Strava announced a slew of features, including AI to weed out leaderboard cheats, a new ‘family’ subscription plan, dark mode and more.
© 2024 TechCrunch. All rights reserved. For personal use only.
Apple is refreshing its iPad lineup on Tuesday, which is a bit overdue, as the previous iPad Air was announced more than two years ago. The most important addition to the new iPad Air is that it now comes in two sizes: 11 and 13 inches. This display is still an LED display; if you […]
© 2024 TechCrunch. All rights reserved. For personal use only.
At its iPad-focused event on Monday, Apple announced a new and improved Magic Keyboard, its keyboard accessory for iPad.
The Magic Keyboard has been “completely redesigned” to be much thinner and lighter, Apple says, and now includes a function row for quick access to controls like screen brightness, volume adjustment and play/pause. Beyond that, the new Magic Keyboard features aluminum palm rests and a larger trackpad. Plus it’s more responsive, Apple says, with haptic feedback and a USB-C port for charging.
It’s the first major revision of the Magic Keyboard since its launch in 2020. And — with the addition of the function row — it’s now on par, feature-wise, with its counterpart the Magic Keyboard Folio.
The new Magic Keyboard comes in two sizes — one for the 11-inch iPad Pro and one for the 13-inch model — and in two colors, black and white. It can be pre-ordered today for the same price as the previous-gen Magic Keyboard, $299 for the 11-inch and $349 for the 13-inch, and will be available in stores next week.
In other keyboard accessory news, there’s a new Smart Folio for iPad Air. It attaches magnetically and supports multiple viewing angles for “greater flexibility” than the old model.
The Smart Folio is available in charcoal gray, light violet, denim and sage and priced at $79 for the 11-inch iPad Air version and $99 for the 13-inch version.
Other announcements at Tuesday’s event included a new iPad Air with an M2 chip and first-ever 13-inch size; a new iPad Pro with completely new M4 chip and stacked OLED screens for higher-fidelity display; a Pro version of the Apple Pencil featuring new sensors; and a new version of the Magic Keyboard. You can catch the full video here:
The era of stationary spacecraft may soon be over.
Portal Space Systems, a company headed by propulsion expert Jeff Thornberg, is looking to help usher in a new renaissance in in-space transportation with its ultra-mobile Supernova satellite bus. Think of it as maneuverability-as-a-service — tech that could propel transfers from low Earth orbit to geostationary orbit in a matter of hours.
The 500-kg Supernova is designed to be payload agnostic and to survive on orbit for at least five years while being continuously maneuvered, the company said. It’s a far cry from legacy spacecraft, which are built to carry only as much propulsion as will be needed to maintain their orbit. Portal’s designed a novel propulsion system, called solar-thermal propulsion, that will produce a massive delta-V of 6 kilometers per second, to enable rapid transfer to cislunar space, or as maneuverable assets to support national security missions.
Portal was founded by space industry veterans Jeff Thornburg, COO Ian Vorbach and VP of engineering Prashaanth Ravindran. Thornburg’s career reads like a history of the space industry writ large. Spanning nearly 30 years, it includes stints at basically ever major aerospace organization: as a military officer in the Air Force; working on liquid rocket engine tech at Aerojet; five years at SpaceX, where he eventually became the second VP of propulsion engineering; working with storied entrepreneur Paul Allen at Stratolaunch; and as a director at Amazon’s Project Kuiper.
“Then I decided I didn’t really want the big company lifestyle and really wanted to dig in on some problems that were not being addressed by any other business, and so my cofounders and I founded portal space systems in November of 2021,” he said. “It’s the only time I feel like I’ve been able to predict the future, even close.”
Given that he has spent most of his career in launch vehicle development and propulsion, he started thinking about what would come next in the industry after SpaceX cracked the code on affordable, rapid, reliable launch. He noticed other trends, too: with more spacecraft launched to orbit than ever before, on-orbit collisions become more likely. On the defense side, Thornburg also started seeing more of a national security interest in responsive space capabilities and the ability for satellites to maneuver ‘without regret.’
But satellites aren’t designed for moving around a lot in orbit; in general, they are launched with enough on-board propulsion to keep them in their intended orbit, not to dodge other objects, and certainly not with out delta-V capability to go to higher orbit. Supernova is part of a new generation of spacecraft that are looking to change that.
The company has developed Supernova’s propulsion system in-house. The solar-thermal system leverages legacy tech designed by NASA and the the Department of Defense, while innovating specific subsystems to optimize for mission performance. For example, the company is bringing a proprietary heat exchanger to that solar thermal propulsion system to deliver even higher performance over a longer lifespan.
“I think the big forcing function here for commercial is just the proliferation of these LEO constellations that they have to maneuver around, and the forcing function for the DOD is China and the future engagement that people believe we’re going to have with them over Taiwan and other issues. Both of those things have come together at the same time, which has created the opportunity that I was hoping for, but didn’t quite predict it being this significant.”
Portal has raised an undisclosed amount of venture funding from unnamed backers, and the first Supernova development is fully funded, the company said. The team has also won over $3 million across five awards from the Space Force and the DOD, including a direct-to-phase II SBIR that’s focused specifically on developing responsive space operations.
Portal is aiming to conduct an in-flight demonstration in late 2025 or early 2026. But before that mission even launches, Thornburg said the startup is looking to scale its team — which stands at around 25 people, but anticipates will swell to up to 200 in 25 years — and grow manufacturing to support the production of multiple spacecraft per year after that first demonstration.
Welcome, folks, to Week in Review (WiR), TechCrunch’s weekly news recap. The weather’s getting hotter — but not quite as hot as the generative AI space, which saw a slew of new models released this week, including Meta’s Llama 3.
In other AI news, Hyundai-owned robotics company Boston Dynamics unveiled an electric-powered humanoid follow-up to its long-running Atlas robot, which it recently retired. As Brian writes, the new robot — also called Atlas — has a kinder, gentler design than both the original Atlas and more contemporary robots like the Figure 01 and Tesla Optimus.
Turning our attention to YouTube for a moment, Dom and Amanda wrote about how Marques Brownlee (MKBHD), the famed gadget reviewer, shouldn’t be blamed for the fate of AI startup Humane AI, whose product, the Ai Pin, Brownlee gave a scathing review of earlier this week. They point out that Humane is a well-funded company with plenty of funds in the bank to burn, and find that critics of Brownlee — who accuse him of being unfairly harsh — have misplaced their rage.
Lots else happened. We recap it all in this edition of WiR — but first, a reminder to sign up to receive the WiR newsletter in your inbox every Saturday.
X charges for posting: X CEO Elon Musk is planning to charge new X users a small fee to enable posting on the social network in an effort to curb what he describes as a “bot problem.”
Change ransomware: An extortion group has published a portion of what it claims are the private and sensitive patient records on millions of Americans stolen during the ransomware attack on Change Healthcare in February.
Tesla adjusts prices: In more Tesla news, the automaker ditched EV inventory price discounts in what CEO Elon Musk characterized as a move to “streamline” sales and delivery. Tesla also dropped the price of its advanced driver assistance package, Full Self-Driving, to $99 per month in the U.S.
Mars free-for-all: Devin reports that space startups are licking their lips over NASA’s decision to convert its $11 billion, 15-year mission to collect and return samples from Mars into essentially a commercial free-for-all.
Waymo problems: Six Waymo robotaxis blocked traffic moving onto an on-ramp in San Francisco on Tuesday. It’s not the first time Waymo vehicles have caused a road blockage, notes Rebecca — but this is the first documented incident involving a freeway.
Google Cloud bets on generative AI: Ron writes about how Google Cloud is investing heavily in generative AI, as evidenced by the string of announcements during Google’s Cloud Next conference earlier in the month.
Generative AI in health: Generative AI is coming for healthcare — but not everyone’s thrilled. Some experts don’t think the tech is ready for prime time.
Airchat, for talking: Anthony breaks down the hype over Airchat, an app launched by former AngelList founder Naval Ravikant and ex-Tinder product exec Brian Norgard that focuses on voice, not text.
Astranis has taken the wraps off a new generation of communications satellites that will serve broadband to customers on Earth from geostationary orbit, but faster and smaller than any comsat up there. They believe the future of orbital communications is not just in higher orbits, but in the possibility of customers — government and commercial — having their very own private satellite network.
Called Omega, the new class of satellites will each provide some 50 gigabits per second of bandwidth in both civilian and military Ka bands — making it clear from the outset that this is intended to be a dual-use technology.
Astranis builds and operates relatively small broadband satellites in high orbits, and sells that capacity on to telecom and internet service providers. The company has contracts to provide capacity to providers in Mexico, the Philippines, Alaska and southeast Asia.
The startup takes pride in the comparatively diminutive size of its GEO satellites, which are normally huge and, as a result, easy to track and potentially attack.
“We need to move to a more resilient architecture. No more big, fat, juicy targets!” said Astranis CEO John Gedmark at an event at Space Symposium where the news was announced.
The improved bandwidth is thanks to a next-gen Astranis software-defined radio, but the signal is deployed more efficiently; while the previous generation sent down a set of coherent beams, like spotlights, the new generation is more like a big LED array, providing even signal across a much greater area. Gedmark said that although the number of points that can be served depends on the customer and use case, it is theoretically in the millions. The satellites use existing Ka-band receivers rather than a bespoke antenna like Starlink’s.
Speaking of competitors: When asked about how the orbital communications market would develop in the near term, Gedmark was highly optimistic. He said that the appetite for bandwidth is effectively unlimited, at least at the prices they are able to offer, which are well below legacy GEO data connections.
Image Credits: Astranis (opens in a new window)
Notably, Astranis said the satellite will support specific waveforms that are of interest to the DOD, like the Protected Tactical Waveform, so it can still provide capacity even in contested environments. Astranis’ proposal — many small satellites in GEO — is a far cry from legacy tech, which has generally relied on very large, and very expensive, non-maneuverable satellites in GEO. In other words, sitting ducks for adversaries.
Like the company’s current satellites, Omega will have the ability to maneuver in GEO using on-board all-electric propulsion. Astranis said the more efficient thrust will allow it to keep its station for at least 10 years, as well as perform plenty of repositioning and other maneuvers. By that time the next generation will probably be ready to slot into place.
What will perhaps be Astranis’ standout product, however, will be dedicated satellites for customers. Obviously nations have their own dedicated spy satellites and the like, but these cost hundreds of millions of dollars and are often funded by defense budgets. But even multinational corporations don’t tend to have that kind of cash laying around, for that purpose at least — and if they did, they don’t tend to have satellite management departments. Astranis plans to essentially offer “satellite as a service” instead, where for an upfront and monthly fee a satellite can be tasked completely (or in part) to the use of a single customer.
Gedmark declined to name any of the companies that had expressed interest or were being wooed in other ways, but he did suggest that energy and oil and gas companies are an obvious one, with holdings across large geographical areas and demand for a good deal of secure satellite data. He also said that, while there are no official plans as yet to approach the cislunar market, there is a huge opportunity there for future growth.
The company aims to complete the first Omega satellite in 2025 and launch to orbit in 2026. The plan is to launch on the order of six satellites at that time, with as many as 24 per year being launched after that depending on how manufacturing scales up.
Meta, hell-bent on catching up to rivals in the generative AI space, is spending billions on its own AI efforts. A portion of those billions is going toward recruiting AI researchers. But an even larger chunk is being spent developing hardware, specifically chips to run and train Meta’s AI models.
Meta unveiled the newest fruit of its chip dev efforts today, conspicuously a day after Intel announced its latest AI accelerator hardware. Called the “next-gen” Meta Training and Inference Accelerator (MTIA), the successor to last year’s MTIA v1, the chip runs models including for ranking and recommending display ads on Meta’s properties (e.g. Facebook).
Compared to MTIA v1, which was built on a 7nm process, the next-gen MTIA is 5nm. (In chip manufacturing, “process” refers to the size of the smallest component that can be built on the chip.) The next-gen MTIA is a physically larger design, packed with more processing cores than its predecessor. And while it consumes more power — 90W versus 25W — it also boasts more internal memory (128MB versus 64MB) and runs at a higher average clock speed (1.35GHz up from 800MHz).
Meta says the next-gen MTIA is currently live in 16 of its data center regions and delivering up to 3x overall better performance compared to MTIA v1. If that “3x” claim sounds a bit vague, you’re not wrong — we thought so too. But Meta would only volunteer that the figure came from testing the performance of “four key models” across both chips.
“Because we control the whole stack, we can achieve greater efficiency compared to commercially available GPUs,” Meta writes in a blog post shared with TechCrunch.
Meta’s hardware showcase — which comes a mere 24 hours after a press briefing on the company’s various ongoing generative AI initiatives — is unusual for several reasons.
One, Meta reveals in the blog post that it’s not using the next-gen MTIA for generative AI training workloads at the moment, although the company claims it has “several programs underway” exploring this. Two, Meta admits that the next-gen MTIA won’t replace GPUs for running or training models — but instead will complement them.
Reading between the lines, Meta is moving slowly — perhaps more slowly than it’d like.
Meta’s AI teams are almost certainly under pressure to cut costs. The company’s set to spend an estimated $18 billion by the end of 2024 on GPUs for training and running generative AI models, and — with training costs for cutting-edge generative models ranging in the tens of millions of dollars — in-house hardware presents an attractive alternative.
And while Meta’s hardware drags, rivals are pulling ahead, much to the consternation of Meta’s leadership, I’d suspect.
Google this week made its fifth-generation custom chip for training AI models, TPU v5p, generally available to Google Cloud customers, and revealed its first dedicated chip for running models, Axion. Amazon has several custom AI chip families under its belt. And Microsoft last year jumped into the fray with the Azure Maia AI Accelerator and the Azure Cobalt 100 CPU.
In the blog post, Meta says it took fewer than nine months to “go from first silicon to production models” of the next-gen MTIA, which to be fair is shorter than the typical window between Google TPUs. But Meta has a lot of catching up to do if it hopes to achieve a measure of independence from third-party GPUs — and match its stiff competition.
On Tuesday Amazon launched a new service called Deadline Cloud that lets customers set up, deploy and scale up graphics and visual effects rendering pipelines on AWS cloud infrastructure. The new service, which is geared toward the media and entertainment industry, was timed for the National Association of Broadcasters conference in Las Vegas that kicks off later this month.
Using Deadline Cloud, customers in media and entertainment as well as architecture and engineering can use AWS compute to render content for TV shows, movies, ads, video games and digital blueprints, said AWS GM of creative tools Antony Passemard.
In other words, AWS is betting on increasing demand for tools that help media, entertainment and other execs navigate the ins and outs of cloud-based rendering.
“We’re at a tipping point in the industry where demand for rendering quality VFX and the amount of content created using generative AI are outpacing customers’ [compute] capacity,” Passemard added in a blog post. “AWS Deadline Cloud meets any customer’s rendering requirements by providing a scalable render farm without having to manage the underlying infrastructure.”
A startup wizard in Deadline Cloud walks customers through the process of setting up a render farm, including providing the size and duration of their projects to determine instance type and configuring permissions. Deadline Cloud then provisions Amazon Elastic Compute Cloud instances and manages the network and compute infrastructure. And — for customers with on-premises compute — Deadline Cloud integrates with this compute and uses it to execute rendering jobs.
Deadline Cloud’s dashboard provides a view to analyze logs, preview in-progress render jobs and review and control costs. With Deadline Cloud, customers can link their own third-party software licenses with the service or leverage usage-based licensing for rendering with existing rendering tools (e.g., Autodesk Maya, Foundry Nuke, and SideFX Houdini) and engines.
“[With Deadline Cloud,] creative teams can embrace the velocity of content pipelines and respond quickly to opportunities to accept more projects, while meeting tight deadlines and delivering high-quality content,” Passemard continued.
Deadline Cloud is now available in the U.S. East (Ohio, North Virginia), U.S. West (Oregon), Asia Pacific (Singapore, Sydney, Tokyo), and Europe (Frankfurt, Ireland) AWS server regions.
Cloud-based rendering is nothing new. Back in 2015, Google made a splash in the space with the acquisition of Zync, whose technology has since been used to launch Google Cloud–powered visual effects tooling in partnership with Sony’s animation studio, Sony Pictures Imageworks. Elsewhere, platforms like Arch and Chaos Cloud have provided on-demand cloud-based VFX infrastructure for years.
But the COVID-19 pandemic accelerated VFX workloads’ move to the cloud as the cost of maintaining hardware — and the space to store it — increased while work simultaneously dwindled, the result of work-from-home mandates and health-related shutdowns of productions. As Passemard alluded to, the rise of generative AI has fueled the demand for rendering hardware, too, and has led to the creation of entirely new cloud-based, GPU-accelerated providers.
Orbit Fab wants to build “gas stations” for satellites — which means it needs the gas cap, a mechanism for transferring propellant from an orbital tanker to the customer spacecraft. That docking mechanism, called RAFTI, is now flight-qualified and on the market. The price tag for each port? Just $30,000.
The Colorado-based startup (and former TechCrunch Disrupt Battlefield finalist) has been in operation since 2018, and its CEO and co-founder Daniel Faber has been working in the space industry for decades; he’s likely best known for heading up Deep Space Industries (DSI), a company that was targeting asteroid mining. The company, which was founded in 2012, was acquired by Bradford Space seven years later.
“If you want [to talk about] something that’s too early, that’s it,” he joked during a recent interview. As part of the company’s efforts to eventually build tech capable of prospecting a distant asteroid, DSI built satellite thrusters for orbital maneuvering. This work, and subsequent conversations with customers and colleagues, eventually led Faber to believe that the next big opportunity was in-space refueling.
Part of it is simple math: Colleagues and former customers told him that they could squeeze as much as $1 million in marginal revenue from satellite missions from an extra kilogram of propellant.
“Spacecraft are optimized with the amount of fuel they have, and when they get to the end of it, an extra kilo would give them a million dollars of marginal revenue,” Faber said. “We create so much value from that, we just have to do it.”
The 2010s also saw the emergence of a handful of satellite servicing companies, like Astroscale, which are developing technology for space debris removal, satellite life extension or last-mile satellite delivery. Faber calls these capabilities “tow truck applications,” and he realized that there would eventually be a need for orbital gas stations to complement this fleet.
So Orbit Fab was born. In the first year of operating, the company raised a $6 million seed round with contributions from Bolt and Munich Re Ventures, the VC arm of Munich Re Group, one of the biggest underwriters of satellites and rockets. In 2023, the company raised a $28.5 million Series A round.
The startup’s technology is ambitious, but the architecture is fairly simple: The idea is to equip customer satellites with the refueling port (Faber referred to it as a “gas cap,” but it’s officially called RAFTI) while the hardware is still on Earth. RAFTI, which stands for “Rapidly Attachable Fluid Transfer Interface,” can also be used to fuel spacecraft on the ground prior to launch. Once a RAFTI-equipped satellite runs out of propellant, one of Orbit Fab’s tankers would be able to pick up some fuel from orbital depots and deliver it straight to the customer’s satellite for refueling.
The only two things the company sells are fuel and the fueling ports; as one might expect, the real money will come from the fuel sales. On its website, Orbit Fab says its hydrazine delivery service in geostationary orbit will cost $20 million for up to 100 kilograms.
Given the simplicity of the architecture, nailing each part of the hardware is critical; hence why it’s taken years for Orbit Fab to debut the refueling port. There are many variables to consider: the cost to the customer, versus the potential marginal revenue from extra life on orbit; the impact of fueling on the customer spacecraft; and the challenge of developing a docking mechanism that can also transfer propellant.
In addition to all of these challenges, the company had to ensure its component complied with standards from NASA, the Space Force and the American Institute of Aeronautics and Astronautics to ensure it is safe, reliable and able to withstand the harsh environment of space.
“This wasn’t cheap,” Faber said. “It wasn’t quick, but at the end of the day, we have an elegant design that meets those requirements and has a simplicity to it that comes from doing a design well.”
One of the biggest shifts from when he started the company to now, Faber said, is the standing up of the U.S. Space Force and the effects that’s had on the space industry writ large. Orbit Fab ended up pivoting much of its attention to addressing the nascent needs of the Space Force, which was very interested in orbital mobility to avoid space debris or rendezvous with other satellites.
The company anticipates the first RAFTIs will go to orbit on customers’ satellites later this year. That will be followed by the first fuel shuttle going up next year, as part of a contract with the DOD to deliver fuel in geostationary orbit in 2025. Orbit Fab is aiming to sell 100 fueling ports this year, which will put the RAFTI “on a decent percentage of satellites going to orbit,” Faber said. Orbit Fab has an additional agreement with an unnamed commercial customer to deliver “a significant quantity of fuel” in a few years, he added.
Beyond these milestones, Faber intimated that the company already has plans to upgrade RAFTI, and to design variants that could support higher-pressure propellants. The team is also thinking about redesigning the grapple housing for larger spacecraft, should the market indicate that’s where they should go next.
“SpaceX has made rockets reusable, Orbit Fab makes satellites reusable,” Faber said. “In this world today, if you’re running a rocket company, and you’re not working towards reusable rockets, you’re working to a dead end. The same is true of satellites: If you’re not making your satellites reusable, you’re just putting preordained junk into orbit.”
Founded in 2017, Atoms Lanka Solutions is a provider of IT consulting and software development services. Throughout the company’s history, we have worked with over 160 customers from 5+ countries..
