Mashable

2 hours ago

Get all-in-one iPhone scanner and PDF app for life for £20

For a limited time, Mashable readers can get a lifetime subscription to the iScanner app for just £19.87 (regularly £158.20) by using the code SCAN at checkout. This portable scanner and PDF editor allows users to scan documents on their iOS devices and upload them as PDFs in seconds, eliminating the need for public scanners or home printers. In addition to scanning, it also offers complete PDF editing and management features, including signing, editing, merging, and protecting PDF files. The app includes various scanning modes and AI tools such as border recognition and automatic adjustments and recognizes text in over 20 languages. This all-in-one PDF app is useful for professionals in various fields, business owners, and students and is on sale at the Mashable Shop until March 30th.

The Verge

2 hours ago

Lifehacker.Tech

8 hours ago

Lenovo Is Showing off a Bunch of Quirky Laptop Prototypes

Lenovo is showcasing three new concept devices at Mobile World Congress, including a foldable laptop and snap-on laptop monitors. The ThinkBook Flip AIPC features a screen that folds over the top half of the device, creating a tall vertical display or mirroring the front screen to the back for presentations. This design eliminates the need for a rolling mechanism, making it taller than a rollable laptop but also more top-heavy. Additionally, Lenovo is presenting the Magic Bay Dual Display and 2nd Display laptop accessories, which are snap-on monitors that attach to the back of certain ThinkBook laptops via pogo pins, offering extended display capabilities.

Wired

6 hours ago

This Modular Phone Concept Is Xiaomi’s Plan to Kill the Camera Bump

Xiaomi showcased a concept phone with a detachable lens at MWC 2025. The modular optical system combines a modified Xiaomi 15 smartphone with a 35-mm lens module that attaches magnetically. Tiny pogo pins for power and a proprietary LaserLink technology for data transfer allow shooting photos in the phone’s camera app as if using the regular built-in camera. The lens tested was a Xiaomi 35-mm f/1.4 lens with a 100-megapixel Light Fusion X Type 4/3 sensor. The detachable lens was as responsive as the regular camera, and focused on subjects at different distances to achieve a lovely natural bokeh.

The Verge

7 hours ago

The Verge

7 hours ago

Engadget

8 hours ago

Lenovo is updating its ThinkPad lineup with new chips and form factors at MWC 2025

Lenovo is showing business laptops lovers some attention by announcing several new ThinkPads at Mobile World Congress 2025, including a new form factor for the T-series line, and updated internals across the business laptop lineup. The ThinkPad T14s 2-in-1 is by far the most interesting of the bunch, with a new convertible body that’s similar to Lenovo’s Yoga laptops, and supports the magnetic Yoga Pen stylus. Lenovo describes the new ThinkPad X13 Gen 6 as “one of the lightest ThinkPad designs ever,” at only 2.05lbs, but that lightweight doesn’t mean the laptop misses out on the latest internals.

Engadget

8 hours ago

Lenovo’s codename Flip is another wild take on a flexible-screen laptop

Lenovo hasn’t even released its futuristic laptop with a rollable screen, but that hasn’t stopped the company from remixing it into a whole new thing at MWC 2025, officially called the ThinkBook codename Flip AIPC Proof of concept. The laptop uses the same flexible OLED featured in its rollable sibling, except this time instead of disappearing inside its chassis, the display folds outwards, allowing what is normally a 13-inch panel to double in size to 18.1 inches at a moment’s notice. Down below, Lenovo created what it calls a Smart Force Pad, which features a three-layer illuminated dashboard that adds customizable controls and icons to its surface.

Engadget

8 hours ago

Lenovo gave its latest 16-inch ThinkBook a bevy of funky concept screen accessories

The ThinkBook line has become Lenovo’s defacto testing ground for far-out ideas and ambitious concepts. At MWC 2025, the company is pushing things further than ever through the new ThinkBook 16p Gen 6 and four of its display accessories. The ThinkBook 16P sports up to an Intel Core Ultra 9 2785HX chip with as much as 64GB of RAM, 2TB of PCIe storage and an NVIDIA RTX 5070 GPU. Lenovo includes a discrete NPU (separate from its main chip) that offers up to 32 TOPS of additional performance. For people who constantly yearn for extra space to multitask while traveling, Lenovo has two add-ons to choose from.

The Verge

8 hours ago

The Verge

8 hours ago

The Verge

8 hours ago

Mashable

8 hours ago

Forget flip phones, Lenovo’s latest concept is a flip laptop

Lenovo is showing off weird and interesting new product ideas at tradeshows. At MWC it’s doing the same thing, with a host of unique, fun, and down right silly ideas, some of which could inform designs and decisions for future products. The central theme to most of Lenovo’s concept ideas has to do with getting more screen from a traditional laptop. Perhaps the most unique concept for that comes in the form of the ThinkBook Flip AIPC Concept, which has a whole second screen that can fold up from behind the first one. Another concept leverages the Magic Bay connector to add a tiny companion device called Tiko to the top of your screen.

Engadget

8 hours ago

XTRIS is a fast-paced arcade-style Playdate game you won’t be able to put down

XTRIS is a $3 game by Fletch Makes Stuff that’ll have you chasing a moving X around a board made up of either four, five or six tiles. The timer’s set to 60 seconds, and in that time you have to land on the X as many times as possible. The X moves fast, and as you race after your target using the D-pad to control the movement of a black square, things become increasingly intense. The music builds like an alarm that’s telling you time is almost up, and flames start shooting up from the bottom of the screen. There’s a Zen Mode for each level as well, so you can work on your dexterity without the time crunch.

TechCrunch

8 hours ago

HackerNoon

9 hours ago

Next-Gen 3D Detection: Challenges, Innovations & Future Directions

Panopticus is an initial effort toward adaptive omnidirectional 3D detection, and the current implementation operates under the assumption of a single-task execution environment. Future improvements of Panopticus considering the multi-task execution environment could be achieved in several ways, such as monitoring the runtime dynamics caused by resource contention among concurrent tasks, exploring how to co-design multi-branch models for different 3D tasks and optimizing those models given the constraint of application and device, and better utilizing available heterogeneous processors, such as CPUs and NPUs, could enhance the efficiency of the multi-task workload.

HackerNoon

9 hours ago

Advancements in Adaptive and 3D Object Detection for Edge AI

Computer vision systems designed for resource-constrained edge devices have achieved efficient resource utilization by adapting to the video content or computing budgets. Existing adaptive methods fall short in supporting resource-efficient omnidirectional 3D detection. This type of detection requires processing each camera view using different 3D perception capabilities, such as enhanced estimation of objects’ 3D location and velocity. To process each view optimally, Panopticus predicts its expected performance based on short-term future dynamics in the spatial distribution, which is crucial for mobile scenarios.

HackerNoon

9 hours ago

Evaluating Panopticus: Performance, Robustness, and Component Analysis

Omnidirectional 3D detection is crucial for various outdoor scenarios, such as robots navigating urban streets. Datasets are summarized in Table 3, which we detail in the following. For the public dataset, we used the nuScenes with large-scale ground-truth annotations in urban driving scenes. We used six object types in the dataset—car, truck, bus, pedestrian, motorcycle, and bicycle—excluding less common types such as construction vehicle. We built a mobile testbed based on a handheld 360° camera, LiDAR, and IMU sensors and calibrated the sensors’ extrinsic to enable coordinate transformation among them.

HackerNoon

9 hours ago

Edge Devices Are Getting Better at Seeing the World in 3D

The goal of our multi-branch model is to allow each region in a surrounding space can be processed through various capabilities of the BEV-based 3D detectors detailed in Table 1. Instead of simply using all types of models loaded on the memory, we modularized the models and combined the core modules, eliminating redundant ones. Each branch within the model represents a unique execution path for an image, comprising a sequence of module decisions at all stages. At this stage, each image can be processed through one of four backbone networks, each of which is differentiated by its capacity and input resolution.

HackerNoon

9 hours ago

How Panopticus Uses AI to Detect Objects in 3D

Panopticus was implemented using Python and CUDA for GPU-based acceleration. All neural networks were developed using PyTorch and trained on the training set in the nuScenes dataset. The neural networks for 3D object detection are developed based on MMDetection3D. For the camera motion network, we customized and trained to produce consistent relative poses between two consecutive frames for any given camera view. We used the GPU-accelerated XGBoost library and linear model in scikit-learn for performance predictors, and PuLP for the ILP solver.

HackerNoon

9 hours ago

Optimizing AI for Real-Time Object Tracking

Panopticus dynamically schedules the inference configuration of the multi-branch model at runtime to maximize detection accuracy within the latency target. Specifically, the goal of the scheduler is to find the optimal selection 𝑐∗ for 𝑁 image-branch pairs, given the possible inference branches { 𝐵𝑖}𝑀𝑖=0 and multi-view images {𝐼𝑗𝑡}𝑁𝑗=0 at time 𝑡. Panopticus first predicts the expected accuracy and latency of each pair. As described in Section 3, the model’s detection capabilities are influenced by spatial characteristics, such as the number, location, and movement of objects.

HackerNoon

9 hours ago

Efficient 3D Object Detection for Edge Devices

Panopticus is an omnidirectional 3D object detection system designed for resource-constrained edge devices that aims to maximize detection accuracy while meeting the latency objective with a given edge device. To accommodate camera views with varying inference requirements, we propose an omnidirectional 3D detection model equipped with multiple inference branches. At runtime, each of 𝑁 multi-view images at time 𝑡 is processed by the proper branch to generate accurate 3D bounding boxes. The optimal selection of branch and image combinations is determined by a spatial-adaptive execution scheduler.

HackerNoon

9 hours ago

Optimizing Omnidirectional 3D Object Detection for Edge AI

We present in-depth observations on methods to improve the accuracy of the baseline BEVDet model mentioned in Section 2. Based on the results, we derived key insights and challenges in the design of Panopticus. Each variant utilizes one of four combinations of ResNet backbones and input resolutions listed in Table 1, such as ResNet34 backbone with a resolution of 256×448, which we refer to as BEVDet-R34. For evaluation metrics, we used average precision (AP), widely employed to assess detection accuracy, and true positive (TP) errors.

HackerNoon

9 hours ago

How 3D Object Detection Helps AI Avoid Obstacles

3D object detection aims to identify objects in space and predict their properties such as 3D location, size, and velocity. The predicted object information is utilized by application functionalities such as obstacle avoidance for robot navigation. Existing methods for omnidirectional 3D object detection utilize LiDAR sensors or multiple cameras providing a 360° perception range. Recent camera-based detectors aggregate information from multiple camera images into a bird’s-eye-view (BEV) space, providing a top-down representation of the surrounding 3D space.

HackerNoon

9 hours ago

This AI Model Gives Edge Devices Eyes on the Back of Their Heads

The paper introduces Panopticus, a system designed for omnidirectional and camera-based 3D object detection on edge devices. It addresses the challenges of computational limitations by employing an adaptive multi-branch detection scheme that accounts for spatial complexities. Panopticus dynamically adjusts the model’s architecture and operations based on available edge resources and spatial characteristics to optimize accuracy within latency limits. Experiments conducted on edge devices in real-world environments demonstrate that Panopticus improves accuracy by 62% on average, given a strict latency objective of 33ms, and achieves a 2.1x latency reduction compared to baselines.

HackerNoon

11 hours ago

Python: A Deeper Look Into How Django Transactions Work

Django’s transaction management code does the following: opens a transaction when entering the outermost atomic block; requests and acquires locks when a DB operation is encountered; creates a savepoint when entering an inner atomic block; releases or rolls back to the savepoint when exiting an inner block; and when code that doesn’t touch the DB is encountered the code is executed as usual. The locks on the DB are still held while non-DB logic is executed.

HackerNoon

12 hours ago

Text-to-SQL Was Supposed to Be AI’s Killer App. It’s Not.

The author recounts the inspiration behind Fabi.ai, stemming from the idea of using LLMs for text-to-SQL to empower non-technical users with self-service analytics. After validating the concept, the author and their CTO embarked on a journey to create Fabi.ai. The core idea is to create an AI that can answer any questions about an organization’s data and create a “data-driven” organization where every leader is leaning on empirical evidence to make their strategic decisions. They also share core learnings from working with LLMs for data analysis.

HackerNoon

13 hours ago

Zoom M3 MicTrak File Recovery: An Adventure Into the World of Data Recovery

A journey deep into the binary encoding of RIFF files and a data recovery tool. The main approach is two-fold: capture what you can from the dying medium and try to recreate the data. Sometimes, just by adjusting read tolerance, you’re able to capture everything from a disk Windows or Mac reject, as the OS times out faster than the dying medium can respond. At the extreme, you’re left with nothing but some of the raw data. All you can do is carve it out – essentially reading every byte from the medium, identifying known headers like JPG(FFD8) or MKV(1A45DFA3), and proceed to capture all sequential data until the end of the file.

HackerNoon

13 hours ago

Unveiling GitHub’s Secret Blueprint—How to Handle Millions of Transactions Daily

GitHub is a highly scalable distributed system that processes millions of transactions daily. It relies on robust algorithms and architecture to ensure high performance and reliability. This article explores how GitHub processes vast amounts of data, scales to handle millions of transactions, and employs diff algorithms to track file changes efficiently. When tracking changes in a file, GitHub uses diff algorithms to compute the minimum number of edits needed to transform one version of a file into another. A widely used algorithm for this is Myers’ Diff Algorithm.

HackerNoon

14 hours ago

Startups of the Year 2024: General Interview

Cosmonova Broadcast has been nominated in HackerNoon’s annual Startups of The Year awards in Kyiv, Ukraine. Cosmonova Broadcast is a telecommunications company providing worldwide cloud-based storage, management, and delivery solutions for TV channels, rights holders, and online media. From channel creation, content management, and streaming to AI-driven voice translation, we provide a comprehensive, scalable, and future-proof media infrastructure. Key indicators of product-market fit include long-term partnerships with major media companies and global streaming providers, scalability and flexibility, and adoption of AI-driven voice translation.

HackerNoon

14 hours ago

Bun just got 60% faster, Doom now runs on TypeScript, and OpenAI launched GPT 4.5 – This Week in JS

This week in JavaScript includes OpenAI’s release of GPT-4.5 “Orion,” which excels in coding tasks and creative scenarios but comes at a higher cost. Dimitri Mitropoulos ran the classic game Doom within TypeScript’s type system. Bun versions 1.2.3 and 1.2.4 are packed with updates including frontend bundling, built-in routing, SQL enhancements, and a macOS speed boost. Lastly, Firefox’s privacy controversy is also mentioned.

Engadget

8 hours ago

Lenovo’s new AI laptops for MWC include the Yoga Pro 9i Aura edition and the IdeaPad Slim 3x

Lenovo has unveiled new laptops with generative AI features at MWC 2024, including the Yoga Pro 9i Aura edition, which is powered by the Intel Core Ultra processor and NVIDIA GeForce RTX 5070 GPU. The Aura edition laptops have been created in partnership with Intel to deliver personalized, productive, and protected computing experiences. The Yoga Pro 9i Aura edition features a 16-inch 3.2K PureSight Pro display with increased brightness and color accuracy. It also features the Lenovo Creator Zone to make third-party video, photo, and music applications easier to use. The IdeaPad Slim 3x, a Copilot+ PC powered by Snapdragon X chip, can process AI tasks on-device and offers an open SSD slot for expanded storage. The Yoga Pro 9i Aura Edition will be available in the second quarter of 2025, starting at $1,799, while the IdeaPad Slim 3x will be available this month starting at $649.

Engadget

8 hours ago

The Lenovo Solar PC Concept feels like a device whose time has come

At MWC 2025, Lenovo showcased the Yoga Solar PC concept, a 14-inch clamshell laptop integrating solar panels into its lid. Weighing 2.6 pounds and less than 0.6 inches thick, it features an OLED display, up to 32GB of RAM, a 50.2Whr battery, and a 2MP IR webcam. The solar panels use Back Contact Cell technology for up to 24% solar energy conversion, matching high-end home solar systems. Dynamic Solar Tracking optimizes energy gathering, enabling an hour of video playback after 20 minutes in the sun. It can harvest power indoors from as little as 0.3 watts of light. A bespoke app tracks light absorption. Although there are no plans for commercial production, its sturdy chassis and distinctive design make it technologically feasible.

Gizmodo

8 hours ago

The Verge

8 hours ago

TechCrunch

9 hours ago

TechCrunch

9 hours ago

Gizmodo

10 hours ago

Engadget

10 hours ago

FAA reportedly ordered staff to find millions of dollars to fund Starlink deal

The FAA is reportedly considering replacing its $2.4 billion contract with Verizon with a Starlink deal led by Elon Musk. FAA employees have been directed to find tens of millions of dollars for a Starlink deal after claims that the current air traffic control communications system, mistakenly attributed to Verizon, is “breaking down very rapidly”. Musk has offered Starlink terminals at no cost to restore air traffic control connectivity and 4,000 terminals were already sent to help upgrade the FAA’s networks. The FAA has not confirmed the decision about its Verizon deal yet.

The Verge

10 hours ago

Gizmodo

10 hours ago

Gizmodo

11 hours ago

The Verge

13 hours ago

Mashable

13 hours ago

Worlds largest call center using AI to neutralize Indian employees accents

Teleperformance, the world’s largest call center operator, has invested $13 million in AI startup Sanas to “neutralize” accents of its Indian and Filipino agents, addressing difficulties some customers face in understanding them. Sanas’ technology aims to reduce accent-based discrimination and also features background noise removal to mask the call center’s location. Teleperformance Deputy Chief Executive Officer Thomas Mackenbrock said that the AI technology can neutralize the accent of the Indian speaker with zero latency. In addition to accent neutralization, Teleperformance is also using AI for transcribing calls and coaching new employees, with plans to invest an additional $104 million in AI.

HackerNoon

14 hours ago

Lessons From My First Profitable iOS App ($1,400 in 90 Days)

The author shares six key lessons learned from launching a profitable iOS app with no audience, funding, or marketing experience. These lessons include the importance of market research before coding, the necessity of marketing alongside development, planning monetization from day one, prioritizing UX over perfect code, tracking user behavior instead of guessing, and focusing on a niche to build an audience. The author initially made mistakes such as coding before validating and launching without a marketing plan but achieved revenue growth by focusing on app store optimization, targeted advertising, and social media engagement.

HackerNoon

14 hours ago

Proactive Issue Detection in Cloud Software

The software industry lags behind top tech companies in issue detection, relying heavily on integration tests and traditional alerting. Top companies use sophisticated techniques to detect issues before they impact customers. Synthetic monitoring involves using a simulated customer to probe a system and measure the results at a fixed frequency, alerting when failures breach a threshold. Canary testing exposes new features to a small set of users and monitors their experience for deviations in latency and availability, automatically aborting and rolling back changes if metrics degrade.

HackerNoon

14 hours ago

Could the ByBit Hack Have Been Prevented? Yes—Here’s How

The North Korean Lazarus Group stole $1.5 billion from ByBit’s cold wallet by compromising the Safe UI through stolen AWS S3 credentials. The attack was executed via a valid transaction using a delegate call. Modules and guards can extend Safe wallet capabilities and enhance security. Immutable UI is crucial as the entire system’s security relies on the Web2-based UI. Restricting the wallet to ERC-20 transactions and requiring additional signatures for high-value transactions could prevent attacks or minimize losses.

HackerNoon

14 hours ago

3 Noteworthy Hacks For Your SaaS Product Global Expansion

When expanding a Vertical SaaS product into international markets, balancing market dynamics and product functionality is crucial. Key steps include assessing the chosen geography’s market to identify opportunities and challenges. For HR/Workforce Management SaaS, consider business licenses and legal limits for data transfer, potentially needing local servers. For Unified Commerce Platforms, assess customers’ technical architecture and ensure robust incident support for critical systems. Ensuring infrastructure scalability is also vital for delivering global service.

HackerNoon

15 hours ago

Writing Scalable and Maintainable React Code with SOLID Principles

SOLID principles, originally for object-oriented programming, can be applied to React projects to write scalable, maintainable, and robust software. The Single Responsibility Principle (SRP) suggests each component should have one reason to change, promoting reusability, readability, and testability. A well-structured dashboard with separate components for graphs, new books, and reviews exemplifies SRP. React’s component-based architecture naturally encourages SRP by breaking down UI elements into focused components.

HackerNoon

15 hours ago

The HackerNoon Newsletter: The Secret Hidden Power of Questions (3/2/2025)

The HackerNoon Newsletter for March 2, 2025, highlights top tech stories including O.XYZ’s launch of OCEAN, a Cerebras-powered AI search engine, and articles on game design and the role of humans in vibe coding. Other topics cover backend transitions, the hidden power of questions, Tether’s potential as a strategic reserve buyer, and Ethereum Swarm’s decentralized web foundation. The newsletter also encourages readers to consolidate technical knowledge through writing.

HackerNoon

15 hours ago

Streamlining the API Lifecycle: Interview with Startups of The Year 2024 Nominee, Ambassador

Ambassador has been nominated in HackerNoon’s Startup of The Year awards in Denver, Colorado for their work in API lifecycle development. Ambassador offers tools like the Blackbird API Development Platform and Edge Stack API Gateway designed to accelerate development, expedite testing, and optimize the delivery of API resources. Their tools focus on cutting through typical friction points to ease collaboration and enable more self-serve experiences and reduce risks by streamlining innovation and reduce risk.

HackerNoon

18 hours ago

Can We Simulate Quantum Phenomena? The FPGA Coprocessor Experiment

Quantum computation promises significant processing advancements, though current systems are in early stages. Researchers are exploring simulating quantum behavior using classical digital hardware via Field-Programmable Gate Arrays (FPGA)-based digital quantum coprocessors. Unlike physical quantum computers, digital quantum coprocessors simulate quantum behaviors using normal computer hardware like FPGAs. This approach offers a useful tool for developing quantum algorithms without needing an actual quantum computer, which reduces the barriers for researchers.

HackerNoon

18 hours ago

Monitoring Golang Services with Prometheus: Choosing Between Pull and Push Models

Prometheus is a monitoring system that collects numerical data (metrics) from applications to track key indicators like request numbers, response times, memory and CPU usage, and error counts. It helps answer questions about service efficiency, performance bottlenecks, and infrastructure scaling needs. Prometheus gathers data using a pull model, where it actively queries services for metrics, or a push model, where services push metrics to an intermediary. The pull model is preferred for its simplicity, fewer points of failure, and automatic cleanup.

HackerNoon

19 hours ago

Bitlayer Advances The First BitVM Implementation Through Major Strategic Partnerships

Bitlayer is integrating Bitcoin into various blockchain ecosystems, partnering with Base, Starknet, Arbitrum, Sonic, and Plume Network to enhance Bitcoin adoption and liquidity. The BitVM integration aims to connect a multichain ecosystem through the BitVM integration, offering new opportunities for Bitcoin holders. The Base-Bitlayer BitVM Bridge connects Bitcoin and Base, Arbitrum users will be able to bridge asset, Starknet integration will enable instant transactions, Plume Network integration will advance Real World Assets (RWA), and Sonic integration will power Web3 TikTok App Layer.

Gizmodo

13 hours ago

TechCrunch

14 hours ago

Wired

14 hours ago

Exclusive: Honor Debuts a New AI Agent That Can Read and Understand Your Screen

Honor is developing Honor UI Agent, a GUI-based mobile AI agent designed to handle tasks by understanding a device’s graphical user interface. In a demo at Mobile World Congress 2025, the AI agent was shown booking a restaurant through OpenTable. This development is part of Honor’s $10 billion Alpha Plan, which focuses on prioritizing AI development for personal technology devices. The company aims to create an “intelligent phone” and “co-create a new paradigm for civilization.”

The Verge

15 hours ago

TechCrunch

15 hours ago

Engadget

15 hours ago

Firefly’s Blue Ghost lander successfully touched down on the moon

Firefly Aerospace’s Blue Ghost lander successfully touched down on the moon in the Mare Crisium region as part of NASA’s Commercial Lunar Payload Services (CLPS) program. The lander is upright, communicating with Earth, and expected to operate for about 14 Earth days. It has already transmitted over 27 GB of data and will capture images of lunar events and lunar dust behavior. NASA is also testing instruments for drilling and collecting samples. The company shared a photo showing the lander’s shadow on the lunar surface.

Gizmodo

15 hours ago

Gizmodo

15 hours ago

Wired

16 hours ago

Mustard Made Lowdown Locker Review: Stylish Storage

The author, a product reviewer with ADHD and a love for ‘Things,’ tackles the challenge of organizing an ever-growing collection of belongings, exacerbated by receiving numerous items for review. The Mustard Made Lowdown Locker emerges as one of the best solutions found for storage and organization. The flat-pack locker is relatively simple to assemble with clear instructions and sturdy metal pieces. Despite a minor issue with a back panel, the locker proves to be a functional and aesthetically pleasing addition to the author’s space.

Gizmodo

16 hours ago

The Verge

16 hours ago

Mashable

16 hours ago

Hands-on: Xiaomi 15 Ultra has a monster, 200-megapixel periscope camera

The Xiaomi 15 Ultra features a quad camera system, including a 200-megapixel periscope telephoto camera with up to 4.3x optical zoom, and three 50-megapixel cameras: a main wide camera with a variable f/1.6-f/4.0 aperture, a telephoto camera with 3.2x optical zoom, and an ultrawide camera. The phone has a 6.73-inch AMOLED display with peak 3,000 nits and 120Hz refresh rate, a Qualcomm Snapdragon 8 Elite chip, 12-16GB of RAM, and 256/512/1,024GB of storage. It is powered by a 5,410mAh battery with 90W wired, 80W wireless, and 10W reverse wireless charging. The phone comes in white, black, or a silver-chrome combo and features Android 15 with Xiaomi’s HyperOS.

Engadget

16 hours ago

The Xiaomi 15 Ultra is a Leica-powered photo showcase

The Xiaomi 15 Ultra, powered by a Snapdragon 8 Elite chip, boasts top-notch specs including up to 16GB of RAM and 1TB of storage, a huge 6,000mAh battery, and fast charging capabilities. Its 6.7-inch AMOLED display features a 1-120Hz refresh rate and up to 3,200 nits of brightness. The phone features a massive circular camera module housing a 1-inch 50-megapixel Sony sensor, a 50MP ultrawide cam, a 50MP mid-range shooter with a 70mm focal length, and a 200MP periscope array with 4.7x optical zoom. It also comes with preset Leica filters.

Wired

16 hours ago

Xiaomi 15 and 15 Ultra Review: Photography Focus

Xiaomi’s new 15 and 15 Ultra flagships offer incremental improvements over their predecessors, with the 15 Ultra being a photographer’s dream. Both phones have exceptional hardware and polished HyperOS software with AI integration. However, the Xiaomi 15 is identical in design to last year’s Xiaomi 14 and lacks Qi2 support. The Xiaomi 15 Ultra has a large camera on the back. Both phones score an IP68 rating for water resistance and are protected by Xiaomi’s Shield Glass.

The Verge

16 hours ago

The Verge

16 hours ago

The Verge

16 hours ago

Gizmodo

17 hours ago

TechCrunch

17 hours ago

TechCrunch

17 hours ago

Wired

18 hours ago

Rewind Cassette Player Review: The Best Portable Tape Player

The Rewind cassette player is highlighted as the best option for tape enthusiasts, offering great sound and long battery life. It’s portrayed as a sturdy and reliable alternative to expensive vintage Walkmans or cheap, unreliable players. The article also discusses the appeal of cassettes in 2025, despite their limitations compared to CDs and vinyl, emphasizing their fragility and potential for wear and tear.

Gizmodo

18 hours ago

The Verge

18 hours ago

Wired

18 hours ago

Can Algorithms, a Touchscreen, and an Internet Connection Make the Perfect Piece of Toast?

The Revolution R180 Connect Plus Smart Toaster, priced at $400, improves upon its predecessor with features like Wi-Fi connectivity, firmware updates, photo display, and weather updates. A key improvement is addressing the issue of uneven toasting, particularly at the bottom of the bread, by concentrating more coils at the bottom of each slot. While the new model toasts more evenly, questions remain about the necessity of a touchscreen and internet connectivity in a toaster, especially when a simple dial could suffice. Additionally, the toaster’s performance in achieving the desired darkness level was inconsistent.

Gizmodo

19 hours ago

Ars Technica

19 hours ago

An small microbial ecosystem has formed on the International Space Station

The largest study yet of the ISS’s microbes hints we’re may be keeping it too clean.

Wired

19 hours ago

These Plants Photosynthesize Deep in the Arctic Even When There’s No Light

Microalgae can grow and reproduce in the Arctic’s coldest and darkest places with minimal light. A recent study found that these organisms thrive at light levels close to the theoretical minimum needed for photosynthesis, much lower than previously thought possible in nature. This challenges the assumption that life in the polar night shuts down, revealing that even with the barest quantum of light, life can bloom. Researchers spent months on a ship in an ice floe to study the limits of photosynthesis in such extreme conditions.

HackerNoon

1 days ago

The TechBeat: Tether May Become America’s New Strategic Reserve Buyer (3/2/2025)

HackerNoon’s Techbeat summarizes trending tech stories. Key topics include Bybit’s security issues, the launch of O.XYZ’s Cerebras-powered AI engine OCEAN, and the role of human ‘vibe coding’ in a post-AI world. Also featured are articles on seamless backend transitions for businesses, humorous AI CAPTCHA fails, Ethereum Swarm’s decentralized web foundation, fleet management app development, a deep dive into the Bybit hack, and the potential for the poorest people to run Bitcoin nodes. Finally, there’s a philosophical challenge to the notion that AI lacks true understanding.

Wired

20 hours ago

Lenovo’s Latest Laptop Concept Has a Screen That Can Flip and Fold

At Mobile World Congress 2025 in Barcelona, Lenovo presented several new laptop concepts, including a solar-powered laptop, a laptop with a screen that flips and folds, and an attachment that turns one laptop screen into three. One highlight is the ThinkBook Flip, featuring a flexible touchscreen OLED display that folds over the lid, providing a small screen when closed and a large 18.1-inch portrait screen when fully opened. This design aims to increase productivity by offering more vertical space, allowing users to view two full-size browser windows on top of each other. Although the concept is still in early pre-production, Lenovo assures that future versions will be sturdier.

Gizmodo

20 hours ago

Ars Technica

20 hours ago

Driving an EV restomod that costs as much as a house—the JIA Chieftain

There are no reused Tesla parts here, but the price tag is prototype-level.

Ars Technica

21 hours ago

Driving the new Mercedes CLA made me a believer in Mercedes-Benz’s EV future

And if it doesn’t, they’ve got a hybrid version, too.

Mashable

21 hours ago

Learn to practice ethical hacking for just $35

The All-in-One Super-Sized Ethical Hacking Bundle is on sale for $34.97 until March 30. This bundle includes 18 courses with over 100 hours of instruction, covering everything from the basics of ethical hacking to advanced penetration testing techniques. The courses provide hands-on experience with practical exercises and real-world scenarios, making it a valuable resource for those interested in pursuing certifications like Certified Ethical Hacker (CEH) or CompTIA Security+. With lifetime access, users can revisit the material whenever needed.

Mashable

21 hours ago

Retro gamers: Get 72,000 games on this console for $90

The Kinhank SuperConsole X2 is on sale for $89.97, offering access to over 72,000 classic games. Supporting over 60 emulators, it covers everything from NES and SNES to PlayStation and Sega Genesis. It features dual-band Wi-Fi, Bluetooth 5.0, and 4K UHD output, and runs on a dual-system setup with Android 9.0 for streaming apps. The offer is valid until March 30.

Mashable

21 hours ago

Get a three-pack of AirTag alternatives for your wallet (and everywhere else) for just $80

A 3-pack of KeySmart SmartCard wallet tracker cards is on sale for $79.97 with free shipping for a limited time. These credit card-sized trackers work with Apple’s Find My app, allowing users to see their wallet’s real-time location on a map, play a sound to locate it, and get notified when they leave it behind. Unlike AirTags, the SmartCard can be recharged with a Qi-enabled wireless charger. They can be used for wallets, luggage, and passports.