The rambutan (Nephelium lappaceum) is a medium-sized tropical tree in the family Sapindaceae, native to Southeast Asia. The fruit is a round to oval single-seeded drupe, 3 to 6 centimetres (1+1⁄4 to 2+1⁄4 inches) long and 3 to 4 centimetres (1+1⁄4 to 1+1⁄2 inches) wide. The leathery skin is reddish (rarely orange or yellow) and covered with fleshy pliable spines, hence the name rambutan, which is derived from a Malay word meaning 'hair'. The spines (also known as "spinterns") contribute to the transpiration of the fruit, which can affect the fruit's quality. The flesh, known as the aril, is translucent, whitish, or very pale pink, with a sweet, mildly acidic flavor reminiscent of grapes. The single seed is glossy brown, about 1.0 to 1.3 centimetres (3⁄8 to 1⁄2 inch) long, with a white basal scar. This photograph shows two rambutans, one whole and one half-peeled to expose the aril, as well as a rambutan seed. The photograph was focus-stacked from 31 separate images. Photograph credit: Ivar Leidus Recently featured: Galaxea fascicularis Roosilawaty Black Sunday Archive More featured pictures

April 17: Evacuation Day in Syria (1946) Artist's impression of Kepler-186f 1080 – Canute IV became King of Denmark upon the death of his brother Harald III. 1809 – Napoleonic Wars: After a three-day chase, the French ship D'Hautpoul was captured off Puerto Rico by a British squadron under Alexander Cochrane. 1951 – The Peak District was designated the first national park in the United Kingdom. 1975 – The Khmer Rouge captured Phnom Penh, the capital of the Khmer Republic, ending the Cambodian Civil War and establishing the socialist state of Democratic Kampuchea. 2014 – NASA announced the discovery of Kepler-186f (pictured), the first exoplanet with a radius similar to Earth's discovered in the habitable zone of another star. Marino Faliero (d. 1355)Hannah Webster Foster (d. 1840)Karen Blixen (b. 1885)Lucrețiu Pătrășcanu (d. 1954) More anniversaries: April 16 April 17 April 18 Archive By email List of days of the year About

Rare Watercolor by ‘Wuthering Heights’ Author Emily Brontë Will Go on Public Display for the First Time “The North Wind,” painted while Emily and her sister Charlotte were studying in Belgium, is now heading to the Brontë family home in Yorkshire The North Wind, Emily Brontë, 1842 Forum Auctions A rare painting by Emily Brontë, the British author best known for her 1847 novel Wuthering Heights, has sold at auction for $42,000. The Brontë Parsonage Museum, located in the Brontë family home in Haworth, Yorkshire, recently placed the winning bid on the watercolor painting known as The North Wind (1842). Following restoration work, it will go on public view at the site where the Brontë sisters—Emily, Charlotte and Anne—created some of 19th-century England’s greatest literary works. While Wuthering Heights’ influence on literature, film and even music is evident, Emily’s works of visual art are exceedingly elusive. “Emily is probably the most enigmatic of the Brontës,” Ann Dinsdale, the principal curator at the museum, tells Artnet’s Jo Lawson-Tancred. “She died at the age of 30, and very few manuscripts or letters by her have survived. It’s extremely rare to see anything associated with Emily coming on the market, making this painting of great importance.” The North Wind is a fittingly enigmatic work. It depicts a windswept woman with dark hair and a light blue cloak facing away from a breeze. Brontë painted The North Wind while staying with her sister Charlotte at the Pensionnat Heger, a boarding school for girls in Belgium, according to a statement from the Brontë Parsonage Museum. Emily Brontë's painting was based on an engraving of Lady Charlotte Harley that accompanied a collection of Lord Byron's works. Public domain via Wikimedia Commons The work is based on an engraving of Lady Charlotte Mary Harley that accompanied an edition of Finden’s Illustrations of the Life and Works of Lord Byron. As Dinsdale sees it, Emily drawing on a Byronic image is fitting for her development as an artist and writer. “Most writers on the Brontës would agree that Byron was probably the greatest literary influence on Emily’s work,” she says to Artnet. As far as the painting’s title, Emily appears to have been inspired by her younger sister Anne, who wrote a poem by the same name in 1838 in which “a captive girl welcomes the cold because it reminds her of her native northern mountains,” Edith Weir wrote in a 1949 volume of the journal Brontë Society Transactions, per the auction catalog. In letters, Charlotte wrote that Emily was taking drawing lessons and left some of her work in Brussels after the girls returned to England, according to the museum’s statement. For more than 180 years after the watercolor was completed, The North Wind passed between private hands. As a result, Emily’s visual artwork remains underappreciated, particularly when compared to that of her siblings. Her brother, Branwell, painted the only surviving portrait of his three sisters, which now hangs in the National Portrait Gallery in London, and Charlotte illustrated the second edition of her 1847 novel Jane Eyre. Branwell Brontë painted his three sisters—Anne, Emily and Charlotte—around 1834. Public domain via Wikimedia Commons “Unlike Charlotte and Branwell, Emily left few drafts, few exercises, to tell the story of her apprenticeship in painting,” Christine Alexander and Jane Sellars wrote in their 1995 book The Art of the Brontës, per the auction catalog. “Her drawings and rough sketches are as fragmentary, and as elusive of interpretation, as her surviving poetry.” Because Emily’s artworks are so rare, the auction house used “equivalent drawings by her sister Charlotte” to decide on an estimate, Rupert Powell, Forum Auction’s deputy chairman, tells Artnet. Experts expected the painting, which is lightly tarnished with spotting and some abrasions, to sell for around $26,000. “It was very tense when lot 53, Emily Brontë’s painting, came up, as the likelihood was that it would disappear into a private collection,” Dinsdale says to Artnet. During the auction, the price rose far beyond the estimate. But the Brontë Parsonage Museum was determined to finally bring The North Wind to Haworth—and to public display—for the first time. “The bidding seemed to go up very fast. Then there was a very tense pause before the gavel came down, and I knew that the painting would be coming to the Brontës’ former home in Haworth,” Dinsdale adds. “It was a very emotional moment for staff at the museum.” Get the latest stories in your inbox every weekday.

Konvoy's latest report shows the first major money moves in the games industry in 2025, including the tensions between the U.S. and China.Read More

Gemini Live’s feature that lets it see and respond to what’s on your camera and your screen will now be free for all Android users via the Gemini app, Google announced today. The AI-powered feature officially launched earlier this month for everyone on Pixel 9 and Samsung Galaxy S25 using the Gemini app. At the time, Google said the feature would launch “soon” for all Android users though would only be available with a Gemini Advanced subscription. But the company has changed its mind and is now making it available for free. “We’ve been hearing great feedback on Gemini Live with camera and screen share, so we decided to bring it to more people,” Google said on X. The feature will roll out to all Android users with the Gemini app starting today, and the rollout will take place “over the coming weeks.” If you want to get an idea of how the feature works, check out this video from Google. In it, a person holds their phone with the camera open at an aquarium so that Gemini can see the animals and share information. Today, Microsoft announced that its similar AI tool, called Copilot Vision, is available now for free in the Edge browser.

Biophysical modeling serves as a valuable tool for understanding brain function by linking neural dynamics at the cellular level with large-scale brain activity. These models are governed by biologically interpretable parameters, many of which can be directly measured through experiments. However, some parameters remain unknown and must be tuned to align simulations with empirical data, such as resting-state fMRI. Traditional optimization approaches—including exhaustive search, gradient descent, evolutionary algorithms, and Bayesian optimization—require repeated numerical integration of complex differential equations, making them computationally intensive and difficult to scale for models involving numerous parameters or brain regions. As a result, many studies simplify the problem by tuning only a few parameters or assuming uniform properties across regions, which limits biological realism. More recent efforts aim to enhance biological plausibility by accounting for spatial heterogeneity in cortical properties, using advanced optimization techniques like Bayesian or evolutionary strategies. These methods improve the match between simulated and real brain activity and can generate interpretable metrics such as the excitation/inhibition ratio, validated through pharmacological and PET imaging. Despite these advancements, a significant bottleneck remains: the high computational cost of integrating differential equations during optimization. Deep neural networks (DNNs) have been proposed in other scientific fields to approximate this process by learning the relationship between model parameters and resulting outputs, significantly speeding up computation. However, applying DNNs to brain models is more challenging due to the stochastic nature of the equations and the vast number of integration steps required, which makes current DNN-based methods insufficient without substantial adaptation. Researchers from institutions including the National University of Singapore, the University of Pennsylvania, and Universitat Pompeu Fabra have introduced DELSSOME (Deep Learning for Surrogate Statistics Optimization in Mean Field Modeling). This framework replaces costly numerical integration with a deep learning model that predicts whether specific parameters yield biologically realistic brain dynamics. Applied to the feedback inhibition control (FIC) model, DELSSOME offers a 2000× speedup and maintains accuracy. Integrated with evolutionary optimization, it generalizes across datasets, such as HCP and PNC, without additional tuning, achieving a 50× speedup. This approach enables large-scale, biologically grounded modeling in population-level neuroscience studies. The study utilized neuroimaging data from the HCP and PNC datasets, processing resting-state fMRI and diffusion MRI scans to compute functional connectivity (FC), functional connectivity dynamics (FCD), and structural connectivity (SC) matrices. A deep learning model, DELSSOME, was developed with two components: a within-range classifier to predict if firing rates fall within a biological range, and a cost predictor to estimate discrepancies between simulated and empirical FC/FCD data. Training used CMA-ES optimization, generating over 900,000 data points across training, validation, and test sets. Separate MLPs embedded inputs like FIC parameters, SC, and empirical FC/FCD to support accurate prediction. The FIC model simulates the activity of excitatory and inhibitory neurons in cortical regions using a system of differential equations. The model was optimized using the CMA-ES algorithm to make it more accurate, which evaluates numerous parameter sets through computationally expensive numerical integration. To reduce this cost, the researchers introduced DELSSOME, a deep learning-based surrogate that predicts whether model parameters will yield biologically plausible firing rates and realistic FCD. DELSSOME achieved a 2000× speed-up in evaluation and a 50× speed-up in optimization, while maintaining comparable accuracy to the original method. In conclusion, the study introduces DELSSOME, a deep learning framework that significantly accelerates the estimation of parameters in biophysical brain models, achieving a 2000× speedup over traditional Euler integration and a 50× boost when combined with CMA-ES optimization. DELSSOME comprises two neural networks that predict firing rate validity and FC+FCD cost using shared embeddings of model parameters and empirical data. The framework generalizes across datasets without additional tuning and maintains model accuracy. Although retraining is required for different models or parameters, DELSSOME’s core approach—predicting surrogate statistics rather than time series—offers a scalable solution for population-level brain modeling. Here is the Paper. Also, don’t forget to follow us on Twitter and join our Telegram Channel and LinkedIn Group. Don’t Forget to join our 90k+ ML SubReddit. Sana HassanSana Hassan, a consulting intern at Marktechpost and dual-degree student at IIT Madras, is passionate about applying technology and AI to address real-world challenges. With a keen interest in solving practical problems, he brings a fresh perspective to the intersection of AI and real-life solutions.Sana Hassanhttps://www.marktechpost.com/author/sana-hassan/SyncSDE: A Probabilistic Framework for Task-Adaptive Diffusion Synchronization in Collaborative GenerationSana Hassanhttps://www.marktechpost.com/author/sana-hassan/Transformers Can Now Predict Spreadsheet Cells without Fine-Tuning: Researchers Introduce TabPFN Trained on 100 Million Synthetic DatasetsSana Hassanhttps://www.marktechpost.com/author/sana-hassan/A Coding Guide to Build a Finance Analytics Tool for Extracting Yahoo Finance Data, Computing Financial Analysis, and Creating Custom PDF ReportsSana Hassanhttps://www.marktechpost.com/author/sana-hassan/Traditional RAG Frameworks Fall Short: Megagon Labs Introduces ‘Insight-RAG’, a Novel AI Method Enhancing Retrieval-Augmented Generation through Intermediate Insight Extraction

Author(s): Luis Ramirez Originally published on Towards AI. For industrial equipment the default approach is preventive maintenance, which involves servicing equipment on a fixed schedule, such as monthly or semi-annually. While better than reactive maintenance (fixing equipment after failure), this one-size-fits-all strategy has significant drawbacks. Equipment units experience different operational conditions, loads, and degradation rates, yet receive identical maintenance schedules. This leads to two suboptimal outcomes: either overspending on unnecessary maintenance (including operational downtime costs) or not maintaining equipment frequently enough, risking costly failures. A better solution is an intelligent maintenance strategy tailored to each piece of equipment based on its actual condition and predicted degradation; this is precisely where predictive maintenance comes into play. Starting simple Predictive maintenance often requires complex machine learning models that can be difficult to implement and interpret. In this post, we take a different approach, focusing on a visual analysis built on top of PCA projections. By focusing on visualization first, we can: Build intuitive understanding of degradation patterns. Establish a baseline for more advanced models. Create a common language for maintenance teams. Identify different failure patterns visually. Achieve quick wins while developing more sophisticated solutions. Source: Image by the author. The Dataset For this analysis, we’ll use the NASA turbofan engine dataset, which contains run-to-failure data of 100 engines. Each engine starts in good condition and develops a fault over time until failure. The dataset includes: Multiple sensor readings (temperatures, pressures, rotation speeds Operating condition indicators The cycle number Besides the initial features, I have added some aggregated data of mean tendency metrics, minimums, maximums and standard deviations that were generated using a length-5 rolling window. This preprocessing helps capture temporal patterns that might indicate degradation. Source: Image by the author. Step 1: Basic Health Visualization Understanding PCA Projections Our first goal is to create an intuitive visualization of equipment health status. We’ll use Principal Component Analysis (PCA) to reduce the many sensor dimensions into a two-dimensional plot that can be easily interpreted. PCA works by: 1. Finding the directions (components) of maximum variance in the data. 2. Projecting the data onto these components.3. Ordering components by how much variance they explain.Source: Image by the author. For this analysis we will only use the first two components, the result is a two-dimensional plot where similar operating conditions cluster together, besides the two main components we will use a gradient to represent the Remaining Useful Life (RUL). We can already see some patterns emerging in this visualization. As the data points move from left to right along PC1, they show increasing engine degradation. However, the linear scale makes it difficult to distinguish between different stages of low RUL values, which are the most critical for maintenance decisions. Source: Image by the author. Let’s apply a log scale to the RUL gradient to highlight these important end-of-life patterns. Now we’re seeing clearer separation between health states. In this enhanced visualization, we can distinguish: Healthy equipment clustered in the center left Moderately degraded equipment in the middle Severely degraded equipment spread in the right side Source: Image by the author. How To Use This Visualization? This basic visualization already provides valuable insights for maintenance teams: 1. Current Health Assessment: By plotting new data points from an operating machine, we can immediately see where it falls in the health spectrum 2. Early Warning System: As a machine’s position shifts toward the degraded regions, we receive an early warning before failure occurs 3. Fleet Comparison: Multiple machines can be plotted simultaneously, allowing us to prioritize maintenance efforts on the most degraded equipment We could even draw rough decision boundaries to separate the sections: Source: Image by the author. While this visual separation into low, medium, and high-risk zones roughly represents RUL > 100, 100 > RUL > 40, and RUL < 40, respectively. Now we have a better grasp of what this PCA represents, but to make decisions based on this, we need to polish our approach. Step 2: Defining In-Sample Regions with Gaussian Mixture Models Identify out of sample regions The previous classification has an important limitation: it extends the classification space infinitely in all directions. This means a point with extreme values, e.g., (-20, 25) or (-20, -15), would still be labeled as “low risk”, or a point like (5, -15) would be labeled as “medium risk”, simply because it falls on that side of the decision boundary. In reality, such extreme values likely represent sensor malfunctions or completely novel operating conditions outside our training data; they can be closer to a failure than to normal operation. Source: Image by the author. To address this limitation, one powerful enhancement is implementing Gaussian Mixture Models (GMMs) to define out-of-sample regions in the PCA space. GMMs model the actual distribution of “normal” data, allowing us to identify when new observations fall outside the regions of known behavior: Source: Image by the author. By using a GMM with a single component (which results in an ellipse), we establish a basic boundary. Although this offers an initial approximation, we can notice that the elliptical shape is not the best match for the distribution of data we have. Source: Image by the author. To improve the quality of the region definition, we can use a GMM with multiple components. This allows us to model the complex, non-elliptical distribution of machine states. This approach allows us to identify unusual degradation patterns that fall outside normal behavior. Source: Image by the author. Step 3: Feature Selection for Better Results Using Mutual Information The PCA algorithm is a deterministic process that maximizes the explained variance kept from the original data in the projected components, but not all sensor variance correlates with equipment health. For example, some sensors might fluctuate due to ambient conditions rather than degradation. To address this, we’ll use mutual information to identify which features most strongly correlate with Remaining Useful Life: Source: Image by the author. Source: Image by the author. This approach allows us to focus on the most relevant indicators of degradation, simplifying the analysis while keeping the number of variables manageable. Improved Visualizations Now, if we repeat the PCA plot using only the selected features, we can see that the shape of the projection has changed: Source: Image by the author. However, the core behavior remains, with healthy machines (datapoints) on the left and degraded ones on the right. Let’s apply the log scale again to better visualize the end-of-life period. Now we can see a clearer picture, there is a better distinction between the different RUL values and we have a delimiting curve/shape for the out of sample data. Source: Image by the author. Step 4: Risk Classification SVM in action After applying feature selection and creating an improved PCA projection, we can implement the classification of the risk regions, from our previous hand draw borders to a more robust method. For that we can use Support Vector Machines with a linear kernel, which will find the better line to separate the region. As the SVM are a supervised method, we need to define some labels to depict the level of risk. For the purposes of this analysis and based on the results we have got with the PCA, we will choose as thresholds: Source: Image by the author. With the definition we can now train our SVM classifier Source: Image by the author. This classification allows us to divide the PCA space into clear regions representing different risk levels. The color of the region represents the SVM classification while the coloring of the dots represents the actual category, based on the labels we have defined. Source: Image by the author. Why use linear SVM and not Neural Networks or boosting trees? I chose to keep the analysis simple yet well-founded, and linear SVMs offer a good trade-off by providing robust yet easily interpretable results. They essentially draw the optimal straight lines that separate the classes — similar to my hand-drawn approach earlier, but with mathematical precision. Now, let’s change the plot to focus on analyzing classification errors. This approach can further refine our understanding. The plot shows that 11.6% of the data points have been classified incorrectly. Among these errors, the most common misclassification occurs between low and medium risk, with errors between medium and high risk being the second most frequent. Fortunately, in these training results, there are no errors between low and high risk categories, which would represent the worst case scenario. Source: Image by the author. Actionable Maintenance Decision Points This classification approach provides clear, actionable information: 1. Low Risk: — Action: Continue normal operation — Monitoring: Routine checks during scheduled maintenance — Frequency: According to standard maintenance schedule2. Medium Risk: — Action: Plan intervention during next scheduled maintenance — Monitoring: Increase monitoring frequency — Preparation: Order potential replacement parts — Documentation: Begin documentation for upcoming maintenance3. High Risk: — Action: Schedule immediate maintenance intervention — Monitoring: Continuous or daily monitoring — Preparation: Ensure all replacement parts are available — Planning: Coordinate with production to minimize impact4. Outside In-Sample Region (Dashed Line): — Action: Investigate unusual behavior — Indication: Potential novel failure mode or sensor issue — Response: Engineering analysis requiredStep 5: Trajectory Analysis Individual Equipment Tracking Over Time Perhaps the most insightful visualization is the degradation trajectory plot, showing how individual machines move through the PCA space over time: This visualization reveals: Different paths to failure, suggesting distinct failure modes. Non-linear degradation patterns. Acceleration of the degradation after the crossing the medium range area. Source: Image by the author. Velocity Field Visualization Building on trajectory analysis, we can create a velocity field visualization showing the average direction and speed of degradation throughout the PCA space: The arrows in this visualization show both the direction and rate of degradation in different regions: – Brighter arrows indicate areas where degradation occurs more rapidly– The overall flow pattern reveals typical degradation paths– Areas with divergent arrows may indicate different failure modesSource: Image by the author. Predicting Future Degradation Paths For maintenance teams, trajectory and velocity field analysis provides great insights: 1. Degradation Rate Estimation: By tracking a machine’s movement through the PCA space over time, teams can estimate how quickly it’s degrading 2. Time-to-Failure Prediction: The velocity field helps predict how long before a machine enters the high-risk region 3. Maintenance Planning: Understanding typical degradation paths allows teams to plan interventions at optimal points 4. Failure Mode Identification: Different trajectory patterns often correspond to different failure modes, helping teams prepare the right replacement parts and procedures For example, if a turbofan engine follows a trajectory similar to Equipment 3 in the visualization, maintenance teams know to focus on specific components that typically fail in that pattern. If it follows a different path, they’d prepare for a different type of intervention. Conclusion In this blog, we have gone through the core steps to create a robust PCA analysis for Predictive Maintenance that can be replicated for other datasets to: Create intuitive representations of equipment health Identify different degradation patterns visually Establish a common language for discussing equipment state Build a foundation for more sophisticated modeling It’s important to keep in mind that PCA is a starting point, not a substitute for more complex models. A natural next step from here is to dive deeper into the dataset, understand feature behavior and interpretation, and build more complex models either for classification or as a forecast having RUL as the target. The second point to consider is that the data used here is exceptionally clean, with 100 run-to-failure assets. In a real-case scenario, if we are starting from scratch, the first big challenge will be to build a useful dataset. More often than not, data from industrial applications has missing values, mislabeled events, noisy sensor data, and few actual failures identified to train on. Understanding the nuances of that specific use case and industry will be crucial for success. Here are some options to continue exploring Aker BP’s Valhall oil platform and Awesome Industrial Datasets. Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor. Published via Towards AI

Pulsar made a great impression with the Xboard QS, which featured expansive physical customization and a build quality strong enough to be classified as a weapon. Along with some of my favorite mechanical switches, Pulsar proved to be a real player in the boutique (and expensive) keyboard space. The PCMK 2 HE, however, takes on a different design philosophy with a lighter, slimmer build and makes the move to magnetic switches to stand among some of the best gaming keyboards recently released. It costs a pretty penny at a base price of $160, especially considering its wired-only connectivity limits versatility, but everything else around it is fantastic and shows that Pulsar isn’t just a one-hit wonder.Pulsar PCMK 2 HE - PhotosPulsar PCMK 2 HE – Design and FeaturesThe PCMK 2 HE rocks a simple design, and it’s a clean and attractive aesthetic with the black and white color scheme that contrasts really well. The QWER, ASDF, Escape, and Enter keys come in the opposite color of the rest of the keycaps (depending on which primary color you choose to buy), and the backlit Pulsar logo above the arrow keys shines brightly. That logo is also a magnetic tab that you can remove, and even get customized for a little extra. The legend on it indicates whether or not Caps Lock, Scroll Lock, or Game Mode is active via the backlighting as well.The exposed keycaps on the aluminum top plate lets them pop and the RGB backlighting flow nicely across the board. These double-shot PBT keycaps don’t have transparent lettering so the lighting doesn’t help with visibility, but there’s a decent brightness that helps the customizable colors still jump out. The dampeners and foam layer make up the rest of the keyboard, however, the keystrokes aren’t exactly soft or quiet. I’m fine with it since the keys have a light and bouncy feel that works well for gaming and long-term typing, although it does sound more “clacky” than most other magnetic keyboards. The bottom is encased in a transparent plastic so it’s not a fully aluminum chassis and doesn’t feel quite as sturdy as it could have been, but the see-through look is pretty sweet.Like most of Pulsar’s catalogue, it’s quite pricey, but there’s no doubt that you get a quality product in return.“Pulsar collaborated with Gateron for its magnetic switches, which have a really smooth linear feel – at just 30g +/- 7 on the initial actuation force, it’s one of the lightest switches I’ve used. In moving to magnetic Hall Effect switches, it affords the keyboard another layer customization that you just can’t get from mechanical keyboards. Popular manufacturers like Logitech and Razer are jumping onto it now, and Pulsar is up there with the big names not just with how well the switches themselves perform, but because of how easy it is to get the most out of them.Pulsar PCMK 2 HE – Software and CustomizationAs with other boutique-style keyboard makers, Pulsar uses a web-based configurator instead of a downloadable software suite to customize the keyboard’s settings. It’s called Bibimbap and makes tweaking things a breeze, similar to what Keychron offers as I saw with the K4 HE that I also reviewed. You just go to the Pulsar software URL (it’ll say “download” but it doesn’t actually download a client) and select the keyboard from the Connect menu. From there, you can mess with things like the customizable actuation point, RGB lighting profile, key assignments, macros, and much more.Pulsar PCMK 2 HE - SoftwareAlmost every feature included is configurable on a per-key basis, and while RGB isn’t going to necessarily shine through the keycaps, the degree to which you tweak the color spectrum, effects, and indicator lights on the magnetic tab is extensive. Having adjustable actuation points, thanks to the magnetic switches, comes in clutch as well, and that’s simply done in the Performance tab in the software – it can be as short as 0.1mm or as deep as 4.0mm and anywhere in between in 0.1mm increments.Because this is a magnetic keyboard with features like Rapid Trigger and Quick Tap, I do need to address the ever-present elephant in the room. Quick Tap, being the name for SOCD (simultaneous opposite cardinal direction) input in this case, is a contentious feature that allows immediate input recognition even with the opposing direction still being held. In a shooter, holding A to strafe left while tapping D to strafe right without letting go of A creates a jiggle-strafing movement that is physically impossible otherwise. And this technique makes you an extremely tough target to hit. While every magnetic keyboard has some form of SOCD now, it doesn’t mean you’re safe to use it – you will get kicked from Counter-Strike 2 matches, for example, so be mindful of that. Rapid Trigger isn’t as problematic, it just recognizes any upward movement as a reset point, making repeated tapping faster.Looking to upgrade your mouse, too?Best OverallRazer Deathadder V3 HyperspeedRazer made notable tweaks to its flagship mouse, making the V3 one of the best mice and further proving why Deathadders have been a staple for PC gaming for so long.See it at AmazonCheck out our roundup of the best gaming mice out now!Despite not being as physically sophisticated as the Xboard QS, there is still some onboard customization you can do with the PCMK 2 HE, like swapping out the switches. If you want to change things up and ditch the Gateron x Pulsar magnetic switches, you can do that easily so long as you’re putting in N-Pole or S-Pole switches.Pulsar PCMK 2 HE – PerformanceTo cut to the chase, gaming performance on the PCMK 2 HE is stellar. That’s largely due to the super-light touch to the Gateron x Pulsar magnetic switches and the adjustable actuation points. Even without using Quick Tap, my skill ceiling in Counter-Strike 2 is a bit higher on account of being able to do quicker keystrokes and have those inputs recognized switfly. While I love the K4 HE that I mentioned earlier, I give the slight edge for gaming to something like the PCMK 2 HE since its switches make these kinds of techniques easier to execute for competitive scenarios. Whether I'm quick-strafing to peak corners, switching weapons on a dime, or crouch-jumping through a window, the PCMK 2 HE let me pull these moves off effortlessly.While I like having a super short actuation point for a competitive shooter, setting my keys to a deeper point works wonders for making sure I don't accidently set off the wrong action in my attack rotation for Final Fantasy XIV. As someone who spent hours-on-end raiding, I definitely felt my fingers wearing out the longer the raid went on, too, and the switches on the PCMK 2 HE helps mitigate that kind of exhaustion by virtue of the light actuation force. It's those sorts of capabilities that may seem minor on paper, but make noticeable differences in practice.While much of these perks are due to the magnetic switches, the keyboard is a pleasure to use even if it's louder than most of its contemporaries. That bounciness I mentioned earlier feels great when I'm typing all throughout the work day, making for a comfortable experience outside of gaming. It’s worth noting that the PCMK 2 HE features a 8000Hz polling rate, which I’ve talked about extensively as it pertains to high-end gaming mice. With regards to keyboards, the benefits are much more limited, and I would go as far to say negligible. Where the continuous and miniscule movements of a mouse swipe come out smoother with a high polling rate, the binary nature of keyboard inputs just don't really need that. If anything, it's to give you piece of mind that you're getting the best performance that's technically possible (but let's be real, putting 8,000Hz polling rate on a box is also a marketing move).Purchasing GuideThe Pulsar PCMK 2 HE is available for $159.95 on Amazon or directly from the Pulsar store page. It comes in two color schemes – black-white keycaps with a black underplate or white-black keycaps with a white underplate.

Apple Intelligence was announced with iOS 18 and has been available since last October, when Apple released iOS 18.1 to the public. Although most apps provide support for Apple Intelligence features by default, developers can choose not to have them in their apps – and it seems that Meta has decided to do so. As reported by the Brazilian blog Sorcererhat Tech, features such as Writing Tools – which lets users create, change, and proofread text with Apple Intelligence – are no longer available in any of Meta’s apps, including Facebook, WhatsApp, and Threads. Typically, iPhone and iPad users can access Writing Tools by tapping on a text field. However, in Meta’s iOS apps, the option is not available. Meta apps also don’t let users create and share Genmoji, Apple’s custom emoji generated by AI. The report notes that Meta has also removed the ability to add keyboard stickers and Memoji to Instagram Stores – something that was possible before. Unsurprisingly, Meta doesn’t give any details about why Apple Intelligence isn’t available in its iOS apps. But if we had to guess, the company probably wants to motivate people to use Meta AI instead of alternatives like Apple Intelligence. Meta AI is available in pretty much all of Meta’s apps and also lets users create and change text, as well as generate images. A WSJ report last year revealed that Apple and Meta had discussed a potential partnership that would bring Llama, Meta’s AI language model, to Apple Intelligence. However, Apple reportedly decided to scrap the deal because it didn’t agree with the company’s privacy policies. Apple and Meta also frequently fight each other over App Store guidelines. Unfortunately for iOS users, this means that they can’t take advantage of Apple Intelligence in some of the most popular apps in the world. Hopefully the company will change its approach in the future. Gadgets I recommend: Add 9to5Mac to your Google News feed.  FTC: We use income earning auto affiliate links. More.You’re reading 9to5Mac — experts who break news about Apple and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Mac on Twitter, Facebook, and LinkedIn to stay in the loop. Don’t know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

Elon Musk has gambled Tesla's future on the Cybercab — a small, autonomous vehicle intended to spearhead the automaker's charge into the robotaxi business and give it a pathway for growth as other electric vehicle makers increasingly eat its lunch.It was an already questionable move. And now, in a grim irony for Musk, it's looking even dicier thanks to the Trump administration he helped usher into the White House with hundreds of millions of dollars in highly visible support.As , Trump's steep tariffs on Chinese goods have forced Tesla to suspend its plans to import parts for the Cybercab and its electric tractor unit, the Semi, from China. While it's unclear how long the suspension will last — or for that matter Trump's tariffs, which have see-sawed back and forth — it could be a major disruption, given that the development of both vehicles already face significant hurdles.Per the reporting, Tesla was scheduled to start receiving parts shipments over the upcoming months to start trial production of the Cybercab and Semi in October, with the goal of starting mass production in 2026.Ironically, Tesla had been preparing for the tariffs for the past two years by sourcing more of its parts from North America, according to a Reuters source. It was even ready to eat the extra costs of the tariffs on China when they were still at 34 percent. But that's a rate firmly in the rear view mirror now. In the "Liberation Day" fallout, Trump raised the tariffs to 84 percent — then to 125 percent, then a staggering 145 percent, and now to a degree so high that the precise figure has become a bit hazy.At its unveiling last October, Musk told investors that the Cybercab would bring trillions of dollars to the company. It would cost less than $30,000 and enter production in 2026. Both promises are being severely threatened by the tariffs.The situation is more dire for the Semi, the development of which has faced repeated setbacks and is years behind schedule. After announcing a delay, the automaker said it would start deliveries in 2021. Since then, it has only produced a small number of the trucks for an even smaller group of customers.The halt on the parts imports is the latest blowback that Tesla is facing from Trump's aggressive economic policy towards China, which is the automaker's second largest market and the largest market for EVs in the world. Last week, Tesla suspended orders for two of its vehicle models on its Chinese website as China raised retaliatory tariffs on US goods to 84 percent. Amid those fears, Tesla stock continued to dive.There are signs that the punishing tariffs are forming a rift between Trump and Musk. On his website X, Musk raged against the president's senior trade advisor Peter Navarro, who devised the tariff policy, calling him a "moron" and worse. Musk also made comments vowing his support for "free trade," and personally tried to change Trump's mind behind the scenes. So far, there're no clear signs of Trump relenting.More on Tesla: Tesla Stock Hits Dreaded Death CrossShare This Article