An anonymous reader writes from a report via PCWorld: Google says its Tensor Processing Unit (TPU) advances machine learning capability by a factor of three generations. "TPUs deliver an order of magnitude higher performance per watt than all commercially available GPUs and FPGA," said Google CEO Sundar Pichai during the company's I/O developer conference on Wednesday. The chips powered the AlphaGo computer that beat Lee Sedol, world champion of the game called Go. "We've been running TPUs inside our data centers for more than a year, and have found them to deliver an order of magnitude better-optimized performance per watt for machine learning. This is roughly equivalent to fast-forwarding technology about seven years into the future (three generations of Moore's Law)," said Google's blog post. "TPU is tailored to machine learning applications, allowing the chip to be more tolerant of reduced computational precision, which means it requires fewer transistors per operation. Because of this, we can squeeze more operations per second into the silicon, use more sophisticated and powerful machine learning models, and apply these models more quickly, so users get more intelligent results more rapidly." The chip is called the Tensor Processing Unit because it underpins TensorFlow, the software engine that powers its deep learning services under an open-source license.

An anonymous user writes: Back in March, Microsoft's Phil Spencer addressed some of the concerns over the Unified Windows Platform and PC gaming. He noted that Microsoft would "plan to open up VSync off, FreeSync, and G-Sync in May" and the company would "allow modding and overlays in UWP applications" sometime further into the future. Well, it appears that Microsoft is on point with the May UWP update. According to the MS DirectX Developer Blog, a Windows 10 update being pushed out today will enable UWP to support unlocked frame rates and variable refresh rate monitors in both G-Sync and FreeSync varieties. Today's update won't automatically enable these features in UWP games like Gears of War or Quantum Break, they will still need to be updated individually by the developer. Microsoft states that Gears of War and Forza will be the first to see these changes, but there is no mention of Quantum Break here, which is a game that could DEFINITELY benefit from the love of variable refresh rate monitors.PCWorld has more details.

MojoKid shares a video showing the upcoming Doom game on NVIDIA's new GeForce GTX 1080 graphics card using the Vulkan API, quoting this report from HotHardware: At a private briefing with NVIDIA, representatives from id software came out on stage to show off the upcoming game...the first public demonstration of the game using both NVIDIA's new flagship and the next-gen API, which is a low-overhead, cross-platform graphics and compute API akin to DirectX 12 and AMD's Mantle. In the initial part of the demo, the game is running smoothly, but its frame rate is capped at 60 frames per second. A few minutes in, however, at about the :53 second mark...the rep from id says, "We're going to uncap the framerate and see what Vulkan and Pascal can do".

With the framerate cap removed, the framerate jumps into triple digit territory and bounces between 120 and 170 frames per second, give or take. Note that the game was running on a projector at a resolution of 1080p with all in-game image quality options set to their maximum values. The game is very reminiscent of previous Doom titles and the action is non-stop.

MojoKid writes (edited and condensed): NVIDIA has unveiled its next-generation Pascal-based GeForce graphics cards -- known as the GeForce GTX 1080 and GeForce GTX 1070. NVIDIA's Pascal architecture is based on 16nm FinFET technology, similar to that of NVIDIA's high-end data center Tesla P100 processing engine though the GeForce cards are targeted at the consumer gaming market. NVIDIA's GP104 GPU at the heart of the new GeForce cards is comprised of some 8 billion transistors and features a 256-bit memory interface with 8GB of Micron GDDR5X graphics memory on the GeForce GTX 1080. The GTX 1070, however, employs standard GDDR5. The core clock speed of the GeForce GTX 1080 hit 2.1GHz at one point during the demonstration, though GTX 1070 clocks were not disclosed. NVIDIA CEO Jen-Hsun Huang claimed the new GeForce GTX 1080 is faster than a pair of GeForce GTX 980 cards in SLI and faster than the company's very expensive Titan X graphics card but at half the price. The new GeForce GTX 1080 will be offered in two versions, a standard card with an MSRP of $599 or a highly-overclockable Founders Edition for $699. The standard GTX 1070 will arrive at $379, while a Founders Edition will be priced at $449. Availability for the GTX 1080 is slated for May 27th and the GTX 1070 for June 10. Anand Tech has more information.

tedlistens writes: At Motherboard, Alex Pasternack writes: "Star Wars is set in a world of wildly advanced technology. But take a good look at the machinery of Star Wars, and you may be surprised to see how wonderfully analog it all is -- buttons! levers! vector graphics! Yes, there are hyperdrives and lightsabers and hologram Princess Leias and droids that know six million languages (including the language of moisture vaporators, along with various etiquette and diplomatic protocols useful across the galaxy). But it's also a world where sometimes you have to hit a robot to get it to work, like an old dashboard radio, a place where the supercomputers are operated manually and where buttons and control panels and screens seem far removed from our own galaxy: tactile, lo-fi, and elegantly simple." May the 4th be with you.

Reader itwbennett writes: In September 2014, Nvidia sued both Samsung and Qualcomm, asking for shipments of Galaxy phones and tablets with graphics processors from Qualcomm, ARM and Imagination Technologies to be blocked. Samsung fired back, suing Nvidia for patent infringement, and for supposedly false claims by Nvidia that its Tegra K1 was the world's fastest mobile processor. Now, in an agreement announced today, Nvidia and Samsung have agreed to 'settle all outstanding IP litigation'. Under the agreement, the companies will license 'a small number' of patents to each other, though they said there's no broad cross-license agreement. There's also no financial payment. And no mention was made of Qualcomm.

An anonymous reader quotes a report from Network World discussing Nvidia's project called DAVE2, where their engineering team built a self-driving car with one camera, one Drive-PX embedded computer and only 72 hours of training data: Neural networks and image recognition applications such as self-driving cars have exploded recently for two reasons. First, Graphical Processing Units (GPU) used to render graphics in mobile phones became powerful and inexpensive. GPUs densely packed onto board-level supercomputers are very good at solving massively parallel neural network problems and are inexpensive enough for every AI researcher and software developer to buy. Second, large, labeled image datasets have become available to train massively parallel neural networks implemented on GPUs to see and perceive the world of objects captured by cameras. The Nvidia team trained a convolutional neural network (CNN) to map raw pixels from a single front-facing camera directly to steering commands. Nvidia's breakthrough is the autonomous vehicle automatically taught itself by watching how a human drove, the internal representations of the processing steps of seeing the road ahead and steering the autonomous vehicle without explicitly training it to detect features such as roads and lanes.

Will Freeman, reporting for The Guardian (condensed): One industry veteran sees arcades and mobile gaming as almost indistinct. He is Nolan Bushnell, co-founder of Atari. Often referred to as the godfather of video games (a phrase he dislikes), he is just about to make his debut in mobile game development, having established a partnership with Dutch publisher Spil, where he will help deliver at least three as-yet-unnamed titles. "When you look at mobile and arcade gaming, they're identical," Bushnell says. "Mobile has some of the same game constraints for the player, and that 'easy to learn, and difficult to master' metric." [...] "Generally, a tremendous number of mobile games are poorly designed," he says. "They can be so focused on graphics that they forget they have to get the timing right, and they have to have proper scoring constructs. I have been so pissed off with some mobile games I've wanted to throw my phone, even if I'm only going to hurt my phone there, and not the game."

MojoKid quotes a report from HotHardware: It looks as though Lian-Li took the term desktop computer a bit too literally with its new DK-04 system, which is literally a desk that doubles as an enclosure for your PC components. The DK-04 has a clean, industrial design and even has an electrically actuated height adjustment system that than can raise the desktop from a height of 30.7 inches to 46 inches at the touch of a button. The desk itself is made out of aluminum, while the desk surface is made of tempered glass so that you can get work done on top of it and admire your PC components at the same time. Speaking of which, the DK-04 supports up to 10 HDD/SDD bays and both micro-ATX and ATX motherboards. And if you're thinking about throwing in some monster graphics cards into the mix, the DK-04 will support cards up to 320mm in length. Cool air is pulled in via four front-mounted 120mm fans and hot air is exhausted via another four 120mm fans that are mounted at the back of the desk. Liquid cooling is also supported.

An anonymous reader quotes a report from Networkworld: Magic Leap has made progress. A year ago, the MR apparatus wasn't really wearable. This may have been due to the sheer size of the prototype hardware, or the software that adjusts the illusion for head movement and changes in perspective may not have been completed. The complexity of the mathematics of this head-tracking applied to the complexity of the mathematics of Magic Leap's proprietary light-field chip may have required more development time. The Wired report of Kelly's experience indicates the prototype headset he tested is now wearable, perhaps untethered from a high-powered graphics-rendering computer used by the Oculus Rift and HTC Vive. Kelly's report raises the question: Could Magic Leap have implemented hand-tracking, too? He writes, "When I raise a hand, it approaches and extends a glowing appendage to touch my fingertip."

Apple, on Tuesday, announced a refresh for its 12-inch MacBook laptop. The 2016 MacBook comes with an Intel Skylake processor -- sixth-generation dual-core Intel Core M model, offering up to 1.3 GHz clock speed with Turbo Boost speeds of up to 3.1 GHz, faster 1866 MHz memory, and a 'rose gold' color variant. Apple assures 10 hours of wireless Web browsing time, or 11 hours of movie playback on a single charge. The new model will hit retail stores on Wednesday. It starts at $1,299 for the 256GB SSD and 8GB (up from 4GB) version, and goes all the way up to $1,599 for the top-of-the-line model which offers 512GB SSD.

A couple of points: the first-generation MacBook didn't fare well with reviewers and plenty of users alike. Second, today's announcement also hints that the MacBook Air and the MacBook Pro lineups won't be getting the Intel Skylake upgrade for at least a few more months -- which is really sad, because, at present, they come equipped with almost three-year-old processor and graphics chips. No wonder, Oculus executive made fun of Apple's computers.

An anonymous reader writes: Google on Wednesday released the second developer preview of Android N. The update, which comes a month after the release of first Android N developer preview, brings with it a number of features and improvements. On a blog post, Google wrote that it is adding Vulkan, a low-overhead graphics API to the package. This would supposedly offload some CPU-bound processes to GPU. Also in the build are new "human-looking" emojis. Improvements can be found here.

An anonymous reader cites a report on VentureBeat: NVIDIA chief executive Jen-Hsun Huang announced that the company has created a new chip, the Tesla P100, with 15 billion transistors, 16GB high-bandwidth memory for deep-learning computing. It's the biggest chip ever made, Huang said. "We decided to go all-in on A.I.," Huang said. "This is the largest FinFET chip that has ever been done." The chip has 15 billion transistors, or three times as much as many processors or graphics chips on the market. It takes up 600 square millimeters. The chip can run at 21.2 teraflops. Huang said that several thousand engineers worked on it for years. Jim McGregor, writing for Forbes (the link is not accessible to ad-blocking tool users): It features NVIDIA's new Pascal GPU architecture, the latest memory and semiconductor process, and packaging technology -- all to create the densest compute platform to date. In addition, it combines 16GB of die stacked second-generation High-Bandwidth Memory (HBM2). The memory and GPU are combined into a multichip module on a state-of-the-art silicon substrate. The P100 has NVIDIA's NVLink interface technology to connect to multiple Tesla P100 GPU modules.

An anonymous reader quotes a report on Polygon about Sony's alleged PlayStation 4K, which Kotaku reported about earlier: Sony's more powerful version of the PlayStation 4 -- also known as the "PlayStation 4.5" and "PlayStation 4K" -- is expected to be announced prior to the launch of PlayStation VR this October, according to a new report from the Wall Street Journal (paywall link). The upgraded PS4 will reportedly be able to play 4K resolution content, according to the report, and deliver "a richer gaming environment, including a high-end virtual-reality experience."

An anonymous reader writes: The Linux 4.6-rc1 kernel has been released. New to the Linux 4.6 kernel are a significant number of new features including NVIDIA GeForce GTX 900 open-source 3D support when using the closed-source firmware files, Dell XPS 13 Skylake laptop support, a fix for laptops that were limiting their own performance due to incorrectly thinking they were overheating, AHCI runtime power management support, Intel graphics power management features enabled by default, a new file-system (OrangeFS), and a range of other improvements.

An anonymous reader cites an article on The Next Web: Microsoft research manager Shahram Izadi is showing off the company's latest innovation using HoloLens: 'holoportation,' enabling him to appear as if he's there in real-time, anywhere in the world. His image is captured in 3D by cameras placed around the room. This is then stitched together, compressed and transmitted so someone else can see, hear and interact with him as though he's right there with them. You can even playback previous interactions, as though "walking into a living memory," and miniaturize the content to make it easier to consume. "Imagine being able to virtually teleport from one place to another," he says. Well, if you're the owner of a HoloLens, you soon could do. Microsoft's HoloLens is arguably the front-runner in the nascent, but fast-evolving, augmented reality space. The company's technology has previously been seen used by astronauts and scientists to "walk on the Mars surface" without stepping out of their office on Earth. It's fascinating to see how Microsoft continues to further innovate in this field.

An anonymous reader quotes a report from PCWorld: Intel could dump Nvidia for a licensing deal with AMD as the chip giant tries to prop up its patent portfolio. Currently, Intel is under a $1.5 billion licensing agreement with Nvidia, which the two companies signed in 2011. At the time, the two companies had spent years fighting each other in courts over patent licensing, and the agreement put all that litigation to rest. Intel's Nvidia deal is set to expire on March 17, 2017, and a recent report by Bloomberg claimed that Intel is now looking to cut a deal with AMD instead.

An anonymous reader writes: For the first time ever, AMD has provided open-source support for next-generation discrete GPUs ahead of the product's launch. AMD developers published initial open-source Linux driver support for Polaris GPUs with the addition adding over sixty-seven thousand lines of code to the Linux kernel. AMD Polaris graphics cards are expected this summer while AMD released the open-source driver support in advance for preparing their new Linux hybrid driver that relies upon the open-source AMDGPU kernel driver.

Reader MojoKid writes: Some might say there's no such thing as a one-size-fits-all solution when it comes to computing, and that's especially true for workstation graphics professionals who need varying levels of performance and memory space. For that reason, NVIDIA is now offering a version of its Quadro M6000 graphics card with 24GB of GDDR5 memory, twice as much memory as much as the original model. According to NVIDIA, customers rendering datasets larger than 12GB can experience up to 5X faster performance compared to the previous Quadro M6000. Like the 12GB version, the new 24GB Quadro M6000 is based on NVIDIA's Maxwell architecture. It has 3,072 CUDA cores, a 384-bit memory bus, four DisplayPort 1.2 connectors, a single DVI-I connectors, and a maximum power consumption rating of 250W. In addition to the doubling the memory buffer, NVIDIA added a few other features, including more GPU clock options, greater software temperature control to keep the GPU temp below the point where throttling occurs, and a new under-power boot message if the card is ever under powered.

HughPickens.com writes: Bryant Jordan reports at Defense Tech that the Cognitive Science and Applications Team at the US Army's Natick Soldier Research, Development and Engineering Center is creating a computer-generated reality "dome" to immerse warfighters in a virtual environment that not only tests their skills, but allows Army researchers to assess soldier cognitive abilities and study the impact of real-world operational situations on decision-making, spatial memory and wayfinding. The dome is a concave virtual-reality system that provides a full 180-degree horizontal field, using high-density, front-projection to create a high-resolution, visual world where the simulations will be modeled on real-world locations. "The integration of multiple input modalities, along with multisensory feedback, increases the realism, immersion and engagement on behalf of users subjected to prolonged, workload-intensive activities," says Dr. Caroline Mahoney. "These novel integrations provide unprecedented opportunities to monitor and optimize human behavior during real-world task execution, and to evaluate and predict the impact of innovative human-systems technologies on operational performance." In the virtual dome, users can interact and alter the environment through hand-held and weapon-based devices, which control movement, orientation and weapon aiming. Future additions to the dome will include whole-body motion tracking, low-frequency vibration and directional wind. Vibro-tactile collision feedback — which combines vibration and touch to help give participants a physical sense of constraints in a virtual environment — will also be included.