French firm promises humanoid robot in 2007

(via Engadget)

Hoping to be known for more than their expertise in raising the white flag, the French announced information on coming humanoid robot Nao, to be available in 2007.

The project has been launched in the early 2005 and aims at turning available to the public, for an affordable price, a humanoid robot based on mechanical, electronic and cognitive features, compeeting with the research prototypes’ ones (see graphic below).

robot-nao2.gif

Delivered with basic behaviors, the robot will be, as of its market introduction, the ideal support to self initiating with robotics. Later on, improved with many behaviors, it will become an autonomous buddy for all the family. It will grow later on from its buddy role up to true functionalities required for daily tasks assistance (monitoring, remote support, information, reality-virtual bond…).

. . .

The robot is based on a Linux platform and on scripting with Urbi, very easy to learn programming language, possibly in code commands for the experts, or through a graphic interface for the beginners.

Currently in final development phase, the first specimens for sale will be presented at the end of 2006 and available in the second half of 2007.

No mention of high power laser-beam eyes, but perhaps that will be a future upgrade in a home-security model of the Nao.

[tags]Robots, Nao, Humanoid robots[/tags]

Anti-RFID techies are here to protect you

(via Engadget)

With the recently announced RFID passports we’ll be getting here in the states soon, some folks are wisely concerned about their electronic safety in a world were RFID signals are easy to pick-up.  Given a desire to protect RFID identities in the coming age where you will not be able to be sans RFID, the folks at RFID-guardian have worked on a device to protect you and all your RFID-ness.

The Guardian has three main components: an RFID tag emulator (that consists of a receiver part, a transmitter part, and an antenna), an RFID reader (also with antenna), and an embedded computer.

. . .

The Reader that is built into the Guardian is used to perform periodic scans of all tags that the Guardian owner carries around with her. If tags keep turning up in those scans, they will probably belong to the Guardian owner, so they must be protected from interrogation or corruption by unknown RFID readers. The Guardian decides this for itself, but it can (of course) be overridden by its owner.

. . .

The RFID Guardian could eventually be integrated into a PDA or a cellphone, but our research prototype is currently built on a breadboard (the next one will be on a single PCB). The current prototype consists of three separate parts:

  • some electronic circuit boards, one for its tag emulator receiver, one for its tag emulator transmitter, and one for its reader (plus some voltage translator boards)
  • a number of radio frequency antennas
  • an embedded computer that performs all the software tasks. This is an XScale PXA270 (an ARM descendant manufactured by Intel) on a development board, the Triton-270 sold by Karo.

There is no estimate yet on how much this will cost.
[tags]RFID, anti-RFID, privacy[/tags]

Anna Konda – firefighting snake robot

(via Engadget)

When I first read about this earlier in the week, I was going to skip it. It sounds pretty cool, but I wasn’t sure it was Blah-worthy — it just didn’t seem interesting enough to me to makeannakonda.jpg a post about it. But I’ve run into this news item some 6 or 8 times since I first ran across, it. So, when I saw it on Engadget while catching up with my news-reading there, I decided I had better just post it and get it out of my head. That said, here’s some of the details on the newest robot that’s not likely to someday destroy you puny humans in its bid to take over the world.

Try to picture a snake-like robot that can move into places that are too dangerous for humans to enter. The snake can climb up stairs, force past beams and twist itself round corners. Imagine that it has a built-in advanced water tap that not only can be turned on and off, but can allow the direction of the water flow to be altered.

. . .

The snake contains 20 water hydraulic motors that move the robotic joints – and a similar number of valves to control the water flow to each motor. Each module consists of two hydraulic motors and two valves. The outer layer is comprised of a strong steel skeleton containing the joint modules,which can rotate around two orthogonal axes. The joints are controlled by custom-built electronics.

. . .

The energy to move the joints comes from 100 bars of hydraulic water pressure. “This pressure is strong enough to lift a car up off the ground, something that again explains how the snake can in principle break through a wall. But both the hydraulic pressure and the use of pure water without additives in the hydraulic system have posed challenges”, Liljebäck says.

. . .

The steel skeleton and motors are being custom-built at local workshops in Trondheim, partly because the research scientists needed to take a novel approach in the construction of the water hydraulic valves.

“The lack of space has been a major challenge,” says Liljebäck. “We needed power valves that were small, water tolerant and capable of controlling both the direction and the amount of the water flow. The closest thing we found on the market that met the criteria was valves used in Formula One racing cars, but these cost NOK 100,000 each and didn’t tolerate water. As a result, we decided to manufacture our own valves and, in co-operation with a local workshop, we built a prototype from scratch.”

Wow. Lift a car or break through a wall? Maybe this snake will one day crush you puny humans in an attempt to take over the world.

[tags]Robot snake, Firefighting robot[/tags]

OpenDNS – trying to improve your surfing experience

A recent ArsTechnica article provides more details and insight into the OpenDNS project which we have recently mentioned here.  Jeremy starts the article by noting that DNS isn’t something he (nor most other surfers) even thinks much about.

Thus, the announcement of a new, free DNS replacement service took me somewhat by surprise. Why would I want to change my DNS provider? What would I gain by doing such a thing? And more importantly, what was the catch?

According to David Ulevitch, founder of OpenDNS, the experience he gained starting and running EveryDNS (a free DNS service for web site hosts) taught him a lesson about the state of DNS services in general and the increasing problem of spam, phishing scams, and botnets. He noticed that nefarious groups were using EveryDNS as a vector to perpetuate these sorts of attacks. He added code to EveryDNS that would search for and deny these attempts, and shared information about the perpetrators with other DNS services. While he noticed the attacks no longer targeted his system, they simply moved elsewhere to find easier targets.

His idea was to attack the problem from the other end. Instead of trying to fix every web hosting DNS service, he thought about providing a DNS service for end users.

. . .

To attract people to this new service, Ulevitch promises that it will speed up DNS resolution, making for a faster web browsing experience. If that wasn’t enough, the service is designed to catch common typos, delivering the URL you want rather than the one you typed, and putting an end to sites that try and deceive users by typo-squatting.

So how well does it work? The answer depends very much on your current service provider. I’ve never had to spend much time waiting for DNS resolution, mostly because my provider, Shaw, keeps a very large cache of DNS addresses on a high-speed server. However, this can have its drawbacks when sites shift IPs too often and the cache isn’t updated quickly enough.

So, is OpenDNS something worth using?  Well, Chris Pirillo seems to think so.  Jeremy, the author of the above-quoted ArsTechnica article has this to say:

Is the site useful or not? With phishing filters coming in IE 7 and Firefox 2.0, it may be a solution looking for a problem. Still, it’s nice to have a backup option in case your ISP’s DNS services flare out temporarily. The one really nice thing about the service is that it requires no software installation (merely setting your DNS settings for your network connection) and if you decide you don’t like it, it’s trivial to change back to your ISP’s DNS service. Having a no-hassle “opt-out” clause is definitely a good thing to have with any new service you wish to try out.

[tags]OpenDNS, Faster surfing[/tags]

Why is the sky blue?

(via MentalFloss)

Turns out, it isn’t.  That’s the short answer.  The long answer is interesting – it’s violet, but our daylight vision is such that it is seen as blue.

Why is the sky blue? It is a question children ask. Yet it also intrigued Leonardo da Vinci and Isaac Newton, among many other legendary thinkers. As late as 1862, the great astronomer John Herschel called the colour and polarization of skylight “great standing enigmas.” Even today, our perception of sky blue is little understood by laymen.

. . .

In 1873, James Clerk Maxwell realized the truth: The scattering substance must be the molecules of air themselves, none other. Indeed, John Ruskin suggested in 1869 that sky blue comes from light “reflected from the divided air itself.” Lord Rayleigh followed Maxwell’s advice and calculated that the observed scattering of skylight requires molecules whose size accords with that indicated by other physical arguments.

. . .

As you gaze at the clear blue sky, then, you are beholding unambiguous evidence that atoms really exist, something that was widely questioned as recently as the 19th century. This presents a profound but simple answer to our opening question about the sky’s colour. But even so, further explanation is required.

If Tyndall and Rayleigh are right, then the violet wavelengths from the sun, having still shorter wavelengths than blue, should be scattered even more. Given this, shouldn’t the sky be violet, not blue?

Indeed the sky is violet, if you observe not with the naked eye but with an instrument that objectively measures the intensity of the spectrum at different wavelengths. Such a device, a spectrophotometer, shows that, in fact, the highest peak of the intensity of skylight occurs in the violet range.

But why do we see blue, nonetheless? The resolution of the mystery lies in our daytime vision, which happens to be eight times less sensitive to violet than to blue light.

Does that mean it is “incorrect” to call the sky blue? Not really. Our names for colors reflect our common perception, whatever a mechanical instrument might say.

There you have it.  The sky is violet.  But we still call it blue.  Our eyes don’t see violet as well as blue, and the higher intensity of violet is not sufficient to overcome our innate sensitivity to blue.  That means more blue is registered by the cones in our eyes, so our brain just gives us what it has more of to work with.

[tags]Why is the sky blue, Eye color sensitivity[/tags]

On the ATI/AMD deal

OK, I know I’ve just said I don’t care enough about the deal to comment on AMD buying ATI (which, through the magic of future posting, was actually written 24 hours ago), and then I go and find two really good write-ups on the deal. First, hit DubiousQuality and check out Bill’s view. He makes some good comments on the deal – certainly enough for me to start caring about the deal. When we’re done there, we’ll look at Joystiq’s addition to the discussion, BTW.

Continue reading “On the ATI/AMD deal”

ATI announces Radeon X1950 video card

(via TechEBlog)

I’m skipping all the talk about AMD buying ATI — I just don’t care enough to write about it. I’ve never used an ATI card for my rig, so I’m not really interested in who owns ATI. All I care about ATI for is the competition that pushes nVidia to continually push out better and better video cards.

So the latest announced card from ATI, the Radeon X1950, is of interest to me personally only because it means nVidia has to put out something better that much sooner and drop the price of all the currently available cards so I can upgrade soon. I know that ATI makes some great cards now. But I’m an nVidia fanboi, and as long as the company produces great cards, I doubt I’ll stray to another vendor.

That said, here’s the details from DailyTech.

This week ATI sent an advisory out to its OEM partners announcing the details of the new Radeon X1950 and X1900 graphic cards. Both ati-x1950.jpgof these new cards are based on the same R580 core, but with some fundamental differences.

R580, the 48 pixel-shader processor version of the R520 (Radeon X1800), was announced this past January. R580 features a robust memory controller capable of utilizing several different types of memory, including GDDR4 which was not even available when the Radeon X1900 was first announced. Since then Hynix and Samsung have both jumped on the GDDR4 train with revenue shipments beginning several weeks ago. Continue reading “ATI announces Radeon X1950 video card”

NIST makes even more accurate atomic clock

(via Engadget)
Apparently, the possibility of being off 1 second every 70 million years was just too much for the National Institute of Standards and Technology.  So NIST recently announced a new ultra-precise clock based on the oscillations of a mercury ion.  The new clock, tested and measured over the past 5+ years, should have an accuracy such that drift will be less than 1 second over 400 million years.  It will still take some time before this clock becomes the new standard, but the extra precision certainly suggests it will happen.

A prototype mercury optical clock was originally demonstrated at NIST in 2000. Over the last five years its absolute frequency has been measured repeatedly with respect to NIST-F1. The improved version of the mercury clock is the most accurate to date of any atomic clock, including a variety of experimental optical clocks using different atoms and designs.

. . .

“We finally have addressed the issue of systemic perturbations in the mercury clock. They can be controlled, and we know their uncertainties,” says NIST physicist Jim Bergquist, the principal investigator. “By measuring its frequency with respect to the primary standard, NIST-F1, we have been able to realize the most accurate absolute measurement of an optical frequency to date. And in the latest measurement, we have also established that the accuracy of the mercury-ion system is at a level superior to that of the best cesium clocks.”

And if you just want to learn more about atomic clocks and how they work, check out the NIST atomic clock page.

[tags]Atomic clock, Ultra-precision[/tags]

Linux Read/Write NTFS driver

(via OSNews)

This is practically old news now, but the Linux NTFS team announced last Friday, July 14th, that a public release of the NTFS for Linux driver is now available.

Hello,

As part of the Linux-NTFS project, I’m happy to announce my contribution to ntfsmount and libntfs which resulted ntfs-3g, a read-write ntfs driver, capable for unlimited file creation and deletion.

The driver was successfully tested very exhaustively for a longer period of time by many ways and methods, creating and destroying millions of files and directories on newly created images, and on over 40 real, very diverse NTFS images collected over the last four years.

[tags]Linux NTFS driver, Linux, NTFS[/tags]

Simple tips to boost your Intarw3b speeds

Chris Pirillo shares a couple of easy to follow tips for improving network speeds on your computer.

  1. Open yourself to OpenDNS. They have configuration pages for your OS or hardware router. Both Scott Beale (LaughingSquid) and I have seen dramatic differences in speed since going through their DNS servers. Matt’s happy about it, too. My speed difference was probably even more dramatic, considering I’ve already implemented my second recommendation…
  2. If you’re on Windows, install FastCache: “Every time you go to a new website, or a website you haven’t been to in a day or so, your computer needs to locate the server again using DNS. Each DNS request can take anywhere from 50 milliseconds, to even a second or more, but most are in the 100 to 200 millisecond range. Want to save a couple hundred milliseconds? Probably not, but what about a couple million or more? That’s what AnalogX FastCache is all about.” I’ve been running FastCache since Mark released it on the first Helpathon. My stats are somewhat stunted, but I’ve saved 6d 18h 27m 53s worth of time (in 118w). It really adds up.

There you have it. Now hop to, and get that pr0n faster.

[tags]Chris Pirillo, Faster networking[/tags]

Superheated metal moves liquid uphill

I’ve been reading the latest issue of American Scientist in my spare time lately. A really cool article caught my eye, and I thought I’d pass it on for the more geeky, science-loving types in my audience. While I think half my readership isn’t into the science stuff I enjoy, I believe that other reader might be. For him, I pass along this cool story of liquid moving in the wrong direction.

SCIENCE OBSERVER
Going Against the Flow
Sometimes particles prefer to propel themselves uphill
Fenella Saunders

Particles strive for the life of a couch potato—sinking into a spot that has the least energy, where gravity can’t pull them down any farther and movement is at a minimum. Getting a particle moving requires keeping it off kilter, out of equilibrium. But particles in such a state tend to bounce all over; harnessing their movement in a single desired direction is the goal of many nanoscale devices.

One way to do this is with a ratchet effect—a mechanism that uses spatial asymmetry and energy gradients to make movement easier in one direction than another. It turns out that in some cases, ratchets not only control movement, but can also move particles in unexpected directions—away from a minimum energy state, the molecular equivalent of a creek climbing uphill under its own power.

. . .

If a skillet is heated to an extremely hot temperature, between 200 and 300 degrees Celsius, drops of water flicked into the pan will skitter across the surface, remaining intact for a minute or so. A surface not quite so hot will boil away the water droplets instantly, but the superheated surface instead instantly turns the bottom of the droplets into a layer of steam. Vapor is a poor heat conductor, so the steam insulates the drops from further boiling. It also provides them with a means of movement: The water drops bounce around like hovercraft on a cushion of air.

Linke and his colleagues did not use a smooth metal surface, but one covered with a sawtooth pattern. The teeth inclined more steeply in one direction than the other—an asymmetrical surface, and therefore a ratchet mechanism. Millimeter-sized water droplets piped onto the superheated sawtooth surface zip off in one direction like airport passengers on a moving walkway, reaching speeds of up to 5 centimeters per second, even if the surface is tilted so that the droplets have to climb uphill. As the investigators reported in the April 21 issue of Physical Review Letters, the phenomenon works for many other liquids, such as ethanol and liquid nitrogen, although the temperature at which the Leidenfrost effect kicks in varies from 50 to 150 degrees above the boiling temperature of the liquid.

More details on the science behind this in the full article. Also, access the full issue online, where many of the articles are freely readable. Some of the really juicy stuff requires membership ($28-$70 for 1-3 years in the US), but after reading a single issue, I’m considering a subscription.

[tags]American Scientist, Liquid moving uphill, Cool geeky science stuff[/tags]

RPU – Ray Tracing Processing Unit

(via Tingilinde)
I have no details on this beyond what the brief note at Tingilinde, but apparently the August 2006 issue of Scientific American has an introductory article to new ray tracing hardware.  The information in the article was part of a talk at SigGraph 2005.  Looks like I’ll need to head to the bookstore and pick up the magazine (or possibly get the digital edition).
[tags]Scientific American, RPU, Ray Tracing Processing Unit, Ray Tracing[/tags]