Nvidia’s latest AI demo is pretty impressive: a tool that quickly turns a “few dozen” 2D snapshots into a 3D-rendered scene. In the video below you can see the method in action, with a model dressed like Andy Warhol holding an old-fashioned Polaroid camera. (Don’t overthink the Warhol connection: it’s just a bit of PR scene dressing.)
The tool is called Instant NeRF, referring to “neural radiance fields” — a technique developed by researchers from UC Berkeley, Google Research, and UC San Diego in 2020. If you want a detailed explainer of neural radiance fields, you can read one here, but in short, the method maps the color and light intensity of different 2D shots, then generates data to connect these images from different vantage points and render a finished 3D scene. In addition to images, the system requires data about the position of the camera.
Researchers have been improving this sort of 2D-to-3D model for a couple of years now, adding more detail to finished renders and increasing rendering speed. Nvidia says its new Instant NeRF model is one of the fastest yet developed and reduces rendering time from a few minutes to a process that is finished “almost instantly.”
As the technique becomes quicker and easier to implement, it could be used for all sorts of tasks, says Nvidia in a blog post describing the work.
“Instant NeRF could be used to create avatars or scenes for virtual worlds, to capture video conference participants and their environments in 3D, or to reconstruct scenes for 3D digital maps,” writes Nvidia’s Isha Salian. “The technology could be used to train robots and self-driving cars to understand the size and shape of real-world objects by capturing 2D images or video footage of them. It could also be used in architecture and entertainment to rapidly generate digital representations of real environments that creators can modify and build on.” (Sounds like the metaverse is calling.)
In a paper describing the work, Nvidia’s researchers said they were able to export scenes at a resolution of 1920 × 1080 “in tens of milliseconds.” The researchers also shared source code for the project, allowing others to implement their methods. It seems NeRF renders are progressing quickly, and could start having a real-world impact in the years to come.
Update March 25th, 15:50PM ET: Updated story with link to research paper and source code.
I didn’t expect to laugh while playing Time Flies, but I did, out loud on the Summer Game Fest show floor. It’s a deceptively simple game with monochromatic, MS Paint-style visuals and a clear premise: You’re a fly and you have a short time to live a full life in a random house.
There are layers to the game’s main goal, as the fly has a bucket list filled with items like “learn an instrument”, “read a book”, “make a friend” and “get drunk.” Each of these tasks is completed in a delightfully surprising way — for instance, getting drunk means landing on the base of a martini glass and sipping from the small droplet of alcohol there. Afterward, the screen becomes distorted, warped lines making it harder to fly through the house. Making a friend involves joining a trail of ants as they walk single-file through cracks in the kitchen walls. The fly lands on the back of an ant and it can hang out, disappearing into one small hole and reappearing from the other in a continuous, friendly loop.
And then the fly dies. Every round ends with the fly’s death, whether that’s caused by the inevitable progression of time or the player’s direct actions, such as getting too close to a strip of fly paper, touching a light bulb or drowning in the full martini glass. A timer ticks down constantly in the upper-left corner, starting with 80-odd seconds at most, and when it hits zero, the fly drops to the ground like a speck of dust.
The timer itself presents a compelling thought experiment at the beginning of every life cycle. The length of each round is determined by choosing a location from a dropdown menu of all the countries in the world, and it’s based on the life expectancy of each region. Selecting “United States,” for example, gives players 77.4 seconds because people there are expected to live 77.4 years, according to the database used by the game. This mechanic, beginning every round with a self-inflicted geographic death sentence, grounds the game in reality. It adds weight to whatever silly, pixelated mechanics may follow, mirroring the quiet way that existential dread constantly grips us all.
Knowing you’ll die doesn’t mean you can’t have fun while you’re alive — as the fly, that is. The house is packed with personal items like books, art, instruments and furniture, and to a buzzy little fly, it feels nearly endless. It’s possible to land on certain environments and the screen will zoom in to allow players to interact with the objects there, showing additional detail. The fly can flip the power switch on a phonograph and collect coins inside a bulbous light fixture, each of these new areas appearing as the fly buzzes past or into them.
The scene that made me laugh out loud involved a headless mannequin sticking out of the ceiling. Yes, you read that correctly, but this isn’t where I laughed yet. Flying into the dummy’s open neck revealed a network of intestines to escape — funny, but I still hadn’t laughed — with an exit precisely where you’d expect it to be. When the screen shifted from a dark intestinal tract to show the fly popping out of the dangling mannequin’s butt cheeks, I couldn’t help myself. I laughed and heard people watching behind me chuckle, too. Together, we all enjoyed the surprising ridiculousness of this fly’s life, and then it dropped dead.
I had a good time with that fly in particular. I played a few rounds of Time Flies and crossed out a few items on the bucket list, but there’s still so much more to explore in that solitary house. I just need some more time.
Time Flies is scheduled to hit PlayStation, Switch and Steam in 2023, developed by Michael Frei and Raphaël Munoz.
All products recommended by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you buy something through one of these links, we may earn an affiliate commission.
Party games are one of the go-to gaming options when it comes to groups of friends looking to play together — especially during the last year of the COVID-19 pandemic. From titles like Super Mario Party and Overcooked to the ever-popular Jackbox series, there is a growing number of party games out there. A recent arrival on the party game scene aims to offer players a creative writing option in the form of Frantic Fanfic.
Under development since December 2020 and launched in April 2021, Frantic Fanfic is all about getting the creative writing juices flowing for groups of two or more. Self-described as “a fandom-frenzied supercharged mess of exposition,” this party game was born out of a love of pop culture, convention-going, and fandom by developers Zee and Michael.
Springing up out of an effort to bring back some of the energy of going to conventions that was effectively lost during the pandemic, there’s something special to be found in Frantic Fanfic.
Frantic Fanfic’s origin
Frantic Fanfic started out from creator Zee’s initial idea of having a few three-minute presentation nights (a game that really blew up in popularity during quarantine in which you create a PowerPoint on a topic, usually of your choosing, and present it to a group) with friends, with a twist. The concept was that instead of knowing the topic beforehand, Zee and friends would create topics, shuffle them, pull from a list, and then need to put together the presentation beforehand.
This presentation night eventually morphed into the first online version of Frantic Fanfic that made use of Google Docs, a Facebook event, tracking time on a phone, and Zee manually sending out links to friends. In this form, Frantic Fanfic was a game where several friends would write the beginning, middle, and end of a story before all parts would come together and they would read each story out loud. A few days after this event, Zee was hit with the idea of a website or an app.
“I was staring at the ceiling thinking ‘Damn what a great time that game night was … when can we do it again, how can I make it better’?” Zee tells Digital Trends. “And then it hit me. Oh my God. It could be a website. An app. A something. We could develop this. And so we did! We’ve been developing Frantic Fanfic since December 2020 and it’s been a fun ride ever since.”
Alongside wanting to have fun with friends, Zee was also missing the ability to attend conventions during the COVID-19 pandemic. “So I thought really hard about the things that I missed from conventions,” says Zee. “What panels I missed. What the vibe was like. I really missed the bad fan fiction reading panels, which I thought were hilarious. Bad writing plus over-dramatic delivery is very, very funny. If I could combine the presentation parties with the bad fanfic reading panels somehow … I feel like that would be a fun thing to do!”
Michael, Zee’s boyfriend and a software developer, makes up the other half of the two-person team behind Frantic Fanfic. After Zee put together what would evolve into the completed version of the game, she and Michael ultimately decided on creating a website.
“We eventually settled on a website because coding an app was so time-intensive just to get onto all the various app distribution networks,” Zee says. “Then came the concept of the visuals. I’m not a website designer by trade so it was a lot of back and forth and learning what we could and couldn’t do there on the fly. Michael is a back-end developer so he learned all the front-end pieces you see in the game within the last eight months. I made mockups in Photoshop and Michael would translate them to actual working code. The first time we saw text inputted into the page and then carried through to the next section was magical.”
So, what do you get when you pair a love of fan fiction, the humor and chaos you can often find at late-night convention panels, and an idea together? You get Frantic Fanfic.
What is Frantic Fanfic?
Thanks to the work that Zee and Michael have put into the game, Frantic Fanfic is a user-friendly game to get right into with friends. The quick-start guide is chock full of helpful information on how to get started playing quickly with a longer version available, as well as a streamer-specific guide.
But at its most basic, Frantic Fanfic allows a group of friends to each write down character names that will then be shuffled around, names will be chosen by the group, and then each player will be prompted to write a different section of several fan fictions using the chosen characters under a time limit. Once every section of each fan fiction has been written, the group will then be prompted to read each out loud.
Winning a round comes down to the reactions that each fan fiction garners from the group playing. During the part of the game where you read what you’ve written out loud, members of the group can react with as many reactions as they’d like. Frantic Fanfic really leans into the collaborative side of things, even when determining which player has won the game.
With three game modes currently available (Standard, OTP version, and Self Insert), there are a number of different ways to play. Four different timing modes are available to choose from, such as lightning round and a more leisurely option, which can affect gameplay even more.
Zee offered some tips for new players of Frantic Fanfic. “Make the most of it by inviting some people that you know are into fun party games and willing to try something new and get a little creative and silly! We find that the best games are played with about four to eight players, so you can get a really good shuffle in there. But you can play between two to 16 players if you really want to.”
Frantic Fanfic’s future
While Frantic Fanfic is still new on the indie party game scene, that hasn’t stopped Zee and Michael from working on updates for the game. Shared to the game’s blog, a recent update has included some adjustments to the settings screen and, by popular demand, the enabling of a two-player mode.
Since it is a two-person team, updates are done in Zee and Michael’s free time. Zee shared that both work full time, so updates will come as time and energy allow. But, a new mode is coming soon alongside the three game modes that are currently available.
“We’re introducing another mode soon – one where you only see the previous part [the fan fiction section that has been written] and all other parts become hidden while the game is active. We still don’t have the name for this mode yet, there’s a bit of a debate because it’s kinda broken telephone, but not really as nothing is being repeated. It’s more like lost pages out of a diary, trying to continue where someone left off mysteriously sort of thing.”
Alongside a number of future updates that include an interactive audience feature that will allow people to vote on their favorite stories as they view streamed games on platforms like Twitch, Zee’s hopes for Frantic Fanfic are high.
“Hopes for the future include making a physical copy of the game, making a bunch of different modes, and also seeing a “big streamer” play it live with a huge audience would be extremely cool!” says Zee. “I’d also really love to take this game on the road to a bunch of different cons, COVID protocols pending.”
With a very positive response from players so far, even reaching a goal set by both creator’s to see over 500 people playing at once in June 2021, Frantic Fanfic continues to see more and more ridiculous fan fictions being written by players and some of that real-world convention energy coming back online.
“We did something that made a lot of people happier and got them to laugh at silly stuff,” says Zee. “And during the pandemic and coming out of the pandemic, when things just look so grim in the world… that’s a really nice thing to say you’ve done.”
Interview responses have been lightly edited for clarity.
Perhaps the biggest change, though, is in the tone and structure of the game. The World in OlliOlli is Radland, which is divided into five distinct areas. You begin being introduced to a colorful group of characters that wouldn’t be out of place in an episode of Adventure Time. The main cast includes a friendly older man named Dad; a very enthusiastic guy called Gnarly Mike, who sets challenges for each level; a cameraperson named Suze and, my favorite, Chiffon, a pipe-smoking “Skate Wizard” who acts as a checkpoint in levels when you mess up. There are other characters to meet along the way, including “Sloshtar the Fortune Telling Fish,” which is as weird as it sounds.
Only two of the game’s five areas — Sunshine Valley and Cloverbrook — were available in the demo, and only a fraction of the levels in each were playable. The final game, from my rough calculations, is likely to have over a hundred crafted levels. What struck me from the ten or so I played was how different they all feel. None are as claustrophobic as those in the Vita games, but some are more constrained and technical, while others sometimes have the camera dynamically zooming out to the point where your character is almost a spec on the screen. There was also a fun bonus level (unlocked by said fortune-telling fish) that was confined to a skate park, which had you pulling trick after trick to reach a certain score in a given time.
The majority of levels in the demo functioned as tutorials, but things briefly opened out in places and the focus seems to be on replayability and mastery. I captured a few of my favorites in the video below (excuse my poor skill level, I forgot about spins and didn’t even work out I could manual until the last five minutes of my playthrough).
Levels are peppered with a lot of light-hearted conversation (which you can skip easily if you just want to get to skating or are replaying a level), and that comfy tone continues through to every aspect of the game’s design. Levels and characters pop with bright colors and an illustrative 3D style. Bennett said Roll7 wants to ensure OlliOlli World “feels representative of the very inclusive and diverse culture of skateboarding.” There’ll be a character creator — which wasn’t in the demo — to customize your skater’s appearance, and you’ll also be able to pick what style and tricks your character goes for.
OlliOlli is known for its unique soundtracks, which sit somewhere between Wipeout and lo-fi beats to skate to. “[OlliOlli music] sort of straddles IDM [independent dance music], electronica, lo-fi and all these different weird, disparate genres that no one’s ever heard of,” Bennett said. “It’s a really odd thing, because we’ve asked labels to send us tracks that they think will work in the vein of the game that we’re building, and no one seems to quite understand it.” While the soundtrack is still being finalized, Roll7 will be working with music labels from past games, such as Ninja Tune, along with new labels like Cascade Records. “My ambition for it is to repeat some of the stories that we’d had from OlliOlli 2, with people just plugging their Vita, well maybe now their Switch, into their hi-fi system, putting it on the menu screen and just leaving it playing,” Bennett said.
One potentially game-changing aspect I didn’t get to look at is the level creation tool. It functions a little like a “zip code system,” according to Bennett, and allows you to predetermine parameters such as “how long the level is, how difficult it is and what the art style will be” in order to generate a level. The game will return a random seed, and you’ll be able to skate it immediately and share it with friends. Bennett said there will be “millions” of potential levels to skate, and the intent is for players “to quite quickly create something, share it with people and then challenge them on it.”
There’s a lot Roll7 isn’t talking about right now. In a blog post introducing the game, Bennett spoke about “high score leaderboards” and “a global competition,” but, when asked, said Roll7 will be sharing more details on that later as “they don’t want to spoil that.”
Despite the many new features, you can still hop into the action quickly, restarts are mapped to a single button and the feel of skating at break-neck speeds is the same. “We don’t want to keep you from that flow state for too long,” Bennett said. “The previous two games, it was very much, ‘there’s a menu, and then there’s the game,’ right? We wanted to build something much more cohesive around that experience.”
OlliOlli World lands this winter on Switch, PlayStation, Xbox and PC.
All products recommended by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you buy something through one of these links, we may earn an affiliate commission.
A new Satechi stand could help turn your iPad Pro into a mini-iMac, adding not only desk convenience but also a whole host of new ports in the process. The Satechi Aluminum Stand & Hub for the new iPad Pro gives the Apple M1-powered tablet easy connections like HDMI and an SD card reader, for those wanting to switch their desktop over to iPadOS.
That’s an increasingly feasible option, as Apple updates the iPad Pro software and the power inside the latest tablet models matches what you’ll find in the newest MacBook Air and Mac mini. Indeed, with the M1 cropping up most recently in the new iMac 24-inch, there’s a lot to be said for making the switch to all-tablet if your software needs fit.
The Aluminum Stand & Hub addresses one of the lingering issues with that, having enough places to plug things in. It’s equipped with a 4K HDMI port (that’s 60Hz capable) along with USB-C PD charging, USB-A for data, and an SD card reader. There’s also an audio jack.
On the top there’s an adjustable stand, to hold the iPad Pro up. That can be positioned upright, to orient it like a regular display, or pushed back for using the iPad Pro more like a graphics tablet. The stand folds flat when it’s not in use, too, meaning that the whole system is actually fairly portable if you’re hot-desking.
It’s reminiscent of the Kensington StudioDock, which launched earlier this year. That was also designed to dock with the iPad Pro, and then give access to a number of ports including external monitor support. Kensington’s approach, though, relied on a docking cradle: that made it relatively easy to slot the tablet into place, but did present some issues with generational upgrades should Apple change its design.
Satechi’s dock, in contrast, has a more straightforward cable to connect to the iPad Pro’s USB-C port. That may not look quite so aesthetically pleasing – though the brushed aluminum finish of the dock in general is nice, and matches the tablet itself – but it definitely opens the door to using the same dock with different iPad Pro sizes and, most likely future generations of the tablet should Apple decide to change the dimensions at all.
It’s available now, priced at $99.99, which also undercuts Kensington’s dock. As for the flexibility of iPadOS as a desktop platform, we’re expecting to hear more about what Apple has in mind there at WWDC 2021, which kicks off with the opening keynote on Monday, June 7.
Elevate your enterprise data technology and strategy at Transform 2021.
Microsoft delivered more than 100 product and service announcements at its Build conference last week. As CEO Satya Nadella said in his keynote, these announcements support the continued “tech intensity” that has accelerated throughout the 2020 global pandemic, expediting the company’s investments in digital transformation — especially the adoption of cloud computing, artificial intelligence, and big data.
Build, the company’s annual event for developers, focused on the intersection of IT modernization and the development of applications that can leverage data and AI to power the way we work, learn, and communicate. The event also focused a lot on the idea of “creators” — not the kind we hear about in the YouTube or TikTok world, but the kind Microsoft hopes to support with its no-code, low-code, and pro-code offerings designed to allow more members of an organization to concurrently deliver material value to the group, making development a team sport where both coders and business experts contribute.
With so many new capabilities brought to market over the three-day event, it would be hard to cover the whole gamut of announcements. But as an analyst, I was particularly drawn to updates around the Azure ecosystem and wanted to provide some insights on what caught my attention and what it means for Microsoft and its ecosystem.
Key enhancements for Azure Arc
The first thing that caught my eye was the increased support for Azure Arc and the company’s clear positioning that multi-cloud is here and that everything it builds needs not only to support Azure but on-prem, the edge, and other clouds like Google and AWS.
Arc was originally announced in 2019 as the company’s solution for managing resources from across clouds. Still, this year’s event included the announcement of several newly added cloud services, including Azure App Service, Functions, Logic Apps, API Management, and Event Grid, to enable more of Microsoft’s services to run in other clouds.
Here is a quick rundown of each.
Azure App Service: A fully managed service for building, deploying, and scaling web apps.
Functions: Event-driven serverless compute platform designed to solve complexities in applications orchestration.
Logic Apps: New integration platform as a service (iPaaS) built on a containerized runtime to make apps more scalable, portable, and automated across the IT environment.
API Management: Hybrid and multi-cloud management platform allowing developers to deploy API gateways side-by-side no matter the host location, optimizing API traffic flow.
Event Grid: Single service for managing event routing from any source to any destination.
In brief, these services enable users to run Kubernetes clusters, on-premises, multi-cloud, and within edge environments using Azure, unlocking data no matter its location for use on Azure compute services. I believe these offerings are essential to meet the current landscape of IT modernization, which often works in parallel to the development of applications — as time goes on, these functions will continue to work more harmoniously.
Azure forms new Azure Applied AI Services
Microsoft announced a plethora of enhancements for Azure AI, Azure Cognitive Services, and Azure ML. All of these updates are designed to enable users to do more with their data. Moreover, the company was pretty straightforward about wanting to allow developers to better leverage the power of AI in apps being developed on its platform.
The new tools, along with some newly minted general-availability (GA) announcements, include:
Updates to Azure Bot Service: A visual authoring canvas along with open-source tools that let developers add telephony and speech capabilities when testing, debugging, and deploying multi-channel bots without requiring massive changes to developer code.
Azure Metrics Advisor: This service went into preview last September but is now GA. Metrics Advisor is a monitoring platform that offers APIs for data ingestion, diagnostics, and anomaly detection without the requirement of machine learning knowledge.
Azure Video Analyzer: Video analytics are hot, and this new offering combines Live Video Analytics and Video Indexer. This service is currently in preview and is designed to deliver analytics from streaming and stored videos, including the auto-extraction of advanced metadata.
Microsoft, notably announced it will combine these new and updated offerings along with Azure Form Recognizer, Azure Immersive Reader, and Azure Cognitive search to make up what it will call the ‘Azure Applied AI Services Group.’ The importance of these tools and connectors comes down to making AI more usable for developers. Despite the constant chatter about AI’s ability to enhance applications, Microsoft knows the critical path is to shorten the development time and simplify the inclusion of AI while improving and building applications.
Azure, for the foreseeable future, will continue to chase AWS’s massive infrastructure business. Still, it is hard to argue with the vast developer ecosystem that the company has created and the hooks that tie these tools and services together. Azure serves as the foundation for the data ecosystem. The continued evolution of Microsoft’s developer ecosystem being more “Code meets Creator” provides a platform for companies to embrace the massive portfolio of software and solutions that Microsoft offers its customers.
This year’s event addressed the rapid proliferation of hybrid cloud, multi-cloud, and the enterprise requirement for applying data at scale. I expect this trend to continue as Microsoft seeks to cement further the adoption of its applications and the use of its cloud and AI services.
Daniel Newman is the principal analyst at Futurum Research, which provides research, analysis, advising, and/or consulting to high-tech companies in the tech and digital industries.
VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.
Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:
up-to-date information on the subjects of interest to you
gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
During a May 18 raid in the U.K., West Midlands Police expected to find a cannabis growing operation after learning about a site stealing electricity on the Great Bridge Industrial Estate. Instead, they found about 100 computers mining cryptocurrency.
The BBC reports that detectives received a tip prior to the raid. Onlookers said that they saw multiple people visiting the site throughout the day and police drones picked up a lot of heat coming from the building — typically a sign of a growing operation. Western Power Distribution also found that the site was illegally connected to its grid, stealing “thousands of pounds [worth of] of electricity.”
“It had all the hallmarks of a cannabis cultivation setup,and I believe it is only the second such crypto mine we have encountered in the West Midlands,” Sgt. Jennifer Griffin said. The authorities seized the computers but didn’t make any arrests.
Crypto-mining operations are becoming a problem not only in the U.K., but around the world. In 2018, New York allowed some power providers to start charging higher fees to crypto miners. In March 2021, New York introduced a bill that would ban mining for three years while the state evaluated the environmental impact.
These measures come in response to the growing climate problem surrounding mining coins like Ethereum. Some scientists say that increased demand for cryptocurrency has already negated the effects of using electric vehicles.
Recent reports show some mining operations bringing in over $20,000 per month, and Nvidia, whose GPUs are frequently at the heart of these mining operations, earns upwards of $400 million each year from crypto miners.
It’s not just private operations, either. One of Russia’s largest oil producers set up a Bitcoin mining farm in Siberia last year that’s entirely powered by gas.
With the price of multiple cryptocurrencies hitting all-time highs, we’re still not sure if demand — in terms of power and coins — will continue to grow. With volatility in currencies like Bitcoin, some investors think the market is showing a repeat of 2014 and 2018, where high volatility caused mass selloffs and devalued the coin.
Regardless, it looks like cannabis farms aren’t the only thing police have to worry about when it comes to stolen electricity.
Although the PlayStation 5 has broken a few records last month, Nintendo’s Switch still continues to hold a special place in the market and gamers’ hearts. That’s despite that admittedly aged hardware that shouldn’t be able to compete with heavyweights like the Xbox or PlayStation. Part of the Switch’s appeal is its portability and flexibility but another part is the unique experiences it offers, like this new game that practically gamifies game-making.
Most gamers have probably dreamt of making their own adventures but few would ever consider having the knowledge or skills to make that happen. Younger dreamers might even be frightened by the prospects of programming just to make blocks move around. There are, of course, “no programming” platforms these days, and the new Game Builder Garage is one such experience made specifically for the Nintendo Switch.
Unlike something with a predefined set of themes and mechanics like Super Mario Maker 2, Game Builder Garage is more free-form and flexible. It revolves around the concept of connecting nodes called “Nodon” that are presented as creatures with different personalities. Grownups might recognize this as a node-based game maker, just presented in a more playful manner.
Game Builder Garage lands on the Nintendo Switch on June 11 for $29.99. While the game-making game is perfectly usable with the Joy-Cons, Nintendo recommends connecting a mouse to the Switch’s USB-C port. It would probably be easier if it connected via Bluetooth, though.
This isn’t the first game maker for the Switch that you run on the Switch. That distinction belongs to FUZE4, though this experience is closer to the traditional concept of programming where you get to type code using a special version of the beginner-friendly BASIC language.
To quote Google CEO Sundar Pichai: AI is “more profound than fire or electricity.”
To back up this claim with incontrovertible evidence: here’s an AI tool made by Google researchers that turns doodles into weird monsters. What could be more profound?
It’s certainly fun, anyway. The tool is called Chimera Painter and uses machine learning to generate imagery based on users’ rough sketches. This sort of dynamic is becoming a relatively common one in machine learning. Nvidia has done it with landscapes before; MIT and IBM did it with buildings; and now Google is … doing it with monsters.
The team behind Chimera Painter explained their methods and motivations in a blog post, saying the idea was to create a “paintbrush that acted less like a tool and more like an assistant.” Chimera Painter is just a prototype, but if software like this becomes common it could “reduce the amount of time necessary to create high-quality art,” claim the team.
The researchers gave themselves the challenge of creating artwork for a fictional fantasy card game, in which players combine features from different monsters and battle them like mutating Pokémon. They trained a machine learning model on a database of more than 10,000 sample monsters, which were themselves in part procedurally generated using 3D models rendered in Unreal Engine. Each image is paired with a “segmentation map” — an overlay that divides the monsters into anatomical parts like claws, snouts, legs, and so on.
Once the model has been trained on this data, users can then paint their own segmentation map which is then rendered using photorealistic textures. If you load up Chimera Painter you can see some of the preset monsters and they’re impressively cohesive. However, painting one yourself takes more time and effort than you might think. Our attempt below, for example, looks like a knock-off Gruffalo depicted using mud. It’s monstrous, but not necessarily a monster.
Anyone who owns a Macbook, laptop, or a smartphone rocking a USB-C port should head over to Amazon today. Third-party retailer Aukey is selling its 6-in-1 USB/SD card/HDMI hub for $20Remove non-product link with the checkout code MYLQGR2P. Without the secret code, the hub’ll set you back $40.
This hub adds three USB 3.0 Type A ports to your device, as well as SD and microSD card readers. There’s also an HDMI port capable of 4K resolution output at 30Hz. The USB ports are capable of data transfer or charging, but Aukey recommends you don’t use it for charging anything larger than a smartphone.
That’s a pretty sweet little gadget to tote around, especially for anyone dealing with limited ports on their laptop. The bottom of the hub features a non-slip padding to keep it in place, and in the box you get the hub, a travel pouch, and the USB-C cord for connecting to the hub.
Aukey also provides a 45-day money-back guarantee, as well as a two-year replacement warranty. If you need an all-purpose hub for a device with a USB-C port, Aukey’s looks like a versatile, well-stocked option, and today it’s going for a great price—if you know the right code, that is.
Note: When you purchase something after clicking links in our articles, we may earn a small commission. Read our affiliate link policy for more details.
Ian is an independent writer based in Israel who has never met a tech subject he didn’t like. He primarily covers Windows, PC and gaming hardware, video and music streaming services, social networks, and browsers. When he’s not covering the news he’s working on how-to tips for PC users, or tuning his eGPU setup.