Blender 4.3 has just been released! The biggest updates are related to Grease Pencil, Sculpting, and Geometry Nodes, but, as usual, there are some improvements to all areas of Blender. In this article, well cover everything that you need to know before upgrading. Lets get into it!Modeling and UVsThe bevel modifier can now use any edge attribute as a bevel weight, which means that you can now have multiple modifiers that use different custom weights. For complex, hard surface objects, or for sharpening up subdivision surfaces, this will help you keep more of your modeling non-destructive. It also means you can control bevels via geometry nodes even though theres not yet a bevel node.Did you know that mesh material slots could also be assigned to the object rather than the mesh data? It makes it so that instances can have different materials. Well, now in 4.3, if you have a setup like that, the Copy Material to Selected command keeps that object relationship like you would expect.When it comes to UV unwrapping, there are some cases where Blenders Angle Based and Conformal operations can leave you with a little too much area stretching. The new Minimum Stretch algorithm, based on the SLIM algorithm, aims to improve that in some situations. Its great for tricky surfaces that need to be straightened or when you dont want to place too many seams.Sculpting and Texture PaintingWorking with sculpting, texturing, and other brushes in Blender has just gotten a bit more convenient now that brushes have been converted into assets, which can be shared between files. That means your favorite custom brushes will now always be there, and you dont have to keep importing them to every project. And now, instead of a large number of brushes in the toolbar, you can find the main types of tools in the toolbar and all of the brush presets for those tools in the Asset Shelf. You can search for and jump to brushes easily, even without the shelf, by clicking on the brush icon in the Tool Settings or using the Shift Space hotkey.All of the settings for the built-in brushes have been tweaked based on user-feedback, so you should find the out-of-the-box sculpting experience much more polished. Its faster too, thanks to many parts of sculpt mode having been rewritten. Entering sculpt mode can be around 5x faster, and brush evaluation itself is about 8 times faster.Also in Sculpt mode, the Voxel Size operator now works in relative mode by default, the lasso tools now have optional stroke stabilization, a polyline gesture can now be finished with double-click, sculpt mask is now accessible in node tools, and a new Mask from Boundary operator was added to modify mask values based on mesh or face set islands.Geometry NodesAnother new zone, called the For Each Element zone, has been added to Geometry Nodes. It allows you to perform operations on each vertex, edge, face, or other type of element of a geometry. This was already possible with the Repeat Zone, but the For Each Element zone simplifies the process and calculates the result significantly faster. Geometry Nodes can work with Grease Pencil objects now! From the perspective of Geometry Nodes, Grease Pencil geometry is made up of a list of layers and each layer is composed of curves. Layers, curves, and curve control points can all have custom attributes. When converting Grease Pencil to curves, each layer is turned into a separate curves instance. Similarly, instances are turned into layers when converting curves to Grease Pencil.Bakes from Geometry Nodes can now be packed into Blender files, which should make sharing files with lots of smaller baked data much easier. Bakes are packed by default, but its suggested to save the bake externally for very large bakes for better performance. Node group inputs in Geometry Nodes can now be associated with gizmos in the 3D viewport, so you can interact with your procedural setups in a much more natural way. The tricky thing about implementing gizmos is that it creates a two-way relationship between the value and the gizmo. Manipulating the gizmo affects the value, and manipulating the value affects the gizmo. So, gizmo connections are represented by two-way links. Instead of plugging the gizmo into the value you want to control (which was my first guess), you connect the group input to both a gizmo and the value. Geometry node groups can also have custom warnings, so that creators can notify users if some combination of inputs might not behave as expected.Geometries can also now be given names using the Set Geometry Name node, for easier organization and navigation in the Spreadsheet editor.A few new utility nodes have been added, such as the Hash Value, Integer Math, and Matrix Determinant nodes.Lastly, the Value to String node no longer has floating point issues when working with integers, node timings are more accurate, the Skip checkbox on the simulation zone is now hidden, and node tools can now work with masks in sculpt mode. Animation and RiggingOn the animation side of things, the properties editor now has an Animation panel for any data block that supports animation. In it, you can select which action the mesh, object, world, or whatever it is that youre animating, is using. That way, you dont have to open up the Action Editor to switch actions.The default motion path theme can now be set in Preferences. The Autokey indicator is now colored red when enabled, for extra clarity. You definitely dont want to forget to turn that off when youre finished! When inserting new keyframes, all other keys are now deselected. That should lead to fewer accidental transformations when working quickly in the dope sheet and graph editor. Bone selectors in the Properties editor now have an eyedropper, so you can easily select them from the 3D View or the Outliner. Grease PencilBlenders Grease Pencil has been completely rewritten for Blender 4.3. The main goal of this v3 was to improve performance and better integrate Grease Pencil objects into the rest of Blender. The first part of this is the integration with Geometry Nodes that Harry already talked about. Combining the powers of proceduralism with the hand-crafted beauty of illustration should lead to some amazing work from the community in the coming months. This new system also comes with some new features. You can now group layers together for better organization and easily managing visibility. The layers dropdown in Draw mode has been promoted to the top bar. Just like with sculpting, the Grease Pencil brushes use the new brush asset system. The Eraser tool now allows you to cleanly cut through strokes anywhere because it can create new points on the parts that it touches. Before, it could only get rid of existing vertices. You can now set the size of the brush in pixels according to the view or in proper scene units. The Simplify option is now measured in screen space pixels rather than in world space units. The Active Smoothing to reduce input jitter has been rewritten and is supposed to feel more accurate and less floaty. The Trim tool can now do multiple trims on the same stroke at the same time. In Edit Mode, theres a new Fill Gradient tool that can be used to adjust fill gradients in the active material. Grease Pencil operations are now multithreaded whenever possible, which should lead to a big boost in performance for heavy scenes. Layer adjustments are now evaluated before modifiers rather than after them. Do be aware that there is forward compatibility for Grease Pencil v3 but no backwards compatibility, so if you save a file in 4.3 you will not be able to use the Grease Pencil objects again in 4.2 or earlier. But anything saved in Blender 4.2 or earlier will be imported just fine into 4.3. Also know that a few features have not made the jump to this new system, including screen space stroke thickness (though you can now recreate that effect with Geometry Nodes), selection order for the interpolation tool in Edit mode no longer works the same, drawing guides have been removed for now but are expected to make a comeback later, and editing the strokes as bezier curves now works only by setting the stroke type to Bezier. RenderingTheres a new shader called the Metallic BSDF, which can be used to render metals according to their real world IOR and extinction coefficient values. I dont expect most artists will use this, because you can get the same aesthetic results just by coloring the principled shader using the calculated PBR base color. But for those who want extreme accuracy, its there if you need it. The Gabor Noise texture is Blenders first new procedural pattern in a while. Its a bit like the Noise and Wave texture had a baby, but with the added ability to somewhat flow along a surface when its orientation input is set to the meshs normal or tangent direction. You can now easily adjust the color temperature and color tint of your renders using the new White Balance panel. Its eyedropper icon makes it really easy to set the balance based on the background or any other image. All real photos have some white balancing going on in the camera, but theres no perfect white, so shifting it slightly might help you get a little closer to photorealism. A new Diffuse Roughness setting in the Principled BSDF mimics the Diffuse BSDF Roughness input which, while not used frequently, blends between the Lambert and the more physically accurate Oren-Nayer reflectance types. In addition, higher diffuse roughness values are now energy preserving. The options are there in Eevee but only make a difference in Cycles. Cycles volumes now have a new Phase Function option, which allows for more accurately replicating common volumes for environments like clouds, atmosphere, underwater, and space dust. The panoramic camera in Cycles now supports central cylindrical projection, which can be used for projecting renders onto giant curved screens. Hardware accelerated ray-tracing is now available for Cycles on Linux, which adds a nice little speed boost if your GPU can support it. Eevee now supports light linking, and it should work exactly the same as in Cycles. Nice! Evees Raytracing options have also been reorganized as they were a bit confusing before. Theres a new Fast GI Approximation panel, and the Max Roughness is what controls where that gets applied instead of the noisier but more accurate Screen Tracing method. CompositingBasic multi-pass compositing in the viewport now works in Eevee, though not yet in Cycles. A white point conversion mode was added to the Color Balance node, so you can convert seamlessly from one temperature or tint to another. A new Save As Render option was added to the File Output node to specify whether the output should have color management applied or not. The GPU compositor now supports Cryptomatte meta-data, which is needed for saving cryptomattes as EXRs. Lastly, the Auto Render option was removed from the Compositor because it was really buggy and now, we can see everything in real time in the viewport anyway. Video EditingThe strip handle grabber mouse icons are now custom and platform specific and look way better on Windows. Strips can be easily linked and unlinked now for faster editing using the Connect Strips command. You can also hold Alt to temporarily ignore linking. The preview area now supports snapping and the toolbar is shown by default. Audio strips can now be adjusted on a sub-frame level with the Slip (hotkey S) operator. The Saturation, Multiply, and Tonemap effects are multithreaded now and several times faster, as is color transformations, rebuilding proxies, loading thumbnails, and drawing several parts of the UI. User InterfaceAs for the UI changes, all icons in the UI are now rendered as SVGs, which makes them extra crisp at any UI scale. Speaking of, the maximum resolution scale has been increased from 3x for interactive elements and from 6x for properties. At this size, you might even notice that the Blender logo in the top left looks a little different.The active editor is now indicated by edge highlighting rather than a tinted header. Splitting editors is a little different now. You can still right click between any editor just like before, but now dragging from any corner initiates a docking operation. It may take some time to get used to, but what it enables is really cool! You can now just rip the outliner from the top right, for example, and place it on the left of the 3D View, or put it back just as easily. The trick to joining adjacent areas is to just drag it ever so slightly and not go far enough to replace the other editor. You can also hold Ctrl while dragging to swap editors or hold Shift while dragging to pop the editor out a new window. These operations are now hinted at in the Status Bar. The status bar also now has better spacing and alignment and can also fit more hotkeys on screen. The hit box for resizing windows is slightly bigger, and theres more visual feedback while doing so. Data block selector lists are now a bit wider, tooltips when hovering over a data block now show a preview for images, movies, and fonts, there are styling options for tooltip text, and periods are no longer appended to the end of all tooltips. Color picking from the viewport while in Render mode now returns the linear color value without the view transform applied. Also, I dont think anyone ever thought it was slow, but color picking performance has been greatly improved. And, the color picker UI is now more compact.This may seem like a small thing, but the forwards and backwards mouse buttons now go forwards and backwards in the file browser rather than an undo or redo step. I cannot tell you how many times I had accidentally done that. Node groups can now have a default node width, and hovering over panels on nodes now shows their description as a tooltip. Inserting nodes with drag to search is much nicer now, because you can now connect the new node by plopping it on the noodle. As with regular auto-insertion, you can avoid this by holding Alt. Save Incremental was moved down slightly to reduce accidental clicks. Windows with a single editor now show descriptive titles. On Mac, Mickey Mouse apparently (the little white glove cursor) now grabs dialogs and areas just like in the rest of the OS. Also on Mac, copy and paste now works for images in the Image Editor. A bit of padding was added to UI lists. A bit of padding was also added to the bottom of the VSE and NLA editors to make room for the Adjust Last Operation panel. Also the cursors for video strip handle manipulation are much nicer now. The control points in custom curve properties are now larger. Editors with horizontal scrollbars now hide them when theres not enough room. Files not found are no longer removed from the Recent Files list automatically but can now be removed with the Clear function. Dialogs now show a move cursor at the top to indicate that they can be dragged. Renaming datablocks has been slightly tweaked, and you can now force Blender to use the name you picked and it will rename the other one rather than the one youre working on. Import and ExportThe Universal Scene Description file format in Blender can now export point clouds, material purposes can be assigned on import, handling of asset paths has been improved, and exporting animated attributes is now more efficient. glTF received a huge number of fixes, as well as some enhancements, which include associating UDIM material names with tile numbers, using matrices as custom attributes, and exporting joint leafs at the tail of leaf bones.Update today!You can grab the 4.3 update from blender.org and read through the official release notes if you're curious about more of the technical details. Don't forget to support the development fund while you're at it, to help make future updates possible!

Imagine sitting down with an AI model for a spoken two-hour interview. A friendly voice guides you through a conversation that ranges from your childhood, your formative memories, and your career to your thoughts on immigration policy. Not long after, a virtual replica of you is able to embody your values and preferences with stunning accuracy. Thats now possible, according to a new paper from a team including researchers from Stanford and Google DeepMind, which has been published on arXiv and has not yet been peer-reviewed. Led by Joon Sung Park, a Stanford PhD student in computer science, the team recruited 1,000 people who varied by age, gender, race, region, education, and political ideology. They were paid up to $100 for their participation. From interviews with them, the team created agent replicas of those individuals. As a test of how well the agents mimicked their human counterparts, participants did a series of personality tests, social surveys, and logic games, twice each, two weeks apart; then the agents completed the same exercises. The results were 85% similar. If you can have a bunch of small yous running around and actually making the decisions that you would have madethat, I think, is ultimately the future, Joon says. In the paper the replicas are called simulation agents, and the impetus for creating them is to make it easier for researchers in social sciences and other fields to conduct studies that would be expensive, impractical, or unethical to do with real human subjects. If you can create AI models that behave like real people, the thinking goes, you can use them to test everything from how well interventions on social media combat misinformation to what behaviors cause traffic jams. Such simulation agents are slightly different from the agents that are dominating the work of leading AI companies today. Called tool-based agents, those are models built to do things for you, not converse with you. For example, they might enter data, retrieve information you have stored somewhere, orsomedaybook travel for you and schedule appointments. Salesforce announced its own tool-based agents in September, followed by Anthropic in October, and OpenAI is planning to release some in January, according to Bloomberg. The two types of agents are different but share common ground. Research on simulation agents, like the ones in this paper, is likely to lead to stronger AI agents overall, says John Horton, an associate professor of information technologies at the MIT Sloan School of Management, who founded a company to conduct research using AI-simulated participants. This paper is showing how you can do a kind of hybrid: use real humans to generate personas which can then be used programmatically/in-simulation in ways you could not with real humans, he told MIT Technology Review in an email. The research comes with caveats, not the least of which is the danger that it points to. Just as image generation technology has made it easy to create harmful deepfakes of people without their consent, any agent generation technology raises questions about the ease with which people can build tools to personify others online, saying or authorizing things they didnt intend to say. The evaluation methods the team used to test how well the AI agents replicated their corresponding humans were also fairly basic. These included the General Social Surveywhich collects information on ones demographics, happiness, behaviors, and moreand assessments of the Big Five personality traits: openness to experience, conscientiousness, extroversion, agreeableness, and neuroticism. Such tests are commonly used in social science research but dont pretend to capture all the unique details that make us ourselves. The AI agents were also worse at replicating the humans in behavioral tests like the dictator game, which is meant to illuminate how participants consider values such as fairness. To build an AI agent that replicates people well, the researchers needed ways to distill our uniqueness into language AI models can understand. They chose qualitative interviews to do just that, Joon says. He says he was convinced that interviews are the most efficient way to learn about someone after he appeared on countless podcasts following a 2023 paper that he wrote on generative agents, which sparked a huge amount of interest in the field. I would go on maybe a two-hour podcast podcast interview, and after the interview, I felt like, wow, people know a lot about me now, he says. Two hours can be very powerful. These interviews can also reveal idiosyncrasies that are less likely to show up on a survey. Imagine somebody just had cancer but was finally cured last year. Thats very unique information about you that says a lot about how you might behave and think about things, he says. It would be difficult to craft survey questions that elicit these sorts of memories and responses. Interviews arent the only option, though. Companies that offer to make digital twins of users, like Tavus, can have their AI models ingest customer emails or other data. It tends to take a pretty large data set to replicate someones personality that way, Tavus CEO Hassaan Raza told me, but this new paper suggests a more efficient route. What was really cool here is that they show you might not need that much information, Raza says, adding that his company will experiment with the approach. How about you just talk to an AI interviewer for 30 minutes today, 30 minutes tomorrow? And then we use that to construct this digital twin of you.

This is today's edition ofThe Download,our weekday newsletter that provides a daily dose of what's going on in the world of technology. Inside Clears ambitions to manage your identity beyond the airport Clear Secure is the most visible biometric identity company in the United States. Best known for its line-jumping service in airports, its also popping up at sports arenas and stadiums all over the country. You can also use its identity verification platform to rent tools at Home Depot, put your profile in front of recruiters on LinkedIn, and, as of this month, verify your identity as a rider on Uber. And soon enough, if Clear has its way, it may also be in your favorite retailer, bank, and even doctors officeor anywhere else that you currently have to pull out a wallet (or wait in line). While the company has been building toward this sweeping vision for years, it now seems its time has finally come. But as biometrics go mainstream, whatand whobears the cost? Read the full story. Eileen Guo LinkedIn Live: Facial verification tech promises a frictionless future. But at what cost? Do you use your face to unlock your phone, or speed through airport security? As biometrics companies move into more and more spaces, where else would you use this technology? The trade off seems simple: you scan your face, you get a frictionless future. But is it really? Join MIT Technology Reviews features and investigations team for a LinkedIn Live this Thursday, November 21, about the rise of facial verification tech and what it means to give up your face. Register for free. Whos to blame for climate change? Its surprisingly complicated. Once again, global greenhouse-gas emissions are projected to hit a new high in 2024. In this time of shifting political landscapes and ongoing international negotiations, many are quick to blame one country or another for an outsize role in causing climate change. But assigning responsibility is complicated. These three visualizations help explain why. Casey Crownhart Cyber Week Sale: subscriptions are half price! Take advantage of epic savings on award-winning reporting, razor-sharp analysis, and expert insights on your favorite technology topics. Subscribe today to save 50% on an annual subscription, plus receive a free digital copy of our Generative AI and the future of work report. Dont miss out. The must-reads Ive combed the internet to find you todays most fun/important/scary/fascinating stories about technology. 1 AI can now translate your voice in real-time during meetings Its part of Microsofts drive to push more AI into its products, but how well it works in the wild remains to be seen. (WP$)+Apple is having less success on that front, at least if its AI notification summaries are anything to go by.(The Atlantic$)2 Anyone can buy data tracking US soldiers in Germany And the Pentagon is powerless to stop it.(Wired$)+Its shockingly easy to buy sensitive data about US military personnel.(MIT Technology Review)3 Bluesky now has over 20 million usersIts user base has tripled in the last three months. (Engadget)+Truth Social, on the other hand, is not doing quite so well.(WP$)+The rise of Bluesky, and the splintering of social.(MIT Technology Review)4 How Google created a culture of concealmentIts been preparing for antitrust action for over a decade, enforcing a policy where employees delete messages by default. (NYT$)+The company reacted angrily to reports it may be forced to sell Chrome. (BBC)5 Project 2025 is already infiltrating the Trump administrationDespite repeated denials, its clearly a blueprint for his next term. (Vox)+A hacker reportedly gained access to damaging testimonies about Matt Gaetz, his pick to be attorney general.(NYT$)6 Quantum computers hit a major milestone for error-free calculationThis is a crucial part of making them useful for real-world tasks. (New Scientist$)7 Technology is changing political speechSlogans are becoming less effective. Now its more about saying different things to different audiences. (New Yorker$)8 Lab-grown foie gras, anyone?This could be the cultivated meat industrys future: as a luxury product for the few. (Wired$)9 Niantic is using Pokmon Go player data to build an AI navigation system If it works, it could unlock some amazing capabilities in AR, robotics and beyond. (404 Media) 10 Minecraft is expanding into the real world It has struck a $110 million deal with one of the worlds biggest theme park operators. (The Guardian)Quote of the day Nobody believes that these cables were severed by accident. Germanys defense minister Boris Pistorius, tells reporters that the severing of two fiber-optic cables in the Baltic Sea was a deliberate act of sabotage, theNew York Timesreports.The big story Are we alone in the universe? ARIEL DAVIS November 2023 The quest to determine if life is out there has gained greater scientific footing over the past 50 years. Back then, astronomers had yet to spot a single planet outside our solar system. Now we know the galaxy is teeming with a diversity of worlds. Were getting closer than ever before to learning how common living worlds like ours actually are. New tools, including artificial intelligence, could help scientists look past their preconceived notions of what constitutes life.Read the full story. Adam Mann We can still have nice things A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or tweet 'em at me.) + How to not only survive but thriveduring the winter. + Fancy working from somewhere new? Here are some of thebest citiesfor a workcation.+ Want to see David Bowie imitating Mick Jagger?Of course you do.+ Its an old(ish) joke butstill funny.

How will we live in Aotearoa in 20, 30, 50 years? Will the quarter of an acre still be the dream of most aspiring homeowners? What will our communities look like and how will climate change affect our choice of housing? Will private ownership still be the most common option? How can collective housing play a role in shaping the future of our housing choices in the future? These are some of the big questions that CoHoHui4, the Collective Housing Conference organised byTHIS is looking at addressing in its 2025 edition, in tautahiChristchurch.Collective housing describes scenarios where residents decide to live in intentional communities, in private, self-contained homes but with some resources that are shared and with the goal of fostering collaboration and mutual support. Examples of collective housing are cohousing, papakinga, community land trusts, ecovillages, housing cooperatives, community housing, co-living, build-to-rent models andmore.CoHoHui this year will be different as, for the first time since its inception in 2019, the event will be offered as an academic conference, with an international call for papers and a special issue Urbanization, Sustainability and Society (USS) Journal,The Future of Housing, published by Emerald Publishing.The event will run across three days, with a networking dinner on the evening of the 15th at Visions, the caf and restaurant on Ara campus. The academic sessions, panels, exhibitions and workshops will be held on the 16th in the Kahukura Building, and site visits on the 17th ofApril.The conference will provide a great opportunity for collective and alternative housing advocates to connect, be inspired and reflect on what key changes are needed for housing to become more affordable, sustainable and truly inclusive in the future. Several themed sessions will run in parallel for academics to present their research, while expert panels and workshops will engage and educate the audience on the latest developments in the collective housingsector.Dr.Thomas Moore, international keynote speaker at next years CoHoHui.Image: University of LiverpoolThomas Moore, one of the international keynote speakers, will be joining the conference from the University of Liverpool, where he works as a Senior Lecturer in the Department of Geography and Planning. He has researched community-led housing models in England since 2007, exploring their growth and development through local case studies, project and funding evaluations, and international comparisons. Tom is interested in the potential for community-led housing models to challenge transactional, market-based logics that characterise many housing systems, as well as the opportunities and limitations of models that rely on citizen leadership and participation. In addition to community-led housing, Tom undertakes research a range of issues related to community planning and housing policy andinequality.Louise Crabtree-Hayesis a Professor at the Institute of Society and Culture at Western Sydney University and is Australias leading expert on housing cooperatives and community land trusts. Louises research focuses on the social, ecological and economic sustainability of community-driven housing developments in Australia; on the uptake of housing innovation in practice and policy; on complex adaptive systems theory in urban contexts; and, on the interfaces between sustainability, property rights, institutional design anddemocracy.Greer ODonnell, co-founder and director ofThe Urban Advisory will give an update on the findings of its New Zealand Housing Survey, a fundamental tool for councils, government organisations, researchers and developers to help build awareness around how people live, want to live and what barriers are in the way to achieving their housingaspirations.On the last day, CoHoHui will take their attendees on a journey to visit several collective housing developments around tautahi: the recently completed Te Pakau Maru stage one for example, aKinga Maha housing development in New Brighton, offering a mixed-tenures approach while featuring Homestar rated, architecturally designed homes that emphasize sustainability, community connection, andresilience.As we look ahead to CoHoHui 2025, the questions this event seeks to answer could not be more urgent: How can we reimagine housing in Aotearoa to create communities that are affordable, sustainable, resilient and inclusive? How can collective housing offer new pathways to address the housing crisis and reshape the way we live, work, and connect with oneanother?Cohohui4 conference is not just a gathering, but a call to action. Its an opportunity to learn from global thought leaders, engage with innovative housing models, and contribute to meaningful change in our housing systems. Whether youre an academic, a policymaker, a practitioner, or simply someone passionate about better housing for all, CoHoHui 2025 invites you to be part of this vitalconversation.Dont miss this chance to connect, collaborate, and be inspired.Register today to join us in tautahi Christchurch, where the future of housing will be envisioned, debated, and set into motion. Together, lets build a better tomorrow, one collective housing community at atime.The biannual CoHoHui conference is organised by The Housing Innovation Society (THIS) in collaboration with the Architecture and Engineering Department, Ara Institute of Canterbury.

Submitted by WA ContentsSnhetta wins competition for cable car stations in Koblenz, Germany Germany Architecture News - Nov 20, 2024 - 14:43 html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"Norwegian architecture firm Snhetta has won a new competition for the new city cable car stations in Koblenz, Germany.The Upper Middle Rhine Valley's designation as a UNESCO World Heritage Site requires extra caution, according to the firm. The expressive stations respectfully integrate into the historical setting as modern statements.Since 2011, Ehrenbreitstein Fortress on the opposite side of the Rhine has been connected to Koblenz's city center via a cable car. The popular, eco-friendly mode of transportation was initially intended to be a temporary structure for the Federal Horticultural Show, but it is now intended to be used permanently.A redesign of the stations was required due to technical requirements and the location's sensitive historical context. Snhetta's proposal for the sculptures for the valley and mountain stations, which form a "cohesive pair" and respond sensitively to the respective context, won the architectural competition held by the cable car operator.The Basilica of St. Castor, which is listed, is close to the valley station. The gently curved shape reinterprets the city's architectural heritage and motifs while paying homage to the historic basilica.According to the jury, "the design exhibits restraint within the World Heritage Site without losing any of its architectural significance." "The St. Castor Church is referenced in the design by the graceful roof shape."A wooden framework supported by V-shaped concrete columns holds up the metal roof shell. While the gently sculpted concrete columns are reminiscent of the church's vaulted stonework and traditional masonry along the river, the metal shingles on the roof are inspired by the city's traditional faade and roof motifs.The reflections and play of light and shadow along the river and the tree-lined riverbank are further enhanced by the subtle green hues and the glossy metal cladding.This subtle effect is further enhanced by the station's curved shape. The station's open area blends in with the Rhine promenade's natural surroundings.The mountain station's technical facilities are also covered by a common, gently chamfered metal roof. Strong, stoic shapes establish a connection to the Ehrenbreitstein stronghold.The metal cladding, whose warmer hues allude to the earthy tones of the fortress walls, supports the wooden structure on slender columns. In front of the station, a "flying" roof incorporates an existing tree, provides a covered waiting area, and spans the ticket office, technical rooms, and kiosk.The historic landscape in front of the fortress is clearly horizontal, as highlighted by the station and the cantilevered canopy. Similar to the bottom station, the top station will keep all of the cable car's technical installations while adding amenities for long-term use."The cultural landscape and identity of the Upper Middle Rhine Valley are characterised by centuries of human intervention," Patrick Lth, Snhetta Partner and Managing Director of the Innsbruck studio."The new stations are part of this natural and cultural context and with our design we continue the dialogue with the significant cultural landscape of Koblenz," Lth added.Snhetta, recently, completed a new expansion and site redesign for the Joslyn Art Museum in Omaha, United States. Moreover, the firm completed the Blanton Museum of Art at the University of Texas, Austin. In addition, the firm also completed Beijing City Library in China, with a giant canopy supported by ginkgo trees and wraped by a fully glazing faade. All images Snhetta.> via Snhetta

Submitted by WA Contentsgmp completes Nanchang East Station with the rhythm of the arcsChina Architecture News - Nov 20, 2024 - 14:16 html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"The team of von Gerkan, Marg and Partners Architects (gmp) has completed a new station with the rhythm of the arcs in Nanchang, China.Named Nanchang East Station, the 100,000-square-meter station has been built in Nanchang, a well-known city that is situated along the middle Yangtze and the southern shore of Poyang Lake.The project was created by Shanghai United Design Group (UDG), China Railway Siyuan Survey and Design Group (T4), and von Gerkan, Marg and Partners Architects (gmp).According to gmp, the station is the centerpiece of a brand-new neighborhood in eastern Nanchang that was created using the concepts of transit-oriented development, or TOD. This neighborhood will have residential units, shops, spaces for creative industries and cultural events, and office space for start-ups. The station houses eight platforms and sixteen tracks.With rail links to Beijing in the north, Hong Kong in the south, Kunming in the west, and Shanghai in the east, it is a significant hub within China's quickly growing high-speed rail network.Additionally, it serves as the beginning of a new route that travels through Jingdezhen to Huangshan, a popular tourist destination.A distinctive feature of the stations design is the three-aisled arc structure, designed by schlaich bergermann partner (sbp) and gmp.The main entrance on the western side is highlighted by a laterally shortened arc. In total, the roof structure consists of a steel construction of 14 rows of triple arc elements.The central, large arc spans 96 meters, creating a spacious, column-free area in the middle for the concourse. The construction time for the structure was significantly shortened by pre-fabricating each arc truss directly on the construction site and then positioning them with the help of a crane runway.The atrium functions as an "urban living room"The main entrance atrium on the west side forms a three-story continuous space that connects the various functional areas across different levels: the main halls, taxi stands, and the subway stations.The atrium functions as an urban living room, linking the station with Nanchangs city center and the TOD area, with eateries and retail outlets that are open to the public.In Chinese urban development, Nanchang East Station is regarded as a prototype transit-oriented development project. The station is a new transportation hub in the eastern part of Nanchang, which is home to about five million people.To facilitate a smooth and quick passenger flow into the city, the station's infrastructure was modified using the Ground Traffic Center (GTC) model, which is utilized in contemporary airports. The station is conveniently located for both public transportation and pedestrians.Employees at surrounding companies can commute to work from farther away thanks to its connection to the high-speed rail network, which enables rapid travel to other major cities.After Tianjin West Station, Hangzhou South Station, and Beijing's Fengtai Station, this project is gmp's fourth significant high-speed rail station in China. Ground floor planSectionRecently, the team of von Gerkan, Marg and Partners Architects (gmp) completed the conversion of a 1940s Santiago Bernabu Stadium, known as Estadio Santiago Bernabu, in Madrid. In addition, the firm completed the refurbishment of a 1969s multi-purpose hall, known as The Hyparschale in Magdeburg, Germany.Another significant gmp project is the Alsterschwimmhalle, often called Hamburgs Swimming Opera, which reopened after an extensive refurbishment.Project factsProject name: Nanchang East StationArchitects:gmpLocation:Nanchang, ChinaCompetition:2020 1st prizeDesign: Meinhard von Gerkan and Nikolaus Goetze with Jan BlaskoCompetition Project Lead: Nils KlinkmannCompetition Team:Tim Andrasko, Deng Zihe, Mariam Hamidou, Li Bo, Liu Jianfeng, Pan Mei, Alp YilmazDetailed Design Project Lead:Liu JianfengDetailed Design Team:Deng Zihe, Lesley-Anne Fischer, Mariam Hamidou, Nils Klinkmann, Luo Dian, Li Jiahui, Lin Qihong, Lorenz Riedinger, Monica Sallowsky, Johann von BothmerProject Management:China Pan MeiPartner Firm in China: China Railway Siyuan Survey and Design Group, UDG GroupStructural Engineering: schlaich bergermann partnerLandscape Architecture:WES LandschaftsArchitekturClient:Jiujingqu Railway JiangxiGFA: 100,000m2All images CreatAR.All drawings gmp.> via gmp

October has seen rising pessimism across workloads and recruitment compared to the previous monthArchitectural staff recruitment fell back in October while overall industry confidence took a hit, according to RIBAs monthly outlook survey.Source: RIBAThe most recent RIBA Future Trends staffing index shows that 81% of surveyed firms expect their number of permanent staff to plateau over the next three months, with 9% of practices expecting an increase and the same number anticipating job cuts.Although small practices returned a -2 staffing figure, medium and large companies expect growth with a +13 index number.In London, staffing outlook remains positive, despite falling from last Septembers +12 balance to +4 in October. The North of England mimics this pattern, dropping from +7 to +3.However, Wales and the West (-6) and the South of England (-3) have not fared so well, becoming negative last month.Despite the increasingly cautious approach to recruitment, views on near-term workload remain just about positive, having returned a +2 balance.The sector toes the line, as 23% of firms expect workloads to decrease versus 24% that anticipate growth. Optimism is strongest in the North of England, the only region anticipating private housing workloads to pick up.Views on private housing turned negative overall, with a ten-point index drop from September (+5) to October (-5).The commercial sector has garnered similar opinions as the workload figure has also experienced a ten-point fall from +8 to -2.Meanwhile, outlooks for the public and community sectors have furtherdeteriorated.The outlooks of small and larger companies on future work continue to differ sharply with smaller firms expressing pessimism whereas bigger practices are hopeful.Current workloads are also 10% lower than twelve months ago.Octobers Future Trends data was collected before the UKs budget announcement on 30 October and saw practices reporting projects being put on hold as the supply chain awaited the news as well as reluctant attitudes towards commissioning and tendering higher-risk building projects.

The design competition for a memorial to Queen Elizabeth II will invite multidisciplinary teams to reimagine a site within St Jamess ParkSource: Malcolm Reading Consultants/Emily Whitfield-WicksSt Jamess ParkThe Queen Elizabeth Memorial Committee (QEMC) has confirmed plans for an international design competition to create a masterplan for a memorial to the late Queen Elizabeth II. The project, located in St Jamess Park near The Mall, will commemorate her life of service and contribution to the United Kingdom, Realms, and Commonwealth.The two-stage competition, to be organised by Malcolm Reading Consultants, is set to open in the coming weeks. It will invite submissions from architects, artists, landscape architects, engineers, and other disciplines, with the aim of developing an integrated masterplan that honours the Queens legacy.Source: Original image by Cecil Beaton. Royal Collection Enterprises Limited 2024, Royal Collection TrustQueen Elizabeth II National Memorial Masterplan Design CompetitionAccording to the QEMC, chaired by the late Queens former private secretary, Lord Janvrin, the initiative has been shaped by public and stakeholder consultation conducted across the UKs four nations. The selected site includes land adjacent to the Mall at Marlborough Gate and the pathway leading to the lake at the Blue Bridge, within the Grade I listed park.The QEMC, established jointly by the UK Government and the Royal Household, comprises members including Baroness Valerie Amos, Dame Amelia Fawcett, and Sir William Shawcross. The committee is tasked with developing proposals for both a permanent memorial and a wider national legacy programme, with recommendations to be presented to the Prime Minister and King Charles III by 2026.The competition will adhere to the Public Contracts Regulations 2015. Prospective participants can register their interest and sign up for updates via the competition website, managed by Malcolm Reading Consultants.

Make Architects, Feilden Clegg Bradley Studios (FCBS) and Fletcher Priest Architects have all confirmed they made staff cuts in recent weeks It is the second time this year that Ken Shuttleworths practice Make has been through a round of redundancies, having laid off around 15 per cent of its staff in March.The practice said it had now made 20 employees redundant and shut down its Sydney office which it set up in 2016.The news comes in the wake of poor financial figures published last month. The company's accounts for the year ending 31 December 2023 revealed a loss of 1.46 million and a drop in turnover from to 19.2 million in 2022 to 18.3 million. The practice, whose key workload is in the office sector, said the downturn had been caused by the economy taking longer to recover than expected and planning decisions [on key projects being] held up.AdvertisementSpeaking about the latest cuts, Shuttleworth, said: Its been a tough time for everyone in the industry.[We] have had to say goodbye to 20 friends and close the Sydney studio, which has been very painful.Looking ahead, the practice founder said: The budget impact on the National Insurance (NI) has been a cruel blow to everyone in a people-based business. It means that costs have increased, which will have an impact on the ability to pay more or employ future staff, when the market changes.However, we believe that we are now very well placed to respond to the challenges and opportunities ahead.The practice will be hoping that one of the schemes to soon move forward is its redevelopment of the former ITV Studios on Londons South Bank.AdvertisementNext week a ruling is expected on a High Court action brought by Save Our South Bank Action Group (SOS), a coalition of local opponents of the 25-storey project who successfully applied for a judicial review.Meanwhile, AJ100 practice Fletcher Priest has confirmed it recently made 10 voluntary redundancies. The company was ranked 28th in this years league table of the UKs largest architectural firms and had a 130-strong workforce at the end of last year.The practices most recent accounts show a 10.5 million turnover. Earlier this year Fletcher Priest unveiled concept plans for a 40-storey tower at 63 St Mary Axe, in the City of London, next door to Foggo Associates Can of Ham, which completed in 2019.Bath-based practice FCBS said it too was reducing the number of employees on its books. The company, which recorded a 19.6 million turnover in its most recent accounts, had 164 staff at the end of last year.FCBS said: Like many practices, we have recently made some redundancies, including voluntary redundancies, across our four offices. This decision was driven by a reduced workload, including a few major projects put on hold or cancelled due to funding.Earlier this month, it emerged the government was looking to scrap funding for FCBSs Liverpool museums plans, along with cash pots promised to other major Northern projects.In its October Budget, it said it was minded to cancel the as-yet unfunded Levelling Up Culture and Capital Projects which were announced in the previous governments final Budget earlier this year.The move throws into question a number of major projects planned in the North, including FCBSs proposed 58 million transformation of Liverpool Slavery and Maritime Museums, a new northern branch of the British Library in Leeds and schemes for the National Railway Museum in York.The AJ understands Hopkins Architects has also carried out a round of redundancy consultations, though the outcome of that process could not be confirmed.In April, Hopkins reported a 30 per cent hike in turnover on the back of increasing international workloads but revealed that income from UK jobs had dropped.According to Hopkins Architects group accounts for the year ended 31 March 2023, the practices fee income rose from 23.4 million to 30.6 million while its headcount remained steady at 180. This increase was down to a significant hike in work outside the UK and Europe, where its turnover more than doubled from 7.9 million to 16.9 million.However, fees from projects in this country fell from 15.3 million to 13.7 million during the reporting period.Last months RIBAs Future Trends survey the institutes monthly bellwether of the professions workload confidence shows that optimism has dropped among the practices regularly surveyed.RIBA's head of economic research and analysis Adrian Malleson said: On balance, practices remain positive about Future Workload, but only just.The recent gains in overall outlook and sectoral confidence have fallen back somewhat this month. Practices are increasingly cautious about recruitment. While the north of England remains positive, the south of England (excluding the capital) has seen marked falls in confidence and recruiting intention.After increasing confidence across the monitored sectors, this month sees all four soften, with a negative outlook.Practices described projects being put on hold ahead of the new governments first budgetHe added: Commentary received from practices described projects being put on hold as practices, clients and contractors awaited the new governments first budget. Practices also reported higher-risk building (HRB) projects being put on hold or progressing slowly, with clients being reluctant to commission HRB projects, and contractors being reluctant to tender for them.While awaiting the budget, practices described ongoing planning delays, a still weak economy, and elevated (though slowly falling) interest rates holding back the sector.2024-11-20Richard Waitecomment and share

With just one week untilThanksgiving, it's time to seriously start planning the menu. If you're dreaming of the perfect turkey and sides,artificial intelligencemight just be the best sous-chef out there.While Thanksgiving might be the most magical day of the year, it's also one of the most stressful. Thanksgivingmemesare hilarious, but there ain't nothing funny about slaving away in a hot kitchen from sunup to sundown only to have your aunt that you see twice a year question the quality of yourmac and cheese. "Who made this?" can be a very loaded sentence on Thanksgiving.If you have the means, it can be a time to revel in the bounty of the season's most succulent offerings. But it's also when relatives dip too deep in the sauce and start arguments about pronouns at the dinner table.One of the ways AI can help is by taking the guesswork out of planning an undeniably delicious Thanksgiving meal, no matter what the holidays throw at you. Here's how to use it.Choose an AI tool and get real about your skills Upgrade your inbox Get cnet insider From talking fridges to iPhones, our experts are here to help make the world a little less complicated. Since I'd like the benefit of choosing certainrecipesfrom around the web, I'm using Microsoft's in-browser, text-prompt AI tool,Copilot. Copilot uses LLM (large language model) processing as well as cited sources from the internet to respond to conversational search requests, and it can analyze web pages as you browse.But make sure you give it the info it needs. As in, if you can barely boil water, don't ask an AI tool to design a multicourse French-Chinese fusion banquet with sous vide steaks and flat noodles made from scratch.I let Copilot know I was an amateur in the kitchen, and it gave me some easy-to-prepare recipes with simple-to-source ingredients and minimal steps and prep. One of the options even sounded fancy -- green bean almondine -- which is literally just green beans with sliced almonds.Give AI a head countLet the robot chef know how many hungry people will be in attendance for your Thanksgiving meal. I used Microsoft's Copilot to convert this New York Times recipe for gravy to accommodate 20 people, and it seamlessly converted the measurements into a larger batch: Screenshot by Rachel Kane/CNETAlter recipes to your needsNot everyone wants raisins in their potato salad, and it's best to avoid foods that might upset people's stomachs or inflame allergies when you'll all be trapped at a dinner table in close proximity.For instance, Copilot was able to give me some critical ingredient substitutions on the fly for removing onions from the green bean casserole. Screenshot by Rachel Kane/CNETTake your budget into accountLet the AI tool know how much you've got to spend on Thanksgiving dinner. Copilot was able to provide me with a list of dishes that were low on cost and high in flavor, including waiting for sales or discounts on turkey, getting a smaller turkey, and using plain potatoes, boxed stuffing mix, gravy mix packets, premade pie crust and canned versions of beans, cream of mushroom soup, pumpkin and cranberry sauce.Copilot estimated my entire Thanksgiving meal to cost between $50 and $75, and included an itemized price breakdown of every ingredient I'd need.Be sure to double-check your local prices, though. I'm not convinced you can still find genuine whipped cream for $2 a container, as a quick search reveals the price to be closer to $4.50. Something called "whipped topping," however, was on sale at Kroger for $1.79.Let AI know what you're working withSometimes bad things happen to good people, and one of those bad things might include losing the use of your oven during the Thanksgiving holiday. Copilot was able to swoop in with some solid suggestions for cooking a whole turkey without the aid of an oven. Just please don't deep-fry a frozen turkey. Screenshot by Rachel Kane/CNET Screenshot by Rachel Kane/CNETI pressed Copilot further, letting the tool know I had only a large pot to cook with, and it gave me a detailed recipe for how to poach a whole turkey.Whatever you're working with on Thanksgiving, AI appears to be a decent kitchen companion.Just don't let your judgmental relatives know a robot helped you season the turkey.