Transcript [MUSIC] GRETCHEN HUIZINGA: Welcome to Abstracts, a Microsoft Research Podcast that puts the spotlight on world-class research in brief. I’m Gretchen Huizinga. In this series, members of the research community at Microsoft give us a quick snapshot – or a podcast abstract – of their new and noteworthy papers.  [MUSIC FADES] On today’s episode, I’m talking to Alex Lu, a senior researcher at Microsoft Research and co-author of a paper called Assessing the Limits of Zero Shot Foundation Models in Single-cell Biology. Alex Lu, wonderful to have you on the podcast. Welcome to Abstracts!  ALEX LU: Yeah, I’m really excited to be joining you today.  HUIZINGA: So let’s start with a little background of your work. In just a few sentences, tell us about your study and more importantly, why it matters.  LU: Absolutely. And before I dive in, I want to give a shout out to the MSR research intern who actually did this work. This was led by Kasia Kedzierska, who interned with us two summers ago in 2023, and she’s the lead author on the study. But basically, in this research, we study single-cell foundation models, which have really recently rocked the world of biology, because they basically claim to be able to use AI to unlock understanding about single-cell biology. Biologists for a myriad of applications, everything from understanding how single cells differentiate into different kinds of cells, to discovering new drugs for cancer, will conduct experiments where they measure how much of every gene is expressed inside of just one single cell. So these experiments give us a powerful view into the cell’s internal state. But measurements from these experiments are incredibly complex. There are about 20,000 different human genes. So you get this really long chain of numbers that measure how much there is of 20,000 different genes. So deriving meaning from this really long chain of numbers is really difficult. And single-cell foundation models claim to be capable of unraveling deeper insights than ever before. So that’s the claim that these works have made. And in our recent paper, we showed that these models may actually not live up to these claims. Basically, we showed that single-cell foundation models perform worse in settings that are fundamental to biological discovery than much simpler machine learning and statistical methods that were used in the field before single-cell foundation models emerged and are the go-to standard for unpacking meaning from these complicated experiments. So in a nutshell, we should care about these results because it has implications on the toolkits that biologists use to understand their experiments. Our work suggests that single-cell foundation models may not be appropriate for practical use just yet, at least in the discovery applications that we cover.  HUIZINGA: Well, let’s go a little deeper there. Generative pre-trained transformer models, GPTs, are relatively new on the research scene in terms of how they’re being used in novel applications, which is what you’re interested in, like single-cell biology. So I’m curious, just sort of as a foundation, what other research has already been done in this area, and how does this study illuminate or build on it?  LU: Absolutely. Okay, so we were the first to notice and document this issue in single-cell foundation models, specifically. And this is because that we have proposed evaluation methods that, while are common in other areas of AI, have yet to be commonly used to evaluate single-cell foundation models. We performed something called zero-shot evaluation on these models. Prior to our work, most works evaluated single-cell foundation models with fine tuning. And the way to understand this is because single-cell foundation models are trained in a way that tries to expose these models to millions of single-cells. But because you’re exposing them to a large amount of data, you can’t really rely upon this data being annotated or like labeled in any particular fashion then. So in order for them to actually do the specialized tasks that are useful for biologists, you typically have to add on a second training phase. We call this the fine-tuning phase, where you have a smaller number of single cells, but now they are actually labeled with the specialized tasks that you want the model to perform. So most people, they typically evaluate the performance of single-cell models after they fine-tune these models. However, what we noticed is that this evaluating these fine-tuned models has several problems. First, it might not actually align with how these models are actually going to be used by biologists then. A critical distinction in biology is that we’re not just trying to interact with an agent that has access to knowledge through its pre-training, we’re trying to extend these models to discover new biology beyond the sphere of influence then. And so in many cases, the point of using these models, the point of analysis, is to explore the data with the goal of potentially discovering something new about the single cell that the biologists worked with that they weren’t aware of before. So in these kinds of cases, it is really tough to fine-tune a model. There’s a bit of a chicken and egg problem going on. If you don’t know, for example, there’s a new kind of cell in the data, you can’t really instruct the model to help us identify these kinds of new cells. So in other words, fine-tuning these models for those tasks essentially becomes impossible then. So the second issue is that evaluations on fine-tuned models can sometimes mislead us in our ability to understand how these models are working. So for example, the claim behind single-cell foundation model papers is that these models learn a foundation of biological knowledge by being exposed to millions of single cells in its first training phase, right? But it’s possible when you fine-tune a model, it may just be that any performance increases that you see using the model is simply because that you’re using a massive model that is really sophisticated, really large. And even if there’s any exposure to any cells at all then, that model is going to do perfectly fine then. So going back to our paper, what’s really different about this paper is that we propose zero-shot evaluation for these models. What that means is that we do not fine-tune the model at all, and instead we keep the model frozen during the analysis step. So how we specialize it to be a downstream task instead is that we extract the model’s internal embedding of single-cell data, which is essentially a numerical vector that contains information that the model is extracting and organizing from input data. So it’s essentially how the model perceives single-cell data and how it’s organizing in its own internal state. So basically, this is the better way for us to test the claim that single-cell foundation models are learning foundational biological insights. Because if they actually are learning these insights, they should be present in the models embedding space even before we fine-tune the model.  HUIZINGA: Well, let’s talk about methodology on this particular study. You focused on assessing existing models in zero-shot learning for single-cell biology. How did you go about evaluating these models?  LU: Yes, so let’s dive deeper into how zero-shot evaluations are conducted, okay? So the premise here is that we’re relying upon the fact that if these models are fully learning foundational biological insights, if we take the model’s internal representation of cells, then cells that are biologically similar should be close in that internal representation, where cells that are biologically distinct should be further apart. And that is exactly what we tested in our study. We compared two popular single-cell foundation models and importantly, we compared these models against older and reliable tools that biologists have used for exploratory analyses. So these include simpler machine learning methods like scVI, statistical algorithms like Harmony, and even basic data pre-processing steps, just like filtering your data down to a more robust subset of genes, then. So basically, we tested embeddings from our two single-cell foundation models against this baseline in a variety of settings. And we tested the hypothesis that biologically similar cells should be similar across these distinct methods across these datasets.  HUIZINGA: Well, and as you as you did the testing, you obviously were aiming towards research findings, which is my favorite part of a research paper, so tell us what you did find and what you feel the most important takeaways of this paper are.  LU: Absolutely. So in a nutshell, we found that these two newly proposed single-cell foundation models substantially underperformed compared to older methods then. So to contextualize why that is such a surprising result, there is a lot of hype around these methods. So basically, I think that,yeah, it’s a very surprising result, given how hyped these models are and how people were already adopting them. But our results basically caution that these shouldn’t really be adopted for these use purposes.  HUIZINGA: Yeah, so this is serious real-world impact here in terms of if models are being adopted and adapted in these applications, how reliable are they, et cetera? So given that, who would you say benefits most from what you’ve discovered in this paper and why?  LU: Okay, so two ways, right? So I think this has at least immediate implications on the way that we do discovery in biology. And as I’ve discussed, these experiments are used for cases that have practical impact, drug discovery applications, investigations into basic biology, then. But let’s also talk about the impact for methodologists, people who are trying to improve these single-cell foundation models, right? I think at the base, they’re really excited proposals. Because if you look at what some of the prior and less sophisticated methods couldn’t do, they tended to be more bespoke. So the excitement of single-cell foundation models is that you have this general-purpose model that can be used for everything and while they’re not living up to that purpose just now, just currently, I think that it’s important that we continue to bank onto that vision, right? So if you look at our contributions in that area, where single-cell foundation models are a really new proposal, so it makes sense that we may not know how to fully evaluate them just yet then. So you can view our work as basically being a step towards more rigorous evaluation of these models. Now that we did this experiment, I think the methodologists know to use this as a signal on how to improve the models and if they’re going in the right direction. And in fact, you are seeing more and more papers adopt zero-shot evaluations since we put out our paper then. And so this essentially helps future computer scientists that are working on single-cell foundation models know how to train better models.  HUIZINGA: That said, Alex, finally, what are the outstanding challenges that you identified for zero-shot learning research in biology, and what foundation might this paper lay for future research agendas in the field?  LU: Yeah, absolutely. So now that we’ve shown single-cell foundation models don’t necessarily perform well, I think the natural question on everyone’s mind is how do we actually train single-cell foundation models that live up to that vision, that can perform in helping us discover new biology then? So I think in the short term, yeah, we’re actively investigating many hypotheses in this area. So for example, my colleagues, Lorin Crawford and Ava Amini, who were co-authors in the paper, recently put out a pre-print understanding how training data composition impacts model performance. And so one of the surprising findings that they had was that many of the training data sets that people used to train single-cell foundation models are highly redundant, to the point that you can even sample just a tiny fraction of the data and get basically the same performance then. But you can also look forward to many other explorations in this area as we continue to develop this research at the end of the day. But also zooming out into the bigger picture, I think one major takeaway from this paper is that developing AI methods for biology requires thought about the context of use, right? I mean, this is obvious for any AI method then, but I think people have gotten just too used to taking methods that work out there for natural vision or natural language maybe in the consumer domain and then extrapolating these methods to biology and expecting that they will work in the same way then, right? So for example, one reason why zero-shot evaluation was not routine practice for single-cell foundation models prior to our work, I mean, we were the first to fully establish that as a practice for the field, was because I think people who have been working in AI for biology have been looking to these more mainstream AI domains to shape their work then. And so with single-cell foundation models, many of these models are adopted from large language models with natural language processing, recycling the exact same architecture, the exact same code, basically just recycling practices in that field then. So when you look at like practices in like more mainstream domains, zero-shot evaluation is definitely explored in those domains, but it’s more of like a niche instead of being considered central to model understanding. So again, because biology is different from mainstream language processing, it’s a scientific discipline, zero-shot evaluation becomes much more important, and you have no choice but to use these models, zero-shot then. So in other words, I think that we need to be thinking carefully about what it is that makes training a model for biology different from training a model, for example, for consumer purposes.  [MUSIC] HUIZINGA: Alex Lu, thanks for joining us today, and to our listeners, thanks for tuning in. If you want to read this paper, you can find a link at aka.ms/Abstracts, or you can read it on the Genome Biology website. See you next time on Abstracts!  [MUSIC FADES] 

James Baldwin Media Library and Refugee House / associerSave this picture!© Pierre-Yves Brunaud•Paris, France Architects: associer Area Area of this architecture project Area:  4406 m² Year Completion year of this architecture project Year:  2025 Photographs Photographs:Pierre-Yves Brunaud Lead Architect: Philippe Madec More SpecsLess Specs Save this picture! Text description provided by the architects. In the heart of the 19th arrondissement of Paris, an isolated site surrounded by blocks of flats and towers has been refurbished. The atelier associer transformed the 1970s Jean Quarré upper secondary hospitality school into a media library and a Refugee House. The high cultural and social value of the chosen brief is unique in the world. It encompasses a neighborhood cultural facility and a place dedicated to the integration of exiled people on the territory of Paris. The media library joins the four media libraries of the city of Paris that possess a "center for the Deaf" to welcome the public communicating with sign language. The Refugee House brings together under one roof all the aspects of refugees' path to integration, in a convivial place of encounter and sharing.Save this picture!Save this picture!Save this picture!Rediscovery of the Existing Structure - The former secondary school structures have been preserved, cleaned, and had asbestos removed. Prefabricated elements like reinforced concrete posts, beams, floors, and façade panels were refurbished. A selective deconstruction of concrete floors, walls, and façade panels was conducted, with cut-off pieces kept on-site for reuse. This process restored the structure's regularity and improved the quality of the exposed reinforced concrete. As a result, ideal spaces for a media library and welcome center have been created. Save this picture!Save this picture!Save this picture!Save this picture!Space Open to the Air and Natural Light - The project is a bioclimatic design, drawing on the resources useful for its proper functioning in its environment. The now refined existing structure enjoys enhanced penetration of natural light. The exposed structure offers soft surfaces and rich textures, and provides inertia favourable to the improved regulation of the interior temperature. The entire project also provides enhanced natural and hygienic ventilation, namely through the creation of interior, double-height hoppers and of an air well with the garden in the patio.Save this picture! A Link made of Wood and Poured Earth - Taking an ecologically responsible approach, the architects have achieved a thorough renovation of the existing situation with a new construction made of biosourced and "geoscourced" material. The buildings, the square media library, and the elongated refugee house are joined by a vertical volume, called "the link", which serves the different levels and spaces of these two establishments. The walls of this unheated wooden construction are made of prefabricated poured earth, guaranteeing thermal inertia and regulated humidity. A wooden mantilla envelops the link whilst also serving as a high-protection sun filter. Hospitality - The tall post and beam structure provides a welcoming space for refugees and residents to enjoy beverages, learn French, and cook. It features spacious areas and diverse places with specially designed acoustics for everyone's comfort. A long terrace facing south offers relaxation moments and access to a shared garden.Save this picture!Save this picture!Save this picture!Landscape and Biodiversity - The project reversed the waterproofing on 70% of the surface of the lot. Open ground reappears everywhere, in the patio at the centre of the media library, on the forecourt, the shaded garden, the shared garden and on the terrace of the Refugee House. The effort to counter the urban heat island effect is organized through increased green surfaces, the implementation of light-coloured ground cover (recycled concrete slabs and stabilised sand), and a shallow pool between the shared garden and the shade garden.Save this picture! Project gallerySee allShow less Project locationAddress:Paris, FranceLocation to be used only as a reference. It could indicate city/country but not exact address.About this officeassocierOffice••• MaterialsWoodConcreteMaterials and TagsPublished on May 22, 2025Cite: "James Baldwin Media Library and Refugee House / associer" 22 May 2025. ArchDaily. Accessed . <https://www.archdaily.com/1030349/james-baldwin-media-library-and-refugee-house-associer&gt ISSN 0719-8884Save世界上最受欢迎的建筑网站现已推出你的母语版本!想浏览ArchDaily中国吗?是否 You've started following your first account!Did you know?You'll now receive updates based on what you follow! Personalize your stream and start following your favorite authors, offices and users.Go to my stream

Amazing sound for everything you listen to. FxSound corrects the limitations of your devices and compressed audio. FxSound is (technically) a 90's kid at heart. It's roots are in a program called DFX Audio Enhancer, born in 1999. This audio processor was a huge hit, and sold millions. As much as we dug all the crazy colorful skins and old-school sound, it was eventually time to move into the future of audio. With a few major overhauls and some polish, FxSound was born. We wanted to build on the pedigree and foundation of DFX, so we kept the best aspects while modernizing the sound and look. The result is the FxSound you see here today. Now, we made FxSound so that you can download it, turn it on, and just watch the cool bars bounce without touching the settings. Out of the box, it's got a great, all-around tone, with a good amount of volume and space added to your sound. But there is so much more under the hood, and with a little active effort FxSound can go from a nice audio quality booster to a fully customizable, powerful audio suite. Read on to learn more about how to use FxSound, and learn a little bit about your audio along the way. Video Games Surround sound to create full immersion into your virtual world. TV and Movies FxSound smooths and improves audio for your favorite shows and movies. Transcription FxSound boosts your clarity to save your sanity. Rewind less, get paid more. EQ Balanced your sound with our 9-band EQ with customizable center frequencies. Visualizer Feel your music in a new way. Effects Boosted bass and volume that will make your neighbors complain. Presets Presets professionally designed for music, gaming, movies, transcription, and more. How does FxSound work? Home stereo systems and PC's are built with a compromise: keep hardware affordable at the cost of quality. FxSound compensates for low quality hardware by targeting and processing sound's timbre, volume, spatial balance, and dynamics. What this means for you is clearer, louder, deeper, and balanced audio. Install our lightweight program, let it run quietly in the background, and enjoy your new audio experience. How does FxSound boost volume? FxSound is built to provide a dynamic yet controlled increase in volume. We prevent harmful peaking to make listening less fatiguing. Along with our equalized boost means you can increase your volume without distortion. (Unlike some other volume boosters out there). How does FxSound improve sound quality for music? Your music files, even ones labeled as "high quality", have some compromises. Some of the original data is removed to make the file a more manageable size. When you stream your music, even more 1's and 0's are cut out to make it fast to deliver. As such, your favorite songs become flat, dull, and lifeless. This is where FxSound comes in. FxSound helps by giving you the highest possible audio passthrough and output. Then, with targeted EQ, you can bring up those compressed areas and return your music to its proper form. How does FxSound help with gaming? Video games are one of the most acoustically rich, audio-saturated situations you'll encounter. Music, sound effects, ambient sounds, spatial processing, audio cues, warnings, indicators, and dialogue all crowd into your headphones or speakers. Unless you're dropping serious coin on a hi-fi gaming setup, you'll lose detail and miss important sounds. And if you're lucky enough to have that high-end gear, you can't always dial in global settings to fit your hearing profile. FxSound will pass your game's audio through at a higher bitrate than almost any other digital sound card. With EQ and effects, you can apply a smoother, richer, more clear experience to your favorite games. When looking to get that edge (and finally joining your high ELO friends), give FxSound a shot. What's New FxSound 1.1.33.0 Beta Changelog Volume normalization supported through RMS normalization Bugfix: Tray icon added back after explorer restart Context menu changes for playback device selection In mini mode, window will be always on top Ukrainian language support added Translation corrections in Arabic, Bosnian, Croatian, Dutch, Hungarian, Japanese, Portuguese and Spanish FxSound 1.1.31.0 Changelog Fixed the unhandled exception and crash in the application after installation Notifications not displayed during output device change is fixed Correction in Swedish translation

Key Takeaways Next Xbox Could Run Windows: A reliable leak suggests Microsoft is working on an Xbox that runs a custom Windows build with Steam support and full backward compatibility. Steam and Xbox Already Work Well Together: Xbox controllers dominate on Steam, and Microsoft continues releasing more games on the platform. Best-in-Class Backward Compatibility: Xbox Series X supports four console generations. A Windows-based Xbox could make this even more straightforward. A Strategic Shift to Stay Competitive: With PlayStation ahead, Microsoft may bet on openness, merging PC and console ecosystems to regain ground. In the never-ending console wars, it looks like Microsoft might be gearing up to throw out the rulebook – and maybe the whole library too.  According to reports, the next-generation Xbox could run a custom version of Windows, support Steam, and offer full backward compatibility.  That’s not just an upgrade. That’s a potential identity crisis. Or maybe, it’s a long-overdue evolution. Why This Isn’t Just Another Rumor First, let’s get this straight. This isn’t a random internet fantasy or Reddit wishlisting.  The source of these claims is a Spanish leaker known as eXtas1s, with a decent track record with Xbox-related news.  In a recent YouTube video, he claimed that Microsoft is working on a new Xbox console that would run a modified version of Windows, support PC storefronts like Steam, and offer full backward compatibility. The details were later summarized in English on ResetEra by a user named P40L0, helping the rumor spread across the broader gaming community.  Even possibly including a dedicated ‘Xbox mode’ launcher on top of Windows.  While Microsoft hasn’t confirmed any of this (and probably won’t anytime soon), it certainly makes sense when you look at the direction Xbox has been heading.  Xbox is no longer just a piece of hardware. It’s Game Pass. It’s cloud gaming. It’s the Xbox app on your PC. So when someone says the next Xbox might just be a living room-friendly Windows PC, the real surprise isn’t the rumor – it’s that it hasn’t already happened. Steam on Xbox? Why Not – They Already Share a Bed Now let’s talk about Steam.  Valve and Microsoft used to be slightly cold toward each other, especially when Windows 8 tried to become a walled garden. But now? They’re practically roommates.  Steam runs perfectly on Windows. Microsoft has been adding more and more games to Steam. The Xbox controller is the go-to option for most Steam players. In some cases, Microsoft even allows you to link your Xbox and Steam accounts.  Source: Xbox And let’s be honest: the Microsoft Store is… not great. Steam is where PC gamers live. Supporting Steam on the next Xbox might be less about generosity and more about pragmatism.  If you can’t beat Valve, join them – and maybe use their massive user base to prop up your console. Backward Compatibility: Microsoft’s Underrated Power Move The Xbox Series X is hands-down the most backward-compatible console on the market. It supports games from four generations – Original Xbox, Xbox 360, Xbox One, and Series X|S.  Source: Xbox.  You can play Knights of the Old Republic, Fable II, or Gears of War 3 without paying for a remake or subscription.  Just insert the disc or download your old digital copy, and you’re good to go. Many older titles look and run better, thanks to features like Auto HDR, higher resolutions, and faster load times. It’s not just support – it’s enhancement. This kind of access is rare. Sony’s PS5 doesn’t support PS3 discs at all. And Nintendo locks most of its classic library behind paid online services.  Microsoft, by contrast, is preserving its gaming legacy in a way that respects the player. If the next Xbox runs Windows and supports Steam, backward compatibility could surpass consoles. You could run your entire game library across decades and platforms with no emulators, just Windows doing what it does best. That’s more than a feature. It’s a reason to stick with Xbox. Or to finally switch.  Is This About PlayStation? Of Course it is Let’s not pretend this is happening in a vacuum.  The PlayStation 5 is dominating this console generation. Sony’s strategy has been aggressive and, frankly, smart. Exclusive titles. Strong global brand. A focus on cinematic single-player experiences that people want to play.  Microsoft tried to play the subscription game with Game Pass. It was a great idea, and it still is – but it hasn’t closed the gap.  And then came the Activision Blizzard acquisition. That was supposed to be a game-changer. Instead, it turned into a regulatory circus and has yet to deliver a big shift in momentum.  Call of Duty is still on PlayStation. Diablo IV didn’t launch as a Game Pass exclusive. So what does Microsoft do when it’s behind? It opens up. This isn’t the first time Microsoft has flipped the script. It open-sourced .NET, embraced Linux, and made Visual Studio Code free. The message behind these moves? We’re not the old Microsoft anymore — we’re the cool one now. Now, that same logic might apply to Xbox.  If people aren’t buying the Xbox for its exclusives, give them a reason to buy it for everything else. Steam, Windows, emulation, mods – these are things gamers want but don’t get from traditional consoles. The Console That Isn’t a Console Imagine a box under your TV that runs Windows, plays Steam games, runs emulators, supports Xbox Game Pass, and doesn’t care where your games come from.  Call it an Xbox, call it a PC, call it ‘Xbox Fusion’ or whatever marketing comes up with. But that’s not a console in the traditional sense – that’s a lifestyle device. And before you laugh at that phrase, remember the Steam Deck? That thing is a handheld Linux PC. People jailbreak it. They install Game Pass on it. They use it to emulate Nintendo games. Valve leaned into that openness, and it worked. If Valve could do it with a $399 handheld, why can’t Microsoft do it with a $499 living room console?  It’s About Ecosystem, Not Exclusives Once upon a time, consoles lived and died by their exclusives.  But Microsoft isn’t trying to sell you a box. It’s trying to sell you into an ecosystem. Game Pass. Xbox Live. Cloud saves. Achievements. Cross-play. Even the Xbox app on your phone. By turning the Xbox into a Windows-powered hybrid, they make it easier to say: ‘You’re already in the Xbox world. You just didn’t know it yet.’ But Will People Buy it? Here’s the catch: People who want a Windows gaming box… already own one. It’s called a PC.  Source: Unsplash.  So the next Xbox must do something a regular gaming PC doesn’t.  It needs to be plug-and-play. It needs to be cheap. It needs to be quiet, reliable, and hassle-free. No driver updates. No BIOS settings. No random crashes because Steam updated, and your mod loader broke.  If Microsoft can pull that off and make a gaming PC that feels like a console, they might not win the generation, but they’ll win back a lot of goodwill. And maybe even some market share.  What This Means for the Industry If the next Xbox truly embraces Windows and Steam, the lines between consoles and PCs will blur even further.  Sony will feel the pressure. Nintendo probably won’t care. But developers? They’ll love it. One build to rule them all, with less fragmentation and more options.  And if Xbox supports things like mouse and keyboard, mods, or even side-loading apps, it could become the best place to play, not because it’s a walled garden, but because it isn’t. Xbox Needs a Bold Move – This Might Be It The console wars are no longer just about graphics or frame rates. They’re about ecosystems. Microsoft tried going all-in on subscriptions, and it helped, but it wasn’t enough. So now, it’s pivoting again. Steam support. Windows under the hood. Full backward compatibility. It sounds wild. But it also sounds like the kind of disruption the gaming industry needs. Don’t be surprised if the next Xbox is a Trojan horse PC for your living room. Microsoft’s been training this horse for years.  Anya Zhukova is an in-house tech and crypto writer at Techreport with 10 years of hands-on experience covering cybersecurity, consumer tech, digital privacy, and blockchain. She’s known for turning complex topics into clear, useful advice that regular people can actually understand and use.  Her work has been featured in top-tier digital publications including MakeUseOf, Online Tech Tips, Help Desk Geek, Switching to Mac, and Make Tech Easier. Whether she’s writing about the latest privacy tools or reviewing a new laptop, her goal is always the same: help readers feel confident and in control of the tech they use every day.  Anya holds a BA in English Philology and Translation from Tula State Pedagogical University and also studied Mass Media and Journalism at Minnesota State University, Mankato. That mix of language, media, and tech has given her a unique lens to look at how technology shapes our daily lives.  Over the years, she’s also taken courses and done research in data privacy, digital security, and ethical writing – skills she uses when tackling sensitive topics like PC hardware, system vulnerabilities, and crypto security.  Anya worked directly with brands like Framework, Insta360, Redmagic, Inmotion, Secretlab, Kodak, and Anker, reviewing their products in real-life scenarios. Her testing process involves real-world use cases – whether it's stress-testing laptops for creative workloads, reviewing the battery performance of mobile gaming phones, or evaluating the long-term ergonomics of furniture designed for hybrid workspaces.  In the world of crypto, Anya covers everything from beginner guides to deep dives into hardware wallets, DeFi protocols, and Web3 tools. She helps readers understand how to use multisig wallets, keep their assets safe, and choose the right platforms for their needs.  Her writing often touches on financial freedom and privacy – two things she strongly believes should be in everyone’s hands. Outside of writing, Anya contributes to editorial style guides focused on privacy and inclusivity, and she mentors newer tech writers on how to build subject matter expertise and write responsibly.  She sticks to high editorial standards, only recommends products she’s personally tested, and always aims to give readers the full picture.  You can find her on LinkedIn, where she shares more about her work and projects.  Key Areas of Expertise: Consumer Tech (laptops, phones, wearables, etc.) Cybersecurity and Digital Privacy PC/PC Hardware Blockchain, Crypto Wallets, and DeFi In-Depth Product Reviews and Buying Guides Whether she’s reviewing a new wallet or benchmarking a PC build, Anya brings curiosity, care, and a strong sense of responsibility to everything she writes. Her mission? To make the digital world a little easier – and safer – for everyone.  View all articles by Anya Zhukova Our editorial process The Tech Report editorial policy is centered on providing helpful, accurate content that offers real value to our readers. We only work with experienced writers who have specific knowledge in the topics they cover, including latest developments in technology, online privacy, cryptocurrencies, software, and more. Our editorial policy ensures that each topic is researched and curated by our in-house editors. We maintain rigorous journalistic standards, and every article is 100% written by real authors.

Hope Dies Last: Visionary People Across the World, Fighting to Find Us a Future by Alan Weisman | Dutton | $32 The expression “rearranging deck chairs on the Titanic” is meant to convey a sense of futility, of meaningless action in the face of catastrophe, but I have been thinking recently of how it might be turned on its head, reimagined as a small gesture of purpose, of good work left to be done aboard that superlative metaphor for capitalist overabundance and technological hubris. I concede, it is not a perfect thought exercise, nor one nearly so elegant as those often deployed in the writing of the environmental journalist Alan Weisman, but I keep returning to it as I read Hope Dies Last, his final installment in a trilogy that also includes The World Without Us (2007) and Countdown (2013). Weisman has an acumen for moving fluidly between a cataclysmic big picture and intimate first-hand accounts of its impact and his rigorously researched storytelling is well-suited to the challenge of writing about a phenomenon as vastly incomprehensible as climate collapse. “We can’t picture 6,780 pyramids of Giza,” Weisman writes, so it is hard to visualize the 40 million tons of carbon humanity produces each year, but we can start to comprehend its impact through the stories of the people on the ground trying to do something about it—engineers at work restoring Mesopotamian wetlands in Iraq or plasma physicists developing commercial-scale fusion power at MIT. There are architects in this book too—including projects in the Korean DMZ by Bijoy Jain, Shigeru Ban, Seung H-Sang, and Minsuk Cho, and a Bjarke Ingels’ proposal for a floating city in Busan inspired by Buckminster Fuller and Shoji Sadao—but Hope Dies Last’s immediate relevance to readers of this publication may have more to do with a statistic familiar to these pages, the 40 percent of global greenhouse emissions for which the construction industry is responsible (a share second only to the oil and gas sector). Our ecological crisis is, at least in part, a byproduct of our built environment. (Courtesy Dutton) Weisman writes about that built environment beautifully. His description of the natural world’s ruthless repossession of a post-human Manhattan in the third chapter of The World Without Us is a small masterpiece of infrastructural prose. Many moments in Hope Dies Last, including a description of the massive network of dams, dikes, locks, levees, and storm surge barriers that comprise the Netherland’s Delta Works, are equally compelling. When COVID-19 locked much of humanity indoors, Weisman was seen as something of a visionary, but what makes his writing so griping is less his sweeping prognostications than his doggedly pragmatic realism. We cannot begin to address our climate crisis without looking rationally at and learning from its myriad constituent parts. The Netherlands’ methodically planned Delta Works, for example, cost less than a tenth of what the U.S. paid to clean up after hurricane Katrina. We need to be smarter about how we build. We also need to build less. Weisman’s second book, Countdown was a study of a central logical fallacy of late capitalism—that technology can solve for the impossibility of infinite population growth on a planet with very finite resources. We don’t like to talk about population control because it can feel coercive or contrary to our values but placing faith in technology to solve for unlimited human growth is lunacy. The same may be said of our hugely energy-intensive development of artificial intelligence infrastructure. A future edition of Countdown might include in its preface the 2023 story of a Belgian man despondent over the climate crisis, who committed suicide after a conversation with a chatbot that arrived at the logical conclusion that most impactful thing he could do to save the earth would be to remove the burden he placed upon it. The environmental outlook is far worse now than when Weisman began his career; the message at the core of Hope Dies Last is more somber than the two books that preceded it. Hope is not an expression Weisman uses lightly. Early in the book he unpacks the foreboding that lingers within a word that conveys an emotion more powerful than optimism precisely because it also carries a tinge of doubt. Hope, he notes, is a word that struggles against itself. That struggle is compounded by the uncomfortable truth of the astonishingly destructive work that has been done in the weeks since Hope Dies Last went to press. Even my advance copy has begun to feel outdated as I watch the current administration slam the door on so many small last best hopes. Yet, there is still work to be done and much of it by the people who think about how we build the world around us. In the words of a naturalist named Rosario whom Weisman meets while she is helping to clear the beach of plastic bottles during a turtle census in the Yucatán: “It’s a lost cause, but we do it anyway.” Justin Beal is an artist and author based in New York. His first book, Sandfuture, was published by MIT Press in 2021. He teaches at Hunter College and the Yale School of Architecture. This post contains affiliate links. AN may have a commission if you make a purchase through these links.

Interviews Andor – Season 2: Mohen Leo (Production VFX Supervisor), TJ Falls (Production VFX Producer) and Scott Pritchard (ILM VFX Supervisor) By Vincent Frei - 22/05/2025 In 2023, Mohen Leo (Production VFX Supervisor), TJ Falls (Production VFX Producer), and Scott Pritchard (ILM VFX Supervisor) offered an in-depth look at the visual effects of Andor’s first season. Now, the trio returns to share insights into their work on the second—and final—season of this critically acclaimed series. Tony Gilroy is known for his detailed approach to storytelling. Can you talk about how your collaboration with him evolved throughout the production of Andor? How does he influence the VFX decisions and the overall tone of the series? Mohen Leo (ML): Our history with Tony, from Rogue One through the first season of Andor, had built a strong foundation of mutual trust. For Season 2, he involved VFX from the earliest story discussions, sharing outlines and inviting our ideas for key sequences. His priority is always to keep the show feeling grounded, ensuring that visual effects serve the story’s core and never become extraneous spectacle that might distract from the narrative. TJ Falls (TJ): Tony is a master storyteller. As Mohen mentioned, we have a great history with Tony from Rogue One and through Season 1 of Andor. We had a great rapport with Tony, and he had implicit trust in us. We began prepping Season 2 while we were in post for Season 1. We were having ongoing conversations with Tony and Production Designer Luke Hull as we were completing work for S1 and planning out how we would progress into Season 2. We wanted to keep the show grounded and gritty while amping up the action and urgency. Tony had a lot of story to cover in 12 episodes. The time jumps between the story arcs were something we discussed early on, and the need to be able to not only justify the time jumps but also to provide the audience with a visual bridge to tell the stories that happened off-screen. Tony would look to us to guide and use our institutional knowledge of Star Wars to help keep him honest within the universe. He, similarly, challenged us to maintain our focus and ensure that the visual tone of the series serviced the story. Tony Gilroy and Genevieve O’Reilly on the set of Lucasfilm’s ANDOR Season 2, exclusively on Disney+. Photo by Des Willie. ©2024 Lucasfilm Ltd. & TM. All Rights Reserved. As you’ve returned for Season 2, have there been any significant changes or new challenges compared to the first season? How has the production evolved in terms of VFX and storytelling? (ML): The return of nearly all key creatives from Season 1, both internally and at our VFX vendors, was a massive advantage. This continuity built immediate trust and an efficient shorthand. It made everyone comfortable to be more ambitious, allowing us to significantly expand the scope and complexity of the visual effects for Season 2. (TJ): We had all new directors this season. The rest of the core creative and production teams stayed consistent from Season 1. We worked to keep the creative process as seamless from Season 1 as we could while working with the new directors and adapting to their process while incorporating their individual skills and ideas that they brought to the table. This season we were able to work on location much more than on Season 1. That provided us with a great opportunity to build out the connective tissue between real world constraints and the virtual world we were creating. In the case with Senate Plaza in Coruscant we also had to stay consistent with what has previously been established, so that was a fun challenge. How did you go about dividing the workload between the various VFX studios? (ML): I can give an answer, but probably better if TJ does. (TJ): We were very specific about how we divided the work on this series. We started, as we usually do, with a detailed breakdown of work for the 12 episodes. Mohen and I then discussed a logical split based on type of work, specific elements, and areas of commonality for particular environments. While cost is always a consideration, we focused our vendor casting around the creative strengths of the studios we were partnering with on the project. ILM is in the DNA of Star Wars, so we knew we’d want to be working with them on some of the most complex work. We chose ILM for the opening TIE Avenger hangar sequence and subsequent escape. We utilized ILM for work in every episode, including the CG KX/K2 work, but their main focus was on Coruscant, and they had substantial work in the ninth episode for the big Senate escape sequence. Hybride‘s chief focus was on Palmo Plaza and the Ghorman environments. They dealt with everything Ghorman on the ground from the street extensions and the truck crash, through the Ghorman massacre, sharing shots with ILM with the KX work. For Scanline VFX, we identified three primary areas of focus: the work on Mina Rau, Chandrila, and Yavin. The TIE Fighter sequence in Season 2 is a standout moment. Can you walk us through the VFX process for that particular sequence? What were some of the technical challenges you faced, and how did you work to make it as intense and realistic as possible? (TJ): This is a sequence I’m particularly proud of as VFX played a central role in the sequence coming together from start to finish. We were intimately involved from the initial conversations of the idea for the sequence. Mohen created digital storyboards and we pitched ideas for the sequence to Tony Gilroy. Once we had a sense of the creative brief, we started working with Luke Hull (Production Designer) and the art department on the physical hangar set and brought it into previz for virtual scouting. With Jen Kitching (our Previz Supervisor from The Third Floor) we had a virtual camera set up that allowed us to virtually use the camera and lenses we would have on our shoot. We blocked out shots with Ariel Kleiman (Director) and Christophe Nuyens (the DoP). This went back through previz and techviz so we could meticulously chart out our plan for the shoot. Keeping with our ethos of grounding everything in reality, we wanted to use as much of the practical set as possible. We needed to be sure our handoffs between physical and virtual were seamless – Luke Murphy, our SFX Supervisor, worked closely with us in planning elements and practical effects to be used on the day. Over the course of the shoot, we also had the challenge of the flashing red alarm that goes off once the TIE Avenger crashes into the ceiling. We established the look of the red alarm with Christophe and the lighting team, and then needed to work out the timing. For that, we collaborated with editor John Gilroy to ensure we knew precisely when each alarm beat would flash. Once we had all the pieces, we turned the sequence over to Scott Pritchard and ILM to execute the work. Scott Pritchard (SP): This sequence was split between our London and Vancouver studios, with London taking everything inside the hangar, and Vancouver handling the exterior shots after Cassian blasts through the hangar door. We started from a strong foundation thanks to two factors: the amazing hangar set and TIE Avenger prop; and having full sequence previs. The hangar set was built about 2/3 of its overall length (as much as could be built on the soundstage), which our environments team extended, adding the hangar doors at the end and also a view to the exterior environment. Extending the hangar was most of the work in the sequence up until the TIE starts moving, where we switched to our CG TIE. As with Season 1, we used a blend of physical SFX work for the pyro effects, augmenting with CG sparks. As TJ mentioned, the hangar’s red warning lighting was a challenge as it had to pulse in a regular tempo throughout the edit. Only the close-up shots of Cassian in the cockpit had practical red lighting, so complex lighting and comp work were required to achieve a consistent look throughout the sequence. ILM London’s compositing supervisor, Claudio Bassi, pitched the idea that as the TIE hit various sections of the ceiling, it would knock out the ceiling lights, progressively darkening the hangar. It was a great motif that helped heighten the tension as we get towards the moment where Cassian faces the range trooper. Once we cut to outside the hangar, ILM Vancouver took the reins. The exterior weather conditions were briefed to us as ‘polar night’ – it’s never entirely dark, instead there’s a consistent low-level ambient light. This was a challenge as we had to consider the overall tonal range of each shot and make sure there was enough contrast to guide the viewer’s eye to where it needed to be, not just on individual shots but looking at eye-trace as one shot cut to another. A key moment is when Cassian fires rockets into an ice arch, leading to its collapse. The ice could very easily look like rock, so we needed to see the light from the rocket’s explosions scattered inside the ice. It required detailed work in both lighting and comp to get to the right look. Again, as the ice arch starts to collapse and the two chase TIE Advanced ships get taken out, it needed careful balancing work to make sure viewers could read the situation and the action in each shot. The world-building in Andor is impressive, especially with iconic locations like Coruscant and Yavin. How did you approach creating these environments and ensuring they felt as authentic as possible to the Star Wars universe? (ML): Our approach to world-building in Andor relied on a close collaboration between the VFX team and Luke Hull, the production designer, along with his art department. This partnership was established in Season 1 and continued for Season 2. Having worked on many Star Wars projects over the decades, VFX was often able to provide inspiration and references for art department designs. For example, for locations like Yavin and Coruscant, VFX provided the art department with existing 3D assets: the Yavin temple model from Rogue One and the Coruscant city layout around the Senate from the Prequel films. The Coruscant model, in particular, involved some ‘digital archaeology.’ The data was stored on tapes from around 2001 and consisted of NURBS models in an older Softimage file format. To make them usable, we had to acquire old Softimage 2010 and XSI licenses, install them on a Windows 7 PC, and then convert the data to the FBX format that current software can read. Supplying these original layouts to the art department enabled them to create their new designs and integrate our real-world shooting locations while maintaining consistency with the worlds seen in previous Star Wars productions. Given that Andor is set approximately twenty years after the Prequels, we also had the opportunity to update and adjust layouts and designs to reflect that time difference and realize the specific creative vision Luke Hull and Tony Gilroy had for the show. StageCraft technology is a huge part of the production. How did you use it to bring these complex environments, like Coruscant and Yavin, to life? What are the main benefits and limitations of using StageCraft for these settings? (SP): Our use of StageCraft for Season 2 was similar to that on Season 1. We used it to create the exterior views through the windows of the Safehouse on Coruscant. As with our work for the Chandrillan Embassy in Season 1, we created four different times of day/weather conditions. One key difference was that the foreground buildings were much closer to the Safehouse, so we devised three projection points (one for each room of the Safehouse), which would ensure that the perspective of the exterior was correct for each room. On set we retained a large amount of flexibility with our content. We had our own video feed from the unit cameras, and we were able to selectively isolate and grade sections of the city based on their view through the camera. Working in context like this meant that we could make any final tweaks while each shot was being set up and rehearsed. While we were shooting a scene set at night, the lighting team rigged a series of lights running above the windows that, when triggered, would flash in sequence, casting a moving light along the floor and walls of the set, as if from a moving car above. I thought we could use the LED wall to do something similar from below, catching highlights on the metal pipework that ran across the ceiling. During a break in shooting, I hatched a plan with colour operator Melissa Goddard, brain bar supervisor Ben Brown, and we came up with a moving rectangular section on the LED wall which matched the practical lights for speed, intensity and colour temperature. We set up two buttons on our iPad to trigger the ‘light’ to move in either direction. We demoed the idea to the DP after lunch, who loved it, and so when it came to shoot, he could either call from a car above from the practical lights, or a car below from the LEDs. (ML): Just to clarify – the Coruscant Safehouse set was the only application of Stagecraft LED screens in Season 2. All other Coruscant scenes relied on urban location photography or stage sets with traditional blue screen extensions. The various Yavin locations were achieved primarily with large backlot sets at Longcross Studios. A huge set of the airfield, temple entrance and partial temple interior was extended by Scanline VFX, led by Sue Rowe, in post, creating the iconic temple exterior from A New Hope. VFX also added flying and parked spaceships, and augmented the surrounding forest to feel more tropical. Andor blends CG with actual real-world locations. Can you share how you balanced these two elements, especially when creating large-scale environments or specific landscapes that felt grounded in reality? (SP): A great example of this is the environment around the Senate. The plates for this were shot in the City of Arts & Sciences in Valencia. Blending the distinctive Calatrava architecture with well-known Star Wars buildings like the Senate was an amazing challenge, it wasn’t immediately clear how the two could sit alongside each other. Our Vancouver team, led by Tania Richard, did an incredible job taking motifs and details from the Valencia buildings and incorporating them into the Senate building on both large and small scales, but still contiguous with the overall Senate design. The production team was ingenious in how they used each of the Valencia buildings to represent many locations around the Senate and the surrounding areas. For example, the Science Museum was used for the walkway where Cassian shoots Kloris (Mon’s driver), the main entrance to the Senate, and the interior of the Senate Atrium (where Ghorman Senator Oran is arrested). It was a major challenge ensuring that all those locations were represented across the larger environment, so viewers understood the geography of the scene, but also blended with the design language of their immediate surroundings. Everything in the Senate Plaza had a purpose. When laying out the overall layout of the Plaza, we considered aspects such as how far Senators would realistically walk from their transports to the Senate entrance. When extending the Plaza beyond the extents of the City of Arts & Sciences, we used Calatrava architecture from elsewhere. The bridge just in front of the Senatorial Office Building is based on a Calatrava-designed bridge in my home city of Dublin. As we reach the furthest extents of the Senate Plaza, we begin blending in more traditional Coruscant architecture so as to soften the transition to the far background. Coruscant is such a pivotal location in Star Wars. How did you approach creating such a vast, densely populated urban environment? What were the key visual cues that made it feel alive and realistic? (ML): Our approach to Coruscant in Season 2 built upon what we established in the first season: primarily, shooting in real-world city locations whenever feasible. The stunning Calatrava architecture at Valencia’s City of Arts and Sciences, for instance, served as the foundation for the Senate exterior and other affluent districts. For the city’s grittier neighborhoods, we filmed in urban environments in London, like the Barbican and areas around Twickenham Stadium. Filming in these actual city locations provided a strong, realistic basis for the cinematography, lighting, and overall mood of each environment. This remained true even when VFX later modified large portions of the frame with Star Wars architecture. This methodology gave the director and DP confidence on set that their vision would carry through to the final shot. Our art department and VFX concept artists then created numerous paintovers based on plates and location photography, offering clear visual guides for transforming each real location into its Coruscant counterpart during post-production. For the broader cityscapes, we took direct inspiration from 3D street maps of cities such as Tokyo, New York, and Hong Kong. We would exaggerate the scale and replace existing buildings with our Coruscant designs while preserving the fundamental urban patterns. When it comes to creating environments like Yavin, which has a very natural, jungle-like aesthetic, how do you ensure the VFX stays true to the organic feel of the location while still maintaining the science-fiction elements of Star Wars? (ML): Nearly all of the Yavin jungle scenes were shot in a large wooded area that is part of Longcross Studios. The greens and art departments did an amazing job augmenting the natural forest with tropical plants and vines. The scenes featuring the two rebel factions in the clearing were captured almost entirely in-camera, with VFX primarily adding blaster fire, augmenting the crashed ship, and painting out equipment. Only the shots of the TIE Avenger landing and taking off, as well as the giant creature snatching the two rebels, featured significant CG elements. The key elements connecting these practical locations back to the Yavin established in A New Hope and Rogue One were the iconic temples. The establishing shots approaching the main temple in episode 7 utilized plate photography from South America, which had been shot for another Disney project but ultimately not used. Other aerial shots, such as the U-Wing flying above the jungle in episode 12, were fully computer-generated by ILM. K-2SO is a beloved character, and his return is highly anticipated. What can you tell us about the process of bringing him back to life with VFX in Season 2? What new challenges did this bring compared to his original appearance? (SP): We had already updated a regular KX droid for the scene on Niamos in Season 1, so much of the work to update the asset to the latest pipeline requirements had already been done. We now needed to switch over to the textures & shaders specific to K2, and give them the same updates. Unique to Series 2 was that there were a number of scenes involving both a practical and a digital K2 – when he gets crushed on Ghorman in episode 8, and then ‘rebooted’ on Yavin in episode 9. The practical props were a lot more beaten up than our hero asset, so we made bespoke variants to match the practical droid in each sequence. Additionally, for the reboot sequence on Yavin, we realised pretty quickly that the extreme movements meant that we were seeing into areas that previously had not required much detail – for instance, underneath his shoulder armour. We came up with a shoulder joint design that allowed for the required movement while also staying mechanically correct. When we next see him in Episode 10, a year has passed, and he is now the K-2SO as we know him from Rogue One. K-2SO has a unique design, particularly in his facial expressions and movement. How did you approach animating him for Season 2, and were there any specific changes or updates made to his character model or animation? (SP): Following Rogue One, Mohen made detailed records of the takeaways learned from creating K-2SO, and he kindly shared these notes with us early on in the show. They were incredibly helpful in tuning the fine details of the animation. Our animation team, led by Mathieu Vig, did a superb job of identifying the nuances of Alan’s performance and making sure they came across. There were plenty of pitfalls to avoid – for instance, the curve to his upper back meant that it was very easy for his neck to look hyperextended. We also had to be very careful with his eyes, as they’re sources of light, they could very easily look cartoonish if they moved around too much. Dialling in just the right amount of eye movement was crucial to a good performance. As the eyes also had several separate emissive and reflective components, they required delicate balancing in the comp on a per-shot basis. Luckily, we had great reference from Rogue One to be able to dial in the eyes to suit both the lighting of a shot but also its performance details. One Rogue One shot in particular, where he says ‘Your behavior, Jyn Erso, is continually unexpected’, was a particularly good reference for how we could balance the lights in his eyes to, in effect, enlarge his pupils, and give him a softer expression. K-2SO also represented my first opportunity to work with ILM’s new studio in Mumbai. Amongst other shots, they took on the ‘hallway fight’ sequence in Episode 12 where K2 dispatches Heert and his troopers, and they did a fantastic job from animation right through to final comp. K-2SO’s interactions with the live-action actors are key to his character. How did you work with the actors to ensure his presence felt as real and integrated as possible on screen, especially in terms of timing and reactions? (ML): Alan Tudyk truly defined K-2SO in Rogue One, so his return for Andor Season 2 was absolutely critical to us. He was on set for every one of K2’s shots, performing on stilts and in a performance capture suit. This approach was vital because it gave Alan complete ownership of the character’s physical performance and, crucially, allowed for spontaneous, genuine interactions with the other actors, particularly Diego Luna. Witnessing Alan and Diego reunite on camera was fantastic; that unique chemistry and humor we loved in Rogue One was instantly palpable. In post-production, our VFX animators then meticulously translated every nuance of Alan’s on-set performance to the digital K-2SO model. It’s a detailed process that still requires artistic expertise. For instance, K2’s facial structure is largely static, so direct translation of Alan’s facial expressions isn’t always possible. In these cases, our animators found creative solutions – translating a specific facial cue from Alan into a subtle head tilt or a particular eye movement for K2, always ensuring the final animation remained true to the intent and spirit of Alan’s original performance. Were there any memorable moments or scenes from the series that you found particularly rewarding or challenging to work on from a visual effects standpoint? (ML): The Plaza sequence in episode 8, which runs for about 23 minutes, stands out as particularly memorable – both for its challenges and its rewarding outcome. Just preparing for it was a daunting task. Its successful execution hinged on incredibly tight collaboration between numerous departments: stunts, creature effects, special effects, the camera department, our tireless greenscreens crew, and of course, VFX. The stunts team, under Marc Mailley, drove the choreography of all the action. Our On-Set VFX Supervisor, Marcus Dryden, was instrumental. He worked hand-in-glove with the director, DP, and assistant directors to ensure we meticulously captured all the necessary elements. This included everything from crowd replication plates and practical effects elements to the performances of stunt teams and creature actors, plus all the crucial on-set data. The shoot for this sequence alone took over three weeks. Hybride, under the leadership of Joseph Kasparian and Olivier Beaulieu, then completed the environments, added the blaster fire, and augmented the special effects in post-production, with ILM contributing the KX droids that wreak havoc in the plaza. (SP): I agree with Mohen here, for me the Ghorman Plaza episode is the most rewarding to have worked on. It required us to weave our work into that of so many other departments – stunts, sfx, costume – to name just a few. When we received the plates, to see the quality of the work that had gone into the photography alone was inspirational for me and the ILM crew. It’s gratifying to be part of a team where you know that everyone involved is on top of their game. And of course all that is underpinned by writing of that calibre from Tony Gilroy and his team – it just draws everything together. From a pure design viewpoint, I’m also very proud of the work that Tania Richard and her ILM Vancouver crew did for the Senate shots. As I mentioned before, it was a hugely challenging environment not just logistically, but also in bringing together two very distinctive architectural languages, and they made them work in tandem beautifully. Looking back on the project, what aspects of the visual effects are you most proud of? (TJ): I’m incredibly proud of this entire season. The seamless collaboration we had between Visual Effects and every other department made the work, while challenging, an absolute joy to execute. Almost all of the department heads returned from the first season, which provided a shorthand shortcut as we started the show with implicit trust and understanding of what we were looking to achieve. The work is beautiful, and the commitment of our crew and vendors has been unwavering. I’m most proud of the effort and care that each individual person contributed to the show and the fact that we went into the project with a common goal and were, as a team, able to showcase the vision that we, and Tony, had for the series. (ML): I’m really proud of the deep integration of the visual effects – not just visually, but fundamentally within the filmmaking process and storytelling. Tony invited VFX to be a key participant in shaping the story, from early story drafts through to the final color grade. Despite the scale and spectacle of many sequences, the VFX always feel purposeful, supporting the narrative and characters rather than distracting from them. This was significantly bolstered by the return of a large number of key creatives from Season 1, both within the production and at our VFX vendors. That shared experience and established understanding of Tony’s vision for Andor were invaluable in making the VFX an organic part of the show. (SP): I could not be prouder of the entire ILM team for everything they brought to their work on the show. Working across three sites, Andor was a truly global effort, and I particularly enjoyed how each site took complete ownership of their work. It was a privilege working with all of them and contributing to such an exceptional series. VFX progression frame Lucasfilm’s ANDOR Season 2, exclusively on Disney+. Photo courtesy of Lucasfilm. ©2025 Lucasfilm Ltd. & TM. All Rights Reserved. How long have you worked on this show? (TJ): This show has been an unbelievable journey. Season 2 alone was nearly 3 years. We wrapped Season 2 in January of 2025. We started prepping Season 2 in February 2022, while we were still in post for Season 1. I officially started working on Season 1 early in 2019 while it was still being developed. So that’s 6 years of time working on Andor. Mohen and I both also worked on Rogue One, so if you factor in the movie, which was shooting in 2015, that’s nearly ten years of work within this part of the Star Wars universe. (ML): I started on the project during early development in the summer of 2019 and finished in December of 2024. (SP): I started on Season 1 in September 2020 and finished up on Season 2 in December 2024. What’s the VFX shots count? (TJ): We had a grand total of 4,124 shots over the course of our 12 episodes. Outside of Industrial Light & Magic, which oversaw the show, we also partnered with Hybride, Scanline, Soho VFX, and Midas VFX. What is your next project? (TJ): You’ll have to wait and see! (SP): Unfortunately, I can’t say just yet either! A big thanks for your time. WANT TO KNOW MORE?ILM: Dedicated page about Andor – Season 2 on ILM website. © Vincent Frei – The Art of VFX – 2025

Touch, I remember touch Power Sink (Winterwire Games, 22nd May, $14.99) - Set in the ruins of a submerged power station, Power Sink blends classic platforming with clever logic-based puzzles as players take on the role of a Diver sent to restore power to a fading underwater colony. Project Zombie (XenoXeno, 24th May, $12.99) - Project Zombie is a survival game set in a post-apocalyptic world, where players must push their physical and mental limits in a constant fight for life. In a world overrun by zombies, quick decisions, clever strategies, and cooperation with other survivors make the difference between life and death. Sauna of the DEAD (KAMIICHI DIGITAL, 22nd May, $7.99) - Step into the world's most unusual sauna game and become a legendary Aufguss Maestro. We believe you have what it takes to be the best heat-bringer in the underworld! Purify zombies in a sauna from hell! Soccer Golf! (Ratalaika Games, 23rd May, $4.99) - Do you love soccer? Do you love golf? Tee off and kickoff in SoccerGolf, a brilliant and challenging new physics game which seamlessly combines your two favorite sports. Choose from 54 challenging levels with 3 unique themes, and get ready for increased obstacles, hazards, and difficulty. Starlight Legacy (eastasiasoft, 28th May, $12.99) - Return to the style of 16-bit JRPGs in Starlight Legacy, a non-linear adventure inspired by genre classics! Set in a post-medieval world in the Evaria Kingdom, explore a vast kingdom created with lush 2D pixel art, a nostalgic turn-based battle system and non-linear story where the four provinces of the kingdom can be explored in any order of your choosing. STEAM-HEART’S Saturn Tribute (City Connection, 28th May, $19.99) - The return of a hardcore and sexy sci-fi vertical scrolling shooter! Players can obtain and power up two types of main shots and six types of sub-weapons. Utilize special techniques such as the "Weapon Crash", which temporarily boosts sub-weapons, and the "Boost" ability, which lets you momentarily accelerate to dodge enemy attacks. Master these skills to suppress the oncoming enemies! Super Waifu Ball (JanduSoft, 22nd May, $5.99) - Super Waifu Ball is an arcade-style adventure where you roll through colorful, obstacle-filled worlds as a cast of adorable waifus. Complete levels, collect stars, and unlock collectible waifu figurines from the gacha machine! Tavern Owner Simulator (GBTUI, 17th May, $12.99) - Welcome to Tavern Owner Simulator, where your dream of managing a medieval pub becomes reality. Step into the shoes of a tavern proprietor and transform a modest establishment into the most popular spot in town. With a mix of strategic planning, creative design, and savvy management, you’ll turn your tiny tavern into a bustling hub of activity. The Operator (Bureau 81, 22nd May, $13.99) - Welcome to the FDI. As our newest Operator, your role is to use your detective skills to assist our field agents and investigate mysterious crimes. Use cutting-edge FDI software to dig for clues, solve puzzles, and uncover the truth. The Town of Nie Iromusubi (HuneX, 22nd May, $39.19) - "The Town of Nie Iromusubi", a fan-disc of the strange and eerie otherworldly BL fantasy "The Town of Nie. "Weaving together extraordinary colors. Trident’s Tale (3DClouds, 22nd May, $24.99) - Set sail for an epic action-packed adventure in Trident’s Tale, where you’ll become Ocean, a bold young captain on a quest to reclaim the legendary Storm Trident—a mythical artifact that grants power over the seas. But beware the ocean is vast, treacherous, and crawling with danger at every turn! Weirdo (CASCHA GAMES, 22nd May, $10.99) - Parallel-Weirdo is a Japanese manga-style visual novel set in DYL High School. It follows Wang Zhe, an ordinary and frail boy, who tries to win the hearts of girls and complete What will you be downloading this week? (63 votes) A Bibelot: Tiret sur Will  0%ADVANCED V.G. Saturn Tribute2%Arcade Archives NebulasRay5%Casual Sport Series: Badminton  0%Cats Visiting Fairy Tales  0%Cattle Country3%Cipheur  0%Cookie Clicker8%Cozy Winter  0%Digger Simulator: Gold Rush  0%Duck Detective: The Ghost of Glamping8%Eggconsole Hydlide Ii Msx2%Extreme! 2-Choice Flag Quiz  0%Food Truck Business Simulator  0%Genopanic  0%Ghost Room Deep  0%Knight's Night!  0%Legends Aligned: Minis in Conflict  0%Merge Numbers  0%Mobile Suit Gundam SEED: Battle Destiny Remastered3%Monster Train 28%Mudness Offroad Car Simulator Runner  0%Not Not 2 - A Brain Challenge2%Onimusha 2: Samurai's Destiny6%Pancho's Mission  0%Pick Race 3D  0%Pipistrello and the Cursed Yoyo19%Planeta 55  0%Please, Touch The Artwork 25%Power Sink3%Project Zombie  0%Sauna of the DEAD  0%Soccer Golf!  0%Starlight Legacy3%STEAM-HEART'S Saturn Tribute2%Super Waifu Ball2%Tavern Owner Simulator  0%The Operator  0%The Town of Nie Iromusubi  0%Trident's Tale  0%Weirdo  0%Nothing for me this week21% So that's your lot for this week's North American Nintendo Download. Go on, be a sport and drop a vote in the poll above, and comment below with your hot picks! Related Games See Also

oops Tesla crushed in Europe as BYD outsells; BEV sales surge 28% Tesla sales fell 49 percent year over year in April. Jonathan M. Gitlin – May 22, 2025 9:52 am | 99 Credit: Getty Images Credit: Getty Images Story text Size Small Standard Large Width * Standard Wide Links Standard Orange * Subscribers only   Learn more The extent of Tesla's meteoric decline in popularity is on vivid display in the latest new car registration numbers coming out of Europe. New car sales were essentially flat in the region last month, with just under 1,400 more cars sold this year than last. But the market is far from static; plug-in sales are booming, with battery electric vehicle registrations up by 28 percent according to the analysts at JATO Dynamics, and plug-in hybrid EV sales increased by 31 percent. Almost every automaker has capitalized on this growth, with a few exceptions—Tesla being the most significant. As the first mainstream BEV-only brand, Tesla led the way in European EV sales and made much of the fact that its Model Y crossover was the best-selling car in Europe for some time. Those days are long gone. Model Y registrations fell by 53 percent last month to just 4,495 units, dropping it to 9th on the list of most-registered BEVs. First place went to the Skoda Elroq, followed by VW's ID.3, ID.7, ID.4, and the new Kia EV3. When you look at sales at the brand level, things get a little worse for the American automaker. Volkswagen sold more EVs than anyone else in Europe last month, increasing by 61 percent to 23,514 units. As for Tesla? It fell to 11th place, with just 7,165 sales in total, a 49 percent decrease year on year. Beating it to 10th place was China's BYD. Barred from the US market by protectionist laws and now heavy new tariffs, BYD has focused instead on Europe. Its PHEVs have been selling strongly there, unaffected by tariffs aimed at BEVs, but even its BEV sales have now eclipsed Tesla, with 7,231 registrations last month. "Although the difference between the two brands' monthly sales totals may be small, the implications are enormous," said Felipe Munoz, global analyst at JATO Dynamics. "This is a watershed moment for Europe's car market, particularly when you consider that Tesla has led the European BEV market for years, while BYD only officially began operations beyond Norway and the Netherlands in late 2022." Jonathan M. Gitlin Automotive Editor Jonathan M. Gitlin Automotive Editor Jonathan is the Automotive Editor at Ars Technica. He has a BSc and PhD in Pharmacology. In 2014 he decided to indulge his lifelong passion for the car by leaving the National Human Genome Research Institute and launching Ars Technica's automotive coverage. He lives in Washington, DC. 99 Comments

Death to Squishies By Bart on May 22, 2025 Behind the Scenes I asked Marcin Cieśla to tell us about his 'Death to Squishies' animation that was featured on BlenderArtists in March. Get ready for the delights and despairs of a one-man animation project which included dancing in front of his phone girl-band style for reference!A stort story about a certain animation. “Death to Squishies” from Ratchet & Clank 3.Ladies and gentlemen. Girls and boys. I rarely write texts like this… but not by choice. Only when my work is done, all set with poise, each detail aligned, no faults or voids. Now and then, when the time employs, it feels like festive days with all their noise. I’m so happy, like a kid with toys! Let’s start the talk - so raise your voice!...and Courtney strikes a daring pose, Perhaps it’s time to end my prose. Alright, enough fooling around. I’m not typing all of this just to waste someone’s time with poor-quality rhymes. So, why are we here? Well, I’d like to tell a story. To write a few words about how I spent almost two years working on a certain project - my biggest one yet. I personally love reading behind-the-scenes stories like this, so I thought someone might find it interesting. And if not, well, too bad - I’ll just write it anyway, and no one can stop me - ha! Look, look, how I just write all of this.Ok man, what’s your point?...you may ask, my dear reader. I will answer brilliantly - I don't know. That's why I think it would be best if I started with a short explanation. In the year of our Lord 2023, I created a small fanart of a character from the Ratchet & Clank series. Specifically, I'm talking about Courtney Gears from Ratchet 3. In the description of this project, when I was finishing it, I mentioned that I might also try to create a 1-minute animation using the character model that I already had at the time. It was quite tempting considering that the goal seemed quite achievable, because in a sense part of the work was already done. At least that's what I thought at the time. After all, I already have a finished character model - why not use it again and create an animation with it? What could go wrong (those familiar with Murphy's law know very well what). Ah, the poor naive author of this text didn’t yet realize what he was signing up for...At the beginning, he created a new project and a scene.The scene, however, was without form, it was a blockout. It is best to start any project like this with a blockout - a kind of 3D “sketch.” You roughly and quickly arrange simple shapes in the scene to get a general idea of where things should be, what size they should have relative to other elements in the environment, and so on. That’s exactly what I did. This stage of work is particularly satisfying because any progress brings immediate, tangible, and easily noticeable results. I drop a character into the file, but there’s nothing around them - just emptiness. So, I create a simple floor and walls. Boom - suddenly, the whole perception of the scene changes completely. It’s a bit dark, so I add a basic lamp. White, colored, spot, or area - whatever it may be, its mere presence makes a huge difference.Although every object in the scene, even in its simple blockout form, had to be manually created, it didn’t take long before the first signs of progress started appearing before my eyes. Those first bursts of dopamine can be incredibly motivating. "Hey, this is actually going pretty smoothly," I thought. And it did. You start a new project and you have only an empty space with a default cube and a camera in front of you. A day or two later, in the same file, you see a whole bunch of colorful, scattered objects. Maybe they are simple, maybe a little sloppy, but it doesn't matter. Compared to what was there at the beginning (an empty 3D file), the progress is huge! You almost feel like patting yourself on the back. And although at this stage it seemed to me that I had already done a lot of work, in reality it was just a false first impression. The real work hadn't even started yet.Lights, action! Oh wait, the camera first.Since the project was supposed to be an animation, I decided that before I went any further, I should roughly outline the camera work. An initial frame layout to see roughly where the close-ups would be, and what the slightly wider shots would look like. Which objects would need more detail, and which could get away with less. It seemed like a reasonable approach - after all, it would be a shame to spend too much time perfecting an object that wouldn’t even be properly visible in the camera.My starting point, of course, was the original animation from Ratchet & Clank. The music video titled "Death to Squishies" contained all the guidelines I needed. In my version, I tried to preserve as much from the original as possible, so initially I simply replicated the camera movements one-to-one. This is where I ran into the first problem. What worked 20 years ago in PlayStation 2 graphics doesn't necessarily translate well to modern visuals. Maybe I just couldn't translate it properly - who knows. The fact is, however, that at the end of the day I slightly modified some shots and cuts. Not drastically, but still. A slightly different frame, maybe a cut half a second later, maybe a different camera movement pace. By the way, it’s incredible how much such small adjustments can change.At this stage of the work, another dose of dopamine kicked in. Smaller than before, but still very motivating. "Not only do I already have the initial version of the scene, but now everything is starting to move! I'm moving at lightning speed, and nothing can stop me!"Very nice, but now get to work.It's probably a confrontation with the later stages of the project, after that first wave of naive joy, what we call short-lived enthusiasm. You start something with joy, see the initial huge progress, and it seems like the work is moving at an express pace, with the finish line already appearing on the horizon. However, reality doesn't really care about it and sooner or later it comes to verify the progress. The first such verification hit when I had to model the final version of the first segment of the scene. What had previously been just a simple, rounded cube was now waiting for its true realization. How dense should the details be, and where should they go? What should be fully modeled, and what could simply be painted onto a texture? What resolutions should the textures have? What colors should they be? You can know all the fundamentals of designing such things, and still get lost. For example, I realized that despite having reference material in the form of the original animation, I could really use designs with more detailed elements. Luckily, Insomniac still makes games from this series today, so I managed to find some modern assets from the latest Ratchet & Clank on PS5 on ArtStation. They worked perfectly as references. My confidence, however, was in a bit of a decline.As it quickly turned out, creating just one element of the scene in the final version takes much longer than the total time I worked on the entire project from the beginning up to that point. My example platform first had to be prepared in high-poly - a version with the highest possible level of detail. Then, the same model had to be created again, this time in low-poly - a geometrically simplified version for optimization. After that, details had to be baked from one version onto the other, textures painted, plus several other technical steps. Then the same thing had to be repeated for each element of the environment that was waiting to be prepared. Glass panels, inner platform segments, the outer ring with lamps, speakers, screens, scaffolding, cages, and so on. All of this only to realize later that while individual segments might look fine on their own, they didn’t necessarily work well together as a whole. For this reason, I went back from time to time to slightly modify already existing parts of the scene. And time kept passing.And he saw that it wasn't very good...Paradoxically, what had helped me earlier was now becoming a source of complexes for me. Using the stunning visuals from Rift Apart—the latest Ratchet & Clank installment - was incredibly helpful at first. Over time, however, as more and more of my own objects filled the scene, a wave of self-doubt started creeping in. "This doesn’t even look half as good as the work of the artists at Insomniac Games." "Why am I even doing this?"For a long time, lighting had been bothering me. Light is one of the most important aspects of 3D graphics - period. In fact, the perceived quality of the final image depends more on the lighting than on the quality of the objects being illuminated. Seriously. Lighting is incredibly important. That said, I kept failing with every new attempt at lighting the scene. Too dark, too bright, the color palette came out ugly, sometimes it felt too flat, other times something got overexposed. I can't count how many times I postponed dealing with it, discouraged by yet another failure. To be honest, I’m still not satisfied with the final lighting. I simply had to draw the line somewhere and move on.But it wasn’t all bad. At this stage of the project, I had already modeled almost all the scene elements - that’s something. A certain milestone. Mrs. Dopamine came for a visit again. A small dose of motivation to push further forward. After all, it would be silly to abandon the project now, with so much work behind me, right?Wait, what was I talking about?You're talking so much nonsense here, Mr. Author, and this was supposed to be about animnantion. I'm asking where is aminantion! So where is it? This amimatnion!A fair question. Maybe a little clumsy, but fair. So, where’s the animation? Honestly - still in its early stages. Normally, a project like this would involve multiple people working in parallel. That way, while the 3D artist is modeling the scene, the animator can already start working on the animation on simplified models, because he doesn't need detailed ones. That alone would cut the work time in half (!). But since I was the only one working on the whole thing, I had a new, another big task in the form of animation. "Well, now it’s finally all downhill from here", I thought. "Hehe, the fool hasn't learned anything", fate probably thought. I think it did. I don’t know - I didn’t ask.The first attempt came together pretty quickly. I wouldn’t say it went well, but at least it went fast. That’s something. I decided to start working on Courtney first. After all, she’s the main character of the entire clip, and the focus is primarily on her. The first shot - when she steps down from the stage. The second - when she… um… dances with her back to the camera. Then the third, fourth, and in one go I did the entire first half. Alright, time to share my progress online and get some feedback from people.And what were the opinions? "Overall, it’s not bad, but…"—which, in short, means "Dude, it’s actually really bad, but I’m trying to be nice." And yeah, it really was bad. The first version of the animation was slow, uninteresting, and just plain awful. It hurts a bit when you realize you’ve done a poor job. But pride doesn’t finish projects. What didn’t work had to go in the trash, and I had to start animating all over again from scratch. This time, slightly humbled by the mediocre feedback, I decided to take things more seriously. What did that mean? Well, when working on animation, it’s always a good idea to record some references first. How do you do that? Very simple: grab your phone and place it on a shelf or desk with recording turned on goof around carefully play in front of the camera the scene you are going to animate live with the hope that no one will ever, ever see this recording ._. With this type of reference, things went a bit better. The second attempt was still far from expectations, but at least it was an improvement over the first one. Thanks to the references, I managed to improve the pace, but also the overall choreography of the shots. Then there was the third attempt, and I stopped there.It's still not over yet?!Although Courtney’s animation was more or less done, the dancers were still waiting. And unfortunately, I don’t mean beautiful women dancing in celebration of the project’s completion. Not at all! There was still a long way to go, and to be honest, at this point, I felt like I had absolutely no idea what I was doing. And it was starting to show - this time in the dancers’ animation. Maybe it wasn’t as bad as Courtney’s first animation attempt, but it definitely wasn’t good. And no amount of goofing around in front of my phone for reference could fix it.How to put it... a bearded dude with all his appeal is not the best reference point for graceful, agile dancing ladies on stage. So, instead of wasting even more time, I decided to spend some of it searching for dance references online. And wow, maybe I just don't know what I'm talking about, but I really didn't know that there were so many terrible dances on the internet. I swear, a good chunk of what I watched were girls either rolling around the floor aimlessly or shaking chaotically, as if they had no idea what they were trying to achieve. And these are things proudly published by dance schools (!). At some point, however, I came across Asian girl bands…Alright, alright - let me explain. Well, you see… oh c’mon, I'm not going to explain myself! What can I say - you can like them or not, but it’s a completely different level. „Western” dance groups aren’t even close to the discipline and precision you see in Asian ones. These dances finally looked like something someone had really thought through. They just made sense! So, finally I had my reference.At this point no one reads this anyway.Since I’ve already scared off whoever I might have scared with my mention of girl bands, I can now openly admit that I really did use them as references. Am I embarrassed about it? Duuude, I was embarrassed when I tried dancing in front of my phone and then watched that pitiful sight. Now, I was just happy to finally push the work forward. And indeed, with the new references, things went a bit smoother. One attempt, then another - I honestly don’t remember how many there were. What I do remember is that this stage of the work was a bit of a relief. The dancers didn't require as much attention as Courtney, I had better references for them, but the fact that the project as a whole was already in "some" state also helped a lot. Maybe it’s silly, but working on something where you can actually see the results, rather than just imagining them vaguely, makes a big difference.Thanks to this, my mood was a bit more positive. The worst was already done, so confidence peeked its sad little eyes out from the depths of despair (pff, look at him, suddenly trying to be a poet). Once I managed to finish animating the dancers after a few attempts, the last thing still waiting for me was the lighting - the task I had been postponing to "hopefully never."Dude, but I don't see anything here.This moment had to come eventually, and now - it had arrived. My eternal nemesis. My weakness. The thing that had haunted me from the very start of this project. Ladies and gentlemen - lighting. (Here, we imagine the sound of a terrified and shocked crowd.)Somewhere in the previous stages of my work, when things weren’t going well, I experimented with lighting. As I’ve mentioned before, I was never satisfied with this aspect of my work. But at least I had some general direction now. A direction that didn’t make me feel "Yes, this is it," but also didn’t scream "Dude, this is a total disaster." So I had a direction that was neutral. Neutral - meaning acceptable. I knew I probably wouldn’t come up with anything better. So, I stuck with relatively simple, uniform blue lighting for the scene. That was supposed to be the background - my foundation for further work. The monotonous, solid color was interrupted by pink screens (with slight variations) and occasional yellow lamps. Nothing spectacular, but I assumed that if it doesn't hurt the eyes, then it's already good enough.While this lighting more or less worked for the scene as a whole, it didn’t quite handle characters in different, dynamic shots. That’s why the final, secret ingredient was character lighting. I experimented with it quite a bit, but in the end, I stuck with the standard three-point lighting setup. The main lamp - to pull the character out of the background, the fill lamp - to eliminate unwanted shadows, and in some shots a contour lamp - to add a bit of drama by lighting the character from behind. This simple set of lights just needed to be placed properly in each shot and voilà - the lighting was done.We go to sleep and the work will do itself.Wait, it’s still not over? How much more…The last thing left to do was rendering. Everything done so far existed only as a 3D project file I had been working on. Now, it was time to turn the fruits of months of effort into a video file. But that’s not as simple as it seems. I mean - it is. But also, it isn’t. It both is and isn’t.On one hand, rendering is simple, since it’s basically a process handled almost entirely by the computer rather than the person. On the other hand, rendering isn’t simple - because of that almost part. It can be a real pain. Theoretically, all that’s left to do is press the "render" button and sit back with a beer in hand, admiring the frames being drawn on the screen, accompanied by the gentle hum of the PC fans under the desk. And, for the most part, that’s true (though I wouldn’t recommend keeping beer too close to the keyboard). The problem, however, is that rendering needs supervision. Especially when the whole process is set to take four months (!). Seriously. That’s a lot of beer cans, but also plenty of time for things to go terribly wrong - and something almost certainly will. Maybe the software closes unexpectedly. Maybe the graphics card runs out of memory. Maybe an unforeseen error pops up. Or, maybe - after 120 rendered frames - the brilliant author of this text realizes that he forgot to enable or hide an object in the first shot (what a silly guy, pff). Since I needed my computer during the day for work, rendering took place at night. And if something went wrong two hours in, well… in the morning, I was greeted by a beautiful error report that had been sitting there for six hours. No problem, that's cool. At least after such a night the electricity bill was unnecessarily a bit higher…Sooner or later, it was finally done. I had all the frames rendered on my disk. In their raw, still-unprocessed state, the entire thing took up (and still does, I need a new disk ._. ) a staggering 1.2TB. Then, I pieced everything together. A bit of post-processing, a pinch of color correction, a handful of final tweaks. And at last - my exhausted eyes beheld it. A file with the friendly name "0100-1800.mp4." Or, in other words - the finished animation. The result of months of my work. I hope it was worth it.Get up, I think he finally finished yapping!Meow, meow, meow - you've talked so much here, man, and honestly, what’s the point of it all? As I eloquently pointed out at the beginning - I don’t know. I don’t know why I wrote this text about my project. If someone actually read it - great, that makes me really happy. Greetings to all three people who did it. I hope it was worth spending some time in my little world of 3D graphics. I know this project means much more to me than it ever will to anyone reading this. And that’s fine. But if reading this wasn’t a waste of time for you, dear reader - that’s something. And with that positive note, I’ll wrap this up. Lights off, time for bed.Until next time!

Now that he’s out of the White House, where does Joe Biden live? The 46th US president has had a lifelong love of real estate, though he hasn’t always had the financial means to express it. “Even as a kid in high school I’d been seduced by real estate,” Biden wrote in his 2007 biography, Promises to Keep. His “idea of Saturday fun” when he was married to his first wife, Neilia, was to “drive around the Wilmington area scouting open houses, houses for sale, land where we could build,” he added. Reportedly, the former Delaware senator even carried issues of Architectural Digest on his daily commute. At one point, he was struggling to pay off three mortgages and a loan from his father-in-law to indulge his house habit. Eventually, Biden slowed down in favor of stability. Since the late 1990s, the Biden family has kept a Wilmington, Delaware, property as their main house. After his vice presidency, Biden was able to make more money through speeches and book deals, in turn allowing him to invest in a long-coveted vacation home.Read on to discover more about President Biden’s real estate portfolio.DuPont mansionIn 1974, the Scranton, Pennsylvania, native paid $185,000 for a rundown 1930s mansion in Wilmington, Delaware, once owned by the prominent DuPont family. Three years later, he married Jill Biden (née Jacobs), and she joined him at the five-bedroom, three-bathroom home, which he nicknamed the Station. The manse served as campaign headquarters during Biden’s 1988 presidential run. The 10,000-square-foot Colonial needed extensive repair, so he sunk ample time and money into renovations. “Whatever he gets, the house eats for breakfast. That house loves cash,” wrote journalist Richard Ben Cramer in his book about the 1988 presidential race, What It Takes.Biden sold the two-acre property for $1.2 million in 1996.“The Lake House”The Bidens bought a four-acre Wilmington plot for $350,000 shortly after offloading their DuPont mansion, then custom-built a 6,850-square-foot manse on the property overlooking a man-made lake built by the DuPonts. According to Delaware Online, Biden designed the home himself. The Colonial-style house, which was completed in 1998, has three bedrooms and four and a half bathrooms and is located in the coveted Greenville neighborhood, also known as Château Country for its concentration of stately Colonials. Reportedly, the estate also hosts a cottage that Biden rented out to the Secret Service for $2,200 a month while he was vice president. Aerial photos reveal that the dwelling, which the Bidens call the Lake House, also boasts a spacious backyard pool.This is still the family’s primary residence.1 Observatory CircleFormer Vice President Kamala Harris and former German Chancellor Angela Merkel outside of the Vice President’s residence in 2021.Photo: Samuel Corum/CNP/Bloomberg via Getty Images