’s upcoming Vega is expected to use HBM rather than GDDR5X, but that chip will target the $300+ segment.

Samsung is proposing a low-cost HBM that would reduce costs in multiple ways. The number of connects per-die would shrink, reducing the number of vias required for each chip. The company wants to replace the large silicon interposer with an organic layer, and believes it can cut costs by removing the on-die buffer as well (how this would impact the overall design remains uncertain). While the resulting HBM variant would have less overall bandwidth than HBM2, Samsung believes it can compensate by increasing the clock rate (presumably without compromising HBM’s overall design, which emphasizes low clock rates and extremely wide buses).
If successful, this low-cost HBM could drive the memory into markets where it can’t currently compete, including low-end graphics cards and the APU market. Right now, Intel has a potent GPU competitor with its Crystal Well, which puts 64-128MB of EDRAM on-package with the CPU. AMD doesn’t really have an answer to Crystal Well at present, and the company’s on-die graphics are already bandwidth-limited. One potential solution is to adopt HBM for APUs and offer a chip with a unified memory pool for CPU and GPU in a single package — but that can only happen if HBM prices drop enough to justify its inclusion. Any push to cut these costs could result in a much improved HBM technology deploying on APUs and other types of SoCs. But it’s not clear how power consumption would compare with other low-power technologies or whether or not we’d see the technology in 15-25W laptops.

HBM3: More capacity, more bandwidth

Samsung’s HBM3 is a straightforward improvement on HBM2 that would debut in 2019 or 2020 and offer higher densities, higher stacks (more RAM per chip, more chips per stack), and 2x the maximum bandwidth of HBM2. The goal is to reduce the core voltage (currently 1.2V) and the I/O signaling power, according to Ars Technica, while improving maximum performance.

Photo courtesy of Ars Technica

HBM3 could allow for 64GB of memory on-die and 512GB/s of memory bandwidth per stack. A four-way stack of HBM3 would offer 2048GB/s of memory bandwidth in aggregate, compared with 1024GB/s with HBM2 and 512GB/s of HBM (all figures assume a four-stack configuration). This kind of bandwidth increase would give graphics cards or other peripherals far more memory than even the highest-end cards offer today and could be critical to driving next-generation VR systems.
The memory industry, however, isn’t as unified on HBM as you might think. As Anandtech details, both Micron and Samsung unveiled proposals for next-generation graphics and desktop memory (DDR5 and GDDR6, respectively). Xilinx is more commonly associated with FPGAs, not RAM. But Samsung used its own presentation to discuss how proper cooling technology is essential to large-scale die stacking and to call for the development of materials that can operate well at higher temperatures.
While many of these proposals are just that — proposals — they point the way to a potential revolution in gaming and high-end applications, while reduced cost and lower power options could extend those revolutions into form factors and power envelopes they currently can’t touch.

http://www.extremetech.com/gaming/234333-hbm-everywhere-samsung-wants-hbm3-low-cost-options-to-blow-the-doors-off-the-memory-market

IBM launches flash arrays for smaller enterprises, aims to court EMC, Dell customers

IBM launched all-flash arrays designed for small and midsized companies.

The company is also aiming a migration program designed to poach customers from the likes of Dell and EMC. IBM's "Flash In" migration program is carried out by its various partners. Via Flash In, IBM is looking to integrate its systems with storage rivals or replace them.
IBM launched the Storwize V7000F and Storwize 5030F as mid-range and entry level flash systems. The systems come with Spectrum Virtualize, which is software designed for data compression, provisioning, and snapshots across various systems.
The Storwize 7000F boosts performance and supports more clustering to manage data. In addition, the Storwize 5030F is an entry-level system.
Overall, Big Blue is hoping that the new systems will be used for workloads for SQL Server to MySQL to the Internet of Things.
IBM said the systems start at $19,000.

http://www.zdnet.com/article/ibm-launches-flash-arrays-for-smaller-enterprises-aims-to-court-emc-dell-customers/

HK scientist develops small lasers that enhance light-based computing

HONG KONG, Aug. 22 (Xinhua) -- Researchers at The Hong Kong University of Science and Technology (HKUST) have fabricated microscopically-small lasers directly on silicon, enabling the future-generation microprocessors to run faster and less power-hungry.
The innovation was made by Kei-may Lau, chair professor of the Department of Electronic and Computer Engineering of HKUST, in collaboration with the University of California, Santa Barbara, and Sandia National Laboratories, and Harvard University.
Silicon forms the basis of everything from solar cells to the integrated circuits at the heart of our modern electronic gadgets. However, the crystal lattice of silicon and of typical laser materials could not match up, making it impossible to integrate the two materials.
Now, Lau's group managed to integrate subwavelength cavities - the essential building blocks of their tiny lasers - onto silicon, allowing them to create and demonstrate high-density on-chip light-emitting elements.
"These whispering gallery mode lasers are extremely attractive light source for on-chip optical communications, data processing and chemical sensing applications," Lau said.
"Putting lasers on microprocessors boosts their capabilities and allows them to run at much lower powers -- a big step towards photonics and electronics integration on the silicon platform and a key solution to the next-generation green information technology."
For years, photonics had been the most energy-efficient and cost-effective method to transmit large volumes of data over long distances, now with these new silicon-based integrated lasers, photonics may be able to use for short-distance data transmission as well, which is set to greatly enhance the speed of data communication.
The finding was recently published as the cover story on Applied Physics Letters.

http://news.xinhuanet.com/english/2016-08/22/c_135624779.htm

California tops electronics theft leaderboard in USPost title

California accounted for more than a third of all electronic cargo thefts in the US over the last couple of years, according to a new report.
The analysis by FreightWatch International covers the period between the third quarter of 2014 and the second quarter of 2016 reveals that California was well ahead of all other US states, with Florida ranked second at 10 per cent of thefts and Tennessee third with 9 per cent.
California's share - 36 per cent of the total in the period - is attributed to the sheer volume of imported electronics flowing through the state and the number of electronics firms based there. Freightwatch also says that the main drivers for thefts appear to be opportunism as well as major commercial hardware releases.
Unsecured parking areas are still the most likely place for thefts to take place, and in a majority of cases (59 per cent) the criminals steal a full truck load (FTL) of the items.
As in previous reports televisions and displays were the most commonly stolen items, although the average value of the thefts (around $131,000) was considerably lower than the average for the sector as a whole ($509,000).
The most valuable stolen shipments were software, components and peripherals ($1.27m) and cell phones and accessories ($1.07m), likely because a lot of relatively high-cost items can be loaded onto a single truck.
FreightWatch warns however that "high value alone will not be the determining factor in theft risk to cargo, as thieves will adjust to the increased risk and modify their efforts accordingly."
"New threats will take shape in the form of new theft methods and new targeted products," it adds. For example, there have been a rising number of incidents involving the use of GPS jamming equipment as well as thefts from moving trucks.
Overall, FreightWatch reports that there were 166 cargo thefts in the second quarter of this year, with food and drink shipments targeted most at 23 per cent of the total and electronics coming in second at 13 per cent. California accounted for 40 per cent of all theft incidents.

https://www.securingindustry.com/electronics-and-industrial/california-tops-electronics-theft-leaderboard-in-us/s105/a2895/#.V7shC6KvKix

Analog Devices boosts cybersecurity portfolio with acquisition

Analog Devices has announced the acquisition of the Cyber Security Solutions (CSS) business of Sypris Electronics.

The CSS business of Sypris Electronics specialises in secure system and software technology, and delivers information security services to customers, including military and government organisations.
With this transaction, ADI says it has enhanced its aerospace and defence capabilities in the area of secure radio communications, bolsters its portfolio of system hardware and software-based cryptographic technologies, and adds a cybersecurity software and services.
ADI senior vice president, Dick Meaney, said: “We have accelerated our ability to offer our customers trust ‘from the sensor to the cloud’ through next-generation cyber-physical security solutions. The competencies and offerings of the CSS business are a great complement to our IoT solutions, as security increasingly becomes paramount to the success of IoT implementations.”

http://www.newelectronics.co.uk/electronics-news/analog-devices-boosts-cybersecurity-portfolio-with-acquisition/144672/

Intel IoT Executive: We're 'Uniquely Positioned' To Partner For The Internet Of Things

Intel executive Venkata “Murthy” Renduchintala said Wednesday that the company is “uniquely positioned” as a partner for tapping into the lucrative Internet of Things market.
“The interconnectedness of everything is something that will be a real game-changer,” he told partners and developers during his keynote at Intel Developer Forum, which takes place in San Francisco this week. “And Intel is uniquely positioned to partner with you to make that happen.”
Intel’s Internet of Things platform – which is made up of reference architectures, as well as processors that enable sensors and gateways at the edge of Internet of Things solutions – will make up the backbone of IoT applications for industrial, enterprise and vertical markets, said Renduchintala, vice president and general manager of Internet of Things at the Santa Clara, Calif.-based company.

Renduchintala, a former Qualcomm executive who Intel hired in November, has been charged with leading tIntel's shift in strategy from a PC-focused company to a connected device, cloud and data center company with a tighter emphasis on the Internet of Things.
Intel during Renduchintala’s keynote showcased different use cases for the Internet of Things in real life – including connected solutions for firefighters.
As part of this solution, Intel partnered with Honeywell to connect a Quark processor to the firefighters' self-contained breathing apparatus so that while they are in dangerous areas others can track how much oxygen they have left in their tank – and ultimately make critical life decisions based on that information.
As part of Intel’s Internet of Things strategy, Renduchintala said the company will focus on the development and deployment of 5G, the next generation of mobile telecommunications standards.
This broadband technology will provide better speeds and will be able to keep up with the explosion of connected devices as part of the Internet of Things. Renduchintala said Intel is aiming for 5G trials in 2018, and commercial deployment of the technology by 2020.
“Given the M&A activity in this space, and Intel’s internal changes, partnerships and stated intent in the IoT market, Intel partners -- including Five Nines IT Solutions – have a compelling story to talk about with our customers,” said Douglas Grosfield, founder and CEO of Five Nines IT Solutions, a Kitchener, Ontario-based strategic service provider.  “Security in the IoT space is playing a game of catch-up now that the industry’s focus is turning to risk mitigation instead of just getting any IoT-enabled product to market as fast as they can.”
For solution providers, Intel’s “muscle” in the Internet of Things will help create opportunities for its channel partners, said Grosfield.

“Ubiquity of Internet-enabled devices is growing, and control over the sanctity of the data these devices share is an important piece of the puzzle, albeit an afterthought until lately. Intel will bring much-needed muscle to the game, and is in a position to effect positive change, which will always spell opportunity for Intel partners,” he said.
Also at the Intel Developer Forum, Intel introduced Joule, a module built for IoT applications by packing a high-performance system-on-module into a low-power package. The maker board enables users to take a concept from prototype into production at a “fraction of the time and development cost,” according to Intel.
Intel unveiled its Knowledge Builder Toolkit as well, which enables developers to create intelligent sensing applications that run directly on Intel’s low-power Curie module solution for wearables. The toolkit will be available in the first quarter of 2017, according to Intel.

http://www.crn.com/news/networking/300081759/intel-iot-executive-were-uniquely-positioned-to-partner-for-the-internet-of-things.htm

Intel to Build 10nm Mobile Chips for ARM

The chip-making rivals have reached a deal that will help Intel grow its foundry business while giving ARM another top manufacturer for its mobile processors.


SAN FRANCISCO—Intel and ARM have been the major players in what has become the central competition in the processor market. Intel has failed to make much headway into a mobile chip space dominated by ARM and its partners, while ARM has been looking to make traction in data centers, where Intel silicon powers more than 95 percent of servers. Now, these two rivals are coming together in a partnership that promises to benefit both companies. At the Intel Developer Forum (IDF) here Aug. 16, officials from both vendors announced that Intel's chip foundry business will build chips for mobile and consumer devices based on ARM's architecture. The agreement will boost Intel's custom manufacturing business, which has been building for the past few years, by adding IP from the company whose architecture powers almost all of the smartphones and tablets in the world. For ARM and its partners—including Qualcomm, Samsung and Apple—the deal gives them access to Intel's powerful chip-manufacturing capabilities and cutting-edge technologies, as well as a foundry option beyond Globalfoundries, Taiwan Semiconductor Manufacture Corp. (TSMC) and Samsung. "Bringing these [vendors] together makes sense," Will Abbey, general manager of ARM's Physical IP Group, said during a technical session at IDF.

Both Abbey and Zane Ball, vice president and general manager of Intel's Custom Foundry business, noted that Intel and ARM had at times worked together in the past—for example, Ball pointed out that field-programmable gate arrays (FPGAs) by its Altera subsidiary are built on ARM's architecture. In a post on the ARM blog, Abbey wrote that "Intel and ARM have worked together for years to help enable the ecosystem, and this is just the latest milestone in that long-standing relationship."
However, this is a significant step for both companies. Intel launched its foundry system several years ago, giving third parties access to its substantial chip-manufacturing technologies. Ball said the company has grown its foundry capabilities from its 22nm and 14nm FinFET processes, and is now doing the same at 10nm. The business "is helping customers around the world by providing access to Intel's technology and manufacturing assets through turnkey services including design, wafer manufacturing, packaging and testing," he wrote on the company blog. "We're enabling new products and experiences on Intel's leading-edge technologies with industry standard design kits, silicon-proven IP blocks, and design services spanning from low-power SOCs [systems-on-a-chip] to high-performance infrastructure devices." ARM will be able to leverage that 10nm design platform for its Artisan physical IP, which includes memory compilers, high-performance and high-density libraries and point of presence (POP) IP for future mobile chips, Ball said. IP development is underway and ARM chips manufactured by Intel are expected to be available in early 2017. Intel famously missed out on the rise of smartphones and tablets, and failed in its efforts over several years to catch up. Under CEO Brian Krzanich, the company in April announced it was canceling development of several low-power chips that had been aimed at mobile devices, opting instead to focus those resources on such emerging growth areas as the cloud, robotics, artificial intelligence (AI), data analytics and the internet of things (IoT). Intel has been working with Qualcomm, Applied Micro, Cavium and other chip makers to drive its low-power architecture into the data center to compete with Intel in such areas as hyperscale environments and cloud service providers. In addition to the ARM announcement, Intel's Ball said LG Electronics also will build chips using the foundry business' 10nm process, while Spectrum, a major mobile chip maker in China, is using Intel's 14nm foundry platform. Achronix Semiconductor and Netronome are using Intel's 22nm process for their respective networking chips, and Altera is using the foundry platform to build what Ball called "the first true 14nm FPGA."

http://www.eweek.com/mobile/intel-to-build-10nm-mobile-chips-for-arm.html

Step-Down Supply Dissipates Low Power, Needs No Inductors

The LTC3256 from Linear Technology Corporation is a highly integrated, high-voltage, low-noise, dual-output power supply, which takes a single positive input and generates 5 V and 3.3 V step-down supplies with high efficiency and no inductors.

The device features a 5.5 V to 38 V input voltage range and includes independently enabled dual outputs: a 5 V 100 mA supply, and a 250 mA 3.3 V low-dropout (LDO) regulator, for a total of 350 mA available output current. This combination of regulators offers much lower power dissipation than a dual-LDO solution. For example, at 12 V VIN and maximum load on both outputs, power dissipation is decreased by over 2 W with the LTC3256, thus reducing significant heat and input current.
The LTC3256 is engineered for diagnostic coverage in ISO26262 systems and incorporates numerous safety and system-monitoring features. The device is suited for applications requiring low noise, low power rails from a high-voltage input such as automotive ECU/CAN transceiver supplies, industrial and telecom housekeeping supplies, and general-purpose, low-power 12 V to 5 V and 3.3 V conversion.
The LTC3256 maximizes efficiency by running the charge pump in 2:1 mode, over as wide an operating range as possible, and automatically switches to 1:1 mode as needed, consistent with VIN and load conditions. Controlled input current and soft switching minimize conducted and radiated EMI. The device offers low quiescent current of only 20 μA with both outputs regulating (no load) and 0.5 μA in shutdown. The integrated watchdog timer, independent power-good outputs, and reset input ensure reliable system operation and enable fault monitoring. A buffered 1.1 V reference output enables system self-testing diagnostics for safety-critical applications.
The LTC3256 also has additional safety features, including over-current fault protection, over-temperature protection and tolerance of 38 V input transients.


http://electronics360.globalspec.com/article/7139/step-down-supply-dissipates-low-power-needs-no-inductors

The Role Of Energy-Efficient Circuits In Wearable Healthcare Applications

As beneficial as they are, health monitors for conditions like high blood pressure, arrhythmia, and epilepsy can be uncomfortable and inconvenient due to all of their protruding wires. This opens up an opportunity for designers of wearable healthcare applications.
“Wearable electronics are needed for proactive healthcare,” said Dr. Jerald Yoo, an associate professor in the Department of Electrical Engineering and Computer Science at the Masdar Institute of Science and Technology in Abu Dhabi. The ability to detect and treat chronic disease early can be a powerful countermeasure. However, this effort calls for the collection and monitoring of voluminous amounts of data, which is where wearable healthcare devices come in.
For wearable (and wireless) healthcare devices to be successful, Yoo noted, they must be energy efficient, minimally obtrusive, and disposable. Yoo discussed the role of energy-efficient circuits in wearable healthcare applications during a recent talk to Cadence employees at the company’s San Jose headquarters.
Healthcare wearables: what goes into the design?
According to the World Health Organization, about 50 million people worldwide have epilepsy. Currently, diagnosing this severe neurological disorder involves doctors interviewing the patient and administering an electroencephalogram (EEG) test, said Yoo. But these methods are hardly conclusive—what is really needed is continuous monitoring, he said.
During his talk, Yoo discussed the challenges and techniques to designing biomedical circuitry. As an example, he highlighted a closed-loop seizure detection microsystem. Creating such a system calls for several key components.
First comes the platform. Here, Yoo considers the introduction of printed fabric circuit boards about six years ago to be quite the revelation. Direct screen-printing of conductive ink on fabric has made many wearable applications possible. The technology also provides an alternative to wet electrodes (which can trigger skin sensitivities if worn for long periods) and dry electronics (which have high electrode impedance and, thus, more noise). Designers creating fabric circuit boards must address a number of challenges, including pad number limits and issues such as heat protection, static and dynamic parameter variation, and high impedance.
Next is the sensor I/F circuit—basically, these designs should use low-noise, energy-efficient implementation circuits. Yoo noted that here, it’s important to have a dedicated DC server loop to remove the electrode offset. Since the servo loop itself elevates noise, Yoo has worked with his students to create a design prototype of a wearable EEG that includes a 500Hz chopper at the servo loop for better noise efficiency.
The digital backend is also critical, providing patient-specific classification and requiring energy efficiency. In this area, there are some distinct EEG seizure detection challenges to be aware of. Namely, intra-patient age-to-age EEG variations and spatial EEG variations are unexpected outcomes. “Many times, the pattern for seizure and non-seizure is very different. The seizure pattern from patient A has almost no correlation to patient B. Other chronic diseases have similar issues,” he explained.
Machine learning addresses variations
Yoo has found that the introduction of machine learning via support vector machines (SVMs) provides a way to resolve these unexpected outcomes. There are two options here: linear SVM (LSVM), which requires limited seizure patterns but offers moderate classification accuracy, and non-linear SVM (NLSVM), which requires sufficient seizure patterns and has high classification accuracy. In his design prototype, Yoo’s choice was to use two LSVMs, one trained for sensitivity and the other trained for specificity. Using this approach in a single system, he found accuracy rates of 95% for sensitivity and 98% for specificity detection performance.
Finally, there is the system-level consideration, which, in this case, consists of the seizure detection system based on a wirelessly powered electrocardiogram (ECG). Yoo and his students used a fully integrated EEG SoC consisting of a 1.8V analog front-end and 1.0V digital backend with 16 channels, SVM and simulation, scalable EEG processing, and machine learning for patient-specific seizure detection. All of their work was implemented using Cadence tools. “System-level consideration for circuit design is very important—you don’t want to burn all of the power from the analog front end and vice versa,” Yoo noted. The next step for Yoo’s work is to test the wearable EEG design on patients.

http://semiengineering.com/the-role-of-energy-efficient-circuits-in-wearable-healthcare-applications/

TSMC posts decreased July revenues

Taiwan Semiconductor Manufacturing Company (TSMC) has announced consolidated revenues of NT$76.39 (US$2.46 billion) for July 2016, down 6.1% sequentially and 5.6% on year.

TSMC's cumulative 2016 revenues through July totaled about NT$501.7 billion, decreasing 1.3% from a year earlier.

TSMC estimated previously that revenues would climb to between NT$254 billion and NT$257 billion in the third quarter of 2016 representing sequential growth of 14.5-16%.

TSMC's target of 5-10% growth in 2016 revenues remains unchanged, the foundry said at its July investors meeting.

http://www.digitimes.com/news/a20160810PM200.html

Micron aims to cram more storage into smartphones with stacked 3D NAND flash

Micron announced its first 3D NAND chip for mobile devices with the goal of cramming more storage into handsets, and maybe reducing reliance on SD card slots.
3D NAND provides more capacity in the same size chip than other technologies can. Micron’s 3D chip has 32GB of capacity and is targeted at mid-range and high-end phones. It’s based on the new UFS 2.1 standard, a fast storage protocol that isn’t in smartphones yet.
Micron believes internal storage capacity in smartphones will need to grow, especially with new applications like virtual reality and streaming video. Low-end smartphones have as little as 4GB of storage, while the iPhone 6S has up to 128GB.
Micron’s goal is to cram even more bits into its 3D NAND mobile chips as it refines the manufacturing process, said Gino Skulick, vice president at Micron’s mobile business unit.
In a few years, internal storage capacity in smartphones could be as great as it is in PCs today, perhaps up to 1TB by 2020, Micron said. But the company didn’t provide a roadmap of the storage capacity it will provide in its 3D mobile flash chips.
3D NAND chips are already in SSDs from companies like Intel, and the density they offer is only growing. Micron has said it can cram more than 3.5TB in gumstick-sized SSDs.
The flash chips pack storage cells vertically, like floors in a skyscraper, allowing for higher capacity and faster communications between chips. Current 2D NAND chips have storage cells placed next to each other.
Micron’s top competitor is Samsung, which makes 3D flash chips used in handsets like its Galaxy S7 and Note 7. Those handsets have a maximum of 64GB of capacity, after which users can expand storage capacity with SD cards. Samsung has also been using 3D flash to increasing the storage capacity of its enterprise SSDs. In March, it shipped an SSD with a whopping 15.36TB of storage.


http://www.pcworld.com/article/3105635/storage/micron-aims-to-cram-more-storage-in-smartphones-with-3d-nand-flash.html

Korean electronics firms rush to open plants in Vietnam

South Korean tech firms are rushing to expand their investment in Vietnam in line with their effort to build a new production base in the Southeast Asian country.

South Korea's Seoul Semiconductor Co. has won a license to invest $300 million to build a semiconductor production factory in northern Vietnam, industry sources said Tuesday.

Last week, the provincial government of Ha Nam gave the green light to Seoul Semiconductor to build the 750,000-square-meter production site that will house a semiconductor factory and a light-emitting diode assembly line, they said.

Under an investment plan released by the Seoul-based firm, a total of $300 million will be injected into building the site by 2019 to conduct research and produce semiconductors, the sources said.

"Vietnam was chosen because it will help us meet the global demand for LEDs and secure a competitiveness in production costs," a Seoul Semiconductor official said.

Industry sources forecast that the firm's move is to provide components to other South Korean tech firms, which already have made inroads into the Southeast Asian country.

According to the sources, South Korean tech giant Samsung Electronics Co. produces between 40 and 50 percent of its smartphones at its two factories located in Bac Ninh province, northern Vietnam, and Thai Nguyen.

In early 2015, Samsung Electronics began building a consumer electronics production complex in the southern city of Ho Chi Mihn.

The complex covers the production of TVs, air conditioners, washing machines, refrigerators and other home appliances. Assembly lines for TVs began operating earlier this year.

Samsung also plans to relocate parts of its TV production lines in Malaysia and kimchi refrigerator lines in South Korea's Gwangju to this complex.

Samsung Electronics will expand its investment size there from $1.4 billion to $2 billion by 2020, the sources said.

In return, the city government of Ho Chi Minh decided last week to give further tax benefits to the tech giant, they said.

Seoul-based LED manufacturer Lumens Co. will begin operating its factory in September located in the southern province of Binh Duong which is currently under construction.

Lumens currently produces LEDs that are used for various products ranging from televisions to smartphones. The company also works with Samsung Electronics.

LG Electronics Inc., South Korea's No. 2 tech giant, also opened a new production base in Haiphong, a northern port city of the country, that will manufacture the company's key products for exports.

The 800,000-square-meter Vietnam Haiphong Campus will allow LG to produce price-competitive electronics, including its globally sold smartphones and TVs, the company said. It plans to spend US$1.5 billion on the complex through 2028.

South Korean flat-screen maker LG Display Co. also plans to build a new production line in Haiphong with $1 billion and begin production next year.

Vietnam stands out as an attractive investment destination as a monthly minimum wage is about 59 percent that of China.

Also, the Southeast Asian country has a population of over 90 million and 60 percent are in their 30s or younger.

"Vietnam has a very good investment condition as labor cost is very cheap and 54 million is a labor population," said Park Byung-book, chief of the Korea Trade-Investment Promotion Agency's branch office in Hanoi. "Tech firms are increasingly making inroads into Vietnam, which used to be favored by textile or sewing firms in the past."

Also, KOTRA said the Vietnamese government is making various efforts to lure more South Korean investments. (Yonhap)

http://www.koreaherald.com/view.php?ud=20160809001008

AMD’s Roy Taylor on how VR will change entertainment

Roy Taylor works as corporate vice president of alliances at Advanced Micro Devices, a chip maker that creates processors and graphics chips for personal computers. But he cares just as much about software as he does about hardware.
That’s because new virtual reality applications will fuel demand for graphics hardware, such as AMD’s new family of Radeon graphics processing units (GPUs). So Taylor has been busy creating alliances with Hollywood entertainment companies, VR startups, and game companies. He sees VR as the dawn of a new entertainment medium.
AMD has opened an office in Hollywood, and it is supporting efforts to improve content creation for VR entertainment. And Taylor said he is looking forward to eye-popping imagery of 4K resolution per eye in future VR technologies — far better than the 1080p resolution per eye in today’s VR headsets.

Read the rest of the article at: 

http://venturebeat.com/2016/08/07/amds-roy-taylor-on-how-vr-will-change-entertainment/

When it Comes to Counterfeit Part Prevention, Semantics Matter

When it Comes to Counterfeit Part Prevention, Semantics Matter

08.02.2016 // Kevin Sink // TTI Insights

The legislation put forth in the 2012 and 2013 National Defense Authorization Acts (NDAA) has gradually solidified into law resulting in the Defense Federal Acquisition Regulation Supplement (DFARS) which is now considered the standard for counterfeit prevention. While these measures have reduced the risk of counterfeit parts flooding the supply chain, the specific terminology in the published DFAR regulations have caused confusion and misinterpretation because they lack some common industry terminology. To address this and other concerns, there is a current revision in process. However, the proposed revision introduces the term “authorized dealer” to refer to the parties with which OEMs must work to prevent counterfeiting. Unfortunately, “dealer” is not a common term within our industry and therefore makes the definition ambiguous. In the case of a regulatory documentation, semantics really do matter.
Another issue with the “authorized dealer” term is that nearly any player in the industry can produce documentation that deems them authorized in some manner and OEMs don’t have a standard to define what qualifies as authorized. With the confusion in the industry, leaders such as ECIA and TTI petitioned for terminology that would more clearly define the “authorized” provider by using phrases such as “authorized distributor” in lieu of “authorized dealer.” This term was introduced in the current open DFAR revision, DFARS 2014-D005 (252.246-7007). The amendment that will officially change this terminology is still under final review.
The other contentious terminology in the DFAR refers to software or firmware installed on components. The terms were likely included to discourage or uncover potential spyware or malware activity against US enterprises by foreign entities. However, given the breadth of functionality and specificity of the manufacturing specs regarding software and firmware, the regulations cause confusion rather than remedy the counterfeit component risk. The current proposal calls for the removal of “embedded software or firmware” from the definition of “electronic part.”
Quality control testing is another area where specificity and standardization are necessary. Because test labs can vary greatly in their services and fees, it was previously difficult for an OEM to understand the extent of testing they were getting. Fortunately, SAE AS6171 provides standardized techniques and requirements. This documentation has leveled the playing field for OEMs that can now use standard practices when sourcing their component testing.
As anti-counterfeit measures are becoming standardized, counterfeiters are also evolving. Some have moved beyond remarking, blacktopping and microblasting wherein used parts are altered and sold as new, to actually manufacturing their own counterfeit components. Due to the short lifecycle of many electronics these days and the expense of manufacturing even sub-quality components, counterfeiters are going after the more expensive parts and that is changing the ballgame for OEMs that frequently source from the open or independent markets.
Another risk in the supply chain is unexpected: the over-reporting of suspected counterfeit parts. While OEMs that place a priority on sourcing only authentic components should keep a keen eye on their supply chain, they should be careful when ascribing a part as counterfeit. There have been cases where an individual has jumped the gun and purported an electronics component counterfeit when it was really a misunderstanding of a simple thing such as a supplier’s lot coding. When a flag is raised about a part being counterfeit, it involves many people and reputations are on the line. By the time the issue is resolved, it is likely that damage is already done.
While standardization of terminology is key, the most critical anti-counterfeiting tool for OEMs continues to be traceability. Risk is substantially reduced when you can identify every step in an electronic component’s supply chain. Purchasing from the authorized supply chain - the manufacturer and their authorized distributors - makes this process significantly easier for OEMs. The supply chain within authorized distributors is short and includes oversight and contractual requirements to handle, store and ensure traceability of parts. So, even as definitions and regulations help to minimize risk, the industry must continue to think defensively and restrict purchases to the safest channels.

http://www.ttiinc.com/object/me-sink-20160802.html

Memory chip industry set for an uptick in 3Q16

The memory chip industry is set for an upturn in the third quarter of 2016 as inventories in the market have almost been cleared in the last two quarters and price hikes have become more apparent, according to industry sources.

Production cuts initiated by the top-three vendors Samsung Electronis, SK Hynix and Micron Technology in the first and second quarters have resulted in tight supply of commodity DRAM chips recently, boosting contract prices, said the sources.

Increasing built-in memory capacity at next-generation smartphones, including new iPhone devices, will further ramp up demand for memory products in the third quarter, said sources, noting that the mobile DRAM capacity of the next-generation iPhone is expected to increase to 3GB from 2GB previously, while those for new Android phones are expected to raise to 6GB from 4GB.

Increasing demand for DRAM products, mostly high margin models, from the server sector is also contributing to the recent uptick of the DRAM industry, added the sources.

Meanwhile, the strong demand for standard DRAM has also weighed on the supply of niche-type memory products, which are widely being used in the LCD TV, STB and GPS sectors, said the sources, adding that significant sales of niche DRAM chips will continue to September.

Prices of NAND flash chips are also bound to increase, buoyed by rising demand, particularly from the next-generation iPhone devices.

Macronix International, a manufacturer of flash and mask ROM memory chips, expects sales of its NAND flash products to reach NT$300 million (US$9.475 million) a month in the third quarter and to NT$400 million in the fourth, according to company president Chih-yuan Lu.

Macronix, which currently accounts for 22% of the global NOR flash market, will aim to further ramp up its share by pushing sales of its NOR flash products in the automobile sector, Lu added.

Backend-service company Powertech Technology (PTI) also expects to see another double-digit growth for its revenues in the third quarter after seeing its sales expand 11.5% on year to NT$21.94 billion in the first half of 2016.

The supply of commodity DRAM has fallen short of demand, while memory density in high-end smartphones and servers continues to rise, commented PTI president Hung Jia-yu recently.

Rising flash memory content per smartphone and greater SSD penetration will encourage PTI to improve technology and boost production capacity for the memory, Hung said.

http://www.digitimes.com/news/a20160731PD200.html

Infineon sees 8% CAGR long-term growth

Infineon had calendar Q2 revenue of €1,632 million for a profit of €254 million.

For calendar Q3 it expects q-o-q revenue increase of 3% (+/ – 2%).
“Revenue, earnings and margin all increased in line with expectations in the third quarter,” says CEO Reinhard Ploss, “demand and was particularly strong for our automotive electronics, renewables and power supply solutions. Despite the current contraction of the semiconductor market and contrary to many of our competitors, Infineon has grown once again compared to the prior year’s corresponding quarter, reflecting its focus on sub-markets with structural growth. We therefore continue to forecast a long-term compound annual growth rate of 8 percent. We are enabling cleaner and safer cars, greener energy and even faster mobile communication.”
“The planned acquisition of Wolfspeed will secure us a decisive technological advantage in the long term and help us grow our system understanding. We are thereby focusing on promising growth areas such as electromobility and the Internet of Things.”
The gross cash position increased to €2,083 million at the end of calendar Q2, compared to €1,803 million at March 31, 2016. The net cash position rose from €27 million to €299 million.

http://www.electronicsweekly.com/news/business/infineon-sees-8-cagr-long-term-growth-2016-08/

Avnet Ups its Bid to Acquire Premier Farnell

Electronics distributor Avnet, Inc. has increased its offer to acquire fellow distributor Premier Farnell, in an attempt to squash a competing bid placed by Datwayler Technical Components UK Limited.
The new bid calls for an all-cash offer of about $765.7 million (about $2.05 per share) for Premier Farnell, a 12.1% premium over the offer from Datwayler (at about $1.83 per share). Based in the U.K., Premier Farnell is a distributor of electronic components and related products to system design companies.
Avnet believes the acquisition will strengthen its digital footprint worldwide, and the combination of Premier Farnell’s online services with Avnet’s supply chain will offer a new level of service for its customers. Avnet says while Premier Farnell’s customers will gain access to larger supply chain services, Avnet customers will be able to do more technical research online through Premier Farnell’s capabilities.
“In addition to deepening our customer base, this acquisition will enhance our go-to-market strategy as we target the industrial Internet of Things with edge-to-enterprise products and solutions,” says Gerry Fay, president of Avnet’s Electronics Marketing, Worldwide.
The acquisition still has many hurdles to go through including approval by Premier Farnell shareholders, the European Union, U.S. and Israel.

http://electronics360.globalspec.com/article/7070/avnet-ups-its-bid-to-acquire-premier-farnell