"Samsung is using artificial intelligence to automate the insanely complex and subtle process of designing cutting-edge computer chips," reports Wired: The South Korean giant is one of the first chipmakers to use AI to create its chips. Samsung is using AI features in new software from Synopsys, a leading chip design software firm used by many companies...

Others, including Google and Nvidia, have talked about designing chips with AI. But Synopsys' tool, called DSO.ai, may prove the most far-reaching because Synopsys works with dozens of companies. The tool has the potential to accelerate semiconductor development and unlock novel chip designs, according to industry watchers. Synopsys has another valuable asset for crafting AI-designed chips: years of cutting-edge semiconductor designs that can be used to train an AI algorithm. A spokesperson for Samsung confirms that the company is using Synopsys AI software to design its Exynos chips, which are used in smartphones, including its own branded handsets, as well as other gadgets...

Chipmakers including Nvidia and IBM are also dabbling in AI-driven chip design. Other makers of chip-design software, including Cadence, a competitor to Synopsys, are also developing AI tools to aid with mapping out the blueprints for a new chip.
But Synopsys's co-CEO tells Wired that Samsung's chip will be "the first of a real commercial processor design with AI."

Lamborghini's exotic Countach sportscar was featured in the opening scene of the 1981 movie Cannonball Run. On the car's 50th anniversary, they've now introduced a 802-horsepower hybrid-electric version. Mashable reports: The reimagined Countach, unveiled in Monterey, California on Friday, is the Italian carmaker's second hybrid. Its drivetrain and performance borrows heavily from the Sián supercar, which was introduced in 2019 as the famous sports car maker's first hybrid. Lamborghini's new Countach (pronounced "coon-tash") has a V12 hybrid engine and 48 volt e-motor, as also seen in the Sián. It can accelerate up to 60 mph in 2.8 seconds and has a 220 mph top speed. Unlike some hybrids, the Countach doesn't have a pure electric mode. Instead, it's always using a blend of the electric motor and gas engine.

The air vents on the back are 3D-printed and the roof is photochromic, meaning it switches from solid to transparent at the push of a button. There are also four exhaust pipes on the backside...

There will only be 112 of Lamborghini's new hybrid Countach ever made...
TechCrunch adds that "Powering the Countach's electric motor is a supercapacitor Lamborghini claims delivers three times more power compared to a lithium-ion battery of the same weight.

"The automaker says it mounted the electric motor directly to the gearbox to preserve the feeling of power transfer you get from a V12 engine."

There's something new in the world of expensive high-performance sportscars (or "hypercars".) Italian carmaker Automobili Pininfarina "has debuted the Battista, the first pure-electric hyper GT, on the streets of California as part of Monterey Car Week," reports Newsweek. (Alternate URL here.) The debut will give U.S. clients the chance to experience the 1,900 horsepower hypercar... [In a video] the Battista, crafted at Automobili Pininfarina's manufacturing facility in Italy, glides smoothly and quickly through California roads at speeds of over 100 miles per hour. The drive shows off the agility of the polished Impulso forged aluminum alloy wheels and exposed bodywork. Pulling off the road, the scissor doors swing open, highlighting the Black Exposed Signature Carbon bodywork.

Supplying the 1,900 hp is a 120-kilowatt-hour battery that powers four electric motors, one in each wheel, that gets about 1696 pound-feet of torque. On a single charge, the car is expected to get a range of over 310 miles.

With an emphasis on the company's "Pure Sound" philosophy and drawing from music theory, the bespoke design of the car is built to have a core frequency of 54 hertz (hz). Wanting to provide an emotional experience for the driver, the organic frequency will rise in multiples of 54 hz as the speed increases.
The company's product platform director of sports cars explains on their web site that "Every driver has an emotional bond with a car and the sound of Battista will nurture this connection, not by replicating a familiar car sound, but with one that radiates the beauty of Battista's design both inside and out. This way, the Battista will not only impress with its aesthetic appeal and performance, but also on a new emotional level enhanced through the sound."

The company's web site also calls it "the most powerful road-legal Italian sports car ever produced."

Ars Technica reports on a paper investigating how much each power plant contributes to global emissions, using data from 2018. "The study finds that many countries have many power plants that emit carbon dioxide at rates well above either the national or global average.

"Shutting down the worst 5 percent of this list would immediately wipe out about 75 percent of the carbon emissions produced by electricity generation." It should surprise nobody that all the worst offenders are coal plants. But the distribution of the highest polluting plants might include a bit of the unexpected.

For example, despite its reputation as the home of coal, China only has a single plant in the top-10 worst (bottom-10?). In contrast, South Korea has three on the list, and India has two. In general, China doesn't have many plants that stand out as exceptionally bad, in part because so many of its plants were built around the same time, during a giant boom in industrialization. As such, there's not much variance from plant to plant when it comes to efficiency. In contrast, countries like Germany, Indonesia, Russia, and the US all see a lot of variance, so they're likely to have some highly inefficient plants that are outliers.

Put a different way, the authors looked at how much of a country's pollution was produced by the worst 5 percent when all of the country's power plants were ranked by carbon emissions. In China, the worst 5 percent accounted for roughly a quarter of the country's total emissions. In the US, the worst 5 percent of plants produced about 75 percent of the power sector's carbon emissions. South Korea had similar numbers, while Australia, Germany, and Japan all saw their worst 5 percent of plants account for roughly 90 percent of the carbon emissions from their power sector. When it comes to carbon emissions, the worst 5 percent of power plants account for 73 percent of the total power sector emissions globally. That 5 percent also produces over 14 times as much carbon pollution as it would if the plants were merely average...

Simply boosting each plant's efficiency to the average for the country would drop power sector emissions by a quarter and up to 35 percent in countries like Australia and Germany. Switching them to natural gas, which produces less carbon dioxide per amount of energy released, would drop global emissions by 30 percent, with many countries (including the US) seeing drops of over 40 percent. Again, because China doesn't see a lot of variance among its plants, these switches would have less of an impact, being in the area of 10 percent drops in emissions. But the big winner is carbon capture and storage. Outfitting the worst of the plants with a capture system that was 85 percent efficient would cut global power sector emissions in half and total global emissions by 20 percent. Countries like Australia and Germany would see their power sector emissions drop by over 75 percent.

Overall, these are massive gains, considering that it's not unreasonable to think that the modifications could be done in less than a decade. And they show the clear value of targeting the easiest wins when it comes to lowering emissions.

"A start-up founded by scientists at the Massachusetts Institute of Technology says it is nearing a technological milestone that could take the world a step closer to fusion energy, which has eluded scientists for decades," reports the New York Times: Researchers at M.I.T.'s Plasma Science and Fusion Center and engineers at the company, Commonwealth Fusion Systems, have begun testing an extremely powerful magnet that is needed to generate immense heat that can then be converted to electricity. It would open the gates toward what they believe could eventually be a fusion reactor... Though a fusion energy breakthrough remains elusive, it is still held out as one of the possible high-technology paths to ending reliance on fossil fuels. And some researchers believe that fusion research could finally take a leap forward this decade. More than two dozen private ventures in the United States, Europe, China and Australia and government-funded consortia are now investing heavily in efforts to build commercial fusion reactors. Total investment by people such as Bill Gates and Jeff Bezos is edging toward $2 billion. The federal government is also spending about $600 million each year on fusion research, and there is a proposed amendment to add $1 billion to the Biden administration's infrastructure bill, said Andrew Holland, chief executive of the Fusion Industry Association...

Commonwealth's new magnet, which will be one of the world's most powerful, will be a crucial component in a compact nuclear fusion reactor known as a Tokamak, a design that uses magnetic forces to compress plasma until it is hotter than the sun... Commonwealth Fusion executives claim that the magnet is a significant technology breakthrough that will make Tokamak designs commercially viable for the first time. They say they are not yet ready to test their reactor prototype, but the researchers are finishing the magnet and hope it will be workable by 2025...

Commonwealth, which has raised more than $250 million so far and employs 150 people, received a significant boost last year when physicists at M.I.T.'s Plasma Science and Fusion Center and the company published seven peer-reviewed papers in the Journal of Plasma Physics explaining that the reactor will work as planned. What remains to be proved is that the Commonwealth prototype reactor can produce more energy than it consumes, an ability that physicists define as Q greater than 1. The company is hoping that its prototype, when complete, will produce 10 times the energy it consumes.
Commonwealth's chief executive (also a plasma physicist) explains to the Times how fusion energy is different than other sources: because it really doesn't require any resources. "You add up all the costs, the cost of normal stuff like concrete and steel, and it will make as much power as a gas plant, but without having to pay for the gas."

Google's Pixel 6 phone will be powered by a Tensor processor which PCMag UK believes is "clearly designed to accelerate machine learning and AI." But does it have bigger implications? Tensor is a signpost, not a destination. Google has never sold huge numbers of Pixel phones and isn't signaling a change in strategy there. Rather, it's saying that it would like Android as a whole to shift toward more on-device processing for AI and ML. That could give a big boost to Google's two core businesses, advertising and data. It could also create problems for the future of 5G...

The more your phone can handle its own ML and AI, the less it needs the cloud. For example, a Tensor-enabled phone could potentially analyze your photos and share data locally with on-device advertising APIs, letting Google proclaim that its cloud services never access your raw data. That would help bridge the gap between the privacy you want and the targeted ads Google needs to survive. But a lot of consumer 5G app ideas assume your phone will offload processing or rendering onto the network.

Phones that are mostly self-sufficient aren't going to need high-bandwidth, low-latency networks or "mobile edge computing." Sure, the rise of more Tensor-like ML-focused chips would take pressure off the carriers' still-shaky 5G builds, but it'll also keep raising the question of why consumers need 5G in the first place. That could reduce carriers' willingness to keep investing in their consumer 5G networks.
Although the first sence of the article's last paragraph adds, "Maybe I'm overstating the issue here..."

The Register's Richard Speed commemorates the 40th anniversary of the introduction of the IBM Model 5150: IBM was famously late to the game when the Model 5150 (or IBM PC) put in an appearance. The likes of Commodore and Apple pretty much dominated the microcomputer world as the 1970s came to a close and the 1980s began. Big Blue, on the other hand, was better known for its sober, business-orientated products and its eyewatering price tags. However, as its customers began eying Apple products, IBM lumbered toward the market, creating a working group that could dispense with the traditional epic lead-times of Big Blue and take a more agile approach. A choice made was to use off-the-shelf hardware and software and adopt an open architecture. A significant choice, as things turned out.

Intel's 8088 was selected over the competition (including IBM's own RISC processor) and famously, Microsoft was tapped to provide PC DOS as well as BASIC that was included in the ROM. So this marks the 40th anniversary of PC DOS, aka MS-DOS, too. You can find Microsoft's old MS-DOS source code here. The basic price for the 5150 was $1,565, with a fully loaded system rising to more than $3,000. Users could enjoy high resolution monochrome text via the MDA card or some low resolution graphics (and vaguely nauseating colors) through a CGA card (which could be installed simultaneously.) RAM landed in 16 or 64kB flavors and could be upgraded to 256kB while the Intel 8088 CPU chugged along at 4.77 MHz.

Storage came courtesy of up to two 5.25" floppy disks, and the ability to attach a cassette recorder -- an option swiftly stripped from later models. There was no hard disk, and adding one presented a problem for users with deep enough pockets: the motherboard and software didn't support it and the power supply was a bit weedy. IBM would resolve this as the PC evolved. Importantly, the motherboard also included slots for expansion, which eventually became known as the Industry Standard Architecture (ISA) bus as the IBM PC clone sector exploded. IBM's approach resulted in an immense market for expansion cards and third party software. While the Model 5150 "sold like hotcakes," Speed notes that it was eventually discontinued in 1987.

Today, Samsung launched the Galaxy Watch 4 and Galaxy Watch 4 Classic -- two new wearables that are "the fruits of Samsung's smartwatch collaboration with Google," writes Becca Farsace via The Verge. From the report: The Watch 4 Classic starts at $349 for the Bluetooth model, rising to $399 for the LTE model, while the Watch 4 is a little less expensive with a starting price of $249 (or $299 with LTE). Both are available to preorder today, and ship August 27th. The big difference between the two models is that the Watch 4 Classic has one of those physical rotating bezels that we've liked so much on Samsung's previous smartwatches, while if you opt for the standard Watch 4, there's a touch-sensitive bezel accessible by swiping at the edges of the screen. The Watch 4 Classic is also made of a more premium stainless steel rather than the aluminum you'll find on the Watch 4. On the right of both watches are a pair of control buttons.

External differences aside, internally both watches share a lot of the same specs. They're both powered by the same 5nm Exynos W920 processor Samsung detailed yesterday, paired with 1.5GB of RAM and 16GB of storage. Battery capacity varies between sizes, but Samsung reckons you'll average around 40 hours of battery life regardless of model. There's LTE on select models, but if you were hoping for 5G, you'll be disappointed -- Samsung says it doesn't think it's worth it because the amount of data smartwatches process is too small.

But the biggest departure from Samsung's previous smartwatches is that the Watch 4 and Watch 4 Classic aren't running its own Tizen operating system. Instead, their software is the result of a collaboration between Samsung and Google, which was announced in May. Samsung is branding the watches' operating system as "Wear OS Powered by Samsung" although Google has called it Wear OS 3. But either way, the hope is that it combines the best of Tizen with the best of Wear OS. On Samsung's watches specifically, the interface you're looking at is One UI Watch, which is effectively Samsung's skin sitting on top of Wear OS. Think of it as Samsung's One UI software on phones, which works with Google's Android. It gives the Watch 4's interface a similar look and feel to Samsung's previous Tizen-powered watches. Google has promised its collaboration with Samsung will lead to a host of high-level benefits for Wear OS, like improved battery life, faster loading apps, and smoother animations.

labloke11 shares a report from Science News: Windbreaks may sound like a counterintuitive idea for boosting the performance of a wind turbine. But physicists report that low walls that block wind could actually help wind farms produce more power. Scientists already knew that the output of a single wind turbine could be improved with a windbreak. While windbreaks slow wind speed close to the ground, above the height of the windbreak, wind speeds actually increase as air rushes over the top. But for large wind farms, there's a drawback. A windbreak's wake slows the flow of air as it travels farther through the rows of turbines. That could suggest that windbreaks would be a wash for wind farms with many turbines.

But by striking a balance between these competing effects, windbreaks placed in front of each turbine can increase power output, new computer simulations suggest. It comes down to the windbreaks' dimensions. Squat, wide barriers are the way to go, according to a simulated wind farm with six rows of turbines. To optimize performance, windbreaks should be a tenth the height of the turbine and at least five times the width of the blades, physicists report July 30 in Physical Review Fluids. Such an arrangement could increase the total power by about 10 percent, the researchers found. That's the equivalent of adding an additional turbine, on average, for every 10 in a wind farm. In the simulations, the wind always came from the same direction, suggesting the technique might be useful in locations where wind tends to blow one way, such as coastal regions. Future studies could investigate how this technique might apply in places where wind direction varies.

An anonymous reader quotes a report from Ars Technica: Google is cooking up the first major Wear OS release since 2018, and Samsung is abandoning Tizen for smartwatches and going all-in on Wear OS with the Galaxy Watch 4. Last night, Samsung took the wraps off the main SoC for the Galaxy Watch 4, and compared to what Wear OS usually gets, Samsung is shipping a beast of an SoC. The "Samsung Exynos W920" will be a multi-generational leap in performance for Wear OS. Samsung says this is a 5 nm chip with two ARM Cortex A55 cores and an ARM Mali-G68 GPU. For the always-on display mode, there's an additional Cortex M55 CPU, which can keep the watch face ticking along while using minimal power. There's also an integrated LTE modem for on-the-go connectivity.

Compared to Samsung's previous smartwatch chip, the Tizen-only Exynos 9110 (10 nm, 2x Cortex A53), the company is promising "around 20 percent" better CPU performance and "ten times better graphics performance." Remember that the Exynos 9110 is from 2018, so those comparative numbers are inflated, but at 5 nm, this is a more modern chip than Wear OS has ever seen. Wear OS has suffered for years at the hands of Qualcomm, which has been starving the ecosystem of quality SoCs for wearables. Most people's last experience with Wear OS is the Snapdragon Wear 2100 or 3100 SoCs, both of which were ancient Cortex A7 CPUs built on a 28 nm process. Qualcomm introduced a slightly more modern chip, the Wear 4100 in 2020 (a Cortex A53-based, 12 nm chip), but almost no manufacturers actually shipped that chip a year later, and we're still getting Wear 3100 launches today. Qualcomm's answer to Samsung's chip will be the Wear 5100, which isn't due until 2022. We should know more about Wear OS 3.0 tomorrow when Samsung holds its Aug. 11 "Unpacked" event. Not only are they expected to reveal the Galaxy Watch 4 and the big Wear OS revamp, but they're planning to launch at least two new foldable smartphones -- the Galaxy Z Fold 3 and Galaxy Z Flip 3.

While some bitcoin mining operations are now looking to nuclear power, the Associated Press reports, Bill Spence (and his company Stronghold Digital Mining) is creating crypto-mining hubs out of waste coal power plants. (Text-only version here): The plant he had bought was in trouble. It was competing with cheap natural gas on the power grid and losing — endangering the 35 jobs at Scrubgrass Generating Station along with the effort to clean up millions of tons of leaching coal waste left behind by mining companies over the course of decades. The plant couldn't just rely on the grid for revenue anymore, because the grid simply didn't need its power all that often. Mr. Spence started to look for other customers...

Already, some power generators — finding they can make more money supplying electricity to Bitcoin-mining operations than selling it to the grid — are shifting focus. Energy Harbor, which owns the Beaver Valley Nuclear Plant in Beaver County, announced earlier this month that it will supply nuclear power to a Bitcoin-mining data center in Ohio. Talen Energy, owner of the Susquehanna Steam Electric Station in Luzerne County, is doing the same. The company said last month that it will develop a data center to mine digital currency that could use up to 300 megawatts, or 12% of the nuclear plant's capacity. Bitcoin miners, in turn, are hyper cognizant of power prices and availability. Some are taking mobile units into the oil fields, hooking up their machines to run on natural gas, a byproduct of oil product that would otherwise be flared...

Today, Scrubgrass, an 85-megawatt blue box with a black smokestack in the hills of Scrubgrass Township, looks much like it did when it first opened in 1993 — except for the trailers filled with Bitcoin miners in the back... [T]here are about 3,000 cryptocurrency miners packed into retrofitted shipping containers behind the power plant, most of them owned by Stronghold and some that belong to other mining companies that buy power from the plant. Another 5,000 machines are scheduled to arrive next month. According to documents filed with the SEC, Stronghold is planning to operate 57,000 miners by the end of next year. In 2020, when the power plant seldom ran, Stronghold made more money from its Bitcoin operations than by selling Scrubgrass's energy to the grid. During the first three months of this year, the trend reversed. It received almost $2 million from power sales and more than $1 million from its crypto datacenter...

Stronghold is buying another waste coal plant, Panther Creek Energy Facility in Carbon County, with plans to replicate its cryptomining data center there, and is eyeing a third.
The Associated Press notes that the waste-coal plants are powered by those thousands of acres of abandoned (and pollutant-emitting) coal piles left behind by earlier coal-powered plants, finally remediating them into reclaimable land. But with waste coal plants, there's always a trade-off.

"In 2019, the last year with available federal data, Scrubgrass emitted the equivalent of 371,000 tons of CO2 — the greenhouse gas footprint of 80,000 cars driving for a year."

An anonymous reader quotes a report from Live Science: Scientists have created key parts of synthetic brain cells that can hold cellular "memories" for milliseconds. The achievement could one day lead to computers that work like the human brain. In the new study, published in the journal Science on Aug. 6, researchers at the Centre national de la recherche scientifique in Paris, France created a computer model of artificial neurons that could produce the same sort of electrical signals neurons use to transfer information in the brain; by sending ions through thin channels of water to mimic real ion channels, the researchers could produce these electrical spikes. And now, they have even created a physical model incorporating these channels as part of unpublished, ongoing research. At a finer level, the researchers created a system that mimics the process of generating action potentials -- spikes in electrical activity generated by neurons that are the basis of brain activity. To generate an action potential, a neuron starts to let in more positive ions, which are attracted to the negative ions inside of the cell. The electrical potential, or voltage across the cell membrane, causes doorways on the cell called voltage-gated ion channels to open, raising the charge even more before the cell reaches a peak and returns to normal a few milliseconds later. The signal is then transmitted to other cells, enabling information to travel in the brain.

To mimic voltage-gated ion channels, the researchers modeled a thin layer of water between sheets of graphene, which are extremely thin sheets of carbon. The water layers in the simulations were one, two, or three molecules in depth, which the researchers characterized as a quasi-two-dimension slit. [T]he researchers wanted to use this two-dimensional environment because particles tend to react much more strongly in two dimensions than in three, and they exhibit different properties in two dimensions, which the researchers thought might be useful for their experiment. Testing out the model in a computer simulation, the researchers found that when they applied an electric field to the channel, the ions in the water formed worm-like structures. As the team applied a greater electric field in the simulation, these structures would break up slowly enough to leave behind a "memory," or a hint of the elongated configuration.

When the researchers ran a simulation linking two channels and other components to mimic the behavior of a neuron, they found the model could generate spikes in electrical activity like action potentials, and that it "remembered" consistent properties in two different states -- one where ions conducted more electricity and one where they conducted less. In this simulation, the "memory" of the previous state of the ions lasted a few milliseconds, around the same time as it takes real neurons to produce an action potential and return to a resting state. This is quite a long time for ions, which usually operate on timescales of nanoseconds or less. In a real neuron, an action potential equates to a cellular memory in the neuron; our brains use the opening and closing of ion channels to create this kind of memory. The new model is a version of an electronic component called a memristor, or a memory resistor, which has the unique property of retaining information from its history. But existing memristors don't use liquid, as the brain does.

phalse phace shares a report from Los Angeles Times: In a sign of the region's worsening drought, state water officials announced Thursday the shutdown of a major hydroelectric power plant at Lake Oroville in Northern California, citing the lowest-ever recorded water level at the reservoir. It marks the first time that officials have been forced to close the Edward Hyatt Powerplant, which was completed in 1967, on account of low water at the lake. The loss of the hydroelectric power source at Lake Oroville, about 75 miles north of Sacramento, could contribute to rolling blackouts in the state during heat waves in coming months.

Officials had warned that once the water level in Lake Oroville fell to 640 feet above sea level, the plant could no longer produce power; at that level, the water cannot reach the intake pipes that flow toward the underground hydroelectric facility. On Thursday, Lake Oroville was at 641 feet with 863,516 acre-feet of storage, which is 24% of its overall capacity and 34% of its historical average for this time, according to the Department of Water Resources. The Hyatt plant is designed to produce up to 750 megawatts of power but has often generated 100 to 400 megawatts, or slightly less than 1% of the state's average daily peak usage.

Rare earth magnet recycling is about so much more than sustainable data centers. From a report: The U.S. alone generates nearly 17 percent of all used hard disk drives -- the largest share globally -- and researchers have estimated that if all of these data storage devices were recycled, they could supply more than 5 percent of all rare earth magnet demand outside of China, potentially helping meet the demand of the information technology sector as well as clean energy companies. A consortium of U.S. researchers, tech companies, hardware manufacturers, and electronic waste recyclers has recently begun exploring exactly how those rare earths can be re-harvested and given a second life.

In 2019, these stakeholders published a report identifying a host of potential strategies, including wiping and re-using entire hard disk drives, removing and reusing the magnet assemblies, grinding up old hard drive magnets and using the powder to manufacture new ones, and extracting purified rare earth elements from shredded drives. Each of these strategies has its own challenges -- removing magnet assemblies by hand is labor intensive; extracting rare earths from technology can be chemical or energy intensive and produce significant waste -- and for any of them to be scaled up, there needs to be buy-in from numerous actors across global supply chains.

Making even the relatively minor supply chain adjustments needed to place used or recycled rare earth magnets inside new drives "is difficult," said Hongyue Jin, a scientist at the University of Arizona who studies rare earth recycling. "And especially when you've got to start from some small amount with a new technology." Still, some companies have begun taking the first steps. In 2018, Google, hard disk drive manufacturer Seagate, and electronics refurbisher Recontext (formerly Teleplan) conducted a small demonstration project that involved removing the magnet assemblies from six hard disk drives and placing them in new Seagate drives. This demonstration, said Kali Frost, a doctoral student in industrial sustainability at Purdue University, was the "catalyst" for the larger 2019 study in which 6,100 magnet assemblies were extracted from Seagate hard drives in a Google data center before being inserted into new hard drives in a Seagate manufacturing facility. Frost, who led the 2019 study, believes it is the largest demonstration of its kind ever done.

The results, which will be published in a forthcoming edition of the journal Resources, Conservation, and Recycling, not only showed that rare earth magnets could be harvested and reused at larger scale, but that there were significant environmental benefits to doing so: Overall, re-used magnet assemblies had a carbon footprint 86 percent lower than new ones, according to the study. Frost says that this estimate conservatively took into account the energy mix of the local power grid where the data center operated. Considering Google's near round-the-clock renewable energy usage at this particular data center, the carbon footprint of the reused magnets was even lower.

In addition to new hardware announced today, Google has made a big change to its new line of Nest cameras: they no longer require a monthly subscription fee to record video. Ars Technica reports: We'll get to the new models in a minute (editor's note: no we won't because this isn't a slashvertisement; you'll have to visit the article), but the biggest news is that Google is making the cameras more useful without a monthly subscription. Previously, core camera features like recording video were locked behind a $6-$12 monthly subscription plan called "Nest Aware," but the new cameras can now record local video. You only get three hours' worth of "events" (motion detection, as opposed to 24/7 video), but it's a start. Google has also moved activity zones and some image recognition features from the cloud-based pay-per-month service to on-device processing, so they work without a subscription, too.

If you still want to pay for the "Nest Aware" subscription, it comes in two tiers. There's the $6 "Nest Aware," which gives you 30 days of "event" video history and facial recognition. The free tier can detect and alert you about people, animals, and vehicles, but the subscription adds facial recognition for "familiar faces" so Nest can tell if a loved one or stranger is at the door and alert you accordingly. The $12-per-month tier is "Nest Aware Plus," which provides 60 days of event video history and 10 days of 24/7 video history if you have a wired (not battery-powered) Google Nest Cam (the doorbell can't record continuous video). Another big added quality-of-life feature is that the cameras can now work offline. Local storage and on-device processing mean the cameras can function without the Internet; previously, the cloud was the only way they had to process and store video.

MojoKid writes: AMD is taking the wraps off of its latest integrated processors known as Ryzen 7 5700G and the Ryzen 5 5600G. As their branding suggests, these new products are based on the same excellent AMD Zen 3 core architecture, but with integrated graphics capabilities on board as well, hence the "G" designation. AMD is targeting more mainstream applications with these chips. The Ryzen 7 5700G is an 8-core/16-thread CPU with 4MB of L2 cache and 16MB of L3. Those CPU cores are mated to an 8 CU (Compute Unit) Radeon Vega graphics engine, and it has 24 lanes of PCIe Gen 3 connectivity. The 5700G's base CPU clock is 3.8GHz, with a maximum boost clock of 4.6GHz. The on-chip GPU can boost up to 2GHz, which is a massive uptick from the 1.4GHz of previous-gen 3000-series APUs.

The Ryzen 5 5600G takes things down a notch with 6 CPU cores (12 threads) and a smaller 3MB L2 cache while L3 cache size remains unchanged. The 5600G's iGPU is scaled down slightly as well with only 7 CUs. At 3.9GHz, the 5600G's base CPU clock is 100MHz higher than the 5700G's, but its max boost lands at 4.4GHz with a slightly lower GPU boost clock of 1.9GHz. In the benchmarks, the Ryzen 5 5600G and Ryzen 7 5700G both offer enough multi-threaded muscle for the vast majority of users, often besting similar Intel 11th Gen Core series chips, with highly competitive single-thread performance as well.

An anonymous reader quotes a report from Ars Technica: Today in the US market, a medium-sized battery EV already has 60-68 percent lower lifetime carbon emissions than a comparable car with an internal combustion engine. And the gap is only going to increase as we use more renewable electricity. That finding comes from a white paper (PDF) published by Georg Bieker at the International Council on Clean Transportation. The comprehensive study compares the lifetime carbon emissions, both today and in 2030, of midsized vehicles in Europe, the US, China, and India, across a wide range of powertrain types, including gasoline, diesel, hybrid EVs (HEVs), plug-in hybrid EVs (PHEVs), battery EVs (BEVs), and fuel cell EVs (FCEVs).

The study takes into account the carbon emissions that result from the various fuels (fossil fuels, biofuels, electricity, hydrogen, and e-fuels), as well as the emissions that result from manufacturing and then recycling or disposing of vehicles and their various components. Bieker has also factored in real-world fuel or energy consumption -- something that is especially important when it comes to PHEVs, according to the report. Finally, the study accounts for the fact that energy production should become less carbon-intensive over time, based on stated government objectives. The life cycle emissions of battery EVs in Europe today at 66-69 percent lower than a comparable gasoline-powered car. China is at 37-45 percent fewer emissions for BEVs, and India shows 19-34 percent.

As for fuel cell EVs (FCEVs), they "are only abut 26-40 percent less carbon-intensive than a comparable gasoline vehicle," notes Ars. "But if hydrogen was produced using renewable energy rather than steam reformation of natural gas, that number would jump to 76-80 percent -- even better than a BEV's numbers."

Along with other optimizations to benefit the Steam Deck, AMD and Valve have been jointly working on CPU frequency/power scaling improvements to enhance the Steam Play gaming experience on modern AMD platforms running Linux. Phoronix reports: It's no secret that the ACPI CPUFreq driver code has at times been less than ideal on recent AMD processors with delivering less than expected performance/behavior with being slow to ramp up to a higher performance state or otherwise coming up short of disabling the power management functionality outright. AMD hasn't traditionally worked on the Linux CPU frequency scaling code as much as Intel does to their P-State scaling driver and other areas of power management at large. AMD is ramping up efforts in these areas including around the Linux scheduler given their recent hiring spree while it now looks like thanks to the Steam Deck there is renewed interest in better optimizing the CPU frequency scaling under Linux.

AMD and Valve have been working to improve the performance/power efficiency for modern AMD platforms running on Steam Play (Proton / Wine) and have spearheaded "[The ACPI CPUFreq driver] was not very performance/power efficiency for modern AMD platforms...a new CPU performance scaling design for AMD platform which has better performance per watt scaling on such as 3D game like Horizon Zero Dawn with VKD3D-Proton on Steam." AMD will be presenting more about this effort next month at XDC. It's quite possible this new effort is focused on ACPI CPPC support with the previously proposed AMD_CPUFreq. Back when Zen 2 launched in 2019, AMD did post patches for their new CPUFreq driver that leveraged ACPI Collaborative Processor Performance Controls but the driver was never mainlined nor any further iterations of the patches posted. When inquiring about that work a few times since then, AMD has always said it's been basically due to resource constraints that it wasn't a focus at that time. Upstream kernel developers also voiced their preference to seeing AMD work to improve the generic ACPI CPPC CPUFreq driver code rather than having another vendor-specific solution. It's also possible AMD has been working on better improvements around the now-default Schedutil governor for scheduler utilization data in making CPU frequency scaling decisions.

A recovery mission off Vietnam's coast showed how advances in technology have given new reach to the Pentagon's search for American war dead. From a report: On a July morning in 1967, two American B-52 bombers collided over the South China Sea as they approached a target in what was then South Vietnam. Seven crew members escaped, but rescue units from the Air Force, Navy and Coast Guard were unable to find six other men, including a navigator from New York, Maj. Paul A. Avolese. It wasn't until last year that scientists scanning the seafloor found one of the B-52s and recovered Major Avolese's remains. "It was very humbling to be diving a site that turned out as hallowed ground, and realizing that maybe we were in a position to help bring closure back to families that had been missing this lost aviator," said Eric J. Terrill, one of two divers who descended to the wreck. Scientists say the recovery highlights a shift in the Pentagon's ability to search for personnel still missing from the Vietnam War.

For decades, such efforts have mainly focused on land in former conflict zones. But in this case, American investigators looked at an underwater site near Vietnam's long coastline, using high-tech robots. Their use of that technology is part of a larger trend. Robotic underwater and surface vehicles are "rapidly becoming indispensable tools for ocean science and exploration," said Rear Adm. Nancy Hann, who manages a fleet of nine aircraft and 16 research and survey vessels for the National Oceanic and Atmospheric Administration. "They have proven to be a force multiplier when it comes to mapping the seafloor, locating and surveying wrecks and other sunken objects, and collecting data in places not easily accessed by ships and other vehicles," Admiral Hann said.

One reason for the new focus on Vietnam's undersea crash sites is that many land-based leads have been exhausted, said Andrew Pietruszka, the lead archaeologist for Project Recover, a nonprofit organization. The group worked on the recent recovery mission with the Defense POW/MIA Accounting Agency, or D.P.A.A., the arm of the Pentagon tasked with finding and returning fallen military personnel. "Over time, a lot of the really good land cases and sites they've already done, they've already processed them," said Mr. Pietruszka, a former forensic archaeologist for D.P.A.A. who now works for the Scripps Institution of Oceanography at the University of California, San Diego. "Now the majority of sites that haven't been looked at are falling in that underwater realm," he added.

Google announced Monday it will build its own smartphone processor, called Google Tensor, that will power its new Pixel 6 and Pixel 6 Pro phones this fall. From a report: It's another example of a company building its own chips to create what it felt wasn't possible with those already on the market. In this case, Google is ditching Qualcomm. The move follows Apple, which is using its own processors in its new computers instead of Intel chips. And like Apple, Google is using an Arm-based architecture. Arm processors are lower power and are used across the industry for mobile devices, from phones to tablets and laptops.

Google Tensor will power new flagship phones that are expected to launch in October. (Google will reveal more details about those phones closer to launch.) That, too, is a strategy shift for Google, which in recent years has focused on affordability in its Pixel devices instead of offering high-end phones. And it shows that Google is again trying to compete directly in the flagship space against Apple and Samsung. The name Google Tensor is a nod to the name of Google's Tensor Processing Unit the company uses for cloud computing. It's a full system on a chip, or SoC, that the company says will offer big improvements to photo and video processing on phones, along with features like voice-to-speech and translation. And it includes a dedicated processor that runs artificial intelligence applications, in addition to a CPU, GPU and image signal processor. It will allow the phone to process more information on the device instead of having to send data to the cloud. Further reading: Google's New Pixel Phones Features a Processor Designed In-House.