Yesterday, during his quarterly earnings call, Tesla CEO Elon Musk revealed a new piece of hardware that the company is working on to perform all the calculations required to advance the self-driving capabilities of its vehicles. The specialized chip, known as "Hardware 3," will be "swapped into the Model S, X, and 3," reports TechCrunch. From the report: Tesla has thus far relied on Nvidia's Drive platform. So why switch now? By building things in-house, Tesla say it's able to focus on its own needs for the sake of efficiency. "We had the benefit [...] of knowing what our neural networks look like, and what they'll look like in the future," said Pete Bannon, director of the Hardware 3 project. Bannon also noted that the hardware upgrade should start rolling out next year. "The key," adds Elon "is to be able to run the neural network at a fundamental, bare metal level. You have to do these calculations in the circuit itself, not in some sort of emulation mode, which is how a GPU or CPU would operate. You want to do a massive amount of [calculations] with the memory right there." The final outcome, according to Elon, is pretty dramatic: He says that whereas Tesla's computer vision software running on Nvidia's hardware was handling about 200 frames per second, its specialized chip is able to crunch out 2,000 frames per second "with full redundancy and failover." Plus, as AI analyst James Wang points out, it gives Tesla more control over its own future.

Today, Samsung unveiled the successor to the Galaxy Tab S3 from last year. The aptly named Galaxy Tab S4 features a 10.5-inch Super AMOLED display with a 2560 x 1600 resolution, Qualcomm Snapdragon 835 processor with 4GB RAM, 64GB internal storage (expandable via microSD card) and 13-megapixel f1.9 rear-facing camera. Unlike the Tab S3, it includes Samsung Dex software that lets users connect a Samsung mobile device to a monitor and then use the device as a pseudo-desktop. Ars Technica reports: The first Dex dock came out over a year ago and was designed to be used with Samsung smartphones. Users could plug their device into the dock, connect it to a monitor, pair a keyboard and a mouse, and use the setup as they would a full desktop PC. The system ran a version of Android that Samsung modified to better suit a desktop UI, which included a lock screen and a task bar area with app icons. Dex on the Galaxy Tab S4 works just like this, with a couple of extra features that leverage the power of a tablet. When connected to a monitor, both the big screen and the tablet's screen can be used simultaneously. In a short demo, Samsung showed how the device supports up to 20 open windows at once and how features like split screen and drag-and-drop can be used just as they would on a desktop PC. Users can launch Dex when not connected to a monitor as well, and that produces the same modified Android UI on the tablet's 10.5-inch, 2560 x 1600 Super AMOLED display. When connected to a monitor, both the big screen and the tablet's screen can be used simultaneously. In a short demo, Samsung showed how the device supports up to 20 open windows at once and how features like split screen and drag-and-drop can be used just as they would on a desktop PC. Users can launch Dex when not connected to a monitor as well, and that produces the same modified Android UI on the tablet's 10.5-inch, 2560x1600 Super AMOLED display. The tablet carries a $649 price, but includes all the specs mentioned above, as well as support for signature Samsung features like Air Command, translate, and off-screen memos, and a redesigned S Pen.

An anonymous reader quotes a report from VentureBeat: For the first time in seven years, Samsung and Apple have not taken the top two positions in the worldwide smartphone market. All hail Huawei. Although Samsung held onto first place -- the South Korean giant typically dominates the first three quarters of the year, with the American company winning the fourth quarter -- Huawei passed Apple for second place this part quarter. The Q2 2018 figures come from IDC (though Canalys and Strategy Analytics both agree that Huawei passed Apple in the quarter).

MojoKid writes from a report via Hot Hardware: Although it has been over three decades since the first Commodore Amigas were originally released, a fan base for the beloved systems is still going strong. In fact, today's Amiga community seems to be more active now that it has been in years, and a number of exciting new hardware projects have cropped up in recent weeks. Two relatively new projects, led by popular members of the Amiga community Paul Rezendes and John "Chucky" Hertell, are designed to breathe new life into the Amiga 4000 and Amiga 1200.

Both men set out to reverse engineer the motherboards for these systems, not only to continue the possibility of repairing existing machines that are prone to serious damage from leaky batteries and electrolytic capacitors, but to potentially spur additional customizations for the platform in the future. Though Paul and John have only made minor modifications to the Amiga 4000 and Amiga 1200 motherboard PCBs to this point, the possibility now also exists for all new variants to arrive at some point in the future for these machines as well. The first actual working motherboards populated with components based on the Amiga 4000 Replica project or Re-Amiga 1200 haven't been shown off just yet, and they may require additional revisions to work out any kinks. However, both projects are good examples of the passion that still remains for the beloved Amiga from computing glory days gone by.

Logitech announced late Monday night that it is acquiring Blue Microphones for $117 million. The company known for USB condenser microphones, such as the Snowball and Yeti, will join Logitech's existing portfolio of brands. The Verge reports: Founded in 1995, Blue sells microphones that range in price from $60 to over $4,000 (for studio-grade hardware), and they're used by podcasters, musicians, and any other consumers who need higher-fidelity audio than what they get from the built-in microphones on their devices. Now, after dropping a heap of cash on the company, Logitech will do its best to make sure Blue's devices become just as essential as its own wireless keyboards and mice. "For Logitech, this is a new space," the company wrote in a blog post. "But, at the same time, it's not at all. Gamers are already using our Logitech G webcams to stream. People are video calling with friends and family thanks to Logitech every day. And in business, our audio and video know-how is apparent every time a video meeting takes place at the office. Joining up with Blue and their microphone lines is a logical adjacent opportunity with great synergies."

An anonymous reader quotes a report from VentureBeat: In a forthcoming paper ("Dexterous In-Hand Manipulation"), OpenAI researchers describe a system that uses a reinforcement model, where the AI [known as Dactyl] learns through trial and error, to direct robot hands in grasping and manipulating objects with state-of-the-art precision. All the more impressive, it was trained entirely digitally, in a computer simulation, and wasn't provided any human demonstrations by which to learn. The researchers used the MuJoCo physics engine to simulate a physical environment in which a real robot might operate, and Unity to render images for training a computer vision model to recognize poses. But this approach had its limitations, the team writes -- the simulation was merely a "rough approximation" of the physical setup, which made it "unlikely" to produce systems that would translate well to the real world. Their solution was to randomize aspects of the environment, like its physics (friction, gravity, joint limits, object dimensions, and more) and visual appearance (lighting conditions, hand and object poses, materials, and textures). This both reduced the likelihood of overfitting -- a phenomenon that occurs when a neural network learns noise in training data, negatively affecting its performance -- and increased the chances of producing an algorithm that would successfully choose actions based on real-world fingertip positions and object poses.

Next, the researchers trained the model -- a recurrent neural network -- with 384 machines, each with 16 CPU cores, allowing them to generate roughly two years of simulated experience per hour. After optimizing it on an eight-GPU PC, they moved onto the next step: training a convolutional neural network that would predict the position and orientation of objects in the robot's "hand" from three simulated camera images. Once the models were trained, it was onto validation tests. The researchers used a Shadow Dexterous Hand, a robotic hand with five fingers with a total of 24 degrees of freedom, mounted on an aluminum frame to manipulate objects. Two sets of cameras, meanwhile -- motion capture cameras as well as RGB cameras -- served as the system's eyes, allowing it to track the objects' rotation and orientation. In the first of two tests, the algorithms were tasked with reorienting a block labeled with letters of the alphabet. The team chose a random goal, and each time the AI achieved it, they selected a new one until the robot (1) dropped the block, (2) spent more than a minute manipulating the block, or (3) reached 50 successful rotations. In the second test, the block was swapped with an octagonal prism. The result? The models not only exhibited "unprecedented" performance, but naturally discovered types of grasps observed in humans, such as tripod (a grip that uses the thumb, index finger, and middle finger), prismatic (a grip in which the thumb and finger oppose each other), and tip pinch grip. They also learned how to pivot and slide the robot hand's fingers, and how to use gravitational, translational, and torsional forces to slot the object into the desired position.

Uber is shuttering its self-driving unit, reports TechCrunch. The company will reportedly stop development of self-driving trucks and instead focus its efforts on self-driving cars. "We recently took the important step of returning to public roads in Pittsburgh, and as we look to continue that momentum, we believe having our entire team's energy and expertise focused on this effort is the best path forward," Eric Meyhofer, head of Uber Advanced Technologies Group, said in an emailed statement. From the report: Uber Freight, a business unit that helps truck drivers connect with shipping companies, is unaffected by this decision. "Rather than having two groups working side by side, focused on different vehicle platforms, I want us instead collaborating as one team, according to an email reviewed by TechCrunch that was sent by Meyhofer to employees. "I know we're all super proud of what the Trucks team has accomplished, and we continue to see the incredible promise of self-driving technology applied to moving freight across the country. But we believe delivering on self-driving for passenger applications first, and then bringing it to freight applications down the line, is the best path forward. For now, we need the focus of one team, with one clear objective." The company will pivot employees focused on self-driving trucks to other work that revolves around self-driving technology.

According to supply chain blog Macotakara, the new iPad Pro models that will be introduced later this year will be slimmer, feature Face ID, and have no headphone jacks. 9to5Mac reports the details: First off, the report offers additional details on the 2018 iPad Pro dimensions. The 10.5-inch model is said to come in at 247.5mm (H) x 178.7mm (W) x 6mm (T), compared to the current dimensions of 250.6mm x 174.1mm x 6.1 mm. Meanwhile, the 12.9-inch iPad Pro is said to stack up at 280mm (H) x 215mm (W) x 6.4mm (T), which compares to the current-generation model at 305.7 x 220.6 x 6.9 mm. With these dimensions, it seems that Apple is focused more on reducing the overall footprint of the 12.9-inch model, fitting the same size display into a considerably smaller body. The report goes on to explain that Apple is likely to ditch the headphone jack with this year's iPad Pro models, a move the company first made with the iPhone 7. While Apple includes a Lightning to 3.5mm headphone adapter to ease the blow for iPhone users, it will not do the same for iPad Pro users, according to today's report.

Today's report corroborates that this year's iPad Pro models will feature Face ID, but it notes that there is no support for landscape Face ID as earlier reports had indicated. This presents an interesting problem for the iPad Pro, which is used commonly in landscape mode with accessories such as the Smart Keyboard. Macotakara notes, however, that Apple is moving the Smart Connector on this year's models to "the lower rear side -- close to the Lightning connector." What exactly this means is unclear, but the report explains that "the next iPad Pro Smart Keyboard may be changed to vertical position specifications." This is seemingly implying that the iPad Pro would dock vertically into the Smart Keyboard, but how that would work is vague at the moment.

TechCrunch's Lucas Matney describes the Lumin operating system that will power Magic Leap's upcoming Magic Leap One mixed reality headset: Alright, first, this is what the Magic Leap One home screen will apparently look like, it's worth noting that it appears that Magic Leap will have some of its own stock apps on the device, which was completely expected but they haven't discussed much about. Also worth noting is that Magic Leap's operating system by and large looks like most other operating systems, they seem to be well aware that flat interfaces are way easier to navigate so you're not going to be engaging with 3D assets just for the sake of doing so. The company seems to be distinguishing between two basic app types for developers: immersive apps and landscape apps. Landscape apps like what you see in the image above, appear to be Magic Leap's version of 2D where interfaces are mostly flat but have some depth and live inside a box called a prism that fits spatially into your environment. It seems that you'll be able to have several of these running simultaneously. Immersive apps, on the other hand, like the game title, Dr. Grordbort -- which Magic Leap has been teasing for years -- respond to the geometry of the space that you are in and is thus called an immersive app.

Moving beyond apps, the company also had a good deal to share about how you interact with what's happening in the headset. Magic Leap will have a companion smartphone app that you can type into, you can connect a bluetooth keyboard and there will also be an onscreen keyboard with dictation capabilities. One of the big highlights of Magic Leap tech is that you'll be able to share perspectives of these apps in a multi-player experience which we now know is called "casting," apps that utilize these feature will just have a button that you can press to share an experience with a contact.

At the company's second quarter 2018 financial results conference call, Intel chief engineering officer Venkata Renduchintala said the "Cannon Lake" 10mn processors won't appear in products until the 2019 holiday season. "The systems on shelves that we expect in holiday 2019 will be client systems, with data center products to follow shortly after," Renduchintala said. Interim CEO Robert Swan went on to tout the company's "very good lineup" of 14nm products. Digital Trends reports: "Recall that 10nm strives for a very aggressive density improvement target beyond 14nm, almost 2.7x scaling," Renduchintala said during the call. "And really, the challenges that we're facing on 10nm is delivering on all the revolutionary modules that ultimately deliver on that program." Although he acknowledged that pushing back 10nm presents a "risk and a degree of delay" in the company's road map, Intel is quite pleased with the "resiliency" of its 14nm roadmap. He said the company delivered an excess of 70 percent performance improvement over the last few years. Meanwhile, Intel's 10nm process should be in an ideal state to mass produce chips towards the end of 2019.

Intel's Cannon Lake chip is essentially a shrink of its seventh-generation "Kaby Lake" processor design. Given the previous launch window, the resulting chips presumably fell under the company's eighth-generation banner despite the older design. But with mass production pushed back to late 2019, the 10nm chips will fall under Intel's ninth-generation umbrella along with CPUs based on its upcoming "Ice Lake" design. Intel claims that its 10nm chips will provide 25 percent increased performance over their 14nm counterparts. Even more, they will supposedly consume 50 percent less power than their 14nm counterparts.

The Google Play Store is no longer welcoming apps that mine cryptocurrencies on users' devices. The new policy plainly states: "We don't allow apps that mine cryptocurrency on devices." As a result, Google will start to remove any app from the Play Store that violates these terms. It will however still allow mining apps that are used to control cryptocurrency mining operations on remote devices -- such as servers or desktops. Bleeping Computer reports: Preventing cryptocurrency mining on Android is crucial because untethered mining operations can easily overheat or destroy batteries -- or even devices if the battery case deforms, leaks, or catches fire. Such scenarios have already happened. Android malware strains like Loapi and HiddenMalware have been known to cause physical damage to infected smartphones. Google didn't publicly announce the policy change, but developers who had apps removed complained on Reddit. Other policy changes were also made, such as the decision to ban apps with repetitive content. You can read the new "Restricted Content" section of the Play Store Policy Center here, and compare it to an earlier version here.

An anonymous reader quotes a report from Motherboard: On July 25, the Massachusetts Senate approved a Resolution that would create a special commission that would research the feasibility of forcing device manufacturers to treat customers and independent repair shops the same as officially licensed repair outlets. According to the proposed study, that means providing customers and independent repair shops with "repair technical updates, diagnostic software, service access passwords, updates and corrections to firmware, and related documentation." Gay Gordon Byrne, executive director of The Repair Organization, helped push the bill in 2012 and has been working to extend the law to tech companies ever since. "This is just one step in a series of steps that will end Repair Monopolies for technology products. I'm thrilled," Byrne told me in an email about the pending study.

The Resolution to create the study group still needs to pass the Massachusetts House, but the session ends July 31 so right-to-repair watch dogs won't have to wait long to see if it goes forward. The proposed makeup of the study commission shows that the legislature is serious about the issue and also reveals how big tech's repair monopoly is about much more than just being able to open up your iPhone without voiding the warranty. The legislature wants the study commission to include 23 members, including various members of the legislature but also a wealth of experts in various tech fields. They want someone from the Massachusetts Farm Bureau, a medical device manufacturer, an expert on electronic waste recycling, someone who repairs complex medical equipment, an intellectual property lawyer, a cyber security expert, a local farmer, and various other experts and citizens affected or knowledgeable about the right-to-repair.

According to leaked benchmarks found in the SiSoft Sandra database, there is an Intel Core i7-9700K processor that doesn't appear to have hyperthreading available. "This increases the core count from the current six cores in the 8th generation Coffee Lake parts to eight cores, but, even though it's an i7 chip, it doesn't appear to have hyperthreading available," reports Ars Technica. "It's base clock speed is 3.6GHz, peak turbo is 4.9GHz, and it has 12MB cache. The price is expected to be around the same $350 level as the current top-end i7s." From the report: For the chip that will sit above the i7-9700K in the product lineup, Intel is extending the use of its i9 branding, initially reserved for the X-series High-End Desktop Platform. The i9-9900K will be an eight-core, 16-thread processor. This bumps the cache up to 16MB and the peak turbo up to 5GHz -- and the price up to an expected $450. Below the i7s will be i5s with six cores and six threads and below them, i3s with four cores and four threads. Even without hyperthreading, the new i7s should be faster than old i7s. A part with eight cores is going to be faster than the four-core/eight-thread chips of a couple of generations ago and should in general also be faster than the six-core/12-thread 8th generation chips. Peak clock speeds are pushed slightly higher than they were for the 8th generation chips, too.

An anonymous reader quotes a report from The Verge: Some newer Huawei phones are actively being blocked from installing the open-source VLC media player app from Google Play. VLC's developers announced today that they're blacklisting some of Huawei's devices after unhappy users left too many one-star reviews for the app. But the negative reviews stem from a decision on Huawei's part and has nothing to do with VLC. The negative reviews are a result of Huawei's aggressive battery management and tendency to kill background apps, which directly affects VLC's background audio playback feature. Huawei users on VLC's forums are well aware of the issue. It's possible to manually disable these battery optimizations and have the app function properly in the background, but VLC claims that people often don't know how to do that, so they blame the app instead. The devices being blacklisted are the Huawei P8, P10, and P20. Users can still manually download the APK from VLC's website if they're interested in using the player.

One big area that Microsoft is focusing on with its next-generation Xbox is game streaming. According to a report from Thurrott.com, Microsoft is working on two new Xbox consoles. The "Xbox Two" will be a console similar to that of the Xbox One and Xbox 360, with updated hardware and specs. The other Xbox console in development will be limited to streaming games. The Verge reports: The streaming-only console will reportedly include a low amount of local compute for handling tasks like controller input, image processing, and collision detection. These tasks are essential to reducing latency in game streaming, and Microsoft is said to be planning to slice up processing between the game running locally and in the cloud in order to reduce input lag and other image processing delays. Microsoft is currently developing its next-generation Xbox console under the Scarlett codename. The software giant recently revealed it's also working on a game streaming service for Xbox that will work across any device. This is a key part of Microsoft's future plans with Xbox, and part of the company's vision for developing its "Netflix for video games" service, Xbox Game Pass.

Sources familiar with Microsoft's plans tell The Verge that Microsoft is currently "all hands" on creating datacenters capable of powering the company's game streaming service. Referred to as codename "XCloud" internally, Microsoft has been experimenting with combining four lots of custom Xbox consoles into a single server blade for its datacenters. These servers will launch initially with developers in mind to build and develop games in the cloud instead of local debug machines, and then to stream games to consumers.

Qualcomm announced its new QTM052 mmWave antenna modules that will enable 5G networks on select mobile phones. The penny-sized antenna array features four antennas that can accurately point toward the nearest 5G tower. It can even bounce signals off of surrounding surfaces, if needed. The Verge reports: The QTM052 is designed to be small enough that device manufacturers will be able to embed it into the bezel of a phone. Qualcomm's X50 5G modem is already designed to support up to four of the antenna arrays, one for each side of the phone, allowing for 16 total antennas and ensuring that no matter how you hold your phone, the signal won't get blocked. Qualcomm says that the first devices with the QTM052 antennas should be launching as early as the beginning of 2019 -- and hopefully, there'll be some actual 5G networks to use them with by then.

An anonymous reader quotes a report from Motherboard: Someone has uploaded what appear to be 11 of Apple's internal repair videos to YouTube. Apple did not immediately respond to a request for comment, but two sources in the repair community familiar with Apple's repair policies told Motherboard these are indeed genuine Apple how-to videos. The videos themselves have an Apple copyright on them, the host references internal Apple documentation and diagnostic tests, and, most importantly, the videos use proprietary Apple disassembly and repair tools that Motherboard has previously confirmed are manufactured by and are exclusive to Apple.

The videos on how to open an iPhone X and replace its battery are particularly interesting, and show that the DIY repair community has gotten extremely good at reverse-engineering Apple's official procedures. The instructor walks the repair tech through the process of opening the case on the iPhone X in a way that closely mirrors the process that sites such as iFixit have been doing for a few years now. The video starts by instructing the tech to remove the screws near the lighting port, then inserting the iPhone X into a device that uses suction cups to pry the screen away from the body while the tech uses a small tool to cut the adhesive along the seams at the edge of the device. Apple's suction cup tool looks like a bulkier version of iFixit's iSclack tool -- a suction cup device that customers can use to disassemble and repair their own device. The video about replacing the iPhone X's battery is remarkably similar to the iFixit video of the same procedure.

An anonymous reader quotes a report from IEEE Spectrum: Today, a team of researchers at Stanford, led by materials science and engineering associate professor Alberto Salleo and postdoctoral research fellow Onur Parlak, announced in Science Advances that they've developed a wearable patch that can determine how much cortisol someone is producing in seconds, using sweat drawn from the skin under the patch. [Cortisol, a steroid hormone, goes up when a person is under physical or emotional strain.] The stretchy patch pulls in the sweat through perforations to a reservoir. A membrane on top of the reservoir allows charged ions, like sodium and potassium, to pass through. Cortisol, which has no charge, can't pass, and instead blocks the charged ions. Signals sent from an electrical sensor in the patch can be used to detect these backups and determine how much cortisol is in the sweat.

The prototype cortisol detection patch channels sweat into a reservoir; a membrane selectively lets charged ions through, and the amount of these ions detected can be translated into a reading of cortisol levels in the sweat. Parlak tested the prototype on several runners, and reported that the cortisol levels detected by the wearable sensor patch matched those obtained by running samples of the runners' sweat through an ELISA (enzyme-linked immunosorbent assay) test that takes several hours.

A team of researchers at Glasgow University have built a robot that uses machine learning to run and analyze its own chemical reaction. The system is able to figure out every reaction that's possible from a given set of starting materials. Ars Technica reports: Most of its parts are dispersed through a fume hood, which ensures safe ventilation of any products that somehow escape the system. The upper right is a collection of tanks containing starting materials and pumps that send them into one of six reaction chambers, which can be operated in parallel. The outcomes of these reactions can then be sent on for analysis. Pumps can feed samples into an IR spectrometer, a mass spectrometer, and a compact NMR machine -- the latter being the only bit of equipment that didn't fit in the fume hood. Collectively, these can create a fingerprint of the molecules that occupy a reaction chamber. By comparing this to the fingerprint of the starting materials, it's possible to determine whether a chemical reaction took place and infer some things about its products.

All of that is a substitute for a chemist's hands, but it doesn't replace the brains that evaluate potential reactions. That's where a machine-learning algorithm comes in. The system was given a set of 72 reactions with known products and used those to generate predictions of the outcomes of further reactions. From there, it started choosing reactions at random from the remaining list of options and determining whether they, too, produced products. By the time the algorithm had sampled 10 percent of the total possible reactions, it was able to predict the outcome of untested reactions with more than 80-percent accuracy. And, since the earlier reactions it tested were chosen at random, the system wasn't biased by human expectations of what reactions would or wouldn't work. The research has been published in the journal Nature.

Corning just announced its most durable glass yet: Gorilla Glass 6. "The company says that the glass will survive up to 15 drops from a one meter height and can be 'up to two times better' than Gorilla Glass 5," reports The Verge. From the report: As phones get slimmer and have ever sleeker glass displays, reports have appeared that the slimness may actually cancel out the improvements in new iterations of Gorilla Glass, since thinner glass is weaker glass, even if it's become stronger. Still Corning argues that sleek edge-to-edge displays have actually led to stronger smartphones. Sometimes, in smartphones of previous years, the bezel would crack first, then leading to a weakness in the glass. There's also a tradeoff between drop resistance and scratch resistance, which Corning has admitted to in the past. Corning says that Gorilla Glass 6 will have the same amount of scratch resistance as previous generations. So although the company claims the new generation of Gorilla Glass is "better," you shouldn't expect new phones made with the glass to be more scratch-resistant. The first devices to feature Gorilla Glass 6 are expected to arrive near the end of the year.