from EETimes: http://ift.tt/1iLENVh
via Yuichun
Tapping and swiping on smartwatch apps is fiddly – and your fingers get in the way of what’s on the screen, too. To the rescue comes Gierad Laput of Carnegie Mellon University in Pittsburgh, Pennsylvania, with a tiltable watch face that could save you a whole lot of screen jabbing.
At the computer-human interaction conference in Toronto, Canada, this week, Laput revealed how he and his Carnegie Mellon colleagues attached a pair of gamepad motion sensors to a smartwatch-style LCD display so that it can detect when a user pans, twists or tilts the screen. The sensed motions are used to navigate around content such as a map, or play games, saving the user the need to poke at the screen to accomplish such tasks.
Pushing down on the screen activates all the sensors at once and is recognised as an option-choosing “click” of the cursor. The motion-detection worked fine when they tried using it to play a version of Doom, a first-person shooter video game. “Smartwatches – and wearables in general – are really exciting, but interacting with them is terrible. It’s a hard problem, and we have to keep chipping away at it,” Laput says.
“Since our fingers are large, and people want smartwatches to be small, we have to go beyond traditional input techniques. Digitising the mechanical movements of a watch face in this way offers interaction without occluding the screen. It is easy to implement and we’d love to see the research go into future products.”
David Harold, a director of Imagination Technologies, based in Hertfordshire, UK, which develops graphics systems for mobile devices, says the motion-sensitive screen could complement existing watch and phone-control methods such as voice-recognition and icon pressing. “It’s an interesting idea but my absolute gut feeling is that this would best be used in gaming, where you definitely don’t want your fingers covering the screen. There’s a trend towards improving wearable interfaces for different applications and this is very much part of that.”
The idea comes as fresh rumours surfaced about a possible Apple smartwatch. On 19 April, Nike decided to focus on fitness monitoring software and abandon manufacture of the iOS-based Fuel Band hardware that currently runs it.
Apple CEO Tim Cook is on Nike’s board, and so there has been some speculation that Nike – a long-term Apple partner since the early iPod days – may be moving aside to give Apple a clearer run in the wearables market with a watch that will runs the Nike Fuel Band software. Speaking on CNBC this week, Nike CEO Mark Parker guardedly said that he is “excited about what’s to come” from a continuing Apple and Nike partnership. There have been rumours for some time that Apple is planning to launch an iWatch this year.
Syndicated content: Paul Marks, New Scientist
Image: HCII/Carnegie Mellon University
Intel’s domination of the microprocessor market is bring ‘chipped away’ by ARM, says US analyst firm IC Insights.
“Intel continues to dominate the microprocessor business, accounting for nearly two-thirds of the market’s total sales in 2013, but the company’s huge MPU marketshare is being chipped away by strong growth in ARM-built processors for smartphones, tablets, and new high-density microservers, which can lower cost in data centres for high-volume Internet traffic and cloud-computing services,” says IC Insights.
With shipments of standard PCs waning, Intel’s total MPU sales fell by about 2% to $36.3 billion in 2013, following a 1% decline in 2012, primarily due to slowing demand for x86 central processing units (CPUs) in personal computers.
Standard notebook PCs continue to face stiff competition from tablets sold by Apple, Samsung, Amazon, and more than a dozen other suppliers worldwide. New smartphones, which are often built with similar ARM processors as tablets, also continue to infiltrate more PC applications.
AMD’s x86 microprocessor sales have been hit harder by the slowdown in notebook and desktop PCs, resulting in its total MPU revenues plunging 21% in both 2012 and 2013. AMD’s 2013 microprocessor sales (excluding stand-alone graphics processors) fell to $2.8 billion, which represented 4.8% of total MPU sales last year compared to its previous marketshares of 6.4% in 2012 and 8.2% in 2011. AMD remained in fourth place among MPU suppliers in 2013.
The two x86 rivals are now scrambling with different strategies to reverse their slumping fortunes in microprocessors. AMD has joined the ARM camp, preparing 2014 introductions of new 28nm Opteron A1100 server processors (with up to eight ARM-based 64-bit CPU cores) while the company also aims graphics-enhanced x86 designs at tablets and notebook PCs. AMD says it will be the only microprocessor maker to offer both ARM and x86 solutions.
Meanwhile, Intel is accelerating its push to diversify its x86 MPUs and make new processors more competitive with ARM-based solutions in a wide range of platforms—from smartphones and tablets to convertible notebooks that can function like tablets, high-density microservers, new wearable systems, and embedded “Internet of Things” (IoT) applications. Intel continues to emphasize its manufacturing leadership, preparing to introduce the first 14nm-fabricated processors (code-named “Broadwell”) in 2H14. With its newest 22nm Atom SoC processors and existing 22nm Core x86 flagship microprocessors, Intel aims to ship 40 million tablet MPUs in 2014, up fourfold from 2013.
The $58.6 billion microprocessor market was the largest single semiconductor product category in 2013, accounting for 22% of total IC sales.
In 2013, tablet microprocessors represented nearly 6% of worldwide MPU sales compared to 4% in the previous year, while cellphone application processors accounted for 25% of the revenue total, up from 22% in 2012. MPUs used in PCs, server computers, and embedded-processing applications slid to 69% of total microprocessor sales in 2013 compared to 74% in 2012.
Researchers at the University of Manchester found a mechanism that could explain unusual electronic and optical behaviour in graphene, and another route to give it a bandgap.
The work involved spreading mono-layer flakes of graphene on a thin (20nm) layer of boron nitride (BN).
Both materials have a hexagonal two-dimensional structure – there is increasing interest in the interaction of various 2-d materials, phosphorene and molybdenum di-sulphide are others. LINK
Some language is required here: when two layers of graphene are aligned with one another exactly, atoms from the layers attract each other and lock the alignment, and they are said to be ‘commensurate’.
Boron nitride’s lattice constant is 1.8% larger than graphene.
“There has been work done to see if graphene and boron nitride would become commensurate. Everyone had assumed they would stay incommensurate – there would be no way to click together,” project scientist Colin Woods told Electronics Weekly.
Monolayer graphene flakes were known to stick to boron nitride, by the same van der Waals’ forces that hold two aligned graphene layers together, but the lattice mis-match was thought to be enough that they would always remain incommensurate.
Woods’ examined graphene flakes flat on the boron nitride in various rotational orientations, using atomic force microscopy, scanning tunnelling microscopy and Raman spectroscopy.
He discovered that when the lattices were clearly misaligned, they were incommensurate.
However when the axes of the hexagons were aligned exactly, the graphene locally stretched to match the boron nitride and became commensurate, and also developed a bandgap.
“The badgap is small. We don’t know if it exists in un-aligned state, where there is a very small bandgap or no bandgap,” said Woods, “but the commensurate/incommensurate transition definitely opens a bandgap. This is a new mechanism for creating a bandgap in graphene.”
Strain build-up cannot continue forever across the graphene, and periodically carbon atoms within the graphene bunch together in a strain-relieving dislocation. Viewed from above, the dislocations form a sharply-defined hexagonal grid (see image), where the hexagons within are perfectly-aligned graphene/boron nitride.
Maximum size for the hexagonal super-lattice is 14nm, compared with the 0.25nm graphene lattice constant.
Sadly, “the area of bandgap not enough for electronic logic”, he said, but this is new fundamental understanding in the behaviour of graphene, which could explain certain electronic and optical behaviours.
Where flakes are not perfectly aligned, the super-lattice quickly de-focusses to the incommensurate state with increasing rotation.
“It was extremely exciting to see that the properties of graphene can change so dramatically by simply twisting the two crystals only a fraction of a degree,” said Woods.
Having observed the effect, the team is looking to get a theoretical handle on the two-dimensional forces and stretches.
“Generally, the previous model used to describe the sort of interaction which has been observed in our experiments describes only the one-dimensional case, but even there it produces very non-trivial solutions. We hope that our system will push the mathematical development of the model to two-dimensions, where even more exciting mathematics is to be expected,” said Woods.
The work is described in a Nature Physics paper: ‘Commensurate–incommensurate transition in graphene on hexagonal boron nitride‘, and was led by graphene co-discoverer Sir Kostya Novoselov.
ST Microelectronics will start hiring the sacked Micron researchers it agreed to take on last month in Q1 2015, says ST CFO Carlo Ferro.
“We will start to hire these people in the course of the first quarter of 2015,” says Ferro.
“We entered into an agreement with Micron and we took commitment to hire about 140 people, possibly 170, depending on the available profiles . . . . . . .under certain terms, agreed with Micron with . . . . some relief of the cost in the initial period, making at the end what I believe is a win-to-win solution,” says Ferro.
This seems to imply that Micron will be contributing to the cost of ST re-hiring the Micron researchers for the initial period of their re-employment with ST.
The researchers were all employed by ST before being transferred to the Numonyx jv between ST and Intel which was later sold to Micron.
Ferro added that the 140/170 researchers would mostly be re-deployed in ST’s Automotive, Industrial and Power business segment.
Research at the University of Manchester has shown how by stacking two dimensional materials – graphene and boron nitride for example – they can design new properties – a bandgap for graphene, for example.
By controlling the relative orientation of graphene and underlying boron nitride, the team can created local strains.
“It was extremely exciting to see that the properties of graphene can change so dramatically by simply twisting the two crystals only a fraction of a degree,” said PhD student Colin Woods. “Generally, the previous model used to describe the sort of interaction which has been observed in our experiments describes only the 1-dimensional case, but even there it produces very non-trivial solutions. We hope that our system will push the mathematical development of the model to two-dimensions, where even more exciting mathematics is to be expected.”
This demonstration of layer interaction in 2d heterostructures is described in a Nature Physics paper: ‘Commensurate–incommensurate transition in graphene on hexagonal boron nitride’, and was led by graphene co-discoverer Sir Kostya Novoselov.
“Research on heterostructures is gaining momentum, and such possibilities for controlling the properties of heterostructures might become very useful for future applications,” said Novoselov.
The next step is to combine several of these crystals in a 3D stack. Already, tunnelling transistors, resonant tunnelling diodes, and solar cells have been made.
FD-SOI, currently looking to expand its ecosystem, will get a boost from the remarks of KLA Tencor CEO, Rick Wallace, about the struggles his customers are having with finfets.
“In logic and foundry, with the introduction of the new 3-D gate architectures, the yield issues our customers are grappling with today are proving to be the most challenging that the industry have ever faced, and even the smallest variation and process margin can cause significant yield losses for these devices,’ says Wallace.
FD-SOI uses few masks and has less processing steps than finfet.
“At 28nm and 20nm, the lower power consumption and higher performance of FD-SOI compared to planar bulk CMOS gives major competitive advantages to FD-SOI in high volume portable applications,” says Handel Jones, founder, chairman and CEO of International Business Strategies Inc, “the lower cost of FD-SOI die compared to 16nm finfet die provides an overwhelming advantage to utilising FD-SOI for high volume applications at this technology node.”
Some companies could be looking again at what FD-SOI has to offer because even Intel seems to be struggling with 14nm finfet, after a reportedly smooth transition to finfet at 22nm.
Yesterday, the developer of FD-SOI, STMicroelectronics said it has recently signed up a foundry to take FD-SOI. ““We have just signed a strategic agreement with a top-tier foundry for 28nm FD-SOI technology,” said Jean-Marc Chery, Executive Vice President and General Manager, Embedded Processing Solutions at ST, “this agreement expands the ecosystem, assures the industry of high-volume production of ST’s FD-SOI based IC solutions for faster, cooler, and simpler devices and strengthens the business and financial prospects of the Embedded Processing Solutions Segment.”
The uncertainty with finfets has caused manufacturers to stall investment plans, says KLA’s Wallace.
“Issues related to leading edge device yield and high concentration of demand across a consolidated customer base and uncertainty over the timing of follow-on capacity have introduced a degree of variability into our quarterly demand forecast and have made visibility into our customer production plans extremely challenging today,” says Wallace.
At the leading edge, FD-SOI has a big cost advantage over finfet and equivalent leakage, according to Jones.
“At 14nm/16nm, the FD-SOI die cost for a 100mm2 die is 28.2% lower than the bulk FinFET die cost and has higher yield,” says Jones, “the leakage of FD-SOI devices is projected to be comparable to that of finfet devices.”
Digitimes expects 422 million smartphones to be sold in a China in 2014, with 278 million coming from China-based smartphone vendors and 144 million coming from Samsung and Apple.
Digitimes expects global shipments of China-based smartphone vendors to reach 412 million units in 2014 with overseas shipments accounting for about 126 million units.
Lenovo and Huawei are each expected to ship 50 million units.
ZTE and CoolPad are each expected to reach 35.5 million units.
TCL is expected to ship 26 million units.
Gionee and Xiaomi are each expected to ship 20 million units.
Wolfson is to be sold to Cirrus Logic. Cirrus will pay 235p per share compared to the market closing price last night of 134p. It values Wolfson at $488 million.
Wolfson made a $12.6 million loss last year, sacked 10% of its staff and borrowed £25 million from the bank
JACKSON, MS—A Mississippi native who has an LED lighting factory in China wants an unusual incentive in exchange for setting up an additional assembly plant in his home state employing 200 workers to start. Lighting OEM Inc. owner and founder James Scott said he wants an edge in bidding for the state’s LED lighting business.
Toshiba is sampling 32GB and 64GB NAND modules to the universal flash storage 2.0 (UFS) standard.
Data transfer is at 11.6Gbps by integrating two 5.8Gbps MIPI M-PHY HS-G3 I/F data lanes – an optional feature of UFS Ver. 2.0.
The modules have read speeds of 650MB/s and write speeds of 180MB/s.
The modules are sealed in a 153 ball FBGA package measuring 11.5mm x 13.0mm x 1.0mm for the 32GB die and 11.5mm x 13.0mm x 1.2mm for the 64GB die and have a signal layout compliant with JEDEC UFS Ver.2.0.
The modules are rated for an operating temperature of -25°C to +85°C and support memory core voltages of 2.7V to 3.6V.
Demand continues to grow for large density, high-performance chips that support high resolution video, driven by improved data-processing speeds in host chipsets and wider bandwidths for wireless connectivity in a wide range of digital consumer products, including smartphones and tablet PCs.
The JEDEC UFS Ver.2.0 compliant interface handles essential functions, including writing block management, error correction and driver software. It simplifies system development, allowing manufacturers to minimize development costs and speed up time to market for new and upgraded products.
Toshiba reckons its first to sample modules in this format and will move to mass production as demand dictates.
Although SEMI forecasts 23% growth in the 2014 semiconductor equipment market, the three biggest spenders aren’t increasing capex significantly, reports Bill Jewell,’s Semiconductor Intelligence.
“Intel’s guidance ranges from a 2% decline to 7% growth, averaging 3% growth. Samsung expects 2014 semiconductor capital spending to be similar to 2013. TSMC’s guidance ranges from a 2% decline to 3% growth, averaging 1% growth. The average expected growth spending by the three companies is 1%.”
The three companies spend 50% of the industry’s capex.
IC Insights’ March forecast was 8.4% growth in semiconductor industry capital spending.
Gartner’s April forecast was 5.5% growth. Gartner forecast semiconductor capital equipment spending would grow 12.2% in 2014, about twice the rate of overall capital spending but about half the growth expected by SEMI.
Jewell expects capex growth of 10% and equipment spending growth of 20%.
The explanation for the mismatch with companies’ forecasts is, said Jewell, that the companies will increase capex as the year unfolds.
There is little chance of an over-spend if you take the ratio of capex to semiconductor market.
“The ratio was 20% in the overcapacity years of 2001 to 2002,” says Jewell, “the ratio dropped below 10% in 2005. Since 2007 the ratio has been in a steady range of 11% to 13%. The ratio in 2014 and 2015 is based on the SEMI equipment forecast (23% growth in 2014 and 2% in 2015) and the WSTS semiconductor market forecast (4.1% in 2014 and 3.4% in 2015). Under these assumptions, the industry will not experience overcapacity in the next few years.”
Assuming 10% growth for semiconductor sales in both 2014 and 2015 and no change in equipment spending in 2014 and 2015, the four-year-average ratio would drop to 10% in 2015, implying under-capacity.
“However we do not believe this will occur,” says Jewell, “we expect companies to increase their equipment purchases in response to solid market growth.”
Only 2 days left to get the new Warranty VOID if NOT removed T-Shirt!
http://ift.tt/PG4FWm
All videos in this series are HERE
Forum HERE
Tiny robots could help you heal. Acting as mini technicians, they could one day assemble medical devices inside the body.
A veritable construction crew of micro-scale robots already exists, from worm-like bots that can move heavy loads to muscle-powered machines that can walk across a lab bench. But until now, finer control over miniature objects has proved elusive.
Eric Diller and Metin Sitti of Carnegie Mellon University in Pittsburgh have created a simple version of micro-robots using rods made of magnetic materials. Each robot is about 1 millimetre long and has two gripping arms. A magnetic field is used to move the robots and operate the grippers.
Previous gripping bots had to be tethered to an outside controller, making them unsuitable for use inside the human body. Other versions could not move and grip things at the same time. “We can move them while they are closed or open, it doesn’t matter,” says Sitti.
So far, the robots have transported small objects and built bridges out of Y-shaped rods. Sitti hopes future versions could be injected into the body along with parts for micro-machines that would swim in the blood and help wounds clot. The builder bot could then create the more advanced device while inside the bloodstream.
“We need to make things smaller to get inside the body easier, but if they are too small, they are not really useful,” he says. “You want to assemble the robot inside the body.”
Journal reference: Advanced Functional Materials, DOI: 10.1002/adfm.201400275
Syndicated content: Jacob Aron, New Scientist
Net neutrality, the principle that the internet should be open to all on equal terms, is under review in they USA as the FCC Commissioners discuss proposed new rules following its defeat in Verizon v FCC.
The White House remains firm in its support of net neutrality. “Absent net neutrality, the Internet could turn into a high-priced private toll road that would be inaccessible to the next generation of visionaries. The resulting decline in the development of advanced online apps and services would dampen demand for broadband and ultimately discourage investment in broadband infrastructure. An open Internet removes barriers to investment worldwide.”
That was the President’s position in January. The following month, the decision in Verizon v FCC stated that the FCC did not have the power to insist on neutrality.
The carriers want to be able to charge extra for super-services and for certain types of traffic making the Internet a tiered service where those who pay more get better speeds.
In Europe, the European Parliament voted earlier this month to stop carriers charging more for certain types of traffic. To take effect, the vote has to be endorsed by the Council of a Ministers.
Neelie Kroes, who seems to decide these matters, remains wedded to the principle that “Internet users can always choose full Internet access – that is, access to a robust, best-efforts Internet with all the applications you wish.”
But Kroes is prepared to accept the provision of tiered services where it is clearly stated exactly what the user is getting and not getting.
The carriers make their usual argument that they need to be able to charge more to maintain their investment levels in the Internet.
In the USA, these investments are scandalously low judging by the standard of the mobile services provided.
The class action lawsuit against Intel, Apple, Google and Adobe alleging no-hiring agreements for each others’ employees has been settled.
The terms of the settlement will not be revealed until the end of May but one of the plaintiff’s lawyers called the settlement terms “excellent” and the rumour is that the companies agreed to pay a collective $324 million to settle.
64,000 employees were affected and the damages award could have been $9 billion.