In April, Google announced that it would comply with feedback given from the European Commission on the way that Android carries itself with Chrome and Google Search. In order to make the playing field fair for other web browsers and search engine providers, it will kindly and easily offer the user to download and install other web browsers and search engines.
Source: Android Police
The update was set to hit over those coming weeks and we're already starting to see the pop-up prompt at least one user to install extra search engines and mobile browsers.
One of the readers over at Android Police has revealed that the change Google announced was already taking effect, albeit not to everyone in Europe just yet, but it should be rolling out over the coming days.
Once you do get the update, a new prompt will show itself once your Google Play Store is updated and will ask you whether you'd like to install any other search engines or browsers outside of Google's offerings.
Samsung’s A-series of midrange phones have three new members: the Galaxy A50, A20 and A10E. If you want a new Galaxy phone, these are some of your cheapest options, sitting just above Samsung’s entry-level Galaxy J line. The A-series is much cheaper than a top tier Galaxy phone. The Galaxy S10 Plus, for instance, costs $1,000 and the supposedly upcoming Galaxy Fold (Samsung delayed its launch due to a number of screen issues on its pre-production units) is priced at nearly $2,000.
The most advanced of the new trio is the Galaxy A50, which we got an early glimpse of in February at MWC 2019. It costs $350 and features a 6.4-inch display, a 4,000-mAh battery and 64GB of built-in storage with the option to expand up to 512GB. On the back are three cameras, which includes a wide-angle lens as well as a “depth lens,” which is used to take portrait shots with blurry, dramatic backgrounds.
The $250 Galaxy A20 has a 6.4-inch screen, a 13-megapixel rear camera and the same 4,000-mAh battery as the A50. Lastly, the cheapest of the bunch of the Galaxy A10E. At $180, it features a 5.83-inch display, an 8-megapixel camera and a 3,000-mAh battery. Both of the phones have just 32GB of onboard memory, but you can expand storage with a microSD card (not included).
All three phones run Android 9.0, with Samsung’s new One UI layered on top. Though exact availability dates for the Galaxy A20 and A10E have not been released, the Galaxy A50 will be available on June 13.
These days, sleep monitors come in all shapes and sizes, from wearables to bedside pods. None of them are more interesting than those that rest below your head, however. Today's smart pillows are designed to track your sleeping habits, help you avoid snoring problems, and lull you to sleep with a range of sounds without disturbing those around you. Opting for a smart pillow is a no-brainer — it's choosing the right model that's tricky.
Thankfully, through continual testing and ample research, we've rounded up some of the best smart pillows on the market. It begins with the iSense Sleep pillow, which has an excellent app for monitoring the nuances of your sleep cycle, but there are five other models to choose from if you're in the market for something different.
iSense Sleep Smart Pillow
The iSense Sleep Smart Pillow might be one of the least-intrusive options available, yet it still provides a slew of useful info regarding your sleeping habits. The tracker pairs with an app that shows you how well you slept, and provides data pertaining to your breathing, heart rate, and overall quality of sleep. The app does a good job of displaying this info in the form of numbers and charts, too, and the pillow itself use cross-cut memory foam cube filling that can accommodate a range of sleeping positions. It even gives you a simplified sleep score, just in case you're looking for the abridged metrics.
Goodnite Smart Anti-Snore Pillow
If you snore, you likely know how difficult it can be to alleviate the issue. Thankfully, this Goodnite pillow uses sensors to detect your head position and any signs of snoring. If the device thinks you're snoring, the pillow will gently inflate, pushing your head to the side and helping open airways that will reduce snoring. The pillow also works in tandem with a mobile app, which tracks your snoring so you can see if you are actually getting better. If you find that changing your head position helps prevent you from snoring, then Goodnite's aptly named Anti-Snore Pillow is likely right for you.
10Minds Motion Pillow
A supplemental pad and memory foam pillow combination, this 10Minds model is similar to the Goodnight snoring pillow but with a more traditional design. The pillow uses four airbags that inflate and reposition your head if it detects snoring, and the pillow cover is designed to disperse sweat and heat more efficiently so you stay cooler. The accompanying mobile app provides additional functionality, including the ability to record your snoring and adjust inflation levels, as well as charts for tracking your snoring habits over time.
Dreampad
The Dreampad is a pillow built with audio in mind. The pad's speakers are designed to provide quiet, clear audio that only you can hear, even with someone else resting beside you. The embedded speakers connect to the accompanying Dreampad app via Bluetooth — though, cable connections are possible — and you can choose from 10 different audio tracks that are specifically designed to help you fall asleep. You can also adjust their volume and playtime until you've created the perfect program to help you drift away. It's a unique approach to sleep that could be exactly what you need.
Zeeq Smart Pillow
The Zeeq pillow does a little bit of everything. First off, there's a snoring function that detects snoring and prompts vibrations, which encourage you to change your position without waking up. There's also a wireless speaker you can use to play music or sleep tracks, as well as tracking features that are designed to measure your restlessness and pinpoint any disturbances in your sleep cycle. There's even an in-pillow alarm clock and an array of memory foam clusters, the latter of which help accommodate various sleeping positions.
The problem with being a jack-of-all-trades is that it can be difficult to do any them especially well, which is why the Zeeq isn't higher up on our list. Other pillows do a better job with audio, snore control, and sleep tracking. That said, Zeeq's offering is a solid choice for those who want a little bit of everything, or want to experiment with various sleeping aids.
Moona
Moona is a crowdfunded pillow pad that combines a pad, bedside pod, and an app to give you the most accurate sleep data possible. Working together, they monitor your sleep habits, provide gentle wake-up notifications, and generally learn your sleep habits to help you create the healthiest schedule for your cycle. You can even delve into things like temperature profiles, and the pad includes a small water pump that can help cool you down by cycling water through the pad when things get too hot. It's an ambitious project that's certainly worth a look as long as you don't mind replacing your alarm clock.
The Google Pixel 4 just keeps on leaking these days, even if its launch is only expected to happen in October. Following some controversial-looking renders that showed it would have a huge camera hump on the back with a similar design to that of the upcoming iPhones’, today we bring you three exclusive shots of the Pixel 4, courtesy of Shivam Pandya.
We have to warn you that the prototype captured here seems to be inside one of those bulky cases that are meant to obscure as much of its design as possible. Still, we can see the right-aligned oval punch hole cutout in the display for two camera sensors, as well as a display ‘chin’ that’s significantly smaller than what we had in past Pixels.
That means there’s no more front-facing speaker inside that bezel, instead there’s a down-firing one nearby. Overall all four bezels seem to be pretty much the same size, and symmetry from this point of view isn’t something Google’s ever done before.
Note that these shots contradict the previously outed renders somewhat, because those had a visible earpiece lower than the rim of the top side, whereas these don’t. The last picture to the right has the new Pixel next to what seems to be a Pixel 2 XL, so it is very likely that what we’re looking at here is the Pixel 4 XL, not its smaller sibling.
Keep in mind that Google has been rumored to have three separate design teams working independently on the Pixel 4 and 4 XL, without any coordination between them. In the end the company’s management will pick one design and go with that. This is apparently Google’s way of ensuring that it doesn’t end up with any more design flops in its smartphone lineup, following the incredibly divisive looks of the Pixel 3 XL with its gigantic notch and the Pixel 3 with its gigantic bezels in a world that’s almost entirely gone bezel-less now.
Photo by Michael Kovac/Getty Images for Vanity Fair
Twitter co-founder and Medium CEO Ev Williams said he got excited about the Web because he thought that with more information at our fingertips it would make people smarter. But the rising popularity of social media sites also created an environment where people have become wired to respond to short-term feedback such as the numbers of likes on a tweet or an Instagram post.
“We get hooked on that just like we get hooked on sugar,” he said on stage at Recode’s annual Code Conference in Scottsdale, Arizona.
When Williams looked at the cycle before Donald Trump was elected president, he doesn’t think people are becoming more intelligent. More information doesn’t make you smarter, he said, if people don’t know how to digest or contextualize it.
“I think part of it is just the limits of the human attention span,” he said.
Social media sites, including Twitter, have been trying to foster “healthier” conversations on their platforms in what can be a space filled with bullies and trolls. Studies have also linked social media to depression and anxiety.
To undo the unhealthy nature of instant feedback, Williams said that people need to build systems that don’t emphasize likes or followers. Photo-sharing site Instagram has been testing the ability for users to hide their “likes.”
The problem, Williams said, is that people have combined the idea of connecting with others with social competition. But that system can burn out users.
“That social competition at all times is exhausting emotionally,” he said.
On this episode of Digital Trends Live, host Greg Nibler and DT Producer Adrien Warner dig into the trending tech stories of the day, including highlights from E3, the Uber Elevate Summit, Portal's new devices, the Beyond Burger's arrival in stores, and a very special goodbye to the Mars Rover.
Later, Nibler welcomes Todd Weaver, founder and chief executive officer of Purism, to discuss the company's dedication to online personal data privacy and security.
DT Editor-in-Chief Jeremy Kaplan then joins us from New York to speak with Neil Sweeney, founder and CEO of Killi, about how to control and monetize your own digital identity, and how to make money from your first-party data.
We're also joined by Michael Mansouri, senior technology adviser at Hawkeye Systems, Inc., to talk about the company's military-grade 360-degree camera, and taking 3D imaging to the next level using A.I.
Finally, we continue our coverage of E3 with DT correspondent Riley Winn, including a conversation with DJWheat (aka Marcus Graham), director of creator development at Twitch.
The AI and ML deployments are well underway, but for CXOs the biggest issue will be managing these initiatives, and figuring out where the data science team fits in and what algorithms to buy versus build.
IBM is aiming to use artificial intelligence inside of its Watson Studio data science platform to automate data prep and some of the drudgery needed before rolling out enterprise AI.
The data prep and governance tools are included in AutoAI. AutoAI will be core to Watson Studio and its efforts to free up data scientists to focus on models. According to Forrester, 60% of respondents cited data quality as a big hurdle to deploying AI.
Analytics has evolved from the basics — visualizations, historicals and dashboards — to the more complex with recommendations and predictions of outcomes. Now it’s time to step it up and get prescriptive.
Watson Studio is combining AutoAI along with its Watson Machine Learning to automate core steps in deploying AI. AutoAI is available in Watson Studio on IBM Cloud. Key features include:
Automation of data preparation and preprocessing.
Model development and feature engineering.
Hyperparameter optimization to build data science and AI models.
A suite of model types for data science including gradient boosted trees and processes to experiment with machine learning.
IBM Neural Networks Synthesis (NeuNeuS), which is in open beta. NeuNueS allows data scientists to optimize speed or accuracy and track model training.
For IBM, AutoAI is part of a Watson Studio build out as well as a broader data science portfolio that includes IBM Watson Assistant and Discovery and Watson Machine Learning.
Along with the Honor 20-series, the company also announced two tablets for the Indian market named Honor Pad 5 and they come in two flavors – 8″ and 10.1″.
The 8-inch model is the more powerful option with Kirin 710 on board, 1920 x 1200px resolution screen and either 3GB or 4GB of RAM.
The slate’s internal storage is 32GB or 64GB and it has single 8MP cameras on the front and back. It runs on Android 9.0 Pie with Magic 2.0 UI on top and interestingly, it offers stereo loudspeakers as well.
The 10.1-inch tablet, on the other hand, runs on slower and older Kirin 659 SoC with either 3GB RAM and 32GB storage or 4GB/64GB RAM combo.
It packs the same 1920 x 1200px resolution LCD but runs the older Android 8.0 Oreo with EMUI 8.0 on top. Actually, this is a re-branded MediaPad T5 for the Indian market.
Both models are offered only in Glacial Blue. The 8-inch Pad 5 retails for INR 15,499 ($225) for the base 3GB/32GB config while the 4GB/64GB variant asks INR 17,499 ($251). The 10-inch alteration sells for INR 16,999 ($244) and INR 18,999 ($275), respectively. The tablets will be made available via Flipkart starting July.
Uber has announced the first international city its flying taxis one day might be buzzing over.
The Australian city of Melbourne will join Los Angeles and Dallas as the third official pilot city for Uber Air, the rideshare company's ambitious project to transport people in short distances via the skies.
Test flights are expected in 2020, with commercial operations aimed for as soon as 2023. Uber claims trips will be priced the same as an UberX ride over the same distance, but we'll see about that.
“Australian governments have adopted a forward-looking approach to ridesharing and future transport technology,” Uber Australia, New Zealand and North Asia general manager Susan Anderson said in a statement.
One of Uber Air’s Skyports.
“This, coupled with Melbourne's unique demographic and geospatial factors, and culture of innovation and technology, makes Melbourne the perfect third launch city for Uber Air. We will see other Australian cities following soon after.
“The State Government of Victoria, Australia has been highly supportive, and we look forward to partnering with them to progress into this first international trial for Uber Air in Melbourne.”
The pitch for Uber Air is to reduce road congestion. Eric Allison, global head of Uber Elevate, said a trip from Melbourne’s CBD to its airport would usually take between 25 minutes to an hour, but with Uber Air it’ll be reduced to 10 minutes.
On that, it’s worth noting that a train line to Melbourne’s airport doesn’t exist. It’s currently under development, and is expected to start construction in 2022 — with completion due nine years later.
Last year, Uber revealed the international cities it was considering for launching the aviation project. Cities included Tokyo and Osaka in Japan; India with Mumbai, Delhi, and Bangalore; Australia in Melbourne or Sydney; Rio de Janeiro, Brazil; and Paris, France.
The company said it would preference high density metro areas which had a population of more than 2 million people, allowing pooled rides to be viable.
Other preferences include “polycentric” regions (urban areas made up of multiple cities), and cities that already integrate Uber well on the street.
The DJI RoboMaster S1 looks like an autonomous sentry ready to protect the smart home of the future, and maybe someday it will be. At the moment, the company says it’s simply the most advanced robot for students on the market. Made to see, feel and hear the world around it, the S1 has 31 sensors to map its environment, which is more than what you’ll find in one of the company’s drones. Combined with its first-person view (FPV) camera, the rover can be programmed to do things like respond in different ways to claps or hand gestures or recognize and react to other S1 units among other things
The S1 is also a big RC car with a cannon that shoots squishy little gel beads and is just a lot of fun to race and do battle with other S1s. First, though, you have to build it.
46 pieces
The S1 needs to be assembled before you can race it.
Joshua Goldman/CNET
Named for DJI’s annual RoboMaster robotics competition, the S1 (short for Step 1) comes disassembled into 46 parts, which DJI says can typically be assembled in about 2 to 4 hours. Its modular design and six pulse width modulation (PWM) ports allow you to modify the S1 and connect additional third-party accessories like speakers or LEDs.
There are six 100-watt brushless servo motors for high speed and accuracy. The motors and Mecanum wheels allow the S1 to move in any direction, not unlike DJI’s drones do in the air. The S1’s FPV camera, infrared-beam and gel-bead blaster are on one of the company’s motorized gimbals that are typically used for stabilizing cameras. That gives you a nice smooth view from the camera while you’re trying to drive.
Although the camera is for navigation and targeting, it can record full HD-resolution video and take photos. Six sets of target sensors lit with LEDs surround the body, which not only look cool but give you something to aim for when you’re battling other S1s with either the IR beam or its gel-bead blaster.
The beads start off as tiny pellets that you soak in water and then load into a cartridge for the blaster. If you’re familiar with Orbeez, that’s essentially what the beads are. The blaster fire rate and angle pitch are limited to avoid injury and DJI includes eye protection just to be safe. They eventually dehydrate again and can be vacuumed up for easy cleaning.
Solo game modes allow you to practice your driving skills and take aim at targets. There are multiplayer options as well, so you can race your friends and family or battle them in a Free-for-All mode, which lets you shoot at other S1s.
The S1’s gamepad isn’t included.
Joshua Goldman/CNET
The RoboMaster S1 is controlled with a phone or tablet. An optional gamepad will be available that gives you a stick and buttons to drive and fire with, as well as a holder for your mobile device. You use your right thumb to control the blaster/camera. I found it uncomfortable to use and it should really have a second stick on the right. It’s also sold separately. You can also control the S1 with a keyboard and mouse, which really gives it a gaming feel.
Included are a set of 44 vision markers, such as the “1” you can see in the picture above. In race modes you can use a set to act as checkpoints or use them for IR target practice. In multiplayer battles, there’s a Heart marker you can scan that virtually repairs your S1 if you get hit by an opponent. Similarly, Mystery Bonus markers give you one of four boosts to use against your competition: Dizziness, Electromagnetic Interference, Extreme Speed and Invincibility.
It’s about learning, too
The S1 can be programmed with Scratch 3.0.
Joshua Goldman/CNET
In the S1’s app is access to coding tools. Under its “Road to Mastery” options, you’ll find project-based courses on programming languages. The S1 can be programmed with Scratch 3.0 or Python. Video tutorials and programming guides can be found under the app’s RoboAcademy.
DJI says you’ll be able to program unique S1 functions, change how it moves, increase efficiency and even optimize the torque for its four wheels. You can also write your own programs to give the S1 custom skills. For example, you can have it evade an attack and return fire automatically.
The S1 is just the start of DJI’s push into the robotics education field and it has plans for courses, educational materials, events and hardware in the months and years to come, the company said.
The RoboMaster S1 is available in the US for $499 starting June 12. A “PlayMore Kit,” which includes the dedicated controller, additional gel beads, one battery and a gel bead container, will be available for purchase at a later date at some price. The S1 will also be available in China and Japan at launch, but no other regions were announced.
After releasing its high-end Radeon VII in February 2019, AMD turned its (and our) attention to the midrange market with its 7nm Navi graphics cards, now officially named the RX 5000 series. Navi plays to AMD's strength: Making high-end graphical power more affordable than ever.
AMD initially debuted the Navi-based Radeon RX 5000 series at Computex, and the first model in the family is the Radeon RX 5700, a GPU that delivers favorable performance compared to Nvidia's competing RTX 2070 graphics, according to AMD's claims. Soon after announcing the RX 5700, AMD added two new models to the GPU family at E3, unveiling a more premium Radeon RX 5700 XT model as well as a supercharged Radeon RX 5700 XT 50th Anniversary Edition.
Whereas Nvidia is playing up the strength of ray tracing on its RTX and GTX 16-series graphics cards, all three models of AMD's RX 5000 series promise to push beyond 1080p gaming, making it more affordable for gamers to get 1440p performance at an affordable price.
Pricing and availability
AMD's Radeon RX 5700 processors will be available for purchase starting July 7, a date chosen to celebrate the 7nm architecture for which these chips are based upon. At E3, CEO Lisa Su revealed that the Radeon RX 5700 will be priced at $379, giving it a premium of $30 over the $349 Nvidia GeForce RTX 2060. The higher end XT model will be priced at $449, making it more affordable than the $499 GeForce RTX 2070. The top-of-the-line Radeon RX 5700 XT 50th Anniversary Edition will cost an additional $50 more.
A source quoted by Red Gaming Tech suggested that AMD's desire to push Navi out the door to combat Nvidia's 16 series could result in low stock at launch — a similar thing happened with the Radeon VII. We will have to wait until July to see how accurate those early inventory claims are.
In addition to the GPUs, AMD revealed during its E3 keynote that there will be a bundle on launch day that will give gamers early access to Gears 5, a Microsoft title that won't officially launch until September. As an incentive, however, gamers who purchase the Radeon bundle will be able to start playing this game in July.
AMD only compared the RX 5000 series graphics against Nvidia's midrange graphics, so we'll likely have to wait until 2020 to see AMD release a Radeon VII successor with a high-end Navi card that can compete with the likes of the Nvidia's flagship RTX 2080 Ti.
Architecture and performance
All three graphics cards in the Radeon RX 5700 family are based on AMD's 7nm process, and we learned that the new architecture moves beyond AMD's aging Graphics Core Next architecture to Radeon DNA, or RDNA. That, we're told, will form the basis for AMD graphics cards for the next 10 years.
RDNA is said to offer major architectural improvements over GCN and the Vega chips that came before. It features an entirely new "compute unit design" for improved efficiency and instructions per clock (IPC). It also features a multilevel cache hierarchy, which reduces latency and allows for greater bandwidth and a reduction in power draw. There have also been optimizations that help boost clock speed.
All of this means that the Navi RX 5000 series, based on the new RDNA, has improvements in IPC by up to 1.25 times. That equates to a 1.5 times improvement in performance per watt over equivalent GCN GPUs.
At Computex, AMD's demo showed that the RX 5700 performed by as much as 10% better than the competing RTX 2070. That would make it around 25-30% quicker than an AMD Vega 64 and only a few percentage points less capable than a Radeon VII.
AMD showed a limited demo of the Radeon RX 5700 on the title Strange Brigade, a game in which AMD hardware typically performs well thanks to its use of the Vulkan API, at Computex. The company followed up at E3, releasing benchmarks to show how the RX 5700 performs when compared to its Nvidia counterparts on 10 of the most popular game titles. On Battlefield V, for example, the RX 5000 performed approximately 21% better, according to AMD's numbers, delivering 117 FPS on the game's Ultra settings in 1440p compared with just 102 FPS on the RTX 2070.
Additionally, AMD's benchmark results show similar or better performance for its new GPUs on titles such as Assassin's Creed Odyssey, Call of Duty: Black Ops 4, Far Cry New Dawn, Metro Exodus, Shadow of the Tomb Raider, Civilization VI, The Division 2, The Witcher, and Tom Clancy's Ghost Recon: Wildlands.
This strong performance can be attributed to the up to 10.3 billion transistors that are packed on the chip. With 36 compute units on the standard RX 5700, this GPU delivers up to 7.95 teraflops of performance. The XT model goes even further, delivering nearly two more teraflops of performance with 40 compute units, and both models take advantage of speedy 8GB GDDR6 memory. The standard RX 5700 has a clock speed of 1.465GHx that can go as high as 1.725GHz, while the XT model starts at 1.605GHz and goes up to 1.905GHz with boost. These chips' RDNA architecture delivers a 25% performance-per-clock improvement compared to the older Graphics Core Next design.
AMD's E3 presentation was primarily centered around the Radeon family's clock speeds, and the company did not detail other aspects of the chip's design. An earlier leak for an "RX 3080" processor suggested that the chip could come with 40 compute units spread across eight shader engines, each with 64 stream processors, though it's unclear how this information might affect render output units (ROPs). WCCFTech speculated that there is potential for up to 128 ROPs. That would be unprecedented for AMD, however, with all of its cards previously limited to a maximum of 64 ROPs.
The eight shader engines would be big news, as it would suggest AMD managed to overcome a limiting factor of just four shader units per die, something that has held back Graphics Core Next-based AMD GPUs. Considering the RX 5000 series uses RDNA and not GCN, that could be one of the major performance enhancements of this new generation of graphics cards.
Without that limitation, we could see Navi GPUs with a number of stream processors in excess of its previous maximum of 4,096, which could unlock intriguing performance potential for the new generation.
AMD's Navi architecture won't be limited to just PC gamers — the company announced that custom variants of its RDNA-based graphics will also be headed to consoles made by Sony and Microsoft. In April 2019, Sony's lead PlayStation architect, Mark Cerny confirmed that the next-generation PlayStation console would use a custom Navi graphics core alongside an AMD Ryzen 3000 CPU based on its Zen 2 architecture. During an interview, Cerny stated that the new PlayStation will support 8K graphics with ray tracing capabilities along with 3D audio.
It's not yet clear whether this will be a custom APU design, or if it will feature a dedicated Navi graphics card. The PS4 and PS4 Pro used an APU design, but considering AMD's claims of Navi scalability, perhaps it will be separate to allow for future expansion with console iterations.
Microsoft announced at E3 that its next-generation Xbox console, which goes by its Project Scarlett code name, will utilize a custom AMD solution based on the 7nm Zen 2 processor and Navi graphics. The chipset has been co-designed and co-engineered by the two companies to bring 8K graphics and ray-tracing into the living room. Like the PC-based Radeon RX 5700, Microsoft confirmed that its custom solution will also use fast GDDR6 memory, and this chip will be capable of supporting up to 120 FPS games with variable refresh rates. Like Sony's rumored PlayStation 5, Project Scarlett isn't expected to launch until 2020.
Alluring graphics without compromise
Despite the strong performance of the RX 5700 graphics, some features found on rival architectures are noticeably absent from the Navi architecture, like variable rate shading, or VRS, and foveated rendering. Instead of supporting these features on RX 5700, AMD is boasting its own sharpening tool called Radeon Image Sharpening, which helps upscales an image to make it look sharper.
With Radeon Image Sharpening enabled, the software will increase the sharpness and details in a game in an effort to boost image quality. Taking a jab at Nvidia, AMD claimed that it's not launching a feature to hurt a game's performance and do trickery to boost the resolution that ends up only to make scenes appear blurry. Instead. according to AMD, Radeon Image Sharpening won't affect the performance of your game, so you won't suffer from stutters or reductions in frame rates.
In a demo at E3, AMD showed that when Radeon mage Sharpening was enabled on Battlefield V at 1440p resolution, the game appeared like 4K thanks to the sharpening, and it ran at 90 FPS.
Radeon Image Sharpening works by examining high contrast areas of a scene, pulling out highlights from dark shadows without affecting any anti-aliasing softening in the image's brighter areas. The result of this sharpening makes 1080p images look closer to 1440p, and the feature's ability to upscale an image is competitive with deep learning supersampling, or DLSS, on Nvidia's chips. Radeon Image Sharpening is currently available for DirectX 9, DirectX 12, and Vulkan, with DirectX 11 support missing right now.
Cinematic visuals
AMD is pushing for more realism when it comes to gaming, and it wants to bring cinematic rendering quality to gaming. With the RX 5700 GPU, AMD also announced the Radeon Media Engine, which supports 4K encode and decode, as well as precinct HDR (PHDR) support.
Game developers also have access to AMD's FidelityFX APIs, which adds more details and increased sharpness to low contrast areas in a game. The feature works in a similar way to Radeon Image Sharpening, but gives developers more control. Whereas Radeon Image Sharpening can be enabled with just a switch as an all or nothing solution, FidelityFX must be implemented within the game.
The feature was demonstrated at E3 on Borderlands 3, showing more details to the honeycomb pattern in the license plate on a car.
Another feature that is supported on the RX 5700 is Radeon Anti-Lag, which is designed for eSports. This feature helps reduce the input lag that gamers experience when they click a mouse and the action not showing up immediately on the display.
Known more formally as motion-to-photon latency, this lag is typically 4 frames — or about 60 milliseconds from the time you press a button to when the action is registered onscreen — on a game played at 60 fps. When the Anti-Lag feature is enabled, the lag is reduced to about 44 milliseconds on the RX 5700 XT. Latency is reduced by 31% across popular eSports titles, according to AMD, and by as much as 35% on Fortnite.
No ray tracing for now
AMD originally stated that it wouldn't be supporting ray tracing, which has been a hallmark of Nvidia's RTX series, at least not until the technology can be implemented on all of its graphics cards. "Utilization of ray tracing games will not proceed unless we can offer ray tracing in all product ranges from low end to high end," David Wang, AMD's senior vice president of engineering at the Radeon Technologies Group, told 4Gamer in an interview.
At this point, it's unclear when hardware-accelerated ray tracing will arrive, and AMD is instead claiming that it's focusing its efforts on features that developers want, like better image quality through tools like FidelityFX. Though AMD is publicly not announcing its support for ray tracing, we do know that the company's custom Navi cards for the next-generation Box and PlayStation consoles due in 2020 will support this feature.
At CES 2019, CEO Lisa Su did state that AMD was "deep in development" on ray tracing technology. Given the performance hit on games that support this Nvidia-exclusive feature so far, AMD is likely waiting until it can make ray tracing less taxing on the GPU. AMD could also be waiting for more games to support ray tracing before making this feature available on its GPU.
50th Anniversary Radeon Edition
In addition to announcing the Radeon RX 5700 and 5700 XT graphics, AMD also announced a beefed up 50th Anniversary Edition of its new GPU at the tail end of its E3 keynote. The Radeon RX 5700 XT Anniversary Edition will be available exclusively on AMD's website with at a price of $499. This special edition GPU features 40 compute units, with CEO Lisa Su describing this particular variant as "the best of the best."
The 50th Anniversary Edition is clocked at 1.68GHz, but can go as high as 1.98GHz with boost. Like the other 5700 series graphics, this model also features 8GB of GDDR6 memory, giving it 10.14 teraflops of performance.
Updated on June 10, 2019: Added the announcement of new features, benchmarks, and comparisons from AMD's E3 keynote.
Samsung’s smartwatch portfolio got updated last month with the latest One UI adding plenty of new features and giving the OS a brand new look. At that point one particular smartwatch was left out – the LTE model of Galaxy Watch 46mm. Now we are happy to report that this has been rectified and it’s finally receiving the OTA update.
The smartwatch is getting two different firmware versions for the US and Europe – R805FXXU1ESE6 is for the European customers while the R805USQU1BSE3 is aimed at the US market. Features are the same across both platforms, though.
A quick reminder that the changes are not only visual – you can now make use of the daily activity goal as well as new outdoor swimming mode and sleeping mode. In addition, the new firmware should help improve battery life, general workout tracking, device syncing and sleep recording.
The three big tech companies, namely Google, Apple, and Microsoft, can best be described as being best frenemies. They are rivals and critics in the industry but also can't but help cooperate with each other to some extent. Usually that involves making services and apps available on the others' platforms even if they're dying to keep them exclusive. The latest proof of that unavoidable cooperation is the all new iCloud app for Windows that has just became available on Microsoft's app Store.
Of course, this isn't the first incarnation of iCloud on Windows but the last time the app made headlines showed the unspoken rift between the two companies. Back in November, it was reported that the Windows 10 October update broke iCloud shared album syncing to the point that Microsoft had to block updates on Windows computers that had a specific version of the iCloud software installed. That was eventually fixed but it seems this new version offers a longer-lasting solution.
Apple rarely makes its desktop apps available on the Microsoft Store, making the surprise announcement of the iCloud Windows 10 app a surprise indeed. It pretty much functions as one would expect from an app with "iCloud" in its name. That is, it lets you store your files on iCloud Drive to access them on non-Windows devices.
Microsoft is, unsurprisingly, very proud of this new app, particularly in how it uses the same technologies used by its own OneDrive Files On-Demand. This enables the iCloud app to transparently access iCloud Drive files from File Explorer and file dialogs as if it was part of the file system. Users can also share files from File Explorer quickly and easily to other iCloud users.
iCloud for Windows 10 PCs let iOS users access iCloud Photos, Mails, Calendars, Tasks, and even Safari Bookmarks even when they're away from their beloved Apple devices. Whether new setup will prevent disasters like last year's breakage is something we'll have to see in the next Windows 10 update.
Despite lackluster sales, continually declining revenue, and customer support issues, HTC, to the surprise of many, is still making new smartphones. The company has announced two new mid-range devices for its home market of Taiwan – HTC U19e and Desire 19+. This marks a move from a series-based naming system (U11 > U12) to a (presumably) year-based one, with the number 19 for devices launched in 2019 — because obviously, product nomenclature is HTC’s biggest problem right now.
“Imagine this for a second: One man, with total control of billions of people’s stolen data, all their secrets, their lives, their futures. I owe it all to Spectre. Spectre showed me that whoever controls the data, controls the future.”
Zuckerberg never uttered those words. The video was a “deepfake,” a technique that uses AI to create videos of people saying something they didn’t, highlighting the challenges social networks face when it comes to policing manipulated content.
The Zuckerberg video could also be a test for Facebook, which has come under fire after it refused to remove an altered video of House Speaker Nancy Pelosi that was slowed to make her seem drunk, according to Vice, which reported earlier on the video. Zuckerberg called Pelosi but she wasn’t “eager” to hear what he had to say, The Washington Post reported on Tuesday.
Facebook’s policy states that it doesn’t remove fake news but will reduce its reach on the site and show information from fact-checkers. It owns photo sharing site Instagram.
“We will treat this content the same way we treat all misinformation on Instagram,” a spokesperson for the photo sharing site said. “If third-party fact-checkers mark it as false, we will filter it from Instagram’s recommendation surfaces like Explore and hashtag pages.”
Fact-checker Lead Stories, which called the video “art,” said in a post that it’s flagging the video as satire and this will not harm the video’s distribution. Users will see a warning label that it isn’t real.
Artists Bill Posters and Daniel Howe partnered with advertising company Canny to make the video. It was created using CannyAI’s video dialogue replacement (VDR) technology, according to the Instagram post. The creators of the video used footage from remarks Zuckerberg made about Russian election interference in 2017. The video includes a CBS logo on the right side and includes a label that says “Zuckerberg: We’re increasing transparency on ads.”
A spokesperson for CBS, which owns CNET, said in a statement that it asked Facebook to “take down this fake, unauthorized use of the CBSN trademark.”
Canny has also teamed up with Posters to create fake videos of President Donald Trump and Kim Kardashian. Spectre, referenced in the video, is an art exhibit that took place at the Sheffield Doc Fest in the UK. The Zuckerberg video was posted four days ago. Posters didn’t immediately respond to a request for comment.
This isn’t the first time a fake video of Zuckerberg has been created. In 2018, filmmaker Andrew Oleck posted a video on Facebook that showed what appears to be Zuckerberg stating that he was deleting the social network, fooling some users into thinking it was real, according to Gizmodo. The video, titled “A World Without Facebook,” was viewed more than 32 million times, but it’s currently unavailable. Versions of that video, though, are still on Facebook and YouTube.
Meanwhile, lawmakers are concerned about the use of deepfakes in the 2020 US presidential election. On Thursday, the House Intelligence Committee is scheduled to hold a hearing on manipulated media and “deepfakes.”
Originally published June 11 at 2:31 p.m. PT. Update, 6:49 p.m. PT: Includes statement from CBS and information from fact-checker.
“Imagine this for a second: One man, with total control of billions of people’s stolen data, all their secrets, their lives, their futures. I owe it all to Spectre. Spectre showed me that whoever controls the data, controls the future.”
Zuckerberg never uttered those words. The video was a “deepfake,” a technique that uses AI to create videos of people saying something they didn’t, highlighting the challenges social networks face when it comes to policing manipulated content.
The Zuckerberg video could also be a test for Facebook, which has come under fire after it refused to remove an altered video of House Speaker Nancy Pelosi that was slowed to make her seem drunk, according to Vice, which reported earlier on the video. Zuckerberg called Pelosi but she wasn’t “eager” to hear what he had to say, The Washington Post reported on Tuesday.
Facebook’s policy states that it doesn’t remove fake news but will reduce its reach on the site and show information from fact-checkers. It owns photo sharing site Instagram.
“We will treat this content the same way we treat all misinformation on Instagram,” a spokesperson for the photo sharing site said. “If third-party fact-checkers mark it as false, we will filter it from Instagram’s recommendation surfaces like Explore and hashtag pages.”
Fact-checker Lead Stories, which called the video “art,” said in a post that it’s flagging the video as satire and this will not harm the video’s distribution. Users will see a warning label that it isn’t real.
Artists Bill Posters and Daniel Howe partnered with advertising company Canny to make the video. It was created using CannyAI’s video dialogue replacement (VDR) technology, according to the Instagram post. The creators of the video used footage from remarks Zuckerberg made about Russian election interference in 2017. The video includes a CBS logo on the right side and includes a label that says “Zuckerberg: We’re increasing transparency on ads.”
A spokesperson for CBS, which owns CNET, said in a statement that it asked Facebook to “take down this fake, unauthorized use of the CBSN trademark.”
Canny has also teamed up with Posters to create fake videos of President Donald Trump and Kim Kardashian. Spectre, referenced in the video, is an art exhibit that took place at the Sheffield Doc Fest in the UK. The Zuckerberg video was posted four days ago. Posters didn’t immediately respond to a request for comment.
This isn’t the first time a fake video of Zuckerberg has been created. In 2018, filmmaker Andrew Oleck posted a video on Facebook that showed what appears to be Zuckerberg stating that he was deleting the social network, fooling some users into thinking it was real, according to Gizmodo. The video, titled “A World Without Facebook,” was viewed more than 32 million times, but it’s currently unavailable. Versions of that video, though, are still on Facebook and YouTube.
Meanwhile, lawmakers are concerned about the use of deepfakes in the 2020 US presidential election. On Thursday, the House Intelligence Committee is scheduled to hold a hearing on manipulated media and “deepfakes.”
Originally published June 11 at 2:31 p.m. PT. Update, 6:49 p.m. PT: Includes statement from CBS and information from fact-checker.
Everyone knows DJI for its drones and gimbals, but its newest device is a little more grounded. The $499 RoboMaster S1 is a tank-like remote control rover packed a wealth of movement options, a dextrous gimbal, a variety of sensors, and a blaster that shoots gel beads. It's a remote control robot that's meant to be as educational is it is fun – and from the 30 minutes or so I spent playing with it last week, that is a whole lot of fun.
Core to the S1 are its Mecanum wheels, a design that allows a robot to move in any direction, including strafing side to side. The S1 is able to stop on a dime and move back, forth, left, and right over uneven terrain with ease. Riding the wheels is a gimbal equipped with both infrared and gel bead blasters that basically allow you to play a remote-controlled version of laser tag or paintball.
This is all controlled via a mobile app, which taps into the S1's camera and allows you to do anything from play battle and racing games (more on this in a bit), to programming the robot to perform a variety of tasks. The robot features some of the machine learning tech DJI's developed through its drone research and can follow and recognize different visual markers, follow a person, recognize gestures, or follow paths drawn on the ground.
But that's just what's built in. People with coding know-how can program more all sorts of more advanced features using Scratch 3.0 or Python. You can program special maneuvers for your battle sessions, change torque of the wheels, or even add custom accessories via six Pulse Width Modulation ports.
If you know nothing about code – presumably most buyers – DJI includes coding courses within the Robomaster app that teach you to code by helping you program new features into your robot. The company says more courses will be added regularly.
I had the chance try out the S1's battle and racing modes, and it might've been the most fun I've had with a remote control vehicle since the first time I flew a drone.
The battle mode is essentially an IRL version of an FPS game. Using either your device's touch-screen or an optional controller, you're able to control your robot pretty much exactly like you would any FPS game.
Except, you can shoot real projectiles and you can see your robot moving in real life as well as your screen – and that makes everything way more fun. There was a small amount of latency to the video feed, but I was surprised at how quickly I was able to grasp handle moving the robot about. Like any FPS game, your goal is to blast your enemies until they die, and you can use visual markers around the play area as health stations or access special moves that give you an edge in battle.
It's worth noting that the gel beads need to be soaked in water before you place them in the S1, and they're very soft and extremely unlikely to hurt anyone. This is especially true as DJI limits your vertical shooting angle, although DJI includes goggles just to be safe (a few beads did ricochet 'dangerously' close to my face).
Next, I tried racing, which was more like real life Mario Kart. This time the goal was to maneuver around a course while marking each checkpoint by allowing your S1 to recognize the visual marker. Here the only caveat was that I had to slow down a bit for the robot to recognize the markers – I couldn't just whiz past them in a frenzy. Still, it was a ton of fun, including Mario Kart-like weapons to annoy your opponents.
The Robomaster will be available June 12 for $499 at store.dji.com, and you'll be able to buy a 'PlayMore Kit' with a dedicated controller, extra beads and a battery at a later date. I know it's way early to buy holiday gifts, but you might want to start saving up.
For more gear, gadget, and hardware news and reviews, follow Plugged on Twitter and Flipboard.
Source: Android Police
