10. Adapteva: 50 gigaflops / watt
• 64 cores
• 100 gigaflops @ 2 watts of power
• 2D mesh of simple cores with high speed
interconnect
• OpenCL compatible
11. Light in a spin.(A) (Top) Calculated far-field intensity distribution of an optical vortex with
topological charge one, spiral pattern created by the interference of the vortex beam and a
copropagating Gaussian beam, and interference pattern with a dislocated fringe created by
the interference of the vortex beam and a Gaussian beam tilted with respect to the vortex
beam (1).
N M Litchinitser Science 2012;337:1054-1055
Published by AAAS
First observation of particles that are their own antiparticles could be on its wayThe matter that makes up the universe consists of particles such as electrons and protons, as well as their counterparts known as antiparticles. Particles and antiparticles that collide, however, annihilate each other in an intense flash of energy. Nevertheless, the Italian physicist EttoreMajorana proposed that some particles could exist that are their own antiparticles although physicists are yet to observe such particles.phys.org/news/2011-01-particles-antiparticles.html#nRlvArxiv - Chain of Majorana States from Superconducting Dirac Fermions at a Magnetic Domain Wall (5 pages)We study theoretically a strongly type-II s-wave superconducting state of two-dimensional Dirac fermions in proximity to a ferromagnet having in-plane magnetization. It is shown that a magnetic domain wall can host a chain of equally spaced vortices in the superconducting order parameter, each of which binds a Majorana-fermion state. The overlap integral of neighboring Majorana states is sensitive to the position of the chemical potential of the Dirac fermions. Thermal transport and scanning tunneling microscopy experiments to probe the Majorana fermions are
NY Times - A new wave of robots, far more adept than those now commonly used by automakers and other heavy manufacturers, are replacing workers around the world in both manufacturing and distribution. Factories like the one here in the Netherlands are a striking counterpoint to those used by Apple and other consumer electronics giants, which employ hundreds of thousands of low-skilled workers. The dutch factory here has several dozen workers per shift, about a tenth as many as the plant in the Chinese city of Zhuhai.At an automation trade show last year in Chicago, Ron Potter, the director of robotics technology at an Atlanta consulting firm called Factory Automation Systems, offered attendees a spreadsheet to calculate how quickly robots would pay for themselves.In one example, a robotic manufacturing system initially cost $250,000 and replaced two machine operators, each earning $50,000 a year. Over the 15-year life of the system, the machines yielded $3.5 million in labor and productivity savings.
Robot arms liThe falling costs and growing sophistication of robots have touched off a renewed debate among economists and technologists over how quickly jobs will be lost. This year, Erik Brynjolfsson and Andrew McAfee, economists at the Massachusetts Institute of Technology, made the case for a rapid transformation. “The pace and scale of this encroachment into human skills is relatively recent and has profound economic implications,” they wrote in their book, “Race Against the Machine.”ke those at a Philips Electronics factory in the Netherlands can perform the same tasks as hundreds of low-skill workers.
Atomic Clusters of gold nanoparticles could lead to 3D real time telepresenceCentral Florida Professor Jayan Thomas is working with gold nanoparticles and studying their properties when they are shrunk into a small size regime called nanoclusters. Nanoclusters are on the small end and nanocrystals are on the larger end of the nanoregime. Nano clusters are so small that the laws of physics that govern the world people touch and smell aren’t often observed.“Nanoclusters occupy the intriguing quantum size regime between atoms and nanocrystals, and the synthesis of ultra-small, atomically precise metal nanoclusters is a challenging task,” Thomas said. Thomas and his team found that nanoclusters developed by adding atoms in a sequential manner could provide interesting optical properties. It turns out that the gold nanoclusters exhibit qualities that may make them suitable for creating surfaces that would diffuse laser beams of high energy. They appear to be much more effective than its big sister, gold nanocrystal which is the (nano)material used by artists to make medieval church window paintings.http://phys.org/news/2011-01-particles-antiparticles.html#nRlvBecause nanoclusters appear to have a better ability to diffuse high beams of energy, they are a promising area for future development. There is still plenty of applications to be explored using these very interesting atomically engineered materials. Until now, much research has been focused on the larger nanocrystal.Thomas is also exploring the use of these particles in the polymer material used for 3D telepresence to make it more sensitive to light. If successful, it can take the current polymers a step closer to developing real time 3D telepresence.
Schematic illustration of single-atom-thick films with patterned regions of conducting graphene (gray) and insulating boron nitride (purple-blue).Integrated circuits, which are in everything from coffeemakers to computers and are patterned from perfectly crystalline silicon, are quite thin -- but Cornell researchers think they can push thin-film boundaries to the single-atom level.Their materials of choice are graphene, single atom-thick sheets of repeating carbon atoms, and hexagonal boron nitride, similarly thin sheets of repeating boron and nitrogen atoms. Researchers led by Jiwoong Park, assistant professor of chemistry and chemical biology, have invented a way to pattern single atom films of graphene and boron nitride, an insulator, without the use of a silicon substrate. The work is detailed in an article in the journal Nature, published online Aug. 30.The technique, which they call patterned regrowth, could lead to substrate-free, atomically thin circuits -- so thin, they could float on water or through air, but with tensile strength and top-notch electrical performance."We know how to grow graphene in single atom-thick films, and we know how to grow boron nitride," Park said. "But can we bring them together side and side? And when you bring them together, what happens at their
Harvard - Applied physicists at the Harvard School of Engineering and Applied Sciences (SEAS) have created an ultrathin, flat lens that focuses light without imparting the distortions of conventional lenses. Lens are a major part of cameras that can make them heavier and bulkier. Flat lens could remove a roadblock to making cameras smaller, lighter and cheaper. Smaller and cheaper cameras would also enable the affordability multi-gigabit cameras that use a few hundred or a few thousand cameras to combine images.At a mere 60 nanometers thick, the flat lens is essentially two-dimensional, yet its focusing power approaches the ultimate physical limit set by the laws of diffraction.Operating at telecom wavelengths (i.e., the range commonly used in fiber-optic communications), the new device is completely scalable, from near-infrared to terahertz wavelengths, and simple to manufacture.http://pubs.acs.org/doi/abs/10.1021/nl302516v
Leap Motion is developing a cheap ($70), more precise (down to 0.01 mm), and much smaller (think “pack of gum” proportions) motion controller. Currently taking pre-orders, the Leap Motion is scheduled to ship between December and February, and with it will come a new market of third party apps designed to take full advantage of the device. The key to the Leap Motion system is better algorithms. This means it could be adapted to use other kinds of sensing than infrared such as radar or light.LIDAR uses ultraviolet, visible, or near infrared light to image objects and can be used with a wide range of targets, including non-metallic objects, rocks, rain, chemical compounds, aerosols, clouds and even single molecules.Vastly improved 3D sensors at lower cost would accelerate the development, capabilities and adoption of robotics and robotic cars. This trend would also be enhanced with the improvements in voice recognition from systems like Apples SIRI. Leap motion has discussed making their system as small as a coin. This means the system should get cheaper and added to all smartphones and tablets and incorporated into other gadgets.
HPCWire - Chipmaker Adapteva is sampling its 4th-generation multicore processor, known as Epiphany-IV. The 64-core chip delivers a peak performance of 100 gigaflops and draws just two watts of power, yielding a stunning 50 gigaflops/watt. The engineering samples were manufactured by GLOBALFOUNDRIES on its latest 28nm process technology.Epiphany is essentially a stripped down general-purpose RISC CPU that throws out almost everything but the number-crunching silicon. But since it doesn't incorporate features needed by operating systems, like memory management, it relies on a host processor to feed it application kernels in the same manner as a GPGPU. The current implementation supports single precision floating point only, but plans are already in the works for a double precision implementation.The general layout of Epiphany is a 2D mesh of simple cores, which talk to each other via a high-speed interconnect. In that sense, it looks more like Intel's manycore Xeon Phi than a graphics processor, but without the x86 ISA baggage (but also without the benefit of the x86 ecosystem).NVIDIA's new K10 GPU computing card can hit about 20 single precision gigaflops/watt, but that also includes 8GB of GDDR5 memory and a few other on-board components, so it's not an apples-to-apples comparison.Developers can now use standard OpenCL source to program the Epiphany processorhttp://www.hpcwire.com/hpcwire/2012-08-22/adapteva_unveils_64-core_chip.html?featured=top
When it comes to storing information, hard drives don't hold a candle to DNA. Our genetic code packs billions of gigabytes into a single gram. A mere milligram of the molecule could encode the complete text of every book in the Library of Congress and have plenty of room to spare. All of this has been mostly theoretical—until now. In a new study, researchers stored an entire genetics textbook in less than a picogram of DNA—one trillionth of a gram—an advance that could revolutionize our ability to save data.A few teams have tried to write data into the genomes of living cells. But the approach has a couple of disadvantages. First, cells die—not a good way to lose your term paper. They also replicate, introducing new mutations over time that can change the data. To get around these problems, a team led by George Church, a synthetic biologist at Harvard Medical School in Boston, created a DNA information-archiving system that uses no cells at all. Instead, an inkjet printer embeds short fragments of chemically synthesized DNA onto the surface of a tiny glass chip. To encode a digital file, researchers divide it into tiny blocks of data and convert these data not into the 1s and 0s of typical digital storage media, but rather into DNA’s four-letter alphabet of As, Cs, Gs, and Ts. Each DNA fragment also contains a digital "barcode" that records its location in the original file. Reading the data requires a DNA sequencer and a computer to reassemble all of the fragments in order and convert them back into digital format. The computer also corrects for errors; each block of data is replicated thousands of times so that any chance glitch can be identified and fixed by comparing it to the other copies.
A few days ago, Stanford marine biologists were excited to detect a white shark swimming along the California coast north of San Francisco. Although the biologists routinely monitor sharks, this particular moment marked the first step toward a "wired ocean" full of mobile robotic receivers and moored listening stations that can detect ocean wildlife as it swims by.Although similar technologies have been used to monitor the ocean itself, specifically to investigate climate change, this is the first such experiment dedicated to wildlife.In addition to providing researchers with near real-time data of sharks and other animals, the project supports a new iPhone and iPad app designed to give the public a more visceral connection to the ocean and the creatures within.The Wave Glider robot – named Carey in honor of noted large pelagic fish biologist Frank Carey – is probing the Pacific Ocean off the California coast in an initiative led by Stanford marine sciences Professor Barbara Block and her research team to keep tabs on the comings and goings of top marine predators, and to provide better census data of all species in the area.The Blue Serengeti Initiative, as the effort is called, picks up where the decade-long Tagging of Pacific Predators (TOPP) project left off. TOPP, an international collaboration among 75 scientists, involved tagging thousands of marine animals –great white sharks, elephant seals and leatherback turtles – and tracking their movements via satellite.
A 3-D model of Tom Johnson, one of the seven white sharks featured in the Shark Net app.There are now 120 white sharks carrying acoustic tags, along with 27 salmon sharks and five mako sharks. Other groups have tagged more animals – including salmon, sturgeon and lingcod – along the California coast with similar acoustic devices, which can also be picked up by the network of listening buoys.Those buoys are moored in three foraging hot spots – Año Nuevo, the Farallon Islands and Tomales Point. These hot spots were known from ongoing photo ID studies by the team as well as the yearlong records from satellite tags that demonstrated the sharks' return with high fidelity. From now until late November, these three areas are plentiful with sea lions, seals and other prey that attract large numbers of white sharks. "These hot spots are like the Outback Steak Houses for white sharks," Block said.The buoys can detect a shark's transmitter – or any animal with an acoustic tag – from as far away as 2,000 feet. The data is delivered in near real time to mobile devices; the whole process takes about four minutes.