Over the past few years, Samsung has risen to dominate much of the consumer NAND business thanks to a stream of well-reviewed SSDs that combined excellent performance and low prices. The company’s first SSD with three-bit (Triple-level Cell, or TLC) NAND, the Samsung 840, offered modest performance but was quite affordable, and the Samsung 840 Evo upped the ante last year by combining slower TLC with faster SLC NAND. Now, Samsung is moving to combine its TLC NAND production and its 3D vertical NAND in a move that could hit a real consumer sweet spot.

One note is that Samsung is attempting to have its cake and eat it too as far as the NAND is concerned, referring to this as “3-bit multi-level cell (MLC).” That’s not how the term is typically used, and until Samsung demonstrates that it’s built TLC NAND with MLC characteristics in both performance and longevity, it’s a misapplication of terminology for the sake of marketing.

Underneath that marketing, however, there is reason to think that TLC built vertically could be superior to its traditional planar counterpart. Samsung’s 850 Pro, the first SSD to use V-NAND, has substantially better reliability and performance than traditional planar counterparts. If Samsung can keep those characteristics and extend them into a TLC drive, it could create the most attractive consumer drive on the market.

Samsung’s PR also inadvertently highlights the misleading ways that process nodes are used to market to consumers. While it claims that its 3D NAND is more than twice as productive per wafer as its 10nm-class NAND, an editor’s note remarks that “10 nanometer-class means a process technology node somewhere between 10 and 20 nanometers.” Independent analysis from Anandtech has shown that Samsung’s V-NAND has a very small die advantage over the leading 16nm NAND from Micron.

If Samsung’s V-NAND has truly doubled wafer productivity over its old “10nm-class” NAND, either its feature sizes were nowhere near as good as what Micron is building or its wafer yields were absolutely terrible. Or, as is most likely, it’s using alternate meanings of wafer productivity to make the comparison look better.

The reason I’m less concerned about the underlying technology, even if I’m snarky on the marketing, is because Samsung’s 3D NAND has already proven itself as a potent force and the company’s TLC drives have done equally well — even if the 840 Evo family has aperformance issue with older data (there’s a fix for that coming in less than a week, if you own an affected drive). In short, there’s good reason to think the company can take the two products and combine them into something even better — possibly pushing SSDs below the 50 cents per GB line that they’ve been bumping against for a while.

A new study looked at the North American desert-dwelling sidewinder rattlesnake (Crotalus cerastes), a creature better known for its venomous bite than its graceful movements. But this snake can climb up sandy slopes without sliding back down to the bottom a feat that few snake species can accomplish.

Snakelike, or limbless, robots are intriguing to scientists for several reasons. First, their lack of legs, wheels or tracks means they don't often get stuck in ruts or held up by bumps in their path. They could also be used to access areas that other bots can't get to, or to explore places that aren't safe for humans. [Biomimicry: 7 Clever Technologies Inspired by Nature]

The sidewinder shimmy

To get a closer look at their live study subjects, the researchers headed to Zoo Atlanta, where they were able to examine six sidewinder rattlesnakes. They tested the snakes on a specially designed inclined TABLE COVERED with loosely packed sand.

Fifty-four trials were conducted, with each of the six snakes slithering up the sandy table nine times, three times each at varying degrees of steepness. As the snakes worked their way up the makeshift sand dune, high-speed cameras tracked their movements, taking note of exactly where their bodies came into contact with the sand as they moved upward.

The researchers found that sidewinder snakes live up to their name. The slithery creatures moved up the sandy incline in a sideways motion, with their heads pointing toward the top of the incline and the rest of their bodies moving horizontally up the slope. The researchers then looked more carefully at how sidewinders carry out these complex movements.

"The snakes tended to increase the amount of body in contact with the surface at any instant in time when they were sidewinding up the slope and the incline angle increased," said Daniel Goldman, co-author of the study and an associate professor of biomechanics at the Georgia Institute of Technology in Atlanta. Specifically, the snakes doubled the amount of their bodies touching the sand when navigating the slope, he added.

And the parts of the snake's body that were touching the sand during the ascent never slipped back down the slope because the creature applied the right amount of force in its movements, keeping the sand under it from sliding, Goldman told Live Science.

Snake robots

To put their newfound understanding of sidewinding to good use, Goldman and his colleagues got in touch with Howie Choset, a professor at The Robotics Institute at Carnegie Mellon University in Pittsburgh. Choset, who has been developing limbless robots for years, already developed a snakelike bot that performs well both in the lab and in real-life situations. However, his slithering machine has run into one particular problem during field tests.

"These guys have been making a robot sidewind for years over a wide diversity of substrates, but they had a lot of trouble on sandy slopes," Goldman said.

To get the robot moving over sandy dunes, the researchers applied what they now know about the sidewinding rattlesnake's patterns of movement. They programmed the robot so that more of its body would come into contact with the ground as it slides up the slope. They also applied what they had learned about force, which enables the robot to move its weight in such a way that it keeps moving upward over the sand without rolling back down the slope.

Now that Choset's snake robot can move over tough terrain, it'll be better equipped to handle the tasks that it was built to tackle.

"Since these robots have a narrow cross section and they're relatively smooth, they can fit into places that people and machinery can't otherwise access," Choset told Live Science.

For example, these limbless robots could be used during search-and-rescue missions, since the slithery machines can crawl into a collapsed building and search for people trapped inside without disturbing the compromised structure. The snake bot could also be sent into containers that may hold dangerous substances, such as nuclear waste, to take samples and report back to hazmat specialists.

Choset also said these robotic sidewinding abilities could come in handy on archaeological sites. For instance, the robots could one day be used to explore the insides of pyramids or tombs, he said.

The research represents a key collaboration between biologists and roboticists, said Auke Ijspeert, head of the Biorobotics Laboratory at the Swiss Federal Institute of Technology at Lausanne (EPFL), who was not involved in the new study.

"I think its a very exciting project which managed to contribute to the two objectives of biorobotics," Ijspeert told Live Science.

"On one hand, they took inspiration from biology to design better control methods for the robot," Ijspeert said. "By looking at how sidewinding takes place in a snake, especially with slopes, they found out the strategy that the animal uses and, when they tested it on the robot, it could really improve the climbing capabilities of the robot."

The researchers also achieved the second goal of biorobotics, he said, which is to use a robot as a scientific tool. By testing the different speeds at which the robotic snake could successfully climb up the sand, the researchers were able to pinpoint exactly how fast real snakes make their way up these slippery slopes.

"It's a nice example of how robots can help in biology and how biology can help in robotics."

The study was published online Oct. 9 in the journal Science.

It's not only Olympic superstars who misjudge their alcohol consumption.

Earlier this week, Michael Phelps was arrested in Baltimore for driving with a blood alcohol reading of 0.14. He's apologized for his behavior, but the goal is to prevent folks who are over the limit from getting behind the wheel in the first place. Not surprisingly, there are gadgets – and apps – to help stop them.

You may have faced a situation in which you were trying to convince a friend not to drive, or you may have wondered if that last glass of wine was enough to impair your own judgment. A personal breathalyzer, which cost from $30 to $120, can do the trick by demonstrating objectively that you or a companion has had one – or more – too many.

Matchbox-sized personal breathalyzers measure blood alcohol concentration (BAC) based on the level of alcohol in your breath (about 10 percent of the alcohol you drink is released into your breath). From that reading, they estimate the percentage of alcohol in your blood. Generally, these devices use either a semiconductor sensor, which is small and inexpensive, or a fuel cell sensor, which can be more accurate and is used by law enforcement in portable devices. I tested two models that work with smartphones.

The Breathometer costs $50 and plugs into the headphone jack of an iPhone or Android smartphone. It uses a semiconductor sensor and an AAA battery that should last for up to 75 tests. The company recommends waiting until 20 minutes after your last drink for a more accurate test, as it will avoid higher results from the alcohol residue in your mouth. It takes about a minute to warm up the device, and then you’re prompted to blow for four to five seconds through a hole from about two inches away.

I had to retest for accuracy several times. And in one case, after I toted it around in my briefcase, the sensor collected too much lint to be effective. (It was easily blown clear, however.)

But the Breathometer was reasonably accurate. If anything, it tended to err on the side of caution by yielding slightly higher results. One evening, after a glass of wine, I blew a 0.03, eliciting a yellow warning and a note: "You should be sober at 12:17 AM" – roughly 2 hours away. A retest brought the number down. And here’s a bonus feature: The app provides a list of local cab companies.

Breathalyzers need to be calibrated regularly, and the Breathometer is no exception. The company says it should be recalibrated every 250 tests, or after nine months. The mail-in service costs $20

Alcohoot, another smartphone accessory, is a competitor. It costs $100, but it uses a fuel cell sensor, the same technology police officers use, and it has a rechargeable lithium-ion battery that will last for up to 500 tests. It's slightly larger than the Breathometer, and it works almost identically. You blow through a hole where the sensor is. The company recommends using one of the eight washable and reusable plastic mouthpieces that come with the device.

The Alcohoot started up more quickly and delivered consistent results, probably because I used the nozzle. I blew a typical 0.017 after a large glass of wine. The program reminded me that the legal limit is 0.08 BAC. The app charted my reading and the time of day, and it recorded that I was below the red line that denotes the legal limit. It also warned me, "You may begin to feel moderate effects." I noted that I could call an Uber, Lyft or Hailo car service from the app.

In addition to being more accurate, the higher-priced Alcohoot doesn't need to be recalibrated as often as the Breathometer. The company says it lasts for up to 1,000 tests or one year. Its recalibration service costs $30.

I trusted the Alcohoot more than the Breathometer, mainly because of its ergonomics, but keep in mind that while these are FDA registered devices, they are not a license to drink and drive. Both companies repeatedly note that you should not do so.

Furthermore, breathalyzers can be imprecise. Using a mouthwash or breath spray can affect the accuracy of the measurement, and alcohol affects everyone differently. On the other hand, having one of these devices at the ready to convince a friend that he should take a cab or sleep on the couch could end up saving a life.

“It’s insane. The rivets—they’re not flat. It’s hard to see anything be able to move around these understructures of bridges,” she says.

But along with fellow UMD professors Nuno Martins and Richard La, that’s exactly what Bergbreiter will have to do. They’re not coming up with a better harness or weird contraption to make it easier for people to scale bridges, however. Instead, they’re at the beginning of a three-year project, bolstered by an $850,000 grant from the National Science Foundation, to design, build, and deploy a swarm of miniature robots, each no more than several centimeters long, that will aid in bridge inspection by climbing on them, traversing the tricky terrain of rivets and bolts on their undersides, and working together with minimal input from a human controller to store and relay images and measurements related to the strength and stability of the bridges they’re crawling over.

“The vision is you could have a whole bunch of these guys running around a bridge,” says Bergbreiter. “I don’t think that’ll happen in three years, but I think we’ll be at the point where we can make a decision how far this can be commercialized.”

The average age of the American bridge is 42 years old—as good an age as any for regular check-ups. Inspecting any one of the approximately 605,000 bridges in the U.S. today is a tedious, hours-long task usually undertaken by a team of experts. Releasing 50 to 100 bug-sized robots in tandem with human teams would cut down on the amount of time needed to inspect a bridge and prevent the need to shut down a bridge to pedestrian and vehicle traffic. The idea is that these robots, once given a certain command, would be autonomous, moving around a bridge on their own and organizing together to examine various surfaces and take measurements. Designing such a robot would be a feat of ingenuity in dynamic movement as well as adhesion, and both are areas Bergbreiter has dug into since helping establish UMD’s Micro Robotics Lab in 2008.

“The key is being able to stick and crawl over bridge surfaces, whether they’re concrete or steel or painted steel,” she says. Bergbreiter says the adhesion technologies being worked on in the lab draw inspiration from the Spiderman movies as well as Tom CRUISE in Mission Impossible. Picture a robot using electroadhesion to latch onto a bridge, or one the size of an earwig using little hairs, or “gecko adhesion,” to remain affixed to a rivet. “The key there is being able to turn it off and on,” says Bergbreiter. Add a tail or a flexible backbone, and you have a robot that will stick to a bridge and move easily from horizontal surfaces to vertical ones, and over structures like rivets that are far from flat. Bergbreiter says small projects using new adhesion tech and different mechanisms for movement have already been demonstrated in UMD’s Micro Robotics Lab.

As for relaying information useful for bridge inspections, the professors might be pulling from the Micro Robotics Lab’s TinyTerps, quarter-sized robots named after UMD’s mascot, the terrapin (terp for short), that can be thought of as tiny Arduino boards. “They effectively have a radio microcontroller, and you can plug different sensors on and off of it,” says Bergbreiter.

It’s soon to tell what these bridge-inspecting, micro-scale robots will look like or how, exactly, they will work. But Bergbreiter says that’s part of the fun of this new project.

“We’ve been working on big robots since the late ’50s and ’60s, and there’s a lot of amazing things that people have done in that area,” she says. But, she adds wryly, “There’s plenty of room at the bottom.”

Inspecting a ship’s cargo is a dull, tedious, time-consuming task. So a pair of researchers at MIT, including graduate student Sampriti Bhattacharyya and her advisor Harry Asada, created a small robot that resembles a squished foam ball to inspect ship cargo quickly, cheaply, and silently. They presented their findings earlier this month at the International Conference on Intelligent Robots and Systems.

Named EVIE, for Ellipsoidal Vehicle for Inspection and Exploration, the robot can swim as fast as three feet per minute. EVIE uses six jets to move underwater, with an algorithm determining which jets PUMP WATER and when. The half of the body that doesn’t house the jets is watertight and sealed, protecting EVIE’s controls, a battery that lasts for up to 40 minutes, an antenna, and inertial sensors.

In future practice, swarms of EVIE-bots might collectively inspect ships in port, swimming and scanning underwater as a team. Bhattacharyya hopes to get the price of a functioning inspection gadget down to around $600, so that the swarms are cheap enough to be useful.