SoylentNews

Freeman writes:

https://arstechnica.com/gaming/2023/08/getting-aaa-games-working-in-linux-sometimes-requires-concealing-your-gpu/

Linux gaming's march toward being a real, actual thing has taken serious strides lately, due in large part to Valve's Proton-powered Steam Play efforts. Being Linux, there are still some quirks to figure out. One of them involves games trying to make use of Intel's upscaling tools.

Intel's ARC series GPUs are interesting, in many senses of the word. They offer the best implementation of Intel's image reconstruction system, XeSS, similar to Nvidia's DLSS and AMD's FSR. XeSS, like its counterparts, utilizes machine learning to fill in the pixel gaps on anti-aliased objects and scenes. The results are sometimes clear, sometimes a bit fuzzy if you pay close attention. In our review of Intel's A770 and A750 GPUs in late 2022, we noted that cross-compatibility between all three systems could be in the works.

That kind of easy-swap function is not the case when a game is running on a customized version of the WINE Windows-on-Linux, translating Direct3D graphics calls to Vulkan and prodding to see whether it, too, can make use of Intel's graphics boost. As noted by Phoronix, Intel developers contributing to the open source Mesa graphics project added the ability to hide an Intel GPU from the Vulkan Linux driver.

[...] Relying on upscaling to bolster performance, especially at lower resolutions, may be unwise. But nearly every major game release brings with it news of which vendor's upscaling system is included or preferred. It's still impressive how many games simply run at all on an OS for which they were never built, but it might never stop being a tricky challenge.

Related:
Apple Has a Proton-Like Game Porting Toolkit for Getting Windows Games on Mac - 20230612

Original Submission

canopic jug writes:

The site It's FOSS is reporting that India's Defense Services are switching to GNU/Linux, ditching an insecure legacy operating system, with an August 15 deadline. Little is known about their home spun distro except that it seems to be based on Ubuntu.

What's Happening: According to a recent report, the Defence Ministry of India has decided to replace Windows with an in-house developed Linux distro called 'Maya' on all computers that are connected to the Internet.

Also reported at The Hindu, Defence Ministry to switch to locally built OS Maya amid threats, which explains that this move is a reaction to increasingly successful attacks against a certain, pervasive, desktop legacy operating system. x

Currently, Maya is being installed only in Defence Ministry systems and not on computers connected to the networks of the three Services. On this, the official said the three Services had also vetted it and would adopt it on service networks as well soon. The Navy had already cleared it and the Army and the Air Force were currently evaluating it, the official added.

Maya was developed by government development agencies within six months, the official said. Maya would prevent malware attacks and other cyberattacks which had seen a steep increase, the official noted.

However, the attacks in and of themselves are less of a problem than the fact that a large, and increasing, number of them are successful against that aging legacy desktop operating system.

For India to pull this off successfully, they must study how their opponent has maneuvered over the years against GNU/Linux deployments and in particular look at case studies like Kerala, Munich, Lower Saxony, Vaasa, and Turku. India's opponent in this move has had many programmes, years ago one was EDGI, and a long standing mandate that "under NO circumstances lose against Linux".

Previously:
(2018) German Documentary on Relations Between Microsoft and Public Administration Now Available in English
(2018) German State of Lower Saxony Plans to Switch From Linux to Windows
(2017) Munich Switching From Linux to Windows 10
(2017) Linux Champion Munich Takes Decisive Step Towards Returning to Windows
(2016) Draft Report Doesn't Say -Which- Software is Causing Problems in Munich
(2016) Munich: The High Cost of Having Committed to Closed-Source Software
(2014) Another German Town Says It Has Completed Its Switch To FOSS

Original Submission

Freeman writes:

https://arstechnica.com/tech-policy/2023/08/scam-victim-cant-stop-endless-stream-of-unwanted-amazon-packages/

Amazon ships more than a million packages daily, but there's at least one person in a million who frowns when she encounters a smiling box placed on her doorstep.

A Canadian woman, Anca Nitu, told CBC that over the past two months, more than 50 packages have arrived at her home. Each package contained a return slip and a pair of shoes from an Amazon buyer located in North America who wrongly shipped their rejected shoes to Nitu's address.
[...]
Amazon said that typically the company advises any recipient of an unwanted package to fill out a Report Unwanted Package form. That Amazon page says to report any unsolicited packages "immediately," confirming that "third-party sellers are prohibited from sending unsolicited packages to customers."

For Nitu, her worries won't necessarily end, even if the packages ever do stop coming. She told CBC that she has no idea whether the Amazon sellers that are using her information to ship unwanted return items are doing anything else with her information. She also worries that if Amazon doesn't unlink her name from the seller accounts, she could one day be charged Amazon seller fees.

Related:
'It's an absurdity': Oak Park woman says unwanted shoes keep showing up on doorstep

Original Submission

looorg writes:

Other peoples code might look good at a glance but is wrong at least half the time. AI or help sites don't offer as much help as one would think.

"Among other findings, the authors found ChatGPT is more likely to make conceptual errors than factual ones. "Many answers are incorrect due to ChatGPT's incapability to understand the underlying context of the question being asked," the paper found."

""Stack Overflow's annual Developer Survey of 90,000 coders recently found that 77 percent of developers are favorable of AI tools, but only 42 percent trust the accuracy of those tools. OverflowAI developed with community at the core and with a focus on the accuracy of data and AI-generated content."

https://www.theregister.com/2023/08/07/chatgpt_stack_overflow_ai/
https://arxiv.org/abs/2308.02312

Original Submission

PiMuNu writes:

Fermilab's g-2 experiment has released a new round of data. Physicists now have a brand-new measurement of a property of the muon called the anomalous magnetic moment that improves the precision of their previous result by a factor of 2. The result is in tension with Standard Model predictions.

An international collaboration of scientists working on the Muon g-2 experiment at the U.S. Department of Energy's Fermi National Accelerator Laboratory announced the much-anticipated updated measurement on Aug. 10. This new value bolsters the first result they announced in April 2021 and sets up a showdown between theory and experiment over 20 years in the making.

"We're really probing new territory. We're determining the muon magnetic moment at a better precision than it has ever been seen before," said Brendan Casey, a senior scientist at Fermilab who has worked on the Muon g-2 experiment since 2008.

Press release:
https://news.fnal.gov/2023/08/muon-g-2-doubles-down-with-latest-measurement/

Paper:
https://muon-g-2.fnal.gov/result2023.pdf

A seven-minute video that provides additional information about muons and the new result by the Muon g-2 collaboration.

Original Submission

Arthur T Knackerbracket has processed the following story:

"The sun is more surprising than we knew," said Mehr Un Nisa, a postdoctoral research associate at Michigan State University. "We thought we had this star figured out, but that's not the case."

Nisa, who will soon be joining [Michigan State University's] faculty, is the corresponding author of a new paper in the journal Physical Review Letters that details the discovery of the highest-energy light ever observed from the sun.

The international team behind the discovery also found that this type of light, known as gamma rays, is surprisingly bright. That is, there's more of it than scientists had previously anticipated.

Although the high-energy light doesn't reach the Earth's surface, these gamma rays create telltale signatures that were detected by Nisa and her colleagues working with the High-Altitude Water Cherenkov Observatory, or HAWC.

[...] "We now have observational techniques that weren't possible a few years ago," said Nisa, who works in the Department of Physics and Astronomy in the College of Natural Science. "In this particular energy regime, other ground-based telescopes couldn't look at the sun because they only work at night," she said. "Ours operates 24/7."

[...] "After looking at six years' worth of data, out popped this excess of gamma rays," Nisa said. "When we first saw it, we were like, 'We definitely messed this up. The sun cannot be this bright at these energies.'"

[...] The gamma rays that Nisa and her colleagues observed had about 1 trillion electron volts, or 1 tera electron volt, abbreviated 1 TeV. Not only was this energy level surprising, but so was the fact that they were seeing so much of it.

[...] "This shows that HAWC is adding to our knowledge of our galaxy at the highest energies, and it's opening up questions about our very own sun," Nisa said. "It's making us see things in a different light. Literally."

Journal Reference:
A. Albert et al, Discovery of Gamma Rays from the Quiescent Sun with HAWC, Physical Review Letters (DOI: 10.1103/PhysRevLett.131.051201)

Original Submission

Arthur T Knackerbracket has processed the following story:

NASA hopes to launch a near-infrared laser transceiver to test a system that could one day be used to communicate with astronauts on Mars.

The Deep Space Optical Communications (DSOC) experiment will head off into the void with Psyche, an asteroid-chasing probe that is scheduled to blast off on October 5. While they journey toward 16 Psyche, a metal-rich asteroid, the DSOC system will spend two years attempting to communicate via laser with two ground stations in Southern California.

NASA believes DSOC's near-infrared lasers can trounce the data transmission speeds achieved using radios.

"DSOC was designed to demonstrate 10 to 100 times the data-return capacity of state-of-the-art radio systems used in space today," enthused Abi Biswas, DSOC's project technologist working at NASA's Jet Propulsion Laboratory. "High-bandwidth laser communications for near-Earth orbit and for Moon-orbiting satellites have been proven, but deep space presents new challenges."

NASA's most recent Mars rover, Perseverance, can communicate with orbiters at two megabits per second. The Mars Reconnaissance Orbiter can chat to Earth at between 0.5 to 4 megabits per second.

Improving those speeds by a factor of between 10 and 100 with lasers therefore has an obvious benefit – even if the pesky limit that is the speed of light means it won't allow synchronous comms with the Red Planet.

[...] "Every component of DSOC exhibits new technology – from the high-power uplink lasers to the pointing system on the transceiver's telescope and down to the exquisitely sensitive detectors that can count the single photons as they arrive," explained Bill Klipstein, the DSOC project manager at JPL. "The team even needed to develop new signal-processing techniques to squeeze information out of such weak signals transmitted over vast distances."

Original Submission

fliptop writes:

A team of researchers from British universities has trained a deep learning model that can steal data from keyboard keystrokes recorded using a microphone with an accuracy of 95%:

When Zoom was used for training the sound classification algorithm, the prediction accuracy dropped to 93%, which is still dangerously high, and a record for that medium.

Such an attack severely affects the target's data security, as it could leak people's passwords, discussions, messages, or other sensitive information to malicious third parties.

Moreover, contrary to other side-channel attacks that require special conditions and are subject to data rate and distance limitations, acoustic attacks have become much simpler due to the abundance of microphone-bearing devices that can achieve high-quality audio captures.

Originally spotted on Schneier on Security.

Reference: Joshua Harrison, Ehsan Toreini, and Maryam Mehrnezhad, A Practical Deep Learning-Based Acoustic Side Channel Attack on Keyboards. arXiv:2308.01074v1

Original Submission

evilcam writes:

Science Alert has a story on a paper suggesting the use of asteroids for space stations.

The basic idea of turning an asteroid into a rotating space habitat has existed for a while. Despite that, it's always seemed relatively far off regarding technologies, so the concept hasn't received much attention over the years.
...
David W. Jensen, a retired Technical Fellow at Rockwell Collins ... released a 65-page paper that details an easy-to-understand, relatively inexpensive, and feasible plan to turn an asteroid into a space habitat.
...
Dr. Jensen breaks the discussion into three main categories – asteroid selection, habitat style selection, and mission strategy to get there (i.e., what robots to use).
...
After a relatively in-depth selection process, Dr. Jensen decided on one in particular as a good candidate – Atira. This S-type asteroid has an entire class of asteroids named after it. Atira comes in at about a 4.8 km diameter and even has its own moon – a 1 km diameter asteroid that orbits it closely.
...
He eventually settled on a torus as the ideal habitat type and then dives into calculations about the overall station mass, how to support the inner wall with massive columns, and how to allocate floor space. All important, but how exactly would we build such a massive behemoth?
Self-replicating robots are Dr. Jensen's answer. The report's third section details a plan to utilize spider robots and a base station that can replicate themselves. He stresses the importance of only sending the most advanced technical components from Earth and using materials on the asteroid itself to build everything else, from rock grinders to solar panels.

The numbers seem wildly out to me but I'm not an expert by any stretch. Feels as feasible as living on Mars...

Original Submission

Freeman writes:

https://arstechnica.com/space/2023/08/after-a-scare-nasas-mars-helicopter-safely-takes-to-the-skies-again/

By the standard of some of its previous flights, the most recent voyage of NASA's intrepid Ingenuity helicopter on Mars was nothing special. Over a period of 24 seconds, the small helicopter rose to an altitude 5 meters above the red planet's dusty surface and then touched back down in the same spot.

[...] Mission success for Ingenuity was completing five relatively short flights. However, since its first test flight in April 2021, the helicopter has exceeded all expectations by flying more than 50 different sorties across Mars and surviving long and dark winters.

[...] "Since the very first flight, we have included a program called 'LAND_NOW' that was designed to put the helicopter on the surface as soon as possible if any one of a few dozen off-nominal scenarios was encountered," Teddy Tzanetos, team lead emeritus for Ingenuity, said. "During Flight 53, we encountered one of these, and the helicopter worked as planned and executed an immediate landing."

Original Submission

Arthur T Knackerbracket has processed the following story:

US scientists responsible for a historic nuclear fusion breakthrough say they have repeated the feat—this time achieving a greater yield of energy.

The Lawrence Livermore National Laboratory stunned the world in December when it announced it had carried out an experimental nuclear reaction that put out more energy than was put into it, a holy grail of science in the quest for unlimited, clean power to end the era of fossil fuels.

"We can confirm the experiment produced a higher yield than the December 2022 experiment," public information officer Paul Rhien said Monday in an emailed statement, without disclosing specific figures.

He added the California lab planned to report the results at upcoming scientific conferences and in peer-reviewed publications.

[...] Nuclear fusion has been touted by its supporters as a clean, abundant and safe source of energy that could eventually allow humanity to break its dependence on coal, crude oil, natural gas and other hydrocarbons driving a global climate crisis.

However, there is still a long way to go before fusion is viable on an industrial scale, providing power to homes and businesses.

[...] During December's experiment, the lab used 192 ultra-powerful lasers to deliver 2.05 megajoules of energy to a tiny capsule smaller than a pea containing isotopes of hydrogen. It produced 3.15 megajoules of fusion energy output.

While the result was a net energy gain, 300 megajoules of energy was needed from the electrical grid to power the lasers.

Original Submission

Arthur T Knackerbracket has processed the following story:

Just because EV technology is rapidly advancing doesn't mean it's without its flaws. Electric vehicle adoption in the U.S. remains relatively low, and for a good reason — many of the biggest remaining problems are considered deal breakers by buyers and will need to be fully remedied before EVs become the default option for most people. The trouble is, solutions for these problems are not always straightforward, taking years of work and potentially billions of dollars to fix, and that's if they can be fixed at all.

Here are the most important issues that EV designers are still working to solve.

Perhaps the biggest hurdle facing the EV industry is the lack of accessible charging infrastructure available to drivers. America has had over 100 years to develop its network of gas stations, but if EVs are to become widely adopted, an equally comprehensive EV charging network needs to be developed within the space of the next decade. The government does recognize this, and in 2022, Congress passed a bill to fund 500,000 new electric vehicle chargers, with a focus on delivering infrastructure to more remote, rural parts of the country.

[...] A closely related problem to the insufficient charging infrastructure is range anxiety. Even though the longest-range EVs today can travel more than 300 miles on a single charge, a patchy charging network and the sometimes inconsistent range indicators on EVs can make longer trips a stressful experience. Most people, understandably, would rather not add any extra stress to their everyday lives and choose to dismiss buying an electric car altogether as a result.

[...] Even if EV chargers end up becoming as ubiquitous as gas stations, there's still the issue of the amount of time it takes to fully charge a car. A range of chargers are available for both home and public charging, with the fastest of those being able to deliver over 100 miles of range in 10-20 minutes, but that's still significantly slower than a gas station stop. Those top-spec Level 3 chargers can't be installed at home either. To make matters worse, not every EV supports the same level of fast charging, so depending on the model of your car, you might not be able to make use of those higher-speed chargers even if you have one nearby.

There's a transition in car computing hardware from roughly 100 little modules all over the car, each with a tiny amount of computing, to, "domain controllers" that each manage a lot of I/O using more powerful computing. MotorTrend gives an overview here, https://www.motortrend.com/features/software-defined-vehicle-hardware-domain-controller

A few of the dozen or so domains include:

Gateway
This central hub connects to the various domains, serving as a connection port for service and managing much of the data-privacy and cybersecurity functionality by demanding multi-factor authentication and other countermeasures that comprise a firewall against malware attacks and attempts to hack into vehicle controls or data. The gateway continuously monitors the various vulnerability fronts (USB ports, SD card slots, connectivity ports, etc. ).

Vehicle Experience
This domain informs and entertains the vehicle occupants, providing infotainment, augmented and virtual reality, onboard gaming or shopping, etc. Sound systems and screens are managed by this domain, often communicating with offboard entertainment providers via the connectivity domain.

Connectivity
This domain oversees data communication between the car and the cloud, the vehicle manufacturer, telematics systems (e.g., the various global-positioning satellite systems), the road infrastructure and other road users (aka V2X—vehicles, pedestrians, cyclists, etc. ), charging stations, home or office, and myriad other nodes on the global "internet of things." This communication can be via cellular networks, WiFi, Bluetooth, wired connections, and other means. This domain serves as the conduit for over-the-air updates of software and firmware.

Paraphrasing an auto technician training site (can't find reference now):
Mechanical problems are easy/fast to diagnose (they can be seen), but time-consuming / hard to fix.
Electrical/electronic problems are hard/slow to diagnose, but fast/easy to fix.

What are the chances that the car companies will give more than lip service to the questions of service/diagnosis for all this electronics?

Original Submission

Freeman writes:

https://arstechnica.com/science/2023/08/lots-of-earth-mass-rogue-planets-could-be-found-by-nasas-roman-telescope/

Planets that go rogue orbit no star. They wander the vacuum of space alone, having been kicked out of their star systems by gravitational interactions with other planets and stars. Nobody really knows how many rogue planets could be out there, but that may change in a few years.

Researchers from NASA's Goddard Space Flight Center and Osaka University in Japan have used the phenomenon of gravitational microlensing to estimate the number of rogue planets that could be revealed in the heart of the Milky Way. They analyzed data from the Microlensing Observations in Astrophysics (MOA) survey that searched for gravitational microlensing events from 2006 to 2014 to figure out how many more of these events we could expect to find with NASA's upcoming Nancy Grace Roman Space Telescope.

[...] "Gravitational microlensing enables us to study a variety of objects with masses ranging from that of exoplanets to black holes," the researchers said in the first of two studies soon to be published in The Astronomical Journal. The second study can be found here.

Original Submission

owl writes:

https://chipsandcheese.com/2023/08/04/sandy-bridge-setting-intels-modern-foundation/

Processor companies typically iterate off proven designs, and for good reason. Changing too many things at once introduces a lot of risk. A lot of changing parts makes it hard to get a good picture of overall performance, so it also makes tuning difficult. Pentium 4 and Bulldozer made clean breaks from prior architectures, and both were unsuccessful designs. But sometimes, making a clean break does pay off.

Sandy Bridge is one such case. It inherited architectural features from Intel's prior P6 and Netburst architectures, but can't be considered a member of either line. Unlike Netburst, Sandy Bridge was so successful that Intel's high performance cores to this day can trace their lineage to back to it. In the early 2010s, Sandy Bridge was so successful that Intel went completely unchallenged in the high end CPU market. And as a testament to the architecture's solid design, Sandy Bridge CPUs still deliver enough performance to remain usable across a range of everyday tasks.

Original Submission