Day: December 5, 2018

Posted by

Nathan Kirsch |

Tue, Dec 04, 2018 – 5:24 PM

The NVIDIA PhysX physics simulation engine is going open source after being introduced 10 years ago (also, thanks Ageia)! It’s available today as open source under the simple BSD-3 license according to a blog post that made the announcement. NVIDIA PhysX is a widely used physics SDK for many game tiles. In fact, it is integrated into the Unreal Engine (versions 3 and 4) and Unity3D game engines. It will be interesting to see what happens here njow that it goes open source.

The current whereabouts of He Jiankui—the scientist who claims to have engineered the world’s first genetically modified human babies—is unknown. Rumors are now circulating that he’s been detained by the Chinese government.

The last that anyone has seen or heard from He Jiankui was on Wednesday November 28, after he spoke in Hong Kong at the second International Summit on Human Genome Editing, the South China Morning Post reports.

The scientist is currently mired in an intense controversy after claiming to have produced the world’s first gene-edited babies. He said he used the CRISPR gene-editing system to modify the DNA of human embryos, resulting in the birth of twin girls with an alleged immunity to the AIDS virus. Because gene-editing is still in its nascent stage of development, He was criticized for conducting the experiment prematurely and for implanting the modified embryos in the mother’s womb. To make matters worse, the clinical trial was done in secret, and He failed to go through the normal channels, among other alleged improprieties.

As reported by Newsweek, the Hong Kong-based publication Apple Daily claims the embattled scientist was summoned back to Southern University of Science and Technology in Shenzhen, where he works, at the close of the Hong Kong summit. Chen is apparently under house arrest on campus and security guards have been stationed on the university grounds, Apple Daily reports.

These claims, however, have not been corroborated. A spokesperson for the the university told the South China Morning Post that, “Right now nobody’s information is accurate, only the official channels are,” adding that “We cannot answer any questions regarding the matter right now, but if we have any information, we will update it through our official channels.”

He, who is currently on unpaid leave, did not disclose the details of his research to the university. In a statement released last week, the university said He’s work “seriously violated academic and ethics standards,” prompting the establishment of an independent committee and investigation into the matter.

Likewise, the Chinese government has also condemned the work, saying the experiment “crossed the line of morality and ethics adhered to by the academic community and was shocking and unacceptable,” in the words of Xy Nanping, China’s vice minister of science and technology. On Thursday, the Chinese government claimed to shut down the gene-editing project and launched its own investigation.

The Chinese government has a habit of making people disappear for extended periods of time, the most recent example being Chinese actress Fan Bingbing, who went missing this past July. Fan issued a statement in early October, at which time she was formally charged with tax evasion and other offenses. Considered China’s highest-paid actress, Fan was ordered to pay 884 million yuan ($127 million) to avoid criminal prosecution. She made her first public appearance in late October after months of detention.

No evidence exists at this time to support the suggestion that He is being detained by the Chinese government, but it’s a possibility that has to be considered. His rogue experiment is seen by some as a tremendous embarrassment to China and its scientists, who are trying to shake accusations that China is now the unregulated Wild West of biomedical research.

Of course, detaining super-rich celebrities for tax evasion is not the same thing as detaining a scientist who’s gone rogue, but it wouldn’t be out of character for the Chinese government to make an example of He.

[South China Morning Post, Newsweek]

Russian state media published video earlier today of its latest laser weapons system, the Peresvet. The move comes just a day after the U.S. said it would likely pull out of a key arms treaty between the two countries, and the timing of the video’s release doesn’t appear to be a coincidence.

“Peresvet laser systems, based on new physical principles, entered combat service in [a] testing regime with the Russian armed forces,” Russian Defense Ministry’s newspaper said, according to an English translation.

When President Vladimir Putin made his annual address to Russia’s Federal Assembly back in March, he promised to deploy some high-tech weapons in the near future, and the new laser system got what you might call a “soft launch” back in July when the military first unveiled video of the device. But as tensions increase between the U.S. and Russia, it appears that Putin doesn’t want the West to forget that Russia is developing plenty of new weapons should the New Cold War turn hot.

Sputnik, one of the Kremlin’s propaganda media outlets, published video of the new weapon this morning, though it appears to be very similar to the video that was posted back in July.

What can the Peresvet laser system do? So far, Russia isn’t giving us any specifics. But the country wants to assure America that if the U.S. pulls out of the INF Treaty, as the U.S. State Department threatened to do yesterday, Russia is ready to go so far as to attack European countries if it must.

“If the INF treaty is destroyed, we won’t leave it without a response,” General Staff chief Valery Gerasimov said today in Russia while addressing military leaders. “You as military professionals must understand that the target for Russian retaliation won’t be U.S. territory but the countries where the intermediate-range missiles are deployed.”

The INF Treaty bans the deployment of new intermediate-range nuclear weapons with a range between 310 miles and 3,400 miles (500km and 5,500km)—the kind of high-powered missiles that would pose a threat to Russia if placed in nearby European countries like Poland and Romania.

The U.S., reportedly led by Donald Trump’s national security adviser, John Bolton, claims that Russia is already in violation of the INF Treaty and that Russia has 60 days to become compliant before the U.S. nullifies the landmark 1987 treaty. Secretary of State Mike Pompeo made the ultimatum yesterday in a meeting between NATO leaders in Brussels.

But Putin isn’t shying away from what he considers a massive provocation. President Putin called the potential violation of the treaty a “thoughtless step” and said that Russia would “react accordingly”

“Many other countries—about a dozen of them, probably—produce such weapons [in breach of the INF treaty], and Russia and the United States have limited themselves bilaterally,” President Putin said earlier today, according to the Financial Times.

“Now, apparently, our American partners believe that the situation has changed so much that the United States should have such a weapon. What is the answer from our side? Well it is simple: we will also do it then.”

The New Cold War sure looks a lot like the Old Cold War. But at least we have America’s best and brightest working on the problem. Right? Right?

[TASS and Sputnik]

Most lasers are either friendly tabletop devices or so small that you don’t even notice them. But if you want to do something special like image the structure of a delicate protein, you need a very short wavelength and very high-power laser. That means a free electron laser (FEL).

I love FELs, but they’re expensive. A laser produces one beam, and that beam is usually limited to one or two users at a time. So competition for beam time at FELs is fierce. That may be about to change, though. An unexpected experimental result at the Linac Coherent Light Source may be the key to making multiple beams from a single FEL undulator line.

Enough gibberish

OK, time for some physics-to-almost-English translation. Let’s start with some FEL goodness.

Light is emitted by charged particles when they’re slowed down or forced to change direction. In a free electron laser, blobs of electrons are shot through a series of magnets, called an undulator, that make the electrons wiggle or spiral. With each complete cycle of wiggling or spiraling, the electrons emit a photon.

The wavelength of the photon is dictated by the energy of the electrons and the spacing of the magnets. These are both design elements, so an engineer can set the wavelength of the laser by shifting magnets and/or changing the energy of the electrons.

The electrons travel in a bunch, which has to be very short to ensure that you get laser light. If the bunch is spread out, there is a time delay between the electrons. If the light emitted at the back of the bunch is not in phase with the light at the front, you get reduced output power. Anything that spreads the bunch is bad.

Within the bunch are micro-bunches. These are tiny conglomerates of electrons that are spread throughout the main bunch. If all the micro-bunches are lined up nicely abreast like runners at a starting line, you should get nice, high-quality radiation.

Putting this all together, a FEL is a long line of magnets. At one end, an electron gun and an accelerator shoot in regularly spaced bunches of high energy electrons. These electrons exit the other end, very tired after having given it their all. A pulse of X-ray light exits with each electron bunch.

This is also why one undulator gives one FEL beam.

Don’t turn corners

One way to get more FELs out of a single electron beam is to use several undulators. After the electrons exit one, they are diverted to the next undulator and so on until the electron energy is too low to produce useful radiation. This is pretty much the goal of every FEL, and many of them do this.

But the process involves making the electrons go around corners, which spreads the bunch. After that, you need to compress the bunch again. That sounds simple, but it’s not. Every time something in the first undulator changes, the magnets used to pull the bunch back together again need to be changed. Imagine if you had five or six undulators in a row—it would quickly become unmanageable.

The problem is that the micro-bunches don’t go around corners together; they simply slew to take the shortest possible path. The micro-bunches enter line abreast and exit line astern. Exactly the worst possible case.

The simple picture of electrons turning corners is not quite right, though. Researchers gave an electron bunch a little kick between undulator magnets at the Linac Coherent Light Source. The kick caused the electrons to suddenly (but only very slightly) change direction. This should have resulted in a substantial reduction in light from the FEL but didn’t. Lots of light was emitted along the new direction of the electron beam. This should not have happened because the bunch should have been too spread out.

Taking a closer look at the exact circumstances of the kick revealed the trick.

Electrons lack focus

The key turned out to be how the electron beam is prepared and maintained as it goes through the undulator. The beam focuses and defocuses periodically. If the kick happens in the defocus part of the cycle, the micro-bunches follow a trajectory that just happens to keep them almost in a line abreast.

The rotation is not perfect, with the bunch spreading out a little as it goes. However, the small amount of spreading gets corrected by the next undulator. The electrons at the front emit a little more radiation until they are back in the bunch. At the same time, the electrons at the back absorb a little radiation until they are caught up. Then, all the electrons start emitting lots of energy as radiation.

What does this all mean?

The researchers compare their theoretical results to the data from the Linac Coherent Light Source FEL and showed that they obtain pretty good agreement. The researchers also modeled a kind of bent FEL, where the beam is periodically kicked and sent into a new undulator. They showed that up to five FEL beams could be obtained from a single electron beam.

That is an important finding. FEL facilities are expensive, and user time is limited. That’s because you can’t divide a single FEL beam between multiple user stations very easily. This also makes it very difficult for researchers to get time at FEL facilities. There simply aren’t enough hours in the day.

With this latest development, it may be possible to modify existing FEL facilities to support larger numbers of users at the same time. And that’s a win for everyone.

Physical Review X, 2018, DOI: 10.1103/PhysRevX.8.041036 (About DOIs)

Listing image by UCLA Particle Beam Physics Lab

Innodisk has released a graphics adapter in an M.2 form-factor. The card supports resolutions of up to 1920×1080 at 60 Hz or 3840×2160 at 30 Hz and is designed for various ultra-compact and embedded applications that require industrial temperature ranges.

The Innodisk M.2-2280 graphics adapter is based on the Silicon Motion SM768 controller, presumably with 256 MB of embedded memory. For outputs the card offers three headers for HDMI, LVDS, and DVI-D signals. Under the hood the graphics processor has a 128-bit 2D engine, an ARM Cortex-R5 general-purpose core, a video decoder supporting a variety of formats (including H.264 MVC/AVS+, H.263, MPEG-4, MPEG2, M-JPEG, RealVideo, VC-1, & Theora), and a Content Adaptive Technology (CAT) engine for framebuffer compression. The SM768 can use a PCIe 2.0 x1/x2 or USB 3.0 interface, so it can support a variety of applications. Meanwhile, the version of the chip used by Innodisk can handle industrial grade temperatures from -40°C to +85°C.

Innodisk’s M.2 graphics adapter can be used to add a display output to systems that physically cannot accommodate a graphics card and do not have enough connectors for monitors. Keeping in mind that such applications are pretty diverse and can run a variety of OSes, the adapter supports both Microsoft Windows and Linux-based operating systems.

The manufacturer does not disclose MSRP of its M.2 graphics adapter, but companies do not usually reveal pricing of special-purpose hardware in general. The product is not listed yet, so it is unclear whether Innodisk can ship it straight away, or it will take some time before it polishes off its software.

Source: Innodisk (via Hermitage Akihabara)