Later this year, Google Photos is going to get a significant update that has the distinction of first arriving on Chromebooks. According to a Google blog post, Google Photos will get a new movie editor and video editing features this fall as part of an update to Chrome OS. From the sound of things, it’ll let users make videos similar to the highlight clips the app already automatically makes. You’ll be able to select a theme as well as people or pets you want to feature in it; from there, Google Photos will pull together a movie using video clips and images from your library. It’ll be smart enough to scan longer videos and pull out specific clips to include in these new creations as well. 

While it’s no surprise that Google is including an automated tool, the company is also including the ability to start from scratch, adding video clips and photos in any order you like. The app will let you adjust things like brightness and contrast, trim clips as you see fit, add title cards and music and apply Google’s Real Tone filters that work better with non-white hairstyles and darker skin. 

Google isn’t saying yet if these video editing features will come to the mobile apps for iOS and Android, but Google Photos has usually had feature parity regardless of platform, so it wouldn’t be a surprise to see these tools expand past Chromebooks before long. In fact, the video editor will be built in to an optimized version of the Android Google Photos app specifically built for larger screens. The app will also seamlessly work with the Files and Gallery Chrome OS apps, so you can open a video in the Gallery app and immediately move it right over to Google Photos for editing or including in a new creation. 

Google

There are numerous other handy updates planned for Chrome OS coming in the next few months. Another new Google Photos feature will allow Chromebooks to access your library and use those pictures for background wallpaper; like other Chrome OS wallpaper options, you can pick a specific album and set it to change daily. The aforementioned Gallery app is going to get PDF editing features, so you can fill out forms and sign them if you’re using a Chromebook with a stylus. That feature is coming next week. There’s also a new Cursive app for capturing and organizing hand-written notes; those can be copy and pasted into other apps or exported to PDFs for, depending on how you need to share them. 

Chrome OS is also getting a new dark mode, something that’s been rumored for a long time now. As you can on most other devices, you’ll be able to pick one mode or have it automatically switch based on the time of day. Some new wallpapers will also come with light and dark versions that automatically switch depending on which theme you use, too.

Google

Finally, Google is making a few productivity improvements to Chrome OS. Clicking on the date in the Chromebook shelf will pop up a monthly calendar view; you can choose a date to see your Google Calendar events without having to open the app or website. And Chrome OS will let you save virtual desk setups, so if you have a specific set of tabs and apps you use frequently, you can call them up and dismiss them as needed.

Most of these updates should be coming in August, though Google specifically noted the virtual desk update won’t be available until late September. And the Google Photos video editing tools are set to arrive in the “fall” — hopefully sooner than later.

In 1917, Marcel Duchamp submitted a sculpture to the Society of Independent Artists under a false name. Fountain was a urinal, bought from a toilet supplier, with the signature R. Mutt on its side in black paint. Duchamp wanted to see if the society would abide by its promise to accept submissions without censorship or favor. (It did not.) But Duchamp was also looking to broaden the notion of what art is, saying a ready-made object in the right context would qualify. In 1962, Andy Warhol would twist convention with Campbell’s Soup Cans, 32 paintings of soup cans, each one a different flavor. Then, as before, the debate raged about if something mechanically produced – a urinal, or a soup can (albeit hand-painted by Warhol) – counted as art, and what that meant.

Now, the debate has been turned upon its head, as machines can mass-produce unique pieces of art on their own. Generative Artificial Intelligences (GAIs) are systems which create pieces of work that can equal the old masters in technique, if not in intent. But there is a problem, since these systems are trained on existing material, often using content pulled from the internet, from us. Is it right, then, that the AIs of the future are able to produce something magical on the backs of our labor, potentially without our consent or compensation?

The new frontier

The most famous GAI right now is DALL-E 2, Open AI’s system for creating “realistic images and art from a description in natural language.” A user could enter the phrase “teddy bears shopping for groceries in the style of Ukiyo-e,” and the model will produce pictures in that style. Similarly, ask for the bears to be shopping in Ancient Egypt and the images will look more like dioramas from a museum depicting life under the Pharaohs. To the untrained eye, some of these pictures look like they were drawn in 17th-century Japan, or shot at a museum in the 1980s. And these results are coming despite the technology still being at a relatively early stage.

Open AI recently announced that DALL-E 2 would be made available to up to one million users as part of a large-scale beta test. Each user will be able to make 50 generations for free during their first month of use, and then 15 for every subsequent month. (A generation is either the production of four images from a single prompt, or the creation of three more if you choose to edit or vary something that’s already been produced.) Additional 115-credit packages can be bought for $15, and the company says more detailed pricing is likely to come as the product evolves. Crucially, users are entitled to commercialize the images produced with DALL-E, letting them print, sell or otherwise license the pictures borne from their prompts.

Open AI

These systems did not, however, develop an eye for a good picture in a vacuum, and each GAI has to be trained. Artificial Intelligence is, after all, a fancy term for what is essentially a way of teaching software how to recognize patterns. “You allow an algorithm to develop that can be improved through experience,” said Ben Hagag, head of research at Darrow, an AI startup looking to improve access to justice. “And by experience I mean examining and finding patterns in data.” “We say to the [system] ‘take a look at this dataset and find patterns,” which then go on to form a coherent view of the data at hand. “The model learns as a baby learns,” he said, so if a baby looked at a 1,000 pictures of a landscape, it would soon understand that the sky – normally oriented across the top of the image – would be blue while land is green.

Hagag cited how Google built its language model by training a system on several gigabytes of text, from the dictionary to examples of the written word. “The model understood the patterns, how the language is built, the syntax and even the hidden structure that even linguists find hard to define,” Hagag said. Now that model is sophisticated enough that “once you give it a few words, it can predict the next few words you’re going to write.” In 2018, Google’s Ajit Varma told The Wall Street Journal that its smart reply feature had been trained on “billions of Gmail messages,” adding that initial tests saw options like ‘I Love You’ and ‘Sent from my iPhone’ offered up since they were so commonly seen in communications.

Developers who do not have the benefit of access to a data set as vast as Google’s need to find data via other means. “Every researcher developing a language model first downloads Wikipedia then adds more,” Hagag said. He added that they are likely to pull down any, and every, piece of available data that they can find. The sassy tweet you sent a few years ago, or that sincere Facebook post, may have been used to train someone’s language model, somewhere. Even Open AI uses social media posts with WebText, a dataset which pulls text from outbound Reddit links which received at least three karma, albeit with Wikipedia references removed.

Guan Wang, CTO of Huski, says that the pulling down of data is “very common.” “Open internet data is the go-to for the majority of AI model training nowadays,” he said. And that it’s the policy of most researchers to get as much data as they can. “When we look for speech data, we will get whatever speech we can get,” he added. This policy of more data-is-more is known to produce less than ideal results, and Ben Hagag cited Riley Newman, former head of data science at Airbnb, who said “better data beats more data,” but Hagag notes that often, “it’s easier to get more data than it is to clean it.”

Craiyon / Daniel Cooper

DALL-E may now be available to a million users, but it’s likely that people’s first experience of a GAI is with its less-fancy sibling. Craiyon, formerly DALL-E Mini, is the brainchild of French developer Boris Dayma, who started work on his model after reading Open AI’s original DALL-E paper. Not long after, Google and the AI development community HuggingFace ran a hackathon for people to build quick-and-dirty machine learning models. “I suggested, ‘Hey, let’s replicate DALL-E. I have no clue how to do that, but let’s do it,” said Dayma. The team would go on to win the competition, albeit with a rudimentary, rough-around-the-edges version of the system. “The image [it produced] was clear. It wasn’t great, but it wasn’t horrible,” he added. But unlike the full-fat DALLl-E, Dayma’s team was focused on slimming the model down so that it could work on comparatively low-powered hardware.

Dayma’s original model was fairly open about which image sets it would pull from, often with problematic consequences. “In early models, still in some models, you ask for a picture – for example mountains under the snow,” he said, “and then on top of it, the Shutterstock or Alamy watermark.” It’s something many AI researchers have found, with GAIs being trained on those image libraries public-facing image catalogs, which are covered in anti-piracy watermarks.

Dayma said that the model had erroneously learned that high-quality landscape images typically had a watermark from one of those public photo libraries, and removed them from his model. He added that some early results also output not-safe-for-work responses, forcing him to make further refinements to his initial training set. Dayma added that he had to do a lot of the sorting through the data himself, and said that “a lot of the images on the internet are bad.”

But it’s not just Dayma who has noticed the regular appearance of a Shutterstock watermark, or something a lot like it, popping up in AI-generated art. Which begs the question, are people just ripping off Shutterstock’s public-facing library to train their AI? It appears that one of the causes is Google, which has indexed a whole host of watermarked Shutterstock images as part of its Conceptual Captions framework. Delve into the data, and you’ll see a list of image URLs which can be used to train your own AI model, thousands of which are from Shutterstock. Shutterstock declined to comment on the practice for this article.

A Google spokesperson said that they don’t “believe this is an issue for the datasets we’re involved with.” They also quoted from this Creative Commons report, saying that “the use of works to train AI should be considered non-infringing by default, assuming that access to the copyright works was lawful at the point of input.” That is despite the fact that Shutterstock itself expressly forbids visitors to its site from using “any data mining, robots or similar data and/or image gathering and extraction methods in connection with the site or Shutterstock content.”

Alex Cardinelli, CEO at AI startup Article Forge, says that he sees no issue with models being trained on copyrighted texts, “so long as the material itself was lawfully acquired and the model does not plagiarize the material.” He compared the situation to a student reading the work of an established author, who may “learn the author’s styles or patterns, and later find applicable places to reuse those concepts.” He added that so long as a model isn’t “copying and pasting from their training data,” then it simply repeats a pattern that has appeared since the written word began.

Dayma says that, at present, hundreds of thousands, if not millions of people are playing with his system on a daily basis. That all incurs a cost, both for hosting and processing, which he couldn’t sustain from his own pocket for very long, especially since it remains a “hobby.” Consequently, the site runs ads at the top and bottom of its page, between which you’ll get a grid of nine surreal images. “For people who use the site commercially, we could always charge for it,” he suggested. But he admitted his knowledge of US copyright law wasn’t detailed enough to be able to discuss the impact of his own model, or others in the space. This is the situation that Open AI also perhaps finds itself dealing with given that it is now allowing users to sell pictures created by DALL-E.

The law of art

The legal situation is not a particularly clear one, especially not in the US, where there have been few cases covering Text and Data Mining, or TDM. This is the technical term for the training of an AI by plowing through a vast trove of source material looking for patterns. In the US, TDM is broadly covered by Fair Use, which permits various forms of copying and scanning for the purposes of allowing access. This isn’t, however, a settled subject, but there is one case that people believe sets enough of a precedent to enable the practice.

Authors Guild v. Google (2015) was brought by a body representing authors, which accused Google of digitizing printed works that were still held under copyright. The initial purpose of the work was, in partnership with several libraries, to catalog and database the texts to make research easier. Authors, however, were concerned that Google was violating copyright, and even if it wasn’t making the text of a still-copyrighted work available publicly, it was prohibited from scanning and storing it in the first place. Eventually, the Second Circuit ruled in favor of Google, saying that digitizing copyright-protected work did not constitute copyright infringement.

Rahul Telang is Professor of Information Systems at Carnegie Mellon University, and an expert in digitization and copyright. He says that the issue is “multi-dimensional,” and that the Google Books case offers a “sort of precedent” but not a solid one. “I wish I could tell you there was a clear answer,” he said, “but it’s a complicated issue,” especially around works that may or may not be transformative. And until there is a solid case, it’s likely that courts will apply the usual tests for copyright infringement, around if a work supplants the need for the original, and if it causes economic harm to the original rights holder. Telang believes that countries will look to loosen restrictions on TDM wherever possible in order to boost domestic AI research.

The US Copyright Office says that it will register an “original work of authorship, provided that the work was created by a human being.” This is due to the old precedent that the only thing worth copyrighting is “the fruits of intellectual labor,” produced by the “creative powers of the mind.” In 1991, this principle was affirmed by a case of purloined listings from one phone book company by another. The Supreme Court held that while effort may have gone into the compilation of a phone book, the information contained therein was not an original work, created by a human being, and so therefore couldn’t be copyrighted. It will be interesting to see if there are any challenges made to users trying to license or sell a DALL-E work for this very reason.

Rob Holmes, a private investigator who works on copyright and trademark infringement with many major tech companies and fashion brands, believes that there is a reticence across the industry to pursue a landmark case that would settle the issue around TDM and copyright. “Legal departments get very little money,” he said. “All these different brands, and everyone’s waiting for the other brand, or IP owner, to begin the lawsuit. And when they do, it’s because some senior VP or somebody at the top decided to spend the money, and once that happens, there’s a good year of planning the litigation.” That often gives smaller companies plenty of time to either get their house in order, get big enough to be worth a lawsuit or go out of business.

“Setting a precedent as a sole company costs a lot of money,” Holmes said, but brands will move fast if there’s an immediate risk to profitability. Designer brand Hermés, for instance, is suing an artist named Mason Rothschild, who is producing MetaBirkins NFTs. These are styled images on a design reminiscent of Hermés’ famous Birkin handbag, something the French fashion house says is nothing more than an old-fashioned rip-off. This, too, is likely to have ramifications for the industry as it wrestles with philosophical questions of what work is sufficiently transformational as to prevent an accusation of piracy.

Artists are also able to upload their own work to DALL-E and then generate recreations in their own style. I spoke to one artist, who asked not to be named or otherwise described for fear of being identified and suffering reprisals. They showed me examples of their work alongside recreations made by DALLl-E, which while crude, were still close enough to look like the real thing. They said that, on this evidence alone, their livelihood as a working artist is at risk, and that the creative industries writ large are “doomed.”

Article Forge CEO Alex Cardinelli says that this situation, again, has historical precedent with the industrial revolution. He says that, unlike then, society has a collective responsibility to “make sure that anyone who is displaced is adequately supported.” And that anyone in the AI space should be backing a “robust safety net,” including “universal basic income and free access to education,” which he says is the “bare minimum” a society in the midst of such a revolution should offer.

Trained on your data

AIs are already in use. Microsoft, for instance, partnered with OpenAI to harness GPT-3 as a way to build code. In 2021, the company announced that it would integrate the system into its low-code app-development platform to help people build apps and tools for Microsoft products. Duolingo uses the system to improve people’s French grammar, while apps like Flowrite employs it to help make writing blog posts and emails easier and faster. Midjourney, a DALL-E 2-esque GAI for art, which has recently opened up its beta, is capable of producing stunning illustrated art – with customers charged between $10-50 a month if they wish to produce more images or use those pictures commercially.

For now, that’s something Craiyon doesn't necessarily need to worry about, since the resolution is presently so low. “People ask me ‘why is the model bad on faces’, not realizing that the model is equally good – or bad – at everything,” Dayma said. “It’s just that, you know, when you draw a tree, if the leaves are messed up you don’t care, but when the faces or eyes are, we put more attention on it.” This will, however, take time both to improve the model, and to improve the accessibility of computing power capable of producing the work. Dayma believes that despite any notion of low quality, any GAI will need to be respectful of “the applicable laws,” and that it shouldn’t be used for “harmful purposes.”

And artificial intelligence isn’t simply a toy, or an interesting research project, but something that has already caused plenty of harm. Take Clearview AI, a company that scraped several billion images, including from social media platforms, to build what it claims is a comprehensive image recognition database. According to The New York Times, this technology was used by billionaire John Catsimatidis to identify his daughter’s boyfriend. BuzzFeed News reported that Clearview has offered access not just to law enforcement – its supposed corporate goal – but to a number of figures associated with the far right. The system has also proved less than reliable, with The Times reporting that it has led to a number of wrongful arrests.

Naturally, the ability to synthesize any image without the need for a lot of photoshopping should raise alarm. Deepfakes, a system that uses AI to replace someone’s face in a video has already been used to produce adult content featuring celebrities. As quickly as companies making AIs can put in guardrails to prevent adult-content prompts, it’s likely that loopholes will be found. And as open-source research and development becomes more prevalent, it’s likely that other platforms will be created with less scrupulous aims. Not to mention the risk of this technology being used for political ends, given the ease of creating fake imagery that could be used for propaganda purposes.

Of course, Duchamp and Warhol may have stretched the definitions of what art can be, but they did not destroy art in and of itself. It would be a mistake to suggest that automating image generation will inevitably lead to the collapse of civilization. But it’s worth being cautious about the effects on artists, who may find themselves without a living if it’s easier to commission a GAI to produce something for you. Not to mention the implication for what, and how, these systems are creating material for sale on the backs of our data. Perhaps it is time that we examined if it’s necessary to implement a way of protecting our material – something equivalent to Do Not Track – to prevent it being chewed up and crunched through the AI sausage machine.

A lawmaker wants to punish people for ghosting, or abruptly cutting off communications with someone without explaining why. Arnolfo Teves Jr., a member of the Philippine House of Representatives, claimed in a note accompanying his bill that ghosting "can be likened to a form of emotional cruelty and should be punished as an emotional offense."

Teves referred to studies that indicate social rejection activates the same neural pathways as physical pain. He argued that “ghosting is a form of spite that develops feelings of rejection and neglect” and claimed the emotional toll can impact productivity. The bill doesn't propose specific penalties, but Teves suggested in an interview that community service would be appropriate.

The proposed legislation doesn't make a whole lot of sense. It defines ghosting as something that happens when a person is "engaged in a dating relationship." The document goes on to define a dating relationship as one where the parties live together without being married or are "romantically involved over time and on a continuing basis." Teves claimed neither casual acquaintances nor "ordinary socialization" constitutes a dating relationship. If you're being ghosted by someone you're in more than a casual relationship with (especially a domestic partner!), then there are bigger problems to worry about.

Importantly, the bill does not account for blocking someone without explanation if they're being creepy or threatening. As The Washington Post notes, the bill isn't likely to pass as it stands. The legislature likely has more pressing concerns anyway.

The bill suggests ghosting occurs "for no apparent justifiable reason but solely to cause emotional distress to the victim." That's not really the case. People are busy! They don't always have time to keep chatting to everyone who messages them on a dating app. Also, folks often feel they lack the communication skills to have open and honest conversations. Maybe the ghoster just thinks you're boring and wants to let you down relatively gently.

Being ghosted sucks. But expecting someone you don't really know to keep chatting with you isn't smart. Proposing to punish people for ghosting after you've been burned too many times instead of going to therapy probably isn't the answer either.

Grand Theft Auto VI might reflect evolving social attitudes as much as it does gameplay. Bloombergsources claim Rockstar's future open-world title will include the franchise's first playable female lead character. The Latina woman will reportedly be part of a Bonnie and Clyde-inspired duo at the heart of the story. Developers are also trying to shake a past reputation for transphobia and won't make jokes about marginalized groups, the tipsters said.

Rockstar's attempt to avoid crunch time is also affecting both the scope and release date for the game, according to the sources. GTA VI was supposedly meant to cover large portions of North and South America, but was whittled back to a fictional version of the Miami area (presumably Vice City). And while financial analysts are predicting a release sometime between April 2023 and March 2024, Bloomberg's contacts believe the game is "at least" two years away.

The company has apparently overhauled its design and management structures to reduce the need for excessive work hours. It reportedly added team members, created more opportunities for time off and fired leaders who were allegedly abusive or difficult. While the insiders described problems with bureaucratic overhead and some frustration over a lack of progress, many employees were said to be happy with the reduced pressure.

There's enough financial breathing room, at least. GTA V and GTA Online have enjoyed success for years. GTA V only left The NPD Group's top 20 yearly bestsellers in 2020, and that was because NPD stopped tracking digital sales from Rockstar parent company Take-Two. GTAO, meanwhile, has been popular enough that Rockstar launched a GTA+ subscription service this year to profit from the game's biggest fans. GTA VI may arrive more than a decade after its predecessor, but its creators won't suffer much from the long wait.

Spotify's less-than-stellar quarter can be partly blamed on a decision to exit the hardware business. As part of its earnings data, the streaming service revealed that it stopped manufacturing its Car Thing player. In a statement to Engadget, a company spokesperson pinned the decision on "several factors" that included customer demand and supply chain problems. The withdrawal hurt Spotify's gross profits.

Existing Car Thing units will still work as expected, Spotify said. The company said it still "unlocked helpful learnings" from the device despite its brief history, and that the car remained an "important place" for audio. You can still buy the Car Thing for $50 (down from the usual $90) as of this writing.

Spotify first shared word of Car Thing in 2019, but the finished product only reached the broader public early this year following several months of invitation-only sales. It was effectively a Spotify player for cars that didn't have streaming functionality, and served as an informative experiment for the company. Low demand wouldn't be surprising. Many cars can access Spotify through Android Auto, Apple CarPlay or built-in apps — Car Thing was mainly helpful for people with older vehicles who wanted a music streaming upgrade, but didn't want to mount their phone.

The cancellation comes despite otherwise positive signs. Spotify's free and paid listener bases are still growing (to 433 million and 188 million users respectively), and its shifts toward audiobooks and podcasts are paying dividends by lowering the relative cost of music label royalties. Car Thing's demise theoretically helps Spotify focus on those successes and brace itself for a widely expected global recession.

Projectors have come a long way from the clunky, dim models of the past. The latest models are brighter, sharper, more discreet and easier to install than ever. They also deliver a picture size you could never get on a TV without spending a fortune. With all that, they’re highly desirable for home theater buffs who want a true cinematic experience at home.

Over the last year, we’ve seen models bright enough to use in a normal room, short-throw models that take up very little space and have built-in sound bars, and better picture quality than ever with 4K and Dolby Vision features. Better still, you can now get 4K HDR models that can project up to a 150-inch image for as little as $1,000. That’s around the same price and over double the size of a decent 65-inch TV, giving you something close to true movie theater immersion.

There are a lot of different types of projectors, though, ranging from ultra short throw to portable to long throw. There are also a lot of terms to understand, like lens shift, LCD vs. DLP, laser vs. lamp illumination and more. This guide will help you understand all that so that you can make the best buying decision possible.

The technology

Here are the basics: Projectors generally use two types of technology, LCD and DLP. They’re fundamentally different systems, with their own advantages and drawbacks.

The rise of ultra-short-throw projectors and brighter long-throw models, meanwhile, has been powered by falling prices in laser illumination technology. Lasers are a far better solution than lamps, because they’re brighter and last far longer — up to 30,000 hours instead of 6,000. That’s essentially a lifetime of use (about 10 years).

Most projector manufacturers now use DLPs, or digital light processing units, manufactured nearly exclusively by Texas Instruments (TI). The heart of the tech is an optical semiconductor called a digital micromirror device (DMD) that contains millions of aluminum mirrors. Those tilt either toward the light source (on) or away from it (off) at up to 5,000 times per second.

Budget projectors like BenQ’s HT3550i use TI’s 0.47-inch DMD, while higher end models, like the Samsung Premium LSP9T use the 0.66-inch chip. Both use mirrors that tilt by +12 and -12 degrees for white and black, but TI recently unveiled a new 0.47-inch 4K-capable DMD with +/-17 degrees of tilt, which should enhance both brightness and contrast.

DLP projector makers include LG, Optoma, LG, BenQ and Panasonic. The benefits of the tech are portability, high contrast, less fringing and cheaper projectors, especially 4K and ultra-short-throw models. The biggest drawback is the rainbow effect, or bright red/blue/green artifacts that affect some viewers more than others.

LCD tech, meanwhile, uses a prism to split a light source into red, green and blue beams. Those then pass through LCD displays containing the image and converge via another prism before passing through the projector’s lens.

Epson is the primary user of LCD tech, along with Sony, Sanyo and others. LCD projectors tend to be sharper, more efficient and more color accurate, but have lower contrast ratios and can experience image degradation over time. In general, they’re also more expensive.

What to look for in a projector

Ultra-short-throw

Since the last time we updated our guide, ultra-short-throw projectors have become the hot new category, offering several benefits. You can mount them close to the wall like a TV, with no need to run wires through the walls and ceiling, but still get an immersive image as large as 120 inches — something that’s impossible with a TV unless you’re very rich. They use brighter lasers that never need to be replaced — and because laser light is collimated, focusing is eliminated.

They’re also physically less awkward to install than a ceiling-mounted projector, though that doesn’t mean installation is super easy. To get the perfect screen fit and alignment, you must place them an exact height and distance from your wall or screen. This can be quite a pain, as I’ve discovered.

You also need a perfectly flat wall or screen, because ultra-short-throw projectors beam up at an acute angle, so any imperfections will show as shadows. For that reason, you can’t use a roll-down screen because they have slight ripples.

For the best results, particularly with a lot of ambient light, you should use an ambient light-rejecting (ALR) screen. Those have small ridges that reflect light from below back to your eyes, but absorb any light (ie ambient light) that comes from above. For one of those, you’ll need to budget at least $450 and way up. Some projectors, including models from Epson and HiSense, come with ALR screens.

Elite Starling

Brightness and contrast

Home theater projectors generally range in brightness between 2,000 to 4,000 lumens, but you'll need to take those figures with a grain of salt. Some models might actually hit 3,000 lumens or more, but cranking the lamp to that level will hurt the image quality and lifespan of your bulb. Also, some manufacturers tend to exaggerate maximum brightness.

As a point of reference, many 4K flat panel TVs nowadays can hit 1,000 nits of brightness, but the brightest consumer projectors only display between 100 and 150 nits from the screen. That’s not as big a deal as it might seem, because projector images are much larger and meant to be used in dark rooms, where your eyes will automatically adjust to the light and “brighten” the image.

Contrast is also substantially different on projectors. Unlike OLED TVs, projectors don’t allow for zero black levels because of ambient light, reflections and other reasons. You also can’t have local dimming zones found on LED TVs for true blacks. Some projectors do have a dynamic iris to improve the contrast scene-by-scene, but those can often produce a “pumping” effect, with the image dimming or brightening in mid-scene.

Mounting and fan noise

A big advantage of regular long throw projectors is that you can mount the projector and screen on the ceiling, using zero space in your room. If you plan to do that, don’t forget to budget for a mounting bracket and any necessary long cables, including extra power for Google's finicky Chromecast. Also, keep in mind that it's easier to mount a lightweight projector, and DLP models are usually lighter than those with LCD tech.

Some projectors are noisier than others, and usually the more you spend, the less noise you get. Many of the new 4K DLP projectors, when operating in 4K mode, are particularly noisy. There's one other (cool) thing: if your projector is portable or relatively easy to take down and put up, you can take it outside for magical night screenings under the stars.

HDR and resolution

As related to projectors, these things could each take up an entire article. In fact, they have — for a deeper dive, take a look at Projector Central’s excellent takes on HDR and resolution.

On the resolution front, only expensive projectors have native 4K resolution; indeed, most movie theaters still use 2K projectors for various reasons. However, there are many relatively inexpensive DLP projectors that use pixel-shifting to attain 4K resolution. That system emits each pixel four times while moving it to the correct position for a 4K image, all in less than 1/60th of a second. As such, it puts as many pixels on the screen in the same amount of time as a 4K native projector — and visually, it performs nearly as well.

On the other hand, Epson's LCD “4K enhanced” projectors also have 1080p native resolution, but the image is just shifted twice, not four times. So, those projectors are not 4K natively or otherwise, but do produce double the pixel count of a 1080p projector. If you really want a 4K native projector, you’ll have to pay: two of the cheapest ones are Sony's VPL-VW295ES ($5,000) and JVC's DLA-NX5 ($5,000).

HDR is a very different animal on projectors compared to TVs. As mentioned, projectors can’t produce anywhere close to the amount of light required (1,000 nits) to qualify as true HDR. Rather, they use a technique called tone-mapping to fit the entire HDR gamut into a lower brightness range.

For that reason, among others, almost all projectors only support HDR10. Only one uses Dolby Vision (the Xiaomi Laser Cinema 2, only available officially in China), and just a couple of models work with Samsung’s HDR10+ — and those are Samsung’s own Premiere 4K models. However, most support a wider 10-bit color gamut that allows for superior color reproduction.

Samsung

Optics

If you're mounting a short- or long-throw projector between five and 25 feet, you might need to consider the zoom range and whether the projector has a lens shift option. A decent zoom range will make it easier to mount the projector where you want with the screen size that you want.

Lens shift, meanwhile, is used if the projector is mounted higher or lower relative to the screen than recommended by the manufacturer (or any horizontal distance off center). That creates a trapezoidal shaped image, but by dialing in some lens shift, you can optically square it up. Otherwise, you might have to use a "keystone correction," which digitally stretches or shrinks part of the image, resulting in noticeable distortion or pixel artifacts. Digital correction might not work in gaming modes either, for some projector models.

Gaming

If you’re interested in a projector for gaming, you’ll want to look up the refresh rate and input lag figures. Some new projectors from Viewsonic, Optoma and others offer up to 240 Hz 1080p refresh rates and input lag settings down to 4 or 5 milliseconds. However, some projectors designed more for home entertainment have very poor input lag and refresh rates at just 60 Hz.

Portable projectors

Finally, portable projectors have become popular enough to merit discussion this year. They’re relatively cheap, compact and portable and can run on batteries – making them ideal for entertainment outside or while camping. They’re not nearly as bright as other projectors, of course, but are more designed for a fun night of entertainment under the stars.

Engadget picks

As with previous updates, I’m dividing projectors into ultra-short-throw and long-throw categories. As mentioned, ultra-short-throw models have rapidly established themselves in the market due to the extra performance and convenience, and all manufacturers sell at least a couple of models. Within the ultra-short-throw category, We’ll compare two price categories: under $7,000 and $3,500, with three projectors each. In the long-throw category, we’re again looking at projectors under $1,000, $2,000 and $6,000, with three products in each range. Finally, we’ll take a look at the best portable projectors.

Ultra-short-throw projectors under $3,500

LG CineBeam HU715Q

LG’s nice-looking HU715Q is the new king of ultra-short-throw (UST) projectors due to the picture quality, tech on offer and reasonable $3,000 price. With a 20,000-30,000 hour laser light source, it delivers 2,500 ANSI lumens, along with accurate colors across 86 percent of the DCI-P3 gamut – enough for most HDR content. And speaking of that, it supports both HDR10 and HLG, along with frame-by-frame tone-mapping.

Rival models offer similar specs, but the HU715Q trumps them in sound and technology. The 20-watt 2-way stereo speakers are great on their own, but you can supplement them with Bluetooth speakers from LG and nearly any other brand to create surround sound. The standout feature, though, is LG’s webOS smart TV capabilities that give you Netflix, Disney+, Amazon Prime and other apps, all with full 4K HDR support and surround sound, including eARC output for Dolby Atmos and lossless audio. It also supports HDMI 2.1 and better still, full 4K (4,096 x 2,160) rather than just UHD.

Optoma CinemaX P2

Steve Dent / Engadget

Optoma’s CinemaX P2 made our list last year, but it’s still a great choice because the price has dropped considerably. It delivers 3,000 lumen brightness, impressive contrast ratio and accurate colors with 80 percent DCI-P3 coverage. It’s not quite as sharp as the pricier projectors, as it uses TI’s 0.47-inch rather than 0.66-inch DLP tech, though you’ll still get a near-4K image.

The CinemaX P2 may also better match your decor, as it comes in white rather than dark grey like the P1. The 40-watt NuForce Dolby Digital 2.0 soundbar is one of the best on any ultra-short-throw projector, as well. On the downside, it does offer apps but they’re not as good as you’ll find on, say, Google’s Chromecast.

BenQ V7050i

BenQ’s first UST laser projector is at the top end of the price scale at $3,500, but it offers some impressive capabilities. Light output is a bright 2,500 ANSI lumens and it delivers a full 98 percent DCI-P3 coverage for as good an HDR experience as you can get on a projector. You also get a Filmmaker Mode to see colors as the directors intended. It’s powered by Android TV so you get all the streaming services and apps you want, along with apps, games and more. The downside is the lack of decent speakers, as it only offers dual 5-watt speakers with clear sound but limited bass.

Epson EpiqVision Ultra LS300W

If you’re okay with 1080p projection, Epson’s EpiqVision Ultra LS300W is a very interesting option because of the design, excellent sound, built-in Android TV and extreme 3,600 ANSI lumen brightness. That allows for a wide color gamut with no rainbow effect, excellent connectivity and very good sound without the need to buy a soundbar or surround sound system. Best of all, it’s priced at just $2,000, making it one of the cheaper short-throw projectors out there.

Ultra-short-throw projectors under $7,000

Samsung Premium LSP9T

Ultra-short-throw projectors are a hot enough entertainment category that Samsung decided to re-enter the market with two models as part of its Lifestyle TV series. The $5,500 LSP9T is the premium model and it certainly brings some premium specs. With separate red, green and blue lasers, it covers 106 percent of the full Rec.2020 color range (147 percent of DCI-P3) — something we’ve rarely, if ever seen on any TV or even pro monitor.

It delivers a very bright 2,800 lumens and a 1,500:1 ANSI contrast ratio. As it uses TI’s higher-resolution 0.66-inch DLP chip (with pixel shifting), you get as close as you can to true 4K without investing in a native 4K projector. Finally, the LSP9T is the first projector on the market rated for HDR10+ — Samsung’s answer to Dolby Vision. On top of all that, the LSP9T delivers 40 watts of audio using Samsung’s Acoustic Beam technology, while offering Samsung’s well-regarded Tizen-powered Smart TV platform.

Epson EpiqVision Ultra LS500

Epson

If you need the brightest possible image, Epson’s LCD-powered EpiqVision Ultra LS500 ($3,899) delivers. It’s rated at up to 4,000 lumens, making it one of the brightest ultra-short-throw projectors in any price range. It also supports HDR modes in HDR10 and HLG and is sold with both 100-inch and 120-inch ALR screens, making the price effectively lower. The main drawback is that it only offers double the pixels of 1080p, rather than four times like competing DLP tech. It also offers a relatively weak 10-watt built-in speaker system.

HiSense L9G

This is HiSense’s new $4,300 flagship UST that uses a tricolor laser to achieve high brightness (3,000 ANSI lumens) and an incredible 107 percent BT.2020 HDR coverage, topping even Samsung’s formidable LSP9T. It has a powerful 40W Dolby Atmos sound system and built-in Android TV with Google Assistant and Alexa. Best of all, that price includes a 100-inch ALR Daylight screen, or for an extra $500, you can get it with a 120-inch ALR cinema screen.

Projectors under $1,000

Viewsonic PX701-4K

There are very few 4K projectors available under $1,000, and Viewsonic’s $900 PX701-4K is one of the most recent. For the money, it offers impressive performance. You get 4K HDR with 3,200 lumens of brightness via TI’s .47-inch DLP chip. More importantly for gamers, you can get a 1080p signal at up to 240Hz with a 5-millisecond input lag time. The drawbacks are a limited 1.1x optical zoom, so you’ll need to make sure it fits in your space.

BenQ HT2050A

BenQ

For around $700, the BenQ HT2050A is still one of the best budget 1080p projectors. It delivers where it counts with the best contrast (ANSI 1,574:1) and color accuracy in its class, and is reasonably bright as well, with 2,200 lumens in "vivid" mode. On top of that, it comes with a 1.3x zoom and vertical lens shift option for maximum installation flexibility. The drawbacks include slightly excessive fan noise, rainbow effect and red-tinted 3D.

Optoma HD146X

If you’re looking to spend a little less, the Optoma HD146X is your best option. Using DLP tech, it delivers 1080p at up to 3,600 lumens with excellent brightness, color accuracy, contrast and black levels. You also get decent (16.4-millisecond) input lag for gaming. The drawbacks are a single HDMI port, 1.1x optical zoom and poor built-in audio.

Projectors under $2,000

BenQ HT3550i

BenQ’s $1,600 4K HT3550i is an update to last year’s HT3550, but with a huge addition: Android TV. With that, you get multi-platform wireless projection from Android and iOS devices via Chromecast or Airplay — a huge plus compared to rival projectors. As before, it offers reference-quality 4K color reproduction in both HDR and SDR that's a match for projectors costing triple the price. Contrast is excellent thanks to the dynamic iris, though brightness is limited to 2,000 lumens. It has a 1.3x zoom and vertical lens shift option, plus a surprisingly good built-in speaker. The fan noise is still present, but less than before.

If brightness is more important than picture quality, take a look at BenQ’s $1,500 TK850i instead. It also comes with Android TV, but delivers far more brightness (3,000 lumens) in exchange for less color-accurate picture quality.

Optoma UHD38

Optoma

For extra brightness and speed for gaming, the answer is Optoma’s all-new, $1,600 4K-capable UHD38. It cranks the lumens up to 4,000 and like the Viewsonic PX701-4K, offers 240Hz gaming at 1080p with one of the lowest latency figures we’ve seen yet in a projector at 4.2 milliseconds. Otherwise, you can do 4K 60 Hz gaming with 16.7 milliseconds of lag, which is very quick for 4K. It’s optimized more for gaming than entertainment unlike BenQ’s HT3550i, but it can still handle HDR10 and HLG. It supports both zoom (albeit just 1.1x), but also vertical and horizontal lens shift.

Epson Home Cinema 4010 4K Pro

Epson’s $2,000 Home Cinema 4010 4K Pro is the Cadillac of under-$2K home projectors thanks to features like 2,400 lumen brightness, dynamic iris, and motorized zoom (2.1x), focus and lens shift. It delivers in picture quality too, covering 100 percent of the DCI-P3 color space in cinema mode with both HDR10 and HLG. It also offers near-4K quality using 1,920 x 1,080 LCD image chips with pixel shifting. The drawbacks are lack of support for 60Hz 4K due to the HDMI 1.4 ports.

If you need that, want to pay a bit less and don’t care about the motorized focus, Epson’s $1,700 Home Cinema 3080 4K Pro is the way to go. It offers similar features like HDR10 and HLG, but supports 4K 60p thanks to the HDMI 2.0b ports. There’s no motorization and the zoom drops to 1.6x, but it supports generous tilt, shift and zoom ranges.

Projectors under $6,000

Epson Home Cinema LS11000 laser

Fans of Epson’s LCD projectors will be interested in this model, which received a big update in the form of a laser light source. That bumps the brightness to 2,500 ANSI lumens, and the three 1080p LCDs use a pixel shifter to quadruple the resolution to something close to true 4K. It offers very accurate colors with HDR10 and HLG capability, but is also great for gaming thanks to the 120 Hz refresh rate, 20-millisecond input lag and HDMI 2.1 support. You also get a 3-way motorized lens, scene adaptive correction and more for $4,000.

Optoma UHZ65LV

Optoma’s $6,000 UHZ65LV also uses a long-lasting laser light source to deliver a 5,000 lumen image, much brighter than any lamp-powered projector. It also delivers true 4K resolution up to 60p, thanks to the TI 0.66-inch DLP chip. The extra brightness and contrast make it ideal for HDR10 or HLG content. It also comes with desirable features for a long-throw projector, like a 1.6x zoom and vertical lens shift.

LG CineBeam HU810PW 4K

LG

Speaking of long-throw laser projectors, LG’s $3,000 CineBeam HU810PW is another excellent pick at a much lower price point. There are some compromises, as the laser light pushes out a lower 2,700 lumens (that’s still a lot), and it has a smaller 0.47-inch DLP chip that delivers slightly lower perceived resolution. However, it has dual blue and green lasers which help it deliver accurate HDR colors with an excellent 97 percent DCI-P3 coverage. It also offers a 1.6x zoom with lens shift and an HDMI 2.1 port that allows for 4K at 60p with up to 12-bit color depth. It comes with LG’s webOS, so it supports Amazon Prime, Netflix, Hulu and other streaming services without the need for a dongle.

Sony VPL-VW295ES

If you’re looking for a true, native 4K projector, Sony’s $4,500 VPL-VW295ES is the least costly option out there. It’s by far the sharpest 4K projector in this roundup, thanks to Sony’s proprietary 4K SXRD native DCI 4K (4,096 x 2,160) panels. It also delivers extremely accurate colors, with 100 percent DCI-P3 coverage and HDR10/HLG support. You also get niceties like a 2.06 zoom lens with powered zoom, lens shift and focus. The main drawback is a relatively dim 1,500 lumen brightness, but it’s a top pick if picture quality is paramount above all.

Best budget portable projectors

Xgimi MoGo Pro

Xgimi is a relatively new brand, but has started to gain traction thanks to its lineup of portable projectors. The best of those is the MoGo Pro, which can be powered by a battery (with two hours of playtime) and is small enough to fit in a backpack. But it still delivers a detailed and reasonably accurate 1080p image, while offering automatic focus and vertical keystone adjustment. It even comes with Android TV, giving you all the streaming options you can need – all for $650.

BenQ GS2

BenQ

This $470 model is designed specifically for outdoor entertainment, so it’s battery-powered and splash and shock resistant – making it a good choice for backyard movies or sports events, camping and more. It’s also one of the brightest portable projectors out there. It only delivers 720p resolution, but it does come with a streaming app in the form of Aptoide TV.

Anker Nebula Solar HD

This 1080p projector has a pretty rich feature set considering the $600 price including a battery. It delivers 400 lumens for reasonably bright outdoor use, has a reasonably powerful 2x3W speaker system with Dolby Digital Plus, comes with Android TV and has a built-in stand for easy adjustment.

It's not just students and internet debaters who lean on Wikipedia in a pinch. MIT CSAIL researchers have conducted a study revealing that Wikipedia can influence the legal decisions of judges when there are articles covering relevant cases. The existence of a Wiki page for a case increased its citations by over 20 percent, the scientists said. The boost was pronounced when a case supported a judge's argument, and the language of the articles sometimes manifested in the decisions.

The team conducted the study by having law students write over 150 articles on Irish Supreme Court decisions. Half of the pieces were randomly chosen to be uploaded where judges, lawyers and clerks could use them, while the rest were kept offline to help understand what would happen in the absence of a Wikipedia article. The randomized nature showed a true causal link between articles and citations, according to lead researcher Neil Thompson.

CSAIL also noted that the Irish legal system was an ideal testing ground. Higher courts' decisions bind lower courts, as they do in the UK and US, but there aren't nearly as many articles on Irish Supreme Court decisions as there are for its US counterpart. The researchers increased the number of relevant articles "tenfold" just by writing examples for the study.

As to why people might turn to Wikipedia? It might come down to a simple matter of time. The spike in citations mainly came from lower courts (the High Court) rather than the Supreme Court itself or the Court of Appeal. To CSAIL, that suggested judges and clerks were using Wikipedia to cope with busy court dockets — it was easier to find precedent-setting cases through a quick online search.

The findings are potentially problematic. While the cases themselves might be sound, Wikipedia isn't always accurate. There's a risk that a judge might issue a ruling based on a flawed article, or that malicious actors could manipulate entries to skew a trial's outcome. Study co-author Brian Flanagan argued that the legal community should verify that any online analysis, whether it's from Wikipedia or elsewhere, is both comprehensive and comes from expert sources.

If you've been eyeballing a 2nd-generation Google Nest Hub but can't stomach the regular $100 price tag, there's good news. It's now on sale for just $55 or 45 percent off at B&H Photo Video, matching one of the lowest prices we've seen to date. B&H also has other Nest devices on sale including the Google Nest Audio ($60 or $40 off), Nest Hub Max ($169 or $60 off) and Nest Mini at just $30, for a savings of $20. 

The Nest Hub earned a score of 89 in our review for good reason. It has a minimalist design, a bright 7-inch display, clearer and louder audio than its predecessor and faster performance as well. It's just the right size to fit into any room and can act as a second screen for your video doorbell or even track your sleep from your bedside table. While it may not be as accurate as a slumber-focused wearable, the Nest Hub will show you sleep duration, quality, schedules and snoring data (if that's something you're after).

If it's just a smart speaker you're after and don't need a screen, you may want the Google Nest Audio instead, on sale for $60 or 40 percent off. It has an unassuming yet attractive design, plus great audio quality that's made even better if you pair two of them together and use them in stereo mode. It also functions as a Google Assistant device, giving you the forecast, control of smart home gadgets, music capabilities and more. 

Next up is the Nest Hub Max, available for $169 or $60 off. It's the biggest of Google's smart displays with the best sound quality, thanks to a 10-inch display large enough to comfortably watch YouTube or Netflix videos, or even take a Zoom call. Inside are stereo speakers that help the Hub Max sound better than any other Nest smart display, making it a good option if you want an all-in-one device to stream videos, listen to music and control devices like smart lights, cameras, thermostats and more.

And finally if budget is an issue, the Nest Mini is on sale for just $30, for a savings of $20. It offers the same functionality as Google Nest Audio, but without the same sound quality, of course. Still, it gives you a smart speaker with Google Assistant powers on the cheap for a spare or a kids room. 

Follow @EngadgetDeals on Twitter and subscribe to the Engadget Deals newsletter for the latest tech deals and buying advice.

In a piece examining what the NVIDIA Broadcast can do, we praised its noise removal capability for being "impressively adept" at stripping away everything but your voice. Now, AMD has launched a similar tool for computers powered by its chips, and it could vastly improve your gaming and live streaming experience or lead to cleaner recordings. AMD's Noise Suppression tool can reduce background chatter and sounds in real time through the use of a deep learning algorithm. 

The feature works for both input and output devices, which means it can not only suppress noise captured on your mic, but also noise coming from someone else's. It can help you stay focused if you're in a meeting or attending online classes, for instance, because it can block the sounds of somebody's barking dog or crying baby in the background.

That said, Noise Suppression is only available for computers with Ryzen 5000 series and newer processors, as well as for systems with Radeon RX 6000 series and newer graphics cards. If you know your computer has the components needed to access the feature, make sure to upgrade to the latest version of AMD Software before heading to the Settings page. 

Under Audio Settings in the Audio & Video tab, you'll see a toggle to enable the feature. You can also choose to switch it on for either your input or your output device or for both of them at the same time. The next time you fire up a game or a communications app, such as Microsoft Teams or Slack, you can take advantage of the benefits of the noise suppression tool, so long as you select the "AMD Streaming Device" option as your input and/or output device.

Spotify’s second-quarter financial release shows the streaming giant hasn’t yet felt the dread hand of the looming global recession. Unlike Netflix, which had to report a fall in its overall customer base, Spotify has seen both free and paying accounts grow. It now has 433 million users, up from the 422 million reported at the end of the first quarter. 188 million of those are paying for Premium, a leap of six million from three months ago, while a further four million are signed up on an ad-supported basis.

Despite industry-wide fears that household budgets would cut entertainment costs to help free up much-needed cash, Spotify has dodged cost-cutting so far. The company said that while it was keeping an eye on the “uncertain” environment, it was “pleased with the resilience of [its] business.” That said, the company did spend big to help grow its user figures, with marketing campaigns designed to coax back users who let their subscriptions lapse, or who wanted to expand to a family plan.

That marketing spend helped blow a hole in the company’s finances, with Spotify posting a quarterly loss of €194 million ($197 million). The company is banking on sharp increases in revenue both for subscriptions and advertising to help balance those losses out. In addition, its plan to pivot toward cheaper forms of audio content, like podcasts and audiobooks, should see the volume of cash it pays to record labels fall to a more tolerable (for Spotify) level — even if recording artists continue to demonstrate that they’re being starved of an income by the piddling royalties paid out on a per-stream basis.