Month: December 2020

State researchers in China recently developed a quantum computing system that is said to be 10 billion times faster than Google’s “Sycamore” machine. If so, it would be a significant milestone for the field.

Front: As far as we can tell, it’s true. After a quick look at the research paper and a look at the peer response, it is evident that the Chinese team managed to do something extraordinary here.

Here is the jargon from the paper itself:

Quantum computers promise to perform certain tasks that are believed to be impossible for classical computers to solve. Boson sampling is one such task and is considered a strong candidate to demonstrate the benefit of quantum computation. We perform a Gaussian boson scan by sending 50 indistinguishable single-mode squeezes into a 100-mode, low-loss interferometer with full connectivity and random matrix – all of the optical setup is phase-locked – and the output with 100 highly efficient singles -Sample mode samples. Photon detectors.

Basically, the researchers say they built a quantum computer machine that uses light to perform a very specific task (boson scanning) to demonstrate and measure its effectiveness.

Background: The reason this is important is because, in theory, quantum computers can solve really difficult problems. We’re talking about the kind of problems that physicists and computer scientists value that a classic machine would take thousands of years to solve.

In 2018, Google claimed it had developed the first machine that could demonstrate a “quantum advantage.” This just means that supposedly a quantum computer was able to do something that a classical computer either couldn’t do or couldn’t do in a reasonable time.

Google claimed that its system, a 53-qubit machine that shakes a quantum chip called “Sycamore”, could solve a specific problem that a supercomputer couldn’t solve. Unfortunately for Google, IBM was quick to deny this claim. According to Big Blue, it can solve the same problem on one of its classic supercomputers in just a few days – using algorithms that already existed at the time of Google’s announcement.

Take quickly: What China has done is completely different from what Google has done. In essence, China has built a machine that can only perform the experiment that demonstrates quantum superiority. In other words, it doesn’t actually solve any problems, which makes being a computer a bit voluntary.

The Google machine, on the other hand, is “programmable”. This means that in theory it could be tweaked to solve one or more problems.

That doesn’t mean what China did isn’t a breakthrough. Pushing the boundaries of what quantum science can achieve is the goal of everyone working in this field. The various laboratories around the world working on building quantum computing machines use different approaches because while the future holds promise for the field, we are still taking the first theoretical steps towards useful quantum computers.

China’s methods may have made the most recent breakthrough, but as Lu Chaoyang, the professor who led the experiment, told the Financial Times:

Building a quantum computer is a race between humans and nature, not between countries.

You can find more information on quantum computing in our primer here.

Published on December 4, 2020 – 18:47 UTC

As self-driving cars gain momentum in today’s automotive landscape, the issue of legal liability in the event of an accident has become more relevant.

Studies of the interaction between humans and vehicles have shown time and again that even systems for automating driving – such as adaptive cruise control, which keeps the vehicle at a certain speed and distance from the car in front – are anything but error-free.

Recent evidence suggests that drivers have limited understanding of what these systems can and cannot do (also known as mental models), which contributes to system abuse.

In the video above, a webinar on the dangers of advanced driver-assisted systems.

There are many issues worrying the self-driving car world, including imperfect technology and the lukewarm public acceptance of autonomous systems. There is also the question of legal liability. In particular, what are the legal responsibilities of the human driver and the automaker who built the self-driving car?

Trust and Accountability

In a study recently published in Humanities and Social Science Communications, the authors deal with the problem of excessive driver trust and the resulting system abuse from a legal perspective. They are examining what self-driving car manufacturers should legally be doing to ensure drivers understand how to use the vehicles appropriately.

One solution suggested in the study is to require buyers to sign end-user license agreements (EULAs), similar to the terms that an agreement would require when using new computer or software products. To get consent, manufacturers may use the ubiquitous touchscreen installed on most new vehicles.

The problem is that this is far from ideal or even safe. And the interface may not provide enough information to the driver, creating confusion about the nature of the consent forms and their implications.

The problem is, most end users don’t read EULAs: A 2017 Deloitte study shows that 91% of people agree to them without reading. In young people, the percentage is even higher. 97% agree without checking the terms.

In contrast to using a smartphone app, operating a car involves considerable safety risks, regardless of whether the driver is human or software. Human drivers must agree to take responsibility for the software and hardware results.

“Warning drowsiness” and distracted driving are also a cause for concern. For example, a driver who is upset after receiving continuous warnings might choose to just ignore the message. If the message is displayed while driving, it can be a distraction.

Given these limitations and concerns, the manner in which this consent is to be obtained is unlikely to fully protect automakers from their legal liability in the event the system fails or an accident occurs.

Driver training for self-driving vehicles can help drivers fully understand system capabilities and limitations. This needs to go beyond just buying the vehicle – recent evidence shows that even relying on the information provided by the dealer won’t answer questions.

Against this background, the way forward for self-driving cars will not go smoothly.

This article by Francesco Biondi, Assistant Professor of Human Kinetics at the University of Windsor, has been republished by The Conversation under Creative Commons license. Read the original article.

SHIFT is brought to you by Polestar. It’s time to accelerate the transition to sustainable mobility. That’s why Polestar combines electric driving with state-of-the-art design and exciting performance. Find out how.

Published on December 5, 2020 – 16:00 UTC

Have you always wanted to fly drones but have held back because of the costs? Drone prices keep falling as more drones fly. Affordable drones are now within the reach of most people and are no longer just a toy for people with high expendable incomes. These days, you can find good but cheap drones for less than $ 500 (or even less) if you grab one while it’s on sale. To make this easier, we’ve rounded up the best cheap deals on drones right now.

Today’s best drone deals

How to choose a drone

The right drone for you depends on what you want to do. A cheap drone – especially those that are under $ 250 – has the fewest features and may not have some features that may not make it the best solution for you.

In general (and this is not a hard and fast rule) a cheap drone will usually provide 720p video and a relatively low frame rate, typically 30 frames per second. While this will be fine for most of us, the video will lack that cinematic smoothness. Stepping up to a mid-range drone (usually in the $ 250 to $ 750 range) gives you 1080p video and often 60 fps, resulting in higher quality videos.

Most modern high-end drones offer 4K video, but you will likely have to spend more than $ 1,000 to get 4K video at 60 frames per second. If video quality is a primary concern, expect higher costs.

A cheap drone also lacks other useful features, including tracking, video and drone stabilization. In ideal flight conditions this is not a problem. However, you should take advantage of these features when you want to fly in different conditions or work without much user input.

We also found that many cheaper drones either don’t avoid obstacles or are not as good as more expensive drones. If you’re flying in wide open spaces, you don’t have a big problem here. However, if you plan to fly in areas with obstacles nearby, make sure that the drone you choose has sufficient obstacle avoidance capabilities.

Do drones make noises?

All drones make noise. Most describe it as a humming sound, similar to a bee. The propellers are spinning at extremely high speeds, and that is what makes this noise. It is most noticeable when you are closest to the drone, but you will hardly hear it if it is ever higher in the air.

Can you fly a drone at night

Most drones can be flown at night, although we would not recommend doing so until you have a lot of experience. We recommend keeping your drone in view when flying at night. Drones that can be deployed at night have lights so that they can be seen during night flight.

Remember that most drones don’t have night vision capabilities. Therefore, the video you record during the night flight will only be lit by available ambient lights like moonlight, street lights, etc.

Can you fly a drone in the rain

Most drone manufacturers advise against flying your drone in the rain. Most cheap drones are neither waterproof nor waterproof, so even small amounts of water can damage your drone, especially the motor and battery. The moisture creates a short circuit that causes your drone to stop working with little or no warning. As a rule, therefore, do not fly your drone in rain, fog or excessively humid conditions.

If you are trapped in these conditions, land as soon as possible and move your drone to a dry place, disconnect the battery and let it dry. You may also want to gently shake the drone itself to get water out of the inner case and let it dry as well. Use techniques similar to drying out a wet phone to dry out your drone. If this is your first time flying it, take some time to test the drone at low altitude before returning to normal flying.

Do drones have to be registered?

The Federal Aviation Administration requires that you register your drone based on your scheduled flight and tag your drone with your provided registration number. Any drone weighing more than 250 grams and less than 55 pounds must be registered, which covers nearly every drone currently sold in the market. This registration must be renewed every three years.

You must also have proof of registration with you at all times during the flight and you are not allowed to fly for commercial purposes according to the license terms. Drone flights are only permitted below 400 feet and class “G” or uncontrolled airspace. It is your responsibility to follow these rules.

In some states, drone pilots must comply with additional regulations. So be sure to check the laws of the state in which you are flying before you take off.

Do you need a license to fly a drone?

No license is currently required for recreational or hobby drone pilots. However, additional certifications may be required to fly for commercial purposes.

We strive to help our readers find the best deals on quality products and services, and we carefully and independently select what we cover. The prices, details and availability of the products and offers in this post are subject to change at any time. Make sure they are still valid before making a purchase.

Digital Trends can earn commissions on products purchased through our links, which supports the work we do for our readers.

Editor’s recommendations