Nanowires enable faster, low-power 10nm 3D transistors

When it comes to 3D transistors, you’ve probably heard of FinFET — the 3D, “Tri-gate” transistors that are taking Intel (and eventually other silicon foundries) to 22nm and beyond — but now a research team from Harvard and Purdue have forsaken silicon entirely and created a 3D transistor out of nanowires. These new transistors will get us to the 10nm mark, beyond silicon’s theoretical 14nm limit.

In a conventional 3D FinFET (field effect transistor), the individual components are much the same as a standard, flat, MOSFET, but instead of each layer being slapped down on top of each other, the metal gate wraps around a fin of silicon, kind of like a noose or collar. This requires a process calledatomic layer deposition (ALD), where each part of the transistor is effectively grown, from the ground up, atom by atom. In Harvard and Purdue’s transistor, nanowires made of an indium gallium arsenide alloy replace the silicon fins, and ALD is used to deposit a high-k dielectric.

Indium gallium arsenide (IGA) is one of many semiconductors, including our friend graphene, that might one day replace silicon. In this case, the IGA has been fashioned into a nanowire and then coated with an aluminium dioxide dielectric to create a transistor. IGA is already used in high-frequency electronics because, like graphene, its electron mobility is much higher than silicon, meaning it requires less voltage to switch, and can be switched at a higher frequency.

The end result, as with all discoveries of this ilk, is computer chips that require less power and produce less heat. This in turn allows for smaller batteries and lighter computers. The most notable advantage of the Harvard and Purdue nanowire transistor, though — and unlike many other similar discoveries — is that it can be built using conventional chip fabrication processes. In other words, we might actually see nanowire transistors within a few years.

Read more at ScienceDaily (but it’s severely lacking in details, unfortunately)

[Image credit: Nanodevice Laboratory]

Source

Department of Defense seeks app developers for Angry UAVs

DARPA, the US military’s research wing, has a problem on its hands: Satellites, unmanned drones (UAVs), and myriad other worldwide sensors are now so ubiquitous and omnipotent that the Department of Defense (DOD) doesn’t actually know how to make the best use of them. In other words, the hardware is there — oh boy is the hardware there! — but the software isn’t. To tackle this particularly tricky issue, DARPA is looking for smartphone app developers to help build “sophisticated, adaptive applications.”

Yes, DARPA wants to give smartphone developers access to the DOD’s fleet of Hellfire missile-equipped UAVs. Instead of using a single, remote pilot to fly just one UAV, DARPA imagines “an app [...] that allows a swarm of small deployed UAVs to be controlled as a single unit (a hive [mind] so to speak).” The same app could be applied to DARPA’s BigDog robot mule, or in the future an army of ground-based, Cylon-like robot warriors.

It goes further than battalions of automatons, though: DARPA is also looking for app developers who can program novel solutions for organizing and sharing video surveillance, peer-to-peer transfers between sensors, and new, user-friendly interfaces for the computers that soldiers are required to use, both at HQ and on the battlefield. The press release explicitly mentions using smartphone-type sensors such as accelerometers, cameras, and gyroscopes for combat applications, too; an accelerometer could detect when a truck rolls over it, for example.

At the heart of this rather unprecedented outreach is ADAPT, DARPA’s new sensor system. Built on a common hardware and software platform, ADAPT is simply meant to speed up the DoD’s development cycle, and thus its reaction speed in times of war or conflict. The current route, which just about every government in the world uses, relies on proprietary, custom-built, non-interoperable solutions that take up to eight years to develop. Smartphone developers have shown the ability to develop groundbreaking apps in days or weeks, and our militaristic overlords would very much like a slice of that pie.

The end goal, incidentally, is a military app store, where a soldier or officer — equipped with a standard-issue DOD tablet or smartphone — can hit up the Warfighter Market and download Angry UAVs. Funny, in a terrifying kind of way.

Source

Windows 8 beta coming in February, too late for a final release in 2012?

As a casual aside, while discussing the future of the tiletastic Windows Store, a Microsoft vice president announced that the Windows 8 beta would begin in late February 2012. The beta will be feature-complete and will allow developers to begin listing their apps in the Store.

The timing of the beta is curious, and ultimately quite telling. The first public release of Windows 7 was at the PDC conference at the end of October 2008, and the beta followed at the start of January 2009 — just over two months. The first public build of Windows 8, the Developer Preview, emerged in mid-September 2011; by the time the beta rolls around, it will have been ruminating for more than five months.

Now, the Metro experience and Start screen in Windows 8 are definitely a larger overhaul than the tweaks that occurred between Vista and 7, and five months isn’t really a long time in the grand scale of computer operating systems, but it now looks unlikely that Windows 8 will arrive before next winter. Windows 7 was not released until late October 2009, a full 10 months after the beta release. Count forward 10 months from the end of February and Windows 8 might just land in time for Black Friday 2012 — but if it took twice as long to go from Developer Preview to Beta, who’s to say that Windows 8 will even arrive in 2012?

This conclusion is definitely at odds with some of the rumors that have been bandied around in recent months. In a nutshell, the prevailing opinion is that Windows 8 has to arrive right now to fend off Apple’s complete domination of the tablet market, and its growing share of the laptop market. Some rumors even pointed to Windows 8 tablets arriving in the first half of 2012, perhaps manufactured by Nokia, but for that to occur Microsoft would need to have some kind of internally-tested, probably-ARM-specific build ready to go for CES in January — an incredibly unlikely proposition.

In all likelihood, then, we’ll see Ice Cream Sandwich tablets, the iPad 3, and possibly even the next iteration of Android and Mac OS X before Windows 8 arrives on fondleslabs and beige boxes alike. We suspect, though, that a late release won’t even matter. Windows 7 is still doing fantastically — it just passed the 500 million licenses sold mark — and Windows 8 will be so different from other tablet OSes that timeliness shouldn’t be a primary concern.

Of course, with the iOSification of Mac OS X 10.7, there’s always a possibility that the next version of Apple’s desktop OS will also run on tablets. In that case, Microsoft really would have a fight on its hands — but we’ll cross that river when we get there.

If you’re an app developer, check out the latest Building Windows 8 post.

Source

Generating electricity with a shoe

The human body, like a computer, wastes a huge amount of energy — usually in the form of heat. In some cases, this heat can be captured and repurposed — to keep you warm when you wear a jacket, for example, or to heat up your house in the case of data furnaces. Now, sadly this isn’t a story about a piece of clothing that captures your body heat and turns it into electricity — that’d be cool, but perhaps a bit Matrixesque — but it is about a pair of shoes that turn your footsteps into battery power.

A team of engineers from the University of Wisconsin have used a process called electrowetting to create about 10 watts of power from simply walking along. Electrowetting is where a liquid — usually hydrophobic, and a mixture of oil and water in this case — is forced to move by applying an electric current. These shoes reverse the process and force the mixture over some electrodes, creating a current which is stored in a battery.

As you can imagine, it’s actually rather easy to apply such a technology in shoes: Our natural walking gait involves shifting our weight from toe to heel, toe to heel — and the shoes simply have two (presumably very rugged) bladders at both ends. When you put your toe down, the liquid is forced over some electrodes and down to the heel — and then when your heel lands, the flow is reversed. Voila, electricity! If that wasn’t cool enough, the shoes even have a micro USB port on the back (so that you can cross your legs in a coffee shop and charge up your phone, obviously). Tim Krupenkin, one of the inventors, also says that they’ve created a way of wireless transmitting power from the shoe, “like WiFi” — but no details are given, unfortunately.

Sadly, that’s a recurring theme: There’s a rather sexy sketch of the aforementioned shoe with a USB socket, but there’s scant few details on commercial availability or how it operates in real life. Can you feel the liquid squeezing out of the bladder between your toes and into your heel? What if the bladder bursts? What if you leave the USB cable plugged in and trip and fall?

The idea of shoes that generate electricity is not a new one. Nary a year goes by without some kind of (usually piezoelectric) trainer that will solve all of our mobile computing needs. But for some reason, no such clog has yet arrived on the market. Maybe this is the one.

Read more at Popular Science

Source

Data.gov goes open source, first step in government being afraid of the people

Whether or not you voted for President Obama, it can’t be said that he hasn’t made intelligent moves in bringing the US government fully into the information age. In fact, his first executive order that he ever signed created a new information portal on the web, Data.gov, to allow web users access to information made available by the Freedom of Information act. Accessing that before was difficult because of the bureaucratic hoops people had to jump through to get the data they sought. Coming fully online in 2009, Data.gov allows web users to access a range of information, such as who has visited the White House, and be able to represent that data using visual charts. This toolset makes it much easier for US citizens to hold their government accountable for its actions.

With the success of Data.gov behind him, Obama was able to forge ahead and be instrumental in forming the Open Government Partnership (OGP), a consortium of 46 countries that have agreed to create tools to enable governments around the globe to foster three things: transparency in government affairs, higher civic engagement, and a better level of accountability to help end corruption. A tall order that’s going to require more than a press conference with verbal commitments. Putting his money where his mouth is, Obama has ordered the source code behind Data.gov to become open source, allowing any government, organization, or person to download and use.

This is an important first step for people to regain control of their governments. To take a quote from V for Vendetta, “People should not be afraid of their government, government should be afraid of the people.” Information is a powerful weapon in making this happen, and much preferable to armed militias storming capital buildings because they have had enough tyranny in their lives. With foreign aid usually coming in the form of money or soldiers, Obama has displayed to the world a different policy than any of his predecessors. It seems he understands that for real change to come about in the world as demanded in demonstrations like the Occupy movement, throwing money at problems isn’t a solution.

Available now on GitHub, the Data.gov portal allows users to create charts like the one pictured above. It shows the number of visitors to the White House, where their destination was, and how many people were in the meeting. The search and display possibilities are literally endless, as you can add and subtract variables to manipulate how the data is output. The obvious problem, and one that Data.gov has been accused of, is that the data output is only as good as the quality of the data sets used to generate those reports. A lot of the information available is related to geospatial, causing critics to clamor for more relevant data. To improve the quality of the information, citizens can’t sit back and be happy with just a platform, they must continue to demand improvement and offer to help themselves. A key piece is the fact that the platform is built on Drupal, an opensource CMS. This allows anyone to develop special plug-ins and apps that will work with the data sets that are stored inside the software. By developing for the software used with their governments, citizens can keep tabs on what is going on.

The opening of the code for this portal should be seen as just the beginning of an information revolution. The struggle to get access has already been won, the fight now lies in making sure that the information that is fed to the public is accurate, complete, and unbiased.

Remember, remember the 20th of September 2011. The first day of the new information revolution.

Source

Music to the photographer’s ear: Dial in Adobe Lightroom image adjustments with a MIDI controller

Adobe Lightroom quickly became the go-to software for photographers who need to index and manipulate lots of images in a hurry. But the dozens of tiny adjustment sliders make Lightroom the go-crazy software, too. Now, an enterprising developer from Finland has created a Lightroom plug-in, Knobroom, that lets you use a USB MIDI device to control exposure, highlight recovery, brightness, contrast, and so forth quickly and more accurately than mouse-adjusted sliders. It’s a free download from and it’s getting rave notices from users who’ve tried it.

According to developer Jarno Heikkinen, “Knobroom is a direct manipulation add-on for Adobe Photoshop Lightroom 3, allowing access to Lightroom photo development settings with a MIDI-controller.” That’s a fancy way of saying that the plug-in allows photogs to put hardware controls in place of software settings, without Adobe having to get involved.

The plug-in connects a hardware MIDI controller normally used to fine-tune musical settings to a few of the 70-plus sliders (in nine sections) in Lightroom’s Develop module. The user chooses which sliders to control. Each knob on the MIDI controller directly manipulates one slider, or several if your MIDI box can switch among control presets A, B, C and so forth.

From what we’ve seen, people who tried Knobroom are happy. “Editing images is blazingly fast…” says one early adopter on MaxedIn.net. “It’s very easy to change different settings far from each other, for example moving around sliders in the tone curve and then realizing that you want to adjust exposure a bit takes just a fraction of a second and you don’t have to stop fiddling with the tone curve to do it.”

Knobroom is still alpha software so Heikkinen suggests not using it for production work since it might make Lightroom unstable and you’d risk losing your most recent edits. The good news is that you’d never lose the image file since Lightroom is non-destructive, meaning it never touches the original. Also, if the adjustment slider you want to use is hidden or off-screen, Knobroom can’t bring it into view, so you see the effects of the change but not the numbering setting that helped guide you in the past.

So far Knobroom has only been tested on a couple of controllers. The Novation Nocturn, shown above, is about $80.

Knobroom is not the first external controller to speed up Lightroom. The best-known is Kubota RPG SpeedKeys. It’s a wireless keypad (via a USB plug-in) with 16 keys; each key has three functions if you push an option (function) key first. There are four interchangeable keysets. SpeedKeys gives you access to more Lightroom adjustments than the typical Knobroom-MIDI controller combination but to make adjustments you may have to toggle among the Option 2 – Option 3 banks to adjust up and then adjust down. It’s $350 with a set of workflow tools.

Lightroom is now in version 3 with version 4 expected in 2012 as Adobe’s current release cycle appears to be lengthening from 12-18 months to around 24 months if it arrives mid-2012.

See a user manipulate a Lightroom photo using Knobroom and a MIDI controller below…

Source

3D gaming in 2011: Tips, tricks, and buying advice

So you haven’t gone the stereoscopic 3D route yet — but with top-notch, winter-release games like Skyrim and Batman: Arkham City making good use of 3D tech, you’re seriously considering it. The holiday season is upon us, too, and good, PC-centric 3D gear is now very affordable indeed. But where do you start? Is it like active- and passive-3D TVs, where there’s a huge variety in image quality? Is it better to stick with your current LCD panel and buy a separate emitter, or simply jack it all in and buy an all-in-one 3D monitor?

After extensive testing, with technology like Nvidia’s 3D Vision 2 and some of the top games of 2011, here is the best advice we can give you:

Spring for an all-in-one monitor

Given that you’re liable to need a new 120Hz monitor, get one with a built-in 3D emitter. It will save you time, aggravation, and desk space, and they’re not that much more expensive than monitors without it. (Granted, the 27-inch Asus monitor I used lists for $700, but you can get smaller models with all the features you need for about $300 less.)

Besides, any money you save will have to go to buying the glasses (Nvidia’s sell for $99 by themselves; if you need an emitter too, expect to pay $149).

Play the games you want to play

Computer games are a lot like movies in that what you get from one — or even from one studio — is not necessarily indicative of what you’ll get from another. Some titles will just look better in 3D than others. But the 3D is there to serve you, so don’t worry too much about tracking down the “best” games. Find the kind of games you like and give them a shot. If they don’t bowl you over, you can always just flip the switch.

Take breaks

I hate to get all goody-goody ergonomic on you, but don’t forget to give your eyes a break sometimes. Stereoscopic 3D is different than regular game play, and you may find it difficult to adjust to it. I’ve been living with this stuff for years, and after an hour and a half of playing Batman: Arkham City, my eyes were crying out for a spa treatment. When you’re being fed a lot of new information, take the time you need to process it.

Try before you buy (if you can)

If you know someone who’s got a stereoscopic 3D setup, see if you can snag a few minutes to see what you think of it first-hand. Try to play the kinds of games you like, see how your eyes react, and (perhaps most important) see if you think it’s worth the money.

Not every game or input device appeals to every person, so why should stereoscopic 3D? It is, after all, just another peripheral: Treat it that way, and get to know it before taking it home.

Walk away if you need to

Stereoscopic 3D is undoubtedly cool, but remember that it’s a feature, not a requirement. As we discovered with Battlefield 3, sometimes you’ll be better off putting aside the 3D and just enjoying the game in its “natural” state. As long as you like the game, who cares? There are plenty of other good 3D games out there.

Don’t expect miracles

Remember, we tested these games using one of the fastest video cards out there (that costs $700) on a system with a $1,000 processor and tons of memory. Your computer might not be that robust, so you might need a lower resolution or reduced details to get frame rates better than what you see in an old Rankin/Bass Christmas special. (Remember, with stereoscopic 3D your computer is crunching about twice as much graphics data as it usually does: 60 full-screen images for each eye per second, rather than 60 total per second.) The good news is that these games look pretty good even when they’re taking it easy, so don’t feel like you’re missing out, either.

With upwards of 500,000 pairs of 3D Vision glasses out there in the wild, clearly the technology is gaining in popularity. But everyone dealing in 3D games, from AMD and Nvidia to the studios, need to remember that the only thing that makes a good stereoscopic 3D game is a good game; the more of those there are, the better a deal stereoscopic 3D will be. With such unpredictable quality among some of the most popular titles this holiday season, clearly the industry still has a way to go. But I, for one, am astonished and impressed at how far stereoscopic 3D has already come. If it keeps on like it has, the future’s so bright, I gotta wear active-shutter glasses.

Source

The truth about hardware acceleration on Android

When the relatively open Android platform is compared to a tightly controlled system like iOS, one of the first differences people note is the superior smoothness of apps on iOS. This is something Android has struggled withsince its inception, but many of the assumptions made about Android’s graphical system, and how that relates to interface smoothness are wrong. Google engineer Dianne Hackborn explained the ins and outs of Android’s hardware acceleration in a recent Google+ post, and it might not be the magic bullet everyone was hoping for.

Contrary to what we’ve always heard, Android has had some hardware acceleration for drawing 2D UI elements since before 1.0. Many of the animations we see every day on Android have been hardware accelerated the whole time. For instance, menus popping up, dialog boxes, sliding down the notification bar, and transitioning between activities are being drawn using the GPU.

All “window compositing” is done using GPU-based hardware rendering. We can think of this as drawing any new elements on the screen. When the menu button is pressed, the resulting overlay is handled by the GPU. Should that overlay change, like if a button is highlighted or pressed, that change is rendered in software, and most phones are more than able to push those pixels. When the entire window is changed, that is a GPU task.

So what is changing in Ice Cream Sandwich (ICS)? According to Hackborn, what we will see in Android 4.0 is “full” hardware acceleration. All UI elements in windows, and third-party apps will have access to the GPU for rendering. Android 3.0 had the same system, but now developers will be able to specifically target Android 4.0 with hardware acceleration. Google is encouraging developers to update apps to be fully-compatible with this system by adding the hardware acceleration tag in an app’s manifest.

This is likely to take time as developers certify their apps on ICS, but Google has included a switch in Android’s settings to force hardware rendering. Hackborn warns that in untested apps, this has the potential to break things in subtle, or fundamental ways. This is just the first possible problem with hardware acceleration on Android.

By rendering all of an app’s animations and UI with the GPU, the system takes a hit to memory usage. Loading up the OpenGL drivers for each process takes memory usage of roughly 2MB, and boosts it to 8MB. On devices with limited RAM, this can be a real issue. When more RAM is eaten up, the system will necessarily have to close more background tasks to save memory. Some developers might not want to use the GPU for drawing as a result.

The goal of hardware rendering for the interface is to get smoother operation, but if it isn’t handled appropriately, Android devices can actually perform worse. Using the example of Tegra 2, which can render all the pixels on a 1280×800 screen 2.5 times per second at 60FPS, the problem is clear. The GPU touches each pixel once for the background, once for icons and widgets, once more for labels, and then there are animations to deal with. There is no way to reach 60FPS doing things in such a way.

Android makes hardware acceleration work with a number of tricks, and developers should take note. First, separate windows in the interface are copied into unified overlays for more efficient GPU access. Android also renders the background as one large bitmap that does not need to be re-rendered as the user scrolls around.

Developers will need to be cautious in the way hardware acceleration is handled. Simply drawing everything with the GPU could result in poor performance on existing phones. For the future, the higher the resolution on a device, the more related GPU speed is going to be to overall smoothness and proximity to that 60FPS threshold. The limiting factor for a device’s speed will often be the GPU memory bus bandwidth going forward.

Source

Living Earth Simulator will simulate the entire world

Described as a “knowledge collider,” and now with a pledge of one billion euros from the European Union, the Living Earth Simulator is a new big data and supercomputing project that will attempt to uncover the underlying sociological and psychological laws that underpin human civilization. In the same way that CERN’sLarge Hadron Collider smashes together protons to see what happens, the Living Earth Simulator (LES) will gather knowledge from a Planetary Nervous System (PNS — yes, really) to try to predict societal fluctuations such as political unrest, economic bubbles, disease epidemics, and so on.

Orchestrated by FuturICT, which is basically aconsortium of preeminent scientists, computer science centers around the world, and high-power computing (HPC) installations, the Living Earth Simulator hopes to correlate huge amounts of data — including real-time sources such as Twitter and web news — and extant, but separate approaches currently being used by other institutions, into a big melting pot of information. To put it into scientific terms, the LES will analyze techno-socio-economic-environmental (!) systems. From this, FuturICT hopes to reveal the tacit agreements and hidden laws that actually govern society, rather than the explicit, far-removed-from-reality bills and acts that lawmakers inexorably enact.

The scale of the LES, when it’s complete, will be huge. It is hoped that supercomputing centers all over the world will chip in with CPU time, and data will be corralled from existing projects and a new Global Participatory Platform, which is basically open data on a worldwide scale. The project also has commercial backing from Microsoft Research, IBM, Yahoo Research, and others. All told, the system will create useful knowledge in the fields of energy, networks and communication, economics, crime and corruption, migration, health, and crisis management.

The timing of EU’s billion-euro grant is telling, too. As you probably know, the European Union is struggling to keep the plates spinning, and the LES, rather handily, will probably be the most accurate predictor of economic stability in the world. Beyond money, though, it is hoped that the LES and PNS can finally tell us why humans do things, like watch a specific TV show, buy a useless gadget, or start a revolution.

Looking at the larger picture, the Living Earth Simulator is really an admission that we know more about the physical universe than the social. We can predict with startling accuracy whether an asteroid will hit Earth, but we know scant little about how society might actually react to an extinction-level event. We plough billions of dollars into studying the effects and extent of climate change, but what if understood enough of the psychology and sociology behind human nature to actually change our behavior?

Read more at FuturICT

Source

PCI Express 4.0 announced even though PCIe 3.0 has not matured yet

With PCI-Express 3.0 still in diapers so to speak, its parent PCI-SIG has announced the next version in its superfast I/O standard, PCI-Express (PCIe) 4.0. With widespread support of the current standard most likely not coming until next year, 4.0′s announcement may seem premature. However, with the rapid, breakneck pace of the computing industry, having PCI-SIG “set the pace” to where the theoretical speeds are going to be in 2014 or 2015 will be useful for manufacturers.

It should be an exciting thing for consumers, too, since PCIe 4.0 is touting 16GT/s (16 billion transfers) per lane. That comes to around 2GB/s — twice as fast as PCIe 3 — and and in a full 16-lane configuration, we’re looking at a max theoretical bandwidth of 32GB/s. Hard drives (or solid state drives) aren’t likely to require that much bandwidth for the next decade or so, but games like Crysis 3 (or 4) might be able to utilize it.

Further good news includes the fact that it will be backwards compatible with both 3.0 and 2.0, and have no changes to the physical layout of the slots on motherboards. That will certainly help users as they will not have to upgrade right away, because while the actual technology won’t cost more to implement, add-in cards that can take advantage of the new standard certainly will at first.

Another interesting avenue to consider is the mobile market. PCI-SIG has been reported to be working on plans to bring the standard to tablets and smartphones, allowing faster speeds on handheld devices as well. This would certainly be a smart move with the recent growth of the mobile computing sector. The important thing to realize overall about 4.0 is that it isn’t the technology of tomorrow, but rather a standard for years down the line. There is nothing to lose by letting the public know where the next plateau will be.

Read more at PCI-SIG

Source