According to The Information, two top executives left Google's Pixel team following the poor reception of the Pixel 4 and "rare internal criticism" of the phone. Ars Technica reports: Pixel general manager Mario Queiroz and "the mastermind behind Google's Pixel camera," Marc Levoy, have both left the Pixel team in the past year. Both Queiroz and Levoy have been visible members of the Pixel team at launch events, where they usually give lengthy presentations about the new features. Levoy, in particular, is cause for concern, since the Pixel's camera has been one of the major bright spots of the phone line. The Pixel 4 has not been a huge success. It has sold less than the Pixel 3 and Pixel 3a in its first two quarters. Osterloh was not confident about the device in the leadup to launch, either. The report says that "ahead of the [Pixel 4] October launch in New York," Osterloh called an all-hands meeting and shared his misgivings about the phone. He reportedly "did not agree with some of the decisions made about the phone" and that "in particular, he was disappointed in its battery power." Osterloh runs the Pixel team, so this reporting gives some insight into his management style. For a Steve Jobs type, approving these kinds of broad directions would be a big part of his day-to-day decision making. Osterloh, apparently, is more of a hands-off delegator.

The report says that, following "sluggish sales and tepid reviews" for the Pixel 4, "the company's hardware division needs to maintain favor and funding as Google's top brass look to rein in costs because of a recent slowdown in advertising revenue." Seeing Google Hardware in the same sentence as Google's cost-cutting efforts is a bit scary -- Google has been aggressively killing products that don't regularly pull in millions of users, and it's hard to make an argument that the Pixel line has been doing well or is even getting better over its four years of existence. We've already seen some branches of Google Hardware get the ax: Osterloh has admitted the team has quit the tablet business after poor performance from the Pixel Slate.

9to5Mac has learned more exclusive details about Apple's upcoming over-ear headphones, dubbed the "AirPods Studio," including specifications and settings. From the report: One of the key features of regular AirPods is ear detection, which automatically pauses the song when you take the earphones off. We've learned that AirPods Studio will have a similar feature, but it will work in a different way. Instead of ear detection, Apple is working to include sensors that can detect whether the headphones are on your head or neck. Based on this, we assume that AirPods Studio will play or pause content when they detect being placed on your head. Neck detection can be used to keep the headset turned on while the music is paused, just like when you take just one of the AirPods out of the ear.

Another new sensor will be able to detect left and right ears to automatically route the audio channels. That means there's likely no right or wrong side to use AirPods Studio, whereas current headphones have fixed left and right channels. Just like the AirPods Pro, Apple's new headphones will have Active Noise Cancellation and Transparency Mode. Users will be able to easily switch between the two modes to reduce external noise or to hear the ambient sound.

As AirPods Studio are expected to be mainly focused on professional users, pairing the earphones with a Mac or iOS device will unlock custom equalizer settings, with low, medium, and high frequency adjustments available, sources told us. According to a Bloomberg recent report, Apple's own-brand over-ear headphones will be available in at least two variations of the headphones -- one using leather fabrics and another with lighter materials to fitness use cases. Bloomberg also said Apple is testing a new modular design with exchangeable magnetic ear pads. [...] As for the price, rumors suggest that it will cost $349.

Two security researchers have published today details about a vulnerability in the Windows printing service that they say impacts all Windows versions going back to Windows NT 4, released in 1996. From a report: The vulnerability, which they codenamed PrintDemon, is located in Windows Print Spooler, the primary Windows component responsible for managing print operations. The service can send data to be printed to a USB/parallel port for physically connected printers; to a TCP port for printers residing on a local network or the internet; or to a local file, in the rare event the user wants to save a print job for later. In a report published today, security researchers Alex Ionescu & Yarden Shafir said they found a bug in this old component that can be abused to hijack the Printer Spooler internal mechanism. The bug can't be used to break into a Windows client remotely over the internet, so it's not something that could be exploited to hack Windows systems at random over the internet.

Alienware has announced the second generation of its flagship Area-51m gaming laptop, the Area-51m R2. It's largely similar on the outside to the original Area-51m, but the specs are almost completely overhauled on the inside, with 10th Gen Intel processors, new Nvidia GPUs, faster RAM, and better display options. From a report: All of those upgrades will be limited to the new R2 model, though. Despite Alienware's goals for future-proofed upgradable parts, the Area-51m has failed to live up to its potential. One of the biggest draws of the Area-51m was the option for user-replaceable parts. Most modern laptops have nearly all of their internal components sealed in place in an effort to cut down on size. But the Area-51m took a different approach. It was big, easy to disassemble, and made it simple to replace nearly every internal component, from RAM to the CPU to the GPU to even the thermal cooling components. The idea was to make something more portable than any other desktop but more customizable than any laptop. Alienware's parent company, Dell, even built its own system for replaceable GPUs in laptops, called DGFF (Dell Graphics Form Factor) to make it easier to upgrade in the future.

Amazon has built robot that is designed to kill the novel coronavirus with ultraviolet light. From a report: The robot looks a little like a hotel luggage cart, with a tall metal frame attached to a rectangular wheeled bottom. One side of the frame is outfitted with at least 10 ultraviolet tube lights. In a video shared with CBS News' "60 Minutes," the robot rolls down the freezer aisle of a Whole Foods store, aiming UV light at the freezer doors. The robot could be used in warehouses and at Whole Foods stores to kill the virus on surfaces such as food, packaging, and door handles. Amazon spokeswoman Kristen Kish said the company's robotics group has been designing and testing the UV robot.

An anonymous reader quotes a report from AnandTech: Today alongside with the launch of the Xiaomi Redmi K30 5G Racing Edition, Qualcomm is announcing the new Snapdragon 768G SoC which powers the device. The new SoC is a direct follow-up to the Snapdragon 765G announced last December, and the two chips are very likely the same silicon design, with the new variant increasing the clock frequencies.

The new chip features the same Cortex-A76 cores in a 1+1 configuration (one Prime high-clocked core, and one medium clocked core), alongside 6 Cortex-A55 cores. The difference in CPU performance lies in the frequencies of the big cores which are now at up to 2.8GHz and 2.4GHz for the Performance and Middle core -- a more notable uplift from the 2.4 and 2.2GHz clocks of the Snapdragon 765G. GPU clock frequencies have also been increased, resulting in at 15% performance boost over the Snapdragon 765. The rest of the chip is seemingly identical to the Snapdragon 765 series.

An anonymous reader quotes CNN: Although tidal energy is still in its infancy, it could help to reduce Australia's dependence on fossil fuels... The island nation is only beginning to explore tidal power through a number of pilot projects. But this form of energy has one major advantage: its predictability. While the sun may not shine, or the wind may not blow, the sea moves in predictable tidal currents...

Among those harnessing this tidal potential is Sydney-based Mako Energy. The company makes underwater turbines ranging between two and four meters in diameter. One turbine operating in constantly flowing water can produce enough electricity to power up to 20 homes. Their design enables them to generate electricity even in slow-flowing water, meaning they could be used in rivers and irrigation canals as well as the ocean. "We're developing turbines at a scale where they can be deployed easily in remote communities, coastal businesses, island communities and resorts," Douglas Hunt, managing director of Mako Energy, told CNN Business...

So far, Mako's customers have predominantly been large industrial and government sites, but it wants to make its turbines accessible to energy customers big and small... "We want to contribute to an energy mix that is less reliant on fossil fuels, by empowering local businesses and communities to generate their own power from a predictable and abundant source that is hiding in plain sight — often flowing directly past communities," says Hunt.

"AI has ushered in a new golden age of semiconductor innovation," reports Forbes: For most of the history of computing, the prevailing chip architecture has been the CPU, or central processing unit... But while CPUs' key advantage is versatility, today's leading AI techniques demand a very specific — and intensive — set of computations. Deep learning entails the iterative execution of millions or billions of relatively simple multiplication and addition steps... CPUs process computations sequentially, not in parallel. Their computational core and memory are generally located on separate modules and connected via a communication system (a bus) with limited bandwidth. This creates a choke point in data movement known as the "von Neumann bottleneck". The upshot: it is prohibitively inefficient to train a neural network on a CPU...

In the early 2010s, the AI community began to realize that Nvidia's gaming chips were in fact well suited to handle the types of workloads that machine learning algorithms demanded. Through sheer good fortune, the GPU had found a massive new market. Nvidia capitalized on the opportunity, positioning itself as the market-leading provider of AI hardware. The company has reaped incredible gains as a result: Nvidia's market capitalization jumped twenty-fold from 2013 to 2018.

Yet as Gartner analyst Mark Hung put it, "Everyone agrees that GPUs are not optimized for an AI workload." The GPU has been adopted by the AI community, but it was not born for AI. In recent years, a new crop of entrepreneurs and technologists has set out to reimagine the computer chip, optimizing it from the ground up in order to unlock the limitless potential of AI. In the memorable words of Alan Kay: "People who are really serious about software should make their own hardware...."

The race is on to develop the hardware that will power the upcoming era of AI. More innovation is happening in the semiconductor industry today than at any time since Silicon Valley's earliest days. Untold billions of dollars are in play.
Some highlights from the article:

• Google, Amazon, Tesla, Facebook and Alibaba, among other technology giants, all have in-house AI chip programs.

• Groq has announced a chip performing one quadrillion operations per second. "If true, this would make it the fastest single-die chip in history."

• Cerebras' chip "is about 60 times larger than a typical microprocessor. It is the first chip in history to house over one trillion transistors (1.2 trillion, to be exact). It has 18 GB memory on-chip — again, the most ever."

• Lightmatter believes using light instead of electricity "will enable its chip to outperform existing solutions by a factor of ten."

Reuters reports that the Trump administration "is in talks with semiconductor companies about building chip factories in the United States, representatives from two chipmakers said on Sunday." Intel Corp is in discussions with the United States Department of Defense over improving domestic sources for microelectronics and related technology, Intel spokesman William Moss said in an emailed statement. "Intel is well positioned to work with the U.S. government to operate a U.S.-owned commercial foundry and supply a broad range of secure microelectronics", the statement added.

Taiwan Semiconductor Manufacturing Co, on the other hand, has been in talks with the U.S. Department of Commerce about building a U.S. factory but said it has not made a final decision yet. "We are actively evaluating all the suitable locations, including in the U.S., but there is no concrete plan yet", TSMC spokeswoman Nina Kao said in a statement...

The Wall Street Journal had also reported that U.S. officials are looking at helping South Korea's Samsung Electronics, which has a chip factory in Austin, Texas, to expand its contract-manufacturing operations in the United States.

Long-time Slashdot reader UnknowingFool writes: As NVidia and AMD battle it out with their GPUs in consumer gaming, behind the scenes both manufacturers were competing against each for manufacturing capacity with TSMC's lines. It appears that early on in the 7nm line, NVidia lost out as it is rumored their strategy was to use Samsung more only to go back to TSMC when that failed By then TSMC had sold out their capacity to AMD. For their next generation GPUs it appears both NVidia is securing both 5nm early while grabbing as much 7nm as possible.

Long-time Slashdot reader blackbearnh writes: Since the COVID-19 pandemic has made Personal Protective Equipment worth it's weight in gold, Makers have been trying to help bridge the gap. While sewn masks have been the most common solution, the 3D printing community has been pitching in as well. The Montana Mask has been one of the most popular designs...

Thursday, the group Make the Masks announced that the design files and STLs to print the mask have been released under the GNU General Public License v3, allowing anyone to print, sell, remix or improve the design, as long as they conform to the license. Importantly, the GPLv3 includes an international non-exclusive patent grant, meaning that even if the inventors decide to apply for a patent, it will not restrict anyone from using the design.

MojoKid writes: AMD unveiled two new Ryzen 3 processors recently, designed to drive its Zen 2 CPU microarchitecture into more affordable price points. The new Ryzen 3 3100 and Ryzen 3 3300X are entry-level 3rd gen Ryzen CPUs that will run in any socket AM4 motherboard, with price points of only $99 (Ryzen 3 3100) and $120 (Ryzen 3 3300X). They are both quad-core/eight-thread chips, with similar cache configurations (2MB L2 + 16MB L3), and they both include basic Wraith Stealth coolers in their retail boxes. Internally, however, these two processors are somewhat different. The Ryzen 3 3100 is setup in a 2+2 configuration with two cores active per CCX (CPU Core Complex). The Ryzen 3 3300X, however, has a 4+0 configuration, with all of its active cores residing on a single CCX. These differences should result in better overall performance for the Ryzen 3 3300X, over and above just having a higher peek clock speed of 4.3GHz, versus 3.90Ghz for the Ryzen 3 3100. In the benchmarks, these new AMD quad-core chips offer similar or better performance versus competitive Intel Core i5 chips, but at significantly lower price points.

Ammalgam shares a report from Redmond Pie: Microsoft has officially announced the Surface Book 3. Not much has changed for the device in terms of external design, but internal circuitry is where it's at. This newest addition to the Surface family was formally unveiled alongside the Surface Go 2 today. And the detachable PC finally got the long overdue refresh. Microsoft decided to go with the latest Intel 10th generation CPUs, codenamed Ice Lake. There was talk that the company might opt for the Comet Lake processors. Both are 10th generation chips, but Ice Lake is made on a 10nm process, while Comet Lake is 14nm. [...] As for the CPU options, we have the Core i5-1035G7 and Core i7-1065G7 available. The Core i5 models don't offer dedicated graphics, while the 15-inch variant of the Surface Book 3 only comes in the Core i7 flavor. GPU got a real look, too. The 13.5-inch Surface Book 3 gets you the 4GB Max-Q variant of the NVIDIA GeForce GTX 1650, while the 15-inch one comes with a GTX 1660 Ti Max-Q with 6GB.

On the display side, we have a 3000 x 2000 screen on the 13.5-inch model, while the 15-inch one offers a resolution of 3240 x 2160. Microsoft also claims that the Surface Book 3 has the best battery life of any device it has made up until now. It's rated at 17.5 hours on the 15-inch model and 15.5 hours on the 13.5-inch variant. Some neat improvements when it comes to port selection, with both the USB Type-A and Type-C ports now being USB 3.1 Gen 2, meaning you get 10Gbps instead of 5Gbps. There is also a Surface Connect port on both the base and the tablet, to go with the 3.5mm audio jack in the tablet. All these enhancements mean that the starting price of the Surface Book 3 is a bit higher at $1,599. It will be available on May 21.

An anonymous reader quotes a report from Reuters: Electricity generated by renewable sources like solar, wind and hydro has exceeded coal-fired power in the United States for a record 40 straight days, according to a report based on U.S. government data released on Monday. The boost for renewables is due to a seasonal increase in low-cost solar and hydro power generation, alongside an overall slump in electricity demand caused by coronavirus-related stay-at-home orders, according to the Institute for Energy Economics and Financial Analysis. Coal tends to be the first power source to be cut by utilities when demand falls because subsidized renewable sources are cheaper to operate and often backed by state clean-energy mandates.

Every day between March 25 and May 3, solar, wind and hydro plants together produced more electricity than the nation's coal-fired plants -- accounting for about a fifth of the grid's power, IEEFA said. The longest back-to-back stretch previously was nine days in 2019. In total in 2019, renewables beat coal on just 38 days, IEEFA said. IEEFA added it is possible that renewable energy in the United States could exceed coal on an annual basis for the first time this year, a year earlier than it initially forecast, if the power consumption trends caused by the health crisis continue. The report says overall U.S. electricity consumption is projected to fall 3% this year, with coal-fired power demand falling 20%. Meanwhile, EIA forecasts renewable energy generation will grow 11% because it is dispatched by grid operators whenever it is available because of its low operating cost.

Researchers from Rice University have created an efficient, low-cost device that splits water to produce hydrogen fuel. "The platform developed by the Brown School of Engineering lab of Rice materials scientist Jun Lou integrates catalytic electrodes and perovskite solar cells that, when triggered by sunlight, produce electricity," reports Phys.Org. "The current flows to the catalysts that turn water into hydrogen and oxygen, with a sunlight-to-hydrogen efficiency as high as 6.7%." From the report: This sort of catalysis isn't new, but the lab packaged a perovskite layer and the electrodes into a single module that, when dropped into water and placed in sunlight, produces hydrogen with no further input. The platform introduced by Lou, lead author and Rice postdoctoral fellow Jia Liang and their colleagues in the American Chemical Society journal ACS Nano is a self-sustaining producer of fuel that, they say, should be simple to produce in bulk.

"The concept is broadly similar to an artificial leaf," Lou said. "What we have is an integrated module that turns sunlight into electricity that drives an electrochemical reaction. It utilizes water and sunlight to get chemical fuels." Perovskites are crystals with cubelike lattices that are known to harvest light. The most efficient perovskite solar cells produced so far achieve an efficiency above 25%, but the materials are expensive and tend to be stressed by light, humidity and heat. "Jia has replaced the more expensive components, like platinum, in perovskite solar cells with alternatives like carbon," Lou said. "That lowers the entry barrier for commercial adoption. Integrated devices like this are promising because they create a system that is sustainable. This does not require any external power to keep the module running." The research has been published in the journal ACS Nano.

An anonymous reader quotes a report from Bloomberg: Facebook's Oculus division is building a new version of its Quest standalone virtual reality headset, but the device is facing potential delays due to the impact of Covid-19 on product development and the global supply chain. The social-networking giant is working on multiple potential successors to the Quest. Some models in advanced testing are smaller, lighter, and have a faster image refresh rate for more realistic content, according to people familiar with the matter. These headsets also have redesigned controllers, said the people, who asked not to be identified discussing unannounced products.

The company hasn't finalized which new version of the Quest it will release, and the final product may have different features. Facebook originally planned to launch the new model at the end of 2020, around its annual Oculus Connect conference, but the coronavirus pandemic could delay the headset from shipping until 2021, the people said. Some versions in testing are 10% to 15% smaller than the current Quest. That, in addition to being lighter, makes them more comfortable to wear, one of the people said. The existing model weighs about 1.25 pounds, which is too heavy for some users wearing the device for extended periods. The models in testing are closer to a pound, the person said. The company could sell the new model alongside the current Quest, or choose to fully replace it.

As predicted, the proprietary locking system Apple rolled out with its 2018 MacBook Pros is hurting independent repair stores, refurbishers, and electronics recyclers. A combination of secure software locks, diagnostic requirements, and Apple's new T2 security chip are making it hard to breathe new life into old MacBook Pros that have been recycled but could be easily repaired and used for years were it not for these locks. From a report: It's a problem that highlights Apple's combative attitude towards the secondhand market and the need for national right to repair legislation. "The irony is that I'd like to do the responsible thing and wipe user data from these machines, but Apple won't let me," John Bumstead, a MacBook refurbisher and owner of the RDKL INC repair store, said in a tweet with an attached picture of two "bricked" MacBook Pros. "Literally the only option is to destroy these beautiful $3,000 MacBooks and recover the $12/ea they are worth as scrap."

As Motherboard has reported previously, without official Apple diagnostic software, newer MacBooks cannot be repaired or reset. "By default you can't get to recovery mode and wipe the machine without a user password, and you can't boot to an external drive and wipe that way because it's prohibited by default," Bumstead told Motherboard in an email. "Because T2 machines have no removable hard drive, and the drive is simply chips on the board, this default setting means that a recycler (or anyone) can't wipe or reinstall a T2 machine that has default settings unless they have the user password."

An anonymous reader quotes Techcrunch: Arm — the U.K. company behind the designs of chips for everyone from Apple to Qualcomm to Samsung — is hoping to kickstart developing by offering up access to around 75% of its chip portfolio for free to qualified startups.

The move marks an expansion of the company's Flexible Access program. With it, Arm will open access to its IP for early-stage startups. While some of the biggest companies pay the chip designer big bucks for that information, the cost can be prohibitive for those just starting out...

Interested parties can access the full list of available IP here.

New submitter jsicolo shares a report: Nvidia Chief Scientist Bill Dally has released an open-source ventilator hardware design he developed in order to address the shortage resulting from the global coronavirus pandemic. The mechanical ventilator design developed by Dally can be assembled quickly, using off-the-shelf parts with a total cost of around $400 -- making it an accessible and affordable alternative to traditional, dedicated ventilators, which can cost $20,000 or more. The design created by Dally strives for simplicity, and basically includes just two central components -- a solenoid valve and a microcontroller. The design is called the OP-Vent, and in this video you can see how bare-bones it is in terms of hardware compared to existing alternatives, including some of the other more complex emergency-use ventilator designs developed in response to COVID-19.

UnknowingFool writes: Intel released more information about their next generation CPUs, codenamed Comet Lake. Overall, CPUs will get more cores and threads and slight speed boosts. Price wise, Intel is cutting prices to be more competitive with AMD's Rzyen processors. Some of the downsides include requiring new socket (thus new MBs), LGA 1200 and lack of PCIE 4.0 compatibility. No specific benchmarks were released, however Intel claims to have the fastest gaming CPUs. "[T]he top Comet Lake chip is the same price as the top Coffee Lake at $488, and the cheapest Core i3 is $122," reports PC Gamer. They expect the release date to be sometime in May, though no official date has been confirmed.