Cases Corsair Remembered How to Make a Case: Frame 4000D RS ARGB ReviewMay 13, 2025Last Updated: 2025-05-13We analyze Corsair’s FRAME 4000D’s design, specs, build quality, and thermalsThe HighlightsCorsair’s new FRAME 4000D case reprises concepts from the company’s 4000D case but completely overhauls its designThe motherboard tray, power supply shroud, and other components are modular and intended to be replaceable with 3D printable componentsThe FRAME 4000D case puts Corsair on a much better path than its other recent casesOriginal MSRP: $95-$110Release Date: January 7, 2025Table of ContentsAutoTOC Grab a GN15 Large Anti-Static Modmat to celebrate our 15th Anniversary and for a high-quality PC building work surface. The Modmat features useful PC building diagrams and is anti-static conductive. Purchases directly fund our work! (or consider a direct donation or a Patreon contribution!)IntroToday we’re reviewing the Corsair Frame 4000D and the case is modular in some interesting ways. For instance, the motherboard tray and power supply shroud/tray can come out. And that’s what gets us to the name “frame.” It’s a frame of a case and it can be reconfigured.It’s actually pretty well built. There’s some really good build quality to the case. The fan rail system at the front comes off with a pull and has what Corsair calls “3D Y-shaped patterns.” While the front panel looks flimsy, it’s actually strong due to its depth. The steel stamping for the cable management shroud area is also relatively high build quality.The reason we’re excited about this case is because Corsair lately has really sucked with some of its cases, but the Frame 4000D doesn’t and that’s encouraging.Editor's note: This was originally published on May 8, 2025 as a video. This content has been adapted to written format for this article and is unchanged from the original publication.CreditsTest Lead, Host, WritingSteve BurkeTesting, WritingPatrick LathanVideo Editing, CameraMike GaglioneCameraTim PhetdaraWriting, Web EditingJimmy ThangIn addition to the case’s modularity, it also has trademarks:The new InfiniRail(TM) fan mount is joined by Corsair’s 3D Y-pattern airflow pattern, and better yet, Corsair says this case is “50 Series Ready,” which is impressive, because not even NVIDIA was 50-series ready.This is the Corsair Frame 4000D, and overall, we like it. It’s an interesting case. The case is the successor to Corsair's long-lived and highly popular 4000D (and its 4000D Airflow and 4000X variants) which we first reviewed back in 2020 and even liked. In spite of the name, the Frame 4000D is an entirely new design, which means we're going to have to say "Frame 4000D" every single time we mention it and can’t shorten it to 4000D as that’s a different case. We suspect part of the reason for the similarity is to enable the classic reuse-the-old-Newegg-listing trick: those 1,000+ five-egg reviews are almost all for the original 4000D, not the Frame 4000D, but it gets to share them on the same listing. Corsair also occasionally refers to the "Frame 4000 Series" sans-D, so we may see a Frame 4000X at some point.The case ranges from $95 without fans to the $110 RS ARGB with 3x 120 ARGB fans. It’s targeting the modern budget range. There's a $100 middle step as well with fans but no ARGB. Based on discussion with Corsair, it sounds likely that the MSRPs will increase due to tariffs, but we don't have hard numbers for what those increased prices would be.The Frame 4000D is interesting because Corsair has gone all out with the gigantic holes in the front panel: functionally, they're close to having no front panel at all. Corsair Frame 4000D OverviewThe differentiating factor for the Frame 4000D is that it's intended to be modular, with users able to buy or 3D print alternate components. This is reiterated constantly in Corsair's marketing materials, including public blog posts: "FRAME is all about customization and we have some exciting things on the horizon. In addition to parts that will be available for direct purchase, We have modelled some blanks of the PSU shroud and motherboard tray, so you can download [...] these from Printables and customize them however you like."Corsair Frame 4000D Specs4000D Series (2019)FRAME 4000D SeriesDimensions (mm):466 x 230 x 453 mm487 x 239 x 486 mmMaterial:Steel, Tempered GlassSteel, Tempered GlassIncluded Fans:2x SP120 or 3x AF120 RGB ELITENone or 3x RS120 or 3x RS120 ARGBIncluded Controllers:NoneNoneFan Compatibility:Front: 3x 120mm, 2x 140mmTop: 3x 120mm, 2x 140mmSide: NoneBottom: NoneRear: 1x 120mmFront: 3x 120mm, 2x 140mm, 2x 200mmTop: 3x 120mm, 2x 140mm, 2x 160mmSide: 3x 120mm, 2x 140mmBottom: 2x 120mmRear: 1x 120mm, 1x 140mmRadiator Compatibility:Front: 360mm, 280mm, 240mmTop: 280mm, 240mmSide: NoneBottom: NoneRear: 120mmFront: 360mm, 280mm, 240mmTop: 360mm, 280mm, 240mmSide: 360mm, 280mm, 240mmBottom: NoneRear: 120mmExpansion Slots (Main):7 Horizontal (2 Vertical)7 Horizontal (3 Vertical)Motherboard (Main):Mini-ITX, Micro-ATX, ATX, E-ATXMini-ITX, Micro-ATX, ATX, E-ATXStorage:2x 2.5" SSDs2x 3.5" HDDs4x 2.5" SSDs2x 3.5" HDDsClearances:PSU: 220mmCPU Cooler: 170mmGPU: 360mm (335mm w/ fan)PSU: 220mmCPU Cooler: 170mmGPU: 430 mm (405 mm w/ fan)Dust Filters:Front, Top, PSU, SideFront, PSU, SideFront I/O:USB 3.2 Gen1 Type-A (x1)USB 3.2 Gen2 Type-C (x1)Headphone/Microphone (x1)PowerResetUSB 3.2 Gen1 Type-A (x2)USB 3.2 Gen2 Type-C (x1)Headphone/Microphone (x1)PowerResetPower Supply:ATX (Not Included)ATX (Not Included)Warranty:2 Year2 YearThe BuildThe rotatable vertical GPU mount is one of three major features that genuinely live up to Corsair's promise of modularity, the other two being the PSU shroud and the motherboard tray (which can be removed with four screws). This is vital: being able to eventually buy or print individual components is nice, but it doesn't have anything to do with the case's design. With these three features, we're comfortable saying that the Frame 4000D is more modular than a typical case. It's also possible to swap the glass and mesh side panels to either side of the case, but this won't be useful until Corsair sells side panels individually or introduces more case SKUs.The vertical GPU mount is surprisingly quick to set up: by loosening two thumbscrews around the expansion slot covers, the entire set of seven slots can be removed and rotated 90 degrees. A riser cable can then be installed and supported by two standoffs installed on the surface of the PSU shroud. Using the built-in mounting holes, a maximum of three vertical slots are usable. Obviously Corsair intends for you to purchase a Corsair riser cable, but if you want to use your own, the holes are spaced 122mm apart.The Frame 4000D has old school pop-out fill and drain ports at the top and bottom of the chassis, which we also really appreciate. Open-loop compatibility fits well with the modular philosophy that Corsair is pushing.So far, these things are good. If we’re picking antiquated hills to die on, paper manuals will be one of them: We'd like to see a paper manual included with the case in addition to the version on Corsair's blog. The benefit of the online guide is that it can be updated; for example, the online manual now explains what the point of the strip of mylar in the accessory kit is, whereas we had to email Corsair (it covers the reverse-connector holes in the motherboard tray when they're unused). A paper manual can’t be updated, which is a downside as much as it is an upside: The upside is that a company can’t gaslight a customer if something proves incompatible and is retconned. It’s also just more convenient to build a computer with a physical manual.But then again, maybe Corsair doesn’t deserve paper manuals since its own digital manual incorrectly lists the screw count and steps for removing the power supply shroud. With a digital one, they’ll be able to make as many mistakes as they want and the only people who will know about them is everyone who read this article. For the record, it’s 3 screws, not 2. And the diagram is also incorrect.There's another, larger piece of black mylar in the kit, but the manual devotes less explanation to this one, simply saying that if you "don’t want your cables visible through the bottom mesh quarter-panel, swap the translucent insert for the color-matched blank-out insert included in the Accessory Box." The case also has a Lian Li-ish strip of mesh ventilation below the glass side panel and it ships with the translucent insert. If you install fans in the two 120mm slots on top of the PSU shroud, we'd recommend getting rid of the inserts entirely.The Frame 4000D uses the so-called "InfiniRail" system. At the front of the case, there are two metal rails that hold up to 12 plastic clips. The rails slide in and out for 120mm or 140mm mounting, and the clips slide up and down the whole length of the rails. Technically, the rails can fit 200mm fans at their maximum width, but only by forgoing the clips and mounting directly to the rails. At the top of the case, Corsair took the simpler approach of using a single sliding rail and no clips. The system is more complicated than it strictly needs to be, but our only real complaint is that there are exactly twelve clips included with the base model Frame 4000D, with no spares in case one breaks.The only significant fit-and-finish issue with our review sample was that one of the plastic pieces of the ball snap fasteners had popped out of the chassis in shipping. We were able to find it and put it back, but just like the InfiniRail clips, Corsair didn't provide any extras (like some other manufacturers do). As for build quality, construction of the Frame 4000D feels more delicate than the older 4000 series cases. This is partly a tradeoff for the more breathable mesh pattern, flexible fan mounting, and removable components. All of these things get exchanged for rigidity. Corsair apparently intends to sell Frame 4000D components individually as well, but we don't see any on the store as of this writing.On the positives for build quality, the elaborate stamped 3D pattern on the front panel is much sturdier than it looks. We'll have thermal results later in this review, but the Frame 4000D's front panel appears more open than the 4000D Airflow's (watch our review) while retaining most of its rigidity.There's a cable cover at the front of the case that can be installed in one of two positions, but as is frequently the case, we were only able to use it in the forward position because of the ATX power cable. The other position is better suited to back connect motherboards, of which ASUS BTF, MSI Project Zero, and Gigabyte Project Stealth have been validated by Corsair to work. The cable cover is extremely easy to remove, requiring only a single screw, and it's equally easy to slot the side fan mount from the accessory kit in its place. The side fan mount is another feature that will become more interesting if Corsair introduces a Frame 4000X or other variant with a sealed front panel.Cable management is good with tons of velcro straps behind the motherboard tray, but the best route for the 24-pin cable is along the cable bar, which isn't an option if the side fan mount is installed instead. Front I/O cables are neatly sleeved and the whole I/O unit pops off with two screws, so it would have been nice to allow alternate mounting locations (like in the O11D EVO), especially since having the I/O at the bottom of the case has become unusual these days. Even still, we like the modularity of it. They are delivering on that. As for storage, drives can be mounted on two plates, one behind the motherboard tray and one under the PSU shroud. Each plate can mount either two 2.5" drives, one 3.5" drive, or one "iCUE LINK System Hub controller." We'll be interested to see whether Corsair comes up with a modular solution to fit more drives, but for now, that's it.The default location of the GPU anti-sag arm is on the cable cover, too far forward to benefit anything except the longest GPUs, especially if the cable cover is shifted to its forward position.It can be moved a step further back by using the "mini mount" in the accessory kit, but there's no way to do any finer adjustment, and you should look for a different solution if you really think your GPU needs that support. Appearances are subjective. For some, a possible downside of the Frame 4000D's increased modularity is that it doesn't have the clean, uninterrupted lines of the original 4000D cases. We'll leave it at that; you can form your own opinion.Corsair Frame 4000D Thermals Visit our Patreon page to contribute a few dollars toward this website's operation (or consider a direct donation or buying something from our GN Store!) Additionally, when you purchase through links to retailers on our site, we may earn a small affiliate commission.The $95 base model Frame 4000D that we were sent for review didn't include any stock fans, so Corsair sent along a pack of three RS120 ARGB fans. We used these fans to match the stock configuration of the $110 Frame 4000D RS ARGB, which is how the case is marked on our charts. According to Corsair's spec sheets, the ARGB fans have the same specs as regular RS120s, so these results are also representative of the $100 Frame 4000D RS (non-ARGB)'s performance.CPU Full Load Thermals - Noise-NormalizedWe’ll start with noise-normalized thermals when using our hemi-anechoic chamber to establish the noise levels.Under full load with the case fans adjusted to hit our 27 dBA noise normalization target, CPU temperature in the Frame 4000D averaged 43 degrees Celsius above ambient, or 47 degrees on just the P-cores. That's a significant improvement over the original 4000D Airflow's average of 49 degrees (53 on the P-Cores), but that's to be expected since the older case has only two fans and splits them between the front and rear of the case.Relative to the rest of the chart, the Frame 4000D performed fairly well here. The Phanteks XT Pro Ultra is comparable in price and design, but it split the difference between the Corsair cases, with the Frame 4000D still a couple degrees cooler. The Lancool 207 (read our review) remains the budget case to beat with averages of 41 degrees across all cores and 45 on the P-Cores, while the NZXT H5 Flow 2024 (watch our review) and Phanteks G400A (watch our review) performed similarly well. The G400A remains difficult to find in the US.GPU Full Load Thermals - Noise-NormalizedMoving to GPU thermals under full load in the same noise normalized test, the Frame 4000D averaged 45 degrees above ambient, with 49 degrees for the memory and 59 for the unshown GPU hotspot. That's another victory over the original 4000D Airflow, which averaged 49 degrees for the GPU temperature, but again that's with two fans to the Frame 4000D's three.The G400A effectively tied the Frame 4000D in this test, while the H5 and XT Pro UItra didn't do as well. The XT Pro UItra averaged 47 degrees for the GPU, a couple degrees warmer than the Frame 4000D. The Lancool 207 remains near the top of the chart for cooling. To learn more about that one, you can check out our Best Cases round-up from last year or our original review for more info on that case's pros and cons.CPU Full Load Thermals - Full SpeedAt full speed, the Frame 4000D's CPU thermal performance continues to scale fairly well for its price, while the 4000D Airflow falls further behind. The Frame 4000D's noise levels and performance were essentially tied with the Phanteks XT Pro Ultra here, with the Frame 4000D averaging 40 degrees above ambient, or 43 on the P-Cores. The G400A performed similarly as well, but with a lower 37dBA noise level to the Frame 4000D's 43dBA, while the Lancool 207 remains an outlier with both lower temperatures and lower noise levels, making it a lot better here than really most of these other cases.CPU Full Load Thermals - Standardized FansThe standardized fan test has always had limited usefulness, but this is one of the places it’s useful.It’ll help us evaluate the Frame 4000D's unusual front panel by comparing it against other cases with an identical set of fans and placements including the 4000D. Average all-core CPU temperature was 40 degrees above ambient and the P-Cores were 44 degrees. That's reasonably well-matched to established mesh-fronted cases like Fractal's Meshify 2 Compact (watch our review) and North XL (read our review), and extremely close to SilverStone's recent FARA 514X (read our review). The Frame 4000D is also significantly better than the 4000D Airflow when normalizing to the same fans, reinforcing Corsair’s improvements to the chassis design itself.No matter how open the front panel is, though, the fans still have to pull air through Corsair's filter as well, so it makes some sense that some single-layer mesh designs like the Lancool 207, Pop Air RGB (watch our review), and Flux Pro (read our review) perform slightly better here, although the G400A also did fairly well with two layers.GPU Full Load Thermals - Standardized FansIn the same test, GPU temperature in the Frame 4000D averaged 43 degrees above ambient, tying it with most of the cases we just mentioned (Lancool 207, Pop Air RGB, and Flux Pro), while the XT Pro Ultra did unusually poorly with an average of 50 degrees.As a side note, GPU thermals with the 4000D Airflow were abysmal in this particular test relative to the rest of the chart. That differs from what we saw 5 years ago using our old test hardware, which demonstrates the value of retesting these cases now that we've switched to new components and a flow-through GPU design.VRM & RAM Full Load Thermals - Noise NormalizedAs a final note, the VR VCC sensor averaged 30 degrees above ambient in the noise-normalized test with the Frame 4000D. That puts it at the cooler end of what we typically see from cases, with only a couple of outliers like the Lancool 207 and Flux Pro dropping to the 27 degree mark. The same goes for the SPD Hub average of 24 degrees above ambient (which is the RAM temperature), with relatively few cases on the chart dropping below 22 degrees. Both these sensors indicate normal internal case temperatures.Corsair Frame 4000D Conclusion Our fully custom 3D Emblem Glasses celebrate our 15th Anniversary! We hand-assemble these on the East Coast in the US with a metal badge, strong adhesive, and high-quality pint glass. They pair excellently with our 3D 'Debug' Drink Coasters. Purchases keep us ad-free and directly support our consumer-focused reviews!Of the existing Frame 4000D SKUs, the Frame 4000D RS at $100 is a decent deal, especially given how tightly packed the SKUs’ prices are. Going from 0 fans to 3 fans is worth $5, and going from no RGB to ARGB may or may not be worth another $10. We need Corsair to follow through with more modular case components (like the alternate front panels it showed in the trailer) to make it really interesting, or at the absolute least offer some more printable files. The Lancool 207 remains king in terms of min-maxed price-to-performance, but the Frame 4000D is a reasonable alternative to budget Phanteks cases like the G400A and XT Pro Ultra, or potentially some Montech cases (which we'll be adding more of to the charts shortly). As a successor to the existing 4000 series, we like the Frame 4000D functionally. Some of the finer attention to detail elements like the geometric patterns and yellow accents remain, just with a more complex appearance than previously.Corsair has mentioned that it may have to increase prices, but we don’t yet have final details on what that amount might be on this case. We don't encourage panic buying, but if you've decided that the Frame 4000D is the case for you and it's currently in stock at MSRP, we wouldn't recommend waiting around for a discount. Overall, this is a job well done by Corsair. It is a much better return to form as compared to something like the 6500D (read our review) from last year. Source: https://gamersnexus.net/cases/corsair-remembered-how-make-case-frame-4000d-rs-argb-review #corsair #remembered #how #make #case #frame #4000d #argb #review

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More A new report released by the publicly traded market research and intelligence firm SimilarWeb—covering global web traffic patterns for AI-related platforms for 12 weeks through May 9, 2025—offers a helpful look for enterprises and interested users into the current landscape of generative AI usage online. Using proprietary analytics based on site visits, the report tracks trends across sectors including general-purpose AI tools, coding assistants, content generators, and more. It also maps the downstream disruption of traditional sectors like education, search, and digital freelancing. Why this report matters to enterprise technical decision-makers For enterprise AI leaders—particularly those responsible for model deployment, orchestration, or data integration—this report is more than just a consumer trend snapshot. It’s a map of user familiarity and expectation. Tools like ChatGPT, Claude, DeepSeek, and Grok are what employees are already using at home or experimenting with informally at work. If your internal AI apps or copilots don’t match the baseline experience offered by these tools, you risk user rejection and poor adoption. Moreover, selecting the same models with high external traction for your internal orchestration pipelines or RAG deployments can cut onboarding time and training costs, since your team will already know the interface and behavior. In short, aligning your enterprise stack with the tools dominating usage charts may be the fastest path to trust, usability, and value delivery. Here are five key findings from the report: Developer-focused AI tools are surging in adoption, with traffic to the category up 75% over the past 12 weeks. That growth includes Lovable, which exploded with a jaw-dropping +17,600% spike, and Cursor, which grew steadily month over month. As AI becomes more deeply embedded in IDEs and continuous integration workflows, usage patterns suggest these tools are no longer just experimental—they’re now seen as essential infrastructure for modern software teams. It also helps explain why OpenAI reportedly tried to buy Cursor and has also reportedly finalized a deal to purchase rival AI coding platform Windsurf — as the usage of these tools explodes, OpenAI wants to get a larger cut of the action and revenue, especially since its models are often used as the engine behind-the-scenes for these very users. Two of the fastest-growing AI platforms of the year, Grok and DeepSeek, show just how quickly hype can turn into burnout. Grok traffic skyrocketed more than 1,000,000% in March—driven by its branding as an uncensored yet powerfully intelligent platform and Elon Musk association—before falling more than 5,200% by early May. DeepSeek saw a similar arc, peaking at +17,701% growth before crashing -41%. The takeaway: virality can’t replace retention, especially compared to AI leader OpenAI and also legacy tech brand Google. Once among the most accessible use cases for generative AI, writing and content tools are seeing user fatigue. Category traffic fell 11% overall, with platforms like Wordtune (-35%), Jasper (-19%), and Rytr (-23%) all trending downward. Only Originality.ai bucked the trend with steady traffic gains, likely due to its focus on AI detection rather than generation. This plateau suggests content saturation and possibly growing skepticism over quality or usefulness. Separately from the SimilarWeb report, I think that because chatbots already use a default text input/output field as their main user interface, even for multimodal interactions, the fact is that most users who would be interested in this capability will go to model providers like OpenAI’s website for ChatGPT (and mobile apps), or Google’s Gemini, or even Anthropic’s Claude, directly, rather than seek out a separate AI writing tool. Design-focused AI remains a mixed bag. While overall category usage dipped slightly (-6% over the 12-week window), some platforms made outsized gains. Getimg posted a massive +1,532% surge, while Artbreeder jumped +100%. Others, like Stable Diffusion and Looka, saw double-digit declines. The erratic pattern may reflect a crowded landscape of tools that offer similar functionality but compete on novelty or aesthetics. 5. AI is eating up legacy tech such as crowdsourcing and search Traditional digital services are showing slow but steady declines across the board—likely the result of AI substitution. Freelance platforms such as Fiverr (-17%) and Upwork (-19%) are losing traffic, possibly as users turn to AI tools for tasks like design, writing, and code. Search engines such as Yahoo (-12%) and Bing (-14%) continue a multi-quarter drop in visits, while consumer EdTech companies like Chegg (-62%) and CourseHero (-68%) are in free fall. The signs point to early-stage AI disruption beginning to erode the utility of some legacy platforms. It also offers a hint to enterprises that either leverage or create such services — the time may be coming to reduce dependency on them, either from a revenue generation, marketing, or overall business perspective. Daily insights on business use cases with VB Daily If you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from regulatory shifts to practical deployments, so you can share insights for maximum ROI. Read our Privacy Policy Thanks for subscribing. Check out more VB newsletters here. An error occured. Source: https://venturebeat.com/ai/similarwebs-new-ai-usage-report-reveals-5-surprising-findings-including-explosive-growth-in-coding-tools/ #similarwebs #new #usage #report #reveals #surprising #findings #including #explosive #growth #coding #tools

Senior Software Engineer II, Gameplay - AI/Combat - Unannounced Game | Irvine, CABlizzard EntertainmentIrvine California 92618 United States of America32 minutes agoApplyTeam Name:Job Title:Senior Software Engineer II, Gameplay - AI/Combat - Unannounced Game | Irvine, CARequisition ID:R025329Job Description:At Blizzard, we craft genre-defining games and legendary worlds for all to share. Through unparalleled creativity and storytelling, we create immersive universes and iconic characters that are beloved across platforms, borders, backgrounds, and generations - only made possible by building a work environment that nurtures the artistry of game development and unleashes the aspirations of our people.As a Senior AI Gameplay Engineer, you will collaborate with designers and artists to deliver best-in-class enemy combat experiences. You will support the development of combat AI, animation, and combat frameworks, as well as collaborate in driving improvements in the underlying engine technology. You will work collaboratively with the team to define and execute a technical vision that meets the project’s needs.This role is anticipated to be a hybrid work position, with some work on-site and some work-from-home. The potential home studio for this role is Irvine, CA.Responsibilities:Priorities can often change in a fast-paced environment like ours. Responsibilities will include, but are not limited to:Leveraging your experience and knowledge to deliver best-in-class player experiences, from end to end, within the constraints of game simulation.Empowering designers, artists, and other engineers to deliver the vision of the game and ensuring we delight and entertain our players.Collaborating with designers and engineers on gameplay teams to learn and understand their needs and offer technical solutions.Optimizing team productivity by supporting pipelines, tool development, and workflow improvements.Requirements:7+ years of experience in gameplay engineering or a combination of applicable training and experience.Experience developing using the Unreal engine (both C++ and Blueprint).Experience creating compelling combat at AAA-quality level, including combat, high fidelity enemy animation, VFX/SFX, UX etc.Knowledge & SkillsThe ability to understand and write production-grade C++ code.The capacity to make frequent, readable, well-tested code commits.The ability to communicate and articulate vision, ideas, learnings from others, and to give meaningful feedback.Excellent problem-solving and debugging skills.Experience analyzing game features and content for performance and optimizing a game to release quality.Experience working closely in multi-disciplinary teams with people of diverse mindsets.Mentorship mindset and excellent attitude to helping others level up their knowledge.Extra Points:Computer science, mathematics, physics, or related degrees.Experience developing and shipping AAA game titles for PC, Xbox, and PlayStation consoles.Experience working on 3D action multiplayer games.Experience with a large codebase.Your PlatformBest known for iconic video game universes including Warcraft®, Overwatch®, Diablo®, and StarCraft®, Blizzard Entertainment, Inc. (www.blizzard.com), a division of Activision Blizzard, which was acquired by Microsoft (NASDAQ: MSFT), is a premier developer and publisher of entertainment experiences. Blizzard Entertainment has created some of the industry’s most critically acclaimed and genre-defining games over the last 30 years, with a track record that includes multiple Game of the Year awards. Blizzard Entertainment engages tens of millions of players around the world with titles available on PC via Battle.net®, Xbox, PlayStation, Nintendo Switch, iOS, and Android.Our WorldActivision Blizzard, Inc., is one of the world's largest and most successful interactive entertainment companies and is at the intersection of media, technology and entertainment. We are home to some of the most beloved entertainment franchises including Call of Duty®, World of Warcraft®, Overwatch®, Diablo®, Candy Crush™ and Bubble Witch™. Our combined entertainment network delights hundreds of millions of monthly active users in 196 countries, making us the largest gaming network on the planet!Our ability to build immersive and innovative worlds is only enhanced by diverse teams working in an inclusive environment. We aspire to have a culture where everyone can thrive in order to connect and engage the world through epic entertainment. We provide a suite of benefits that promote physical, emotional and financial well-being for ‘Every World’ - we’ve got our employees covered!The video game industry and therefore our business is fast-paced and will continue to evolve. As such, the duties and responsibilities of this role may be changed as directed by the Company at any time to promote and support our business and relationships with industry partners.We love hearing from anyone who is enthusiastic about changing the games industry. Not sure you meet all qualifications? Let us decide! Research shows that women and members of other under-represented groups tend to not apply to jobs when they think they may not meet every qualification, when, in fact, they often do! We are committed to creating a diverse and inclusive environment and strongly encourage you to apply.We are committed to working with and providing reasonable assistance to individuals with physical and mental disabilities. If you are a disabled individual requiring an accommodation to apply for an open position, please email your request to accommodationrequests@activisionblizzard.com. General employment questions cannot be accepted or processed here. Thank you for your interest.We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, gender identity, age, marital status, veteran status, or disability status, among other characteristics.RewardsWe provide a suite of benefits that promote physical, emotional and financial well-being for ‘Every World’ - we’ve got our employees covered! Subject to eligibility requirements, the Company offers comprehensive benefits including:Medical, dental, vision, health savings account or health reimbursement account, healthcare spending accounts, dependent care spending accounts, life and AD&D insurance, disability insurance;401(k) with Company match, tuition reimbursement, charitable donation matching;Paid holidays and vacation, paid sick time, floating holidays, compassion and bereavement leaves, parental leave;Mental health & wellbeing programs, fitness programs, free and discounted games, and a variety of other voluntary benefit programs like supplemental life & disability, legal service, ID protection, rental insurance, and others;If the Company requires that you move geographic locations for the job, then you may also be eligible for relocation assistance.Eligibility to participate in these benefits may vary for part time and temporary full-time employees and interns with the Company. You can learn more by visiting https://www.benefitsforeveryworld.com/ .In the U.S., the standard base pay range for this role is $121,500.00 - $224,880.00 Annual. These values reflect the expected base pay range of new hires across all U.S. locations. Ultimately, your specific range and offer will be based on several factors, including relevant experience, performance, and work location. Your Talent Professional can share this role’s range details for your local geography during the hiring process. In addition to a competitive base pay, employees in this role may be eligible for incentive compensation. Incentive compensation is not guaranteed. While we strive to provide competitive offers to successful candidates, new hire compensation is negotiable. Create Your Profile — Game companies can contact you with their relevant job openings. Apply Source: https://gamejobs.co/Senior-Software-Engineer-II-Gameplay-AI-Combat-Unannounced-Game-Irvine-CA-at-Blizzard-Entertainment-3848 #senior #software #engineer #gameplay #aicombat #unannounced #game #irvine #blizzard #entertainment

A wildfire rages in British Columbia, Canada, in June 2023Imago/Alamy Stock Photo 2023 smashed the record for the hottest year, but it might have been even hotter. The entire northern hemisphere would have been nearly 1°C warmer on average during its summer without the cooling effect of smoke from massive wildfires in Canada, a climate model suggests. The smoke may also have led to the driest August in India on record. “I think it’s really hard to comprehend how gigantic the fires were. It was insane,” says Iulian-Alin Rosu at the Technical University of Crete in Greece, who presented his team’s findings at a meeting of the European Geosciences Union in Vienna, Austria. Advertisement The emissions were around five or six times higher than those during any previously recorded wildfire season in Canada, estimates Rosu. The carbon dioxide from these fires is having an ongoing warming effect, but in 2023 this warming was outweighed by the cooling effect of smoke blocking sunlight. To estimate how much cooling the smoke caused, Rosu and his colleagues ran a series of climate model simulations with and without the Canadian wildfire emissions. The results suggest that between May and September, the smoke caused local cooling of as much as 5.4°C (9.7°F) in small parts of Canada, and that the northern hemisphere as a whole was 0.9°C (1.6°F) cooler. This may seem surprising given that parts of Canada saw record temperatures during that summer. But the heat records were mostly in western regions, says Rosu, whereas the smoke blew east and had the biggest cooling effect on that side of the country. Unmissable news about our planet delivered straight to your inbox every month. Sign up to newsletter The impacts weren’t limited to Canada. In the model, the wildfire emissions led to changes in winds over Asia that weakened the monsoon and led to less rainfall in India – and that is what happened in reality. “The precipitation anomaly that is seen in the data is really, really close to what we see in our model,” says Rosu. That is an indication that the model did a very good job, he says. However, the cooling effect didn’t last long. “When I looked at the data for November and December, there really wasn’t much of any effect left,” says Rosu. The record set in 2023 for hottest year didn’t last long either – 2024 turned out to be even hotter. Reference:EGU General Assembly 2025 DOI: 10.5194/egusphere-egu25-10833 Topics:climate Source: https://www.newscientist.com/article/2479881-massive-wildfires-in-canada-helped-keep-the-world-cooler-in-2023/?utm_campaign=RSS%7CNSNS&utm_source=NSNS&utm_medium=RSS&utm_content=home #massive #wildfires #canada #helped #keep #the #world #cooler

end-of-life planning Office apps on Windows 10 are no longer tied to its October 2025 end-of-support date Windows 10 will stop getting free security updates on October 14, 2025. Andrew Cunningham – May 13, 2025 11:24 am | 26 Microsoft 365 will encompass Teams, OneDrive, and the suite of productivity apps formerly known as Office. Credit: Microsoft Microsoft 365 will encompass Teams, OneDrive, and the suite of productivity apps formerly known as Office. Credit: Microsoft Story text Size Small Standard Large Width * Standard Wide Links Standard Orange * Subscribers only   Learn more For most users, Windows 10 will stop receiving security updates and other official support from Microsoft on October 14, 2025, about five months from today. Until recently, Microsoft had also said that users running the Microsoft Office apps on Windows 10 would also lose support on that date, whether they were using the continually updated Microsoft 365 versions of those apps or the buy-once-own-forever versions included in Office 2021 or Office 2024. Microsoft has recently tweaked this policy, however (as seen by The Verge). Now, Windows 10 users of the Microsoft 365 apps will still be eligible to receive software updates and support through October of 2028, "in the interest of maintaining your security while you upgrade to Windows 11." Microsoft is taking a similar approach to Windows Defender malware definitions, which will be offered to Windows 10 users "through at least October 2028." The policy is a change from a few months ago, when Microsoft insisted that Office apps running on Windows 10 would become officially unsupported on October 14. The perpetually licensed versions of Office will be supported in accordance with Microsoft's "Fixed Lifecycle Policy," which guarantees support and security updates for a fixed number of years after a software product's initial release. For Office 2021, this means Windows 10 users will get support through October of 2026; for Office 2024, this should extend to October of 2029. Some Microsoft support sites still list the old end-of-support dates for Office apps running on Windows 10. We've contacted Microsoft to see whether these sites will be updated to match the new dates. Microsoft is likely extending this Office support date and the Windows Defender definitions to help cover people who buy into the Extended Security Updates program for Windows 10, which will allow individuals and institutions to stay on Windows 10 past the official public end-of-support date. Businesses and other institutions will be able to buy between one and three extra years of security updates, with costs that steadily increase for each year of updates. Individuals will be able to buy a single extra year of updates for a flat $30 per PC. For people who aren't paying for extra Windows 10 updates, Microsoft has stayed firmly committed to both Windows 10's end-of-support date and Windows 11's minimum system requirements, which will prevent many active Windows 10 systems from upgrading to the newer operating system. Perhaps unsurprisingly, Microsoft's solution is that these people should buy a new PC entirely; the company declared 2025 "the year of the Windows 11 PC refresh" back in January. It's possible to install and run Windows 11 on older "unsupported" PCs, and the day-to-day experience is often indistinguishable from running the software on a "supported" PC. But there are additional hoops to jump through when you're installing and upgrading the OS that may keep most non-technical users from wanting to give it a try. Andrew Cunningham Senior Technology Reporter Andrew Cunningham Senior Technology Reporter Andrew is a Senior Technology Reporter at Ars Technica, with a focus on consumer tech including computer hardware and in-depth reviews of operating systems like Windows and macOS. Andrew lives in Philadelphia and co-hosts a weekly book podcast called Overdue. 26 Comments Source: https://arstechnica.com/gadgets/2025/05/microsoft-extends-office-support-on-windows-10-through-2028/ #office #apps #windows #are #longer #tied #its #october #endofsupport #date

Author(s): Marie Originally published on Towards AI. Hallucinations in Healthcare LLMs: Why They Happen and How to Prevent Them Building Trustworthy Healthcare LLM Systems — Part 1 Image generated by the author using ChatGPT TL;DR LLM hallucinations: AI-generated outputs that sound convincing but contain factual errors or fabricated information — posing serious safety risks in healthcare settings. Three main types of hallucinations: factual errors (recommending antibiotics for viral infections), fabrications (inventing non-existent studies or guidelines), misinterpretations (drawing incorrect conclusions from real data). Root causes of hallucinations: probabilistic generation, training that rewards fluency over factual accuracy, lack of real-time verification, and stale or biased data. Mitigation approaches: Retrieval-Augmented Generation (RAG), domain-specific fine-tuning, advanced prompting, guardrails. This series: build a hallucination-resistant pipeline for infectious disease knowledge, starting with a PubMed Central corpus. Hallucinations in medical LLMs aren’t just bugs — they’re safety risks. This series walks through how to ground healthcare language models in real evidence, starting with infectious diseases. Introduction LLMs (large language models) are changing how we interact with medical knowledge — summarizing research, answering clinical questions, even offering second opinions. But they still hallucinate — and in medicine that’s a safety risk, not a quirk. In medical domains, trust is non-negotiable. A hallucinated answer about infectious disease management (e.g., wrong antibiotic, incorrect diagnostic criteria) can directly impact patient safety, so grounding models in verifiable evidence is mandatory. That’s why this blog series exists. This four-part series will show you how to build a hallucination-resistant workflow, step-by-step: Part 1 (this post): what hallucinations are, why they happen and how to build a domain-specific corpus using open access medical literature Part 2: Turn that corpus into a RAG pipeline Part 3: Add hallucination detection metrics Part 4: Put it all together and build a transparent interface to show users the evidence behind the LLM’s responses What Are Hallucinations in LLMs? Hallucinations are model-generated outputs that sound correct and coherent, but are not factually correct. They sound convincing but are often false, unverifiable or entirely made up. Why They Matter in Healthcare These errors can have serious implications, especially in clinical settings where they might lead to improper treatment recommendations. The wrong recommendation in clinical settings could have life or death consequences, which is why it is critical to mitigate these hallucinations by building transparent, evidence-based systems. Image generated by the author using ChatGPT Main Types of Hallucinations 1. Factual Errors Factual errors happen when LLMs make incorrect claims about verifiable facts. Using our infectious disease example, recommending antibiotics for influenza would be a type of factual error. 2. Fabrications Fabrications involve LLMs inventing non-existent entities or information. In the context of healthcare, for example, these could be fictional research studies, medical guidelines that don’t exist or made-up technical concepts. 3. Misinterpretations Misinterpretation happens when LLMs take real information but misrepresents or mis-contextualizes it. For example, a model might reference a study that exists, but draws the wrong conclusions Why LLMs hallucinate Large language models hallucinate because they don’t truly understand facts like humans do simply predict what words should come next based on patterns they’ve observed in their training data. When these AI systems encounter unfamiliar topics or ambiguous questions, they don’t have the ability to say “I don’t know” and instead generate confident-sounding but potentially incorrect responses. This tendency stems from several factors: Their training prioritizes fluent, human-like text over factual caution They lack real-time access to verified information sources They have no inherent understanding of truth versus fiction. Conflicting information in training data can push the model to average contradictory sources. The problem is compounded by limitations in training data that may contain outdated, biased, or inaccurate information, as well as the fundamental auto-regressive nature of how these models generate text one piece at a time. How Can We Address Hallucinations? There are various methods to mitigate or detect hallucinations. Mitigation Strategies Fine-tuning with Domain-Specific Data: The main reason for hallucination lies in knowledge gaps in the model’s training data. This approach helps by introducing domain specific knowledge and can be very powerful to create models that understand better the specialized medical terminology or various nuances in clinical text. Retrieval-Augmented Generation (RAG): This method allows the integration of external knowledge sources by retrieving relevant information before generating the answer. It helps by grounding the model outputs in verified external sources instead of relying only on the model’s training data. This is the method we will be focusing on in this series Other noteworthy strategies: advanced prompting methods like Chain-of-Thoughts or Few-Shot Learning can help mitigate hallucinations by guiding the model’s answer in the right direction. Rules-based guardrails that screen outputs before they reach users add another safety layer. Hallucination Detection Source-attribution scoring: This method compares the LLM answer to the retrieved documents to detect how much of the answer is grounded in the source. Beyond identifying hallucinations, it also allows to highlight the source behind the LLM answer, which helps building trust and transparency. Semantic Entropy Measurement: This method measures uncertainty about the meaning of generated responses and has been developed specifically to address the risk of hallucinations in critical areas involving patient safety for example Consistency-Based Methods: This method involves a self-consistency check, where hallucinations can be detected by prompting the model multiple times with the same query and comparing the outputs for consistency. Some interesting open-access publications to go a bit further: If you’re interesting in reading recent research on this topic, here’s a few research papers worth reading: Code Walkthrough: Downloading Medical Research from PubMed Central To reduce hallucinations in healthcare LLMs, grounding them in reliable medical literature is critical. Let’s start by building a corpus from one of the best sources available: PubMed Central (PMC). This script helps you automate the retrieval of open-access medical papers, making it easy to bootstrap a dataset tailored to your task (e.g., infectious diseases). Here’s how it works: 1. Setup and Environment import requestsimport xml.etree.ElementTree as ETimport jsonimport os, re, timefrom dotenv import load_dotenvload_dotenv()api_key = os.getenv("NCBI_API_KEY")email = os.getenv("EMAIL")base_url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/" You’ll need to set your NCBI API key and email in a .env file. You can still call the NCBI API without an API key, but this unlocks higher rate limits and it is free 2. Search PMC Because we are interested in full texts to build our knowledge base, we should only download articles that are open access. To do so, we need to fetch the articles from PMC: # 1. Search PMCsearch_url = f"{base_url}esearch.fcgi"search_params = { "db": "pmc", "term": query, "retmax": max_results, "retmode": "json", "api_key": api_key, "email": email}print("Searching PMC...")search_resp = requests.get(search_url, params=search_params)search_resp.raise_for_status()ids = search_resp.json()["esearchresult"]["idlist"] This code queries PMC with your search terms (for example “infectious diseases”) and returns a list of document identifiers (PMCIDs). 3. Fetch and Parse Articles Now we can fetch the full texts using the PMCIDs: # 2. Batch fetchfetch_url = f"{base_url}efetch.fcgi"for i in range(0, len(ids), batch_size): = ids[i : i + batch_size] batch_ids fetch_params = { "db": "pmc", "id": ",".join(batch_ids), "retmode": "xml", "api_key": api_key, "email": email, } time.sleep(delay) r = requests.get(fetch_url, params=fetch_params) r.raise_for_status() Our response is an XML object, so the final step is to parse it and create a dictionary with the relevant sections: pmcid, title, abstract, full_text, publication_date, authors: root = ET.fromstring(r.content)for idx, article in enumerate(root.findall(".//article")): # Extract article details article_data = { "pmcid": f"PMC{batch_ids[idx]}", "title": "", "abstract": "", "full_text": "", "publication_date": "", "authors": [], } # Extract title title_elem = article.find(".//article-title") if title_elem is not None: article_data["title"] = "".join(title_elem.itertext()).strip() # Extract abstract abstract_parts = article.findall(".//abstract//p") if abstract_parts: article_data["abstract"] = " ".join( "".join(p.itertext()).strip() for p in abstract_parts ) # Extract publication date pub_date = article.find(".//pub-date") if pub_date is not None: year = pub_date.find("year") month = pub_date.find("month") day = pub_date.find("day") date_parts = [] if year is not None: date_parts.append(year.text) if month is not None: date_parts.append(month.text) if day is not None: date_parts.append(day.text) article_data["publication_date"] = "-".join(date_parts) # Extract authors author_elems = article.findall(".//contrib[@contrib-type='author']") for author_elem in author_elems: surname = author_elem.find(".//surname") given_names = author_elem.find(".//given-names") author = {} if surname is not None: author["surname"] = surname.text if given_names is not None: author["given_names"] = given_names.text if author: article_data["authors"].append(author) # Extract full text (combining all paragraphs) body = article.find(".//body") if body is not None: paragraphs = body.findall(".//p") article_data["full_text"] = " ".join( "".join(p.itertext()).strip() for p in paragraphs ) The data can then be saved into a jsonl that will be used in our next step — building our RAG system. Let’s be mindful of licensing restrictions: While open access literature allows anyone to access and read the content, it doesn’t mean the authors agreed to redistribution of their work. While this blog post and its content are intended for personal and educational use, if you decide to use this function to build a dataset that will be redistributed or commercialized, it is important to comply with the article’s license agreement. To do so, let’s define a function that will help us pull the license data from the downloaded article: def detect_cc_license(lic_elem): """ Inspect <license> … </license> for Creative Commons URLs or keywords and return a normalised string such as 'cc-by', 'cc-by-nc', 'cc0', or 'other'. """ if lic_elem is None: return "other" # 1) gather candidate strings: any ext-link href + full text candidates: list[str] = [] for link in lic_elem.findall(".//ext-link[@ext-link-type='uri']"): href = link.get("{http://www.w3.org/1999/xlink}href") or link.get("href") if href: candidates.append(href.lower()) candidates.append("".join(lic_elem.itertext()).lower()) # 2) search for CC patterns for text in candidates: if "creativecommons.org" not in text and "publicdomain" not in text: continue # order matters (most restrictive first) if re.search(r"by[-_]nc[-_]nd", text): return "cc-by-nc-nd" if re.search(r"by[-_]nc[-_]sa", text): return "cc-by-nc-sa" if re.search(r"by[-_]nc", text): return "cc-by-nc" if re.search(r"by[-_]sa", text): return "cc-by-sa" if "/by/" in text: return "cc-by" if "publicdomain/zero" in text or "cc0" in text or "public domain" in text: return "cc0" return "other" Here’s a short breakdown of what the licenses mean: Image uploaded by author Here’s the full function for PubMed download: def download_pmc_articles(query, ): max_results = 100, batch_size = 20, delay = 0.2, allowed_licenses = {"cc-by", "cc-by-sa", "cc0"}, out_file = "pmc_articles.jsonl" base_url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/" # 1. Search PMC search_url = f"{base_url}esearch.fcgi" search_params = { "db": "pmc", "term": query, "retmax": max_results, "retmode": "json", "api_key": api_key, "email": email } print("Searching PMC...") search_resp = requests.get(search_url, params=search_params) search_resp.raise_for_status() ids = search_resp.json()["esearchresult"]["idlist"] # 2. Batch fetch fetch_url = f"{base_url}efetch.fcgi" skipped, saved = 0, 0 with open(out_file, "w") as f: for i in range(0, len(ids), batch_size): batch_ids = ids[i:i+batch_size] fetch_params = { "db": "pmc", "id": ",".join(batch_ids), "retmode": "xml", "api_key": api_key, "email": email } time.sleep(delay) r = requests.get(fetch_url, params=fetch_params) r.raise_for_status() root = ET.fromstring(r.content) for idx, article in enumerate(root.findall(".//article")): # Check license license = detect_cc_license(article.find(".//license")) if license not in allowed_licenses: skipped += 1 continue # skip disallowed license # Extract article details article_data = { "pmcid": f"PMC{batch_ids[idx]}", "title": "", "abstract": "", "full_text": "", "publication_date": "", "authors": [] } # Extract title title_elem = article.find(".//article-title") if title_elem is not None: article_data["title"] = "".join(title_elem.itertext()).strip() # Extract abstract abstract_parts = article.findall(".//abstract//p") if abstract_parts: article_data["abstract"] = " ".join("".join(p.itertext()).strip() for p in abstract_parts) # Extract publication date pub_date = article.find(".//pub-date") if pub_date is not None: year = pub_date.find("year") month = pub_date.find("month") day = pub_date.find("day") date_parts = [] if year is not None: date_parts.append(year.text) if month is not None: date_parts.append(month.text) if day is not None: date_parts.append(day.text) article_data["publication_date"] = "-".join(date_parts) # Extract authors author_elems = article.findall(".//contrib[@contrib-type='author']") for author_elem in author_elems: surname = author_elem.find(".//surname") given_names = author_elem.find(".//given-names") author = {} if surname is not None: author["surname"] = surname.text if given_names is not None: author["given_names"] = given_names.text if author: article_data["authors"].append(author) # Extract full text (combining all paragraphs) body = article.find(".//body") if body is not None: paragraphs = body.findall(".//p") article_data["full_text"] = " ".join("".join(p.itertext()).strip() for p in paragraphs) f.write(json.dumps(article_data) + "\n") saved += 1 print(f"Saved batch {i//batch_size + 1}") print(f"Downloaded {saved} articles to {out_file}, {skipped} articles removed by license filter") Now you can call your function with your query to create your corpus. For example: # Install packages if neededpip install python-dotenv requestsquery = 'bacterial pneumonia treatment'max_results = 500batch_size = 50download_pmc_articles(query, max_results, batch_size) And that’s it! Now all your articles are saved in a jsonl file and ready to be processed for RAG What’s Next: Preparing the Data for RAG In Part 2, we’ll take the domain-specific corpus you just built and use it to power a Retrieval-Augmented Generation (RAG) system — grounding your LLM in real evidence to reduce hallucinations and improve trust. Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor. Published via Towards AI Source: https://towardsai.net/p/artificial-intelligence/hallucinations-in-healthcare-llms-why-they-happen-and-how-to-prevent-them #hallucinations #healthcare #llms #why #they #happen #and #how #prevent #them

For 25 years the logo of middleware tech company Havok has graced the loading screens of big-budget games from across the industry. Its plugins drive the ragdoll and particle physics of games like Helldivers 2, its cloth simulation technology makes the capes of Destiny 2 flap in the wind, and its pathfinding tech helps the monstrosities of Elden Ring wander the The Lands Between.Despite how far its name has traveled, the company's kept a low profile in that time, letting its clients take the bulk of the credit for the boundary-pushing physics. But economic shifts in the industry seemed to drive the company to surface at the 2025 Game Developers Conference.At the event, Havok announced a new pricing model for developers working on budgets under $20 million. The company's normally high-end offerings haven't been affordable for many developers, and while they aren't ready to support low budget indies just yet, general manager David Cochrane told Game Developer that the company now wants to work with studios with "tens" of employees, not necessarily "hundreds."Developers on those lower budgets can now license tools like Havok Physics and Havok Navigation for the low low price of $50,000 per Havok product. There's no royalty fee, and studios can also use plug-in versions of the tools for Unreal Engine.Related:To see a company known for selling premium tools for big-budget games open up to the lower-tier market is quite a phenomenon—especially in an era where the biggest hits of 2025 include Clair Obscure: Expedition 33 and RuneScape Dragonwilds. Small games are becoming bigger hits—and to hear Cochrane tell it, Havok seems to be following the trend.Havok is the rare middleware company to last 25 yearsCochrane explained to Game Developer that Havok began planning the new pricing scheme after studying how many mid-sized studios were planning more ambitious games using Unreal Engine. Havok is a "good fit" for Unreal he said, and to see developers embracing the kinds of solutions Havok offers—but not being able to afford its offerings—began to shift their thinking.It's the kind of adaptation Havok has made repeatedly over the last 25 years, following what Cochrane described as the "creativity" of developers. But it's not the kind of creativity you might think of at first glance. Cochrane praised the gameplay advancements and new kinds of ideas that emerged from the last two-and-a-half decades, but the company's primary focus has been the creativity of technical problem-solving the industry has produced. "The industry has always found creative ways to deliver games that make the most of the of their chosen platform," he said, alluding to the ways developers adapt to online networking or "diversity of hardware." Most recently, he said Havok has paid close attention to the rise of powerful handheld devices like the Nintendo Switch and Steam Deck—and how developers are making "new kinds of games" suited for those devices.Related:But not every trend Havok's been around for has stuck. He recalled the rise of early Facebook games around 2009-2012. If that trend had stayed around, Havok would have been in a tough spot since Facebook games don't need expensive physics or cloth simulation technology. "There was almost a sense of 'was the whole industry gonna have to pivot hard toward this?'" he recalled.It didn't. "There can be a new style of game that comes along—whether that's a live service, or whether it's a social game, or whatever—when MAU numbers are high, the whole industry will take notice, but the game industry always tends to drift back toward an equilibrium which involves many different types of games across many different platforms."So even with the surge of fast-followers for UGC games like Roblox and Minecraft—few of which need Havok's high-end physics and cloth technology—Cochrane said there will still be an audience for games gushing with plentiful particles.Related:The next space for innovation: AI navigationLike everyone else at GDC, Cochrane had a lot to say about AI—but not the same kind of AI that had everyone else aflutter.Cochrane explained that Havok's most overlooked product its its AI navigation tech—but it's also one the most quietly groundbreaking pieces in the company's portfolio. "What we're really seeing now is an increased diversity in the the scale of different characters that [players] come across in a game," he said. Games have had enormous enemies for years, but newer pathfinding tech allows gigantic creatures to share space with smaller in-game characters instead of having their own dedicated areas.Image via Havok.He referenced FromSoftware's Elden Ring here, a game where players can be fighting twisted knights in ruins one minute, then be ambushed by a dragon the next, often in the same area. It's not the only game in 2025 we've seen playing with new ideas of what NPCs should be doing in a game. Just hours later the folks at Bloober Team showed us the ambitious combat system of Cronos: The New Dawn, and earlier this year we were surprised by how the massive monsters of Eternal Strands were able to smoothly navigate through the game's open environments.He called this evolution a "common theme" across the industry—one that makes for a fascinating pairing with Havok's decision to chase smaller studios. "The reality is...the sizes of worlds that people are dealing with at the moment are stressing some of the historic techniques that would have been used to build [methods] for navigation," he said. "Some of the things we're doing is showing ways in which we can have very memory-efficient representations for navigation. We can allow characters to navigate across large worlds that might include streamed geometry in and out.""Now you won't actually hold the whole game world in memory at one point in time, but you still want NPCs that can actually navigate naturally through those larger environments. That's an area we're investing a lot in."Game Developer and GDC are sibling organizations under Informa Tech. Source: https://www.gamedeveloper.com/programming/25-years-in-havok-unveils-royalty-free-pricing-for-budgets-up-to-20-million #years #havok #unveils #royaltyfree #pricing #for #budgets #million

You don't have to wait until the Google I/O keynote for a peek at the future of Android. Google is hosting a pre-I/O web show that will show off "what’s new on Android."The prerecorded online event, dubbed The Android Show: I/O Edition, starts at 10 a.m. PT / 1 p.m. ET today. A 29-second teaser video Google posted to YouTube featuring Sameer Samat, Google’s Android Ecosystem president, offers no spoilers and little substance. Watch it live in the video above.“Hi, everyone, I’m Sameer, and this year we have something special for you: The Android Show: I/O Edition,” he says while standing in front of a bookshelf featuring various Android phones. “We hope you’ll all tune in on May 13 to see what’s new on Android.” Recommended by Our EditorsThe obvious subject for this production would be the upcoming Android 16 release, which hit its third beta in March and had looked primed for an introduction onstage at I/O. Google PR says that I/O itself will feature “more updates in keynotes and technical sessions for developers.”  Release notes and other hints from Android 16's beta releases and earlier developer previews indicate this update will add features like a return of lock-screen widgets,  AI-generated summaries of notifications from messaging apps, new tools to manage health data securely, and stricter security features.That description may not sound like enough change to deserve a new version number. (I'd like to see Google copy more of Apple’s app-privacy features, such as the option to limit an app’s contacts-list access to only one or a few items.) But you'll find out soon enough what fills out the rest of the Android 16 sales pitch. As Samat says at the end of the video: “Whoah, May 13, that’s coming up fast.” Source: https://me.pcmag.com/en/android/29554/why-wait-google-teases-android-16-sneak-peek-ahead-of-io #how #watch #039the #android #show #edition039 #for #sneak #peek

Learn about the benefits and trade-offs of different ways to access the underlying texture pixel data in your Unity project.Pixel data describes the color of individual pixels in a texture. Unity provides methods that enable you to read from or write to pixel data with C# scripts.You might use these methods to duplicate or update a texture (for example, adding a detail to a player’s profile picture), or use the texture’s data in a particular way, like reading a texture that represents a world map to determine where to place an object.There are several ways of writing code that reads from or writes to pixel data. The one you choose depends on what you plan to do with the data and the performance needs of your project.This blog and the accompanying sample project are intended to help you navigate the available API and common performance pitfalls. An understanding of both will help you write a performant solution or address performance bottlenecks as they appear.For most types of textures, Unity stores two copies of the pixel data: one in GPU memory, which is required for rendering, and the other in CPU memory. This copy is optional and allows you to read from, write to, and manipulate pixel data on the CPU. A texture with a copy of its pixel data stored in CPU memory is called a readable texture. One detail to note is that RenderTexture exists only in GPU memory.The memory available to the CPU differs from that of the GPU on most hardware. Some devices have a form of partially shared memory, but for this blog we will assume the classic PC configuration where the CPU only has direct access to the RAM plugged into the motherboard and the GPU relies on its own video RAM (VRAM). Any data transferred between these different environments has to pass through the PCI bus, which is slower than transferring data within the same type of memory. Due to these costs, you should try to limit the amount of data transferred each frame.Sampling textures in shaders is the most common GPU pixel data operation. To alter this data, you can copy between textures or render into a texture using a shader. All these operations can be performed quickly by the GPU.In some cases, it may be preferable to manipulate your texture data on the CPU, which offers more flexibility in how data is accessed. CPU pixel data operations act only on the CPU copy of the data, so require readable textures. If you want to sample the updated pixel data in a shader, you must first copy it from the CPU to the GPU by calling Apply. Depending on the texture involved and the complexity of the operations, it may be faster and easier to stick to CPU operations (for example, when copying several 2D textures into a Texture2DArray asset).The Unity API provides several methods to access or process texture data. Some operations act on both the GPU and CPU copy if both are present. As a result, the performance of these methods varies depending on whether the textures are readable. Different methods can be used to achieve the same results, but each method has its own performance and ease-of-use characteristics.Answer the following questions to determine the optimal solution:Can the GPU perform your calculations faster than the CPU?What level of pressure is the process putting on the texture caches? (For example, sampling many high-resolution textures without using mipmaps is likely to slow down the GPU.)Does the process require a random write texture, or can it output to a color or depth attachment? (Writing to random pixels on a texture requires frequent cache flushes that slow down the process.)Is my project already GPU bottlenecked? Even if the GPU is able to execute a process faster than the CPU, can the GPU afford to take on more work without exceeding its frame time budget?If both the GPU and the CPU main thread are near their frame time limit, then perhaps the slow part of a process could be performed by CPU worker threads.How much data needs to be uploaded to or downloaded from the GPU to calculate or process the results?Could a shader or C# job pack the data into a smaller format to reduce the bandwidth required?Could a RenderTexture be downsampled into a smaller resolution version that is downloaded instead?Can the process be performed in chunks? (If a lot of data needs to be processed at once, there’s a risk of the GPU not having enough memory for it.)How quickly are the results required? Can calculations or data transfers be performed asynchronously and handled later? (If too much work is done in a single frame, there is a risk that the GPU won’t have enough time to render the actual graphics for each frame.)By default, texture assets that you import into your project are nonreadable, while textures created from a script are readable.Readable textures use twice as much memory as nonreadable textures because they need to have a copy of their pixel data in CPU RAM. You should only make a texture readable when you need to, and make them nonreadable when you are done working with the data on the CPU.To see if a texture asset in your project is readable and make edits, use the Read/Write Enabled option in Texture Import Settings, or the TextureImporter.isReadable API.To make a texture nonreadable, call its Apply method with the makeNoLongerReadable parameter set to “true” (for example, Texture2D.Apply or Cubemap.Apply). A nonreadable texture can’t be made readable again.All textures are readable to the Editor in Edit and Play modes. Calling Apply to make the texture nonreadable will update the value of isReadable, preventing you from accessing the CPU data. However, some Unity processes will function as if the texture is readable because they see that the internal CPU data is valid.Performance differs greatly across the various ways of accessing texture data, especially on the CPU (although less so at lower resolutions). The Unity Texture Access API examples repository on GitHub contains a number of examples showing performance differences between various APIs that allow access to, or manipulation of, texture data. The UI only shows the main thread CPU timings. In some cases, DOTS features like Burst and the job system are used to maximize performance.Here are the examples included in the GitHub repository:SimpleCopy: Copying all pixels from one texture to anotherPlasmaTexture: A plasma texture updated on the CPU per frameTransferGPUTexture: Transferring (copying to a different size or format) all pixels on the GPU from a texture to a RenderTextureListed below are performance measurements taken from the examples on GitHub. These numbers are used to support the recommendations that follow. The measurements are from a player build on a system with a 3.7 GHz 8-core Xeon® W-2145 CPU and an RTX 2080.These are the median CPU times for SimpleCopy.UpdateTestCase with a texture size of 2,048.Note that the Graphics methods complete nearly instantly on the main thread because they simply push work onto the RenderThread, which is later executed by the GPU. Their results will be ready when the next frame is being rendered.Results1,326 ms – foreach(mip) for(x in width) for(y in height) SetPixel(x, y, GetPixel(x, y, mip), mip)32.14 ms – foreach(mip) SetPixels(source.GetPixels(mip), mip)6.96 ms – foreach(mip) SetPixels32(source.GetPixels32(mip), mip)6.74 ms – LoadRawTextureData(source.GetRawTextureData())3.54 ms – Graphics.CopyTexture(readableSource, readableTarget)2.87 ms – foreach(mip) SetPixelData(mip, GetPixelData(mip))2.87 ms – LoadRawTextureData(source.GetRawTextureData())0.00 ms – Graphics.ConvertTexture(source, target)0.00 ms – Graphics.CopyTexture(nonReadableSource, target)These are the median CPU times for PlasmaTexture.UpdateTestCase with a texture size of 512.You’ll see that SetPixels32 is unexpectedly slower than SetPixels. This is due to having to take the float-based Color result from the plasma pixel calculation and convert it to the byte-based Color32 struct. SetPixels32NoConversion skips this conversion and just assigns a default value to the Color32 output array, resulting in better performance than SetPixels. In order to beat the performance of SetPixels and the underlying color conversion performed by Unity, it is necessary to rework the pixel calculation method itself to directly output a Color32 value. A simple implementation using SetPixelData is almost guaranteed to give better results than careful SetPixels and SetPixels32 approaches.Results126.95 ms – SetPixel113.16 ms – SetPixels3288.96 ms – SetPixels86.30 ms – SetPixels32NoConversion16.91 ms – SetPixelDataBurst4.27 ms – SetPixelDataBurstParallelThese are the Editor GPU times for TransferGPUTexture.UpdateTestCase with a texture size of 8,196:Blit – 1.584 msCopyTexture – 0.882 msYou can access pixel data in various ways. However, not all methods support every format, texture type, or use case, and some take longer to execute than others. This section goes over recommended methods, and the following section covers those to use with caution.CopyTexture is the fastest way to transfer GPU data from one texture into another. It does not perform any format conversion. You can partially copy data by specifying a source and target position, in addition to the width and height of the region. If both textures are readable, the copy operation will also be performed on the CPU data, bringing the total cost of this method closer to that of a CPU-only copy using SetPixelData with the result of GetPixelData from a source texture.Blit is a fast and powerful method of transferring GPU data into a RenderTexture using a shader. In practice, this has to set up the graphics pipeline API state to render to the target RenderTexture. It comes with a small resolution-independent setup cost compared to CopyTexture. The default Blit shader used by the method takes an input texture and renders it into the target RenderTexture. By providing a custom material or shader, you can define complex texture-to-texture rendering processes.GetPixelData and SetPixelData (along with GetRawTextureData) are the fastest methods to use when only touching CPU data. Both methods require you to provide a struct type as a template parameter used to reinterpret the data. The methods themselves only need this struct to derive the correct size, so you can just use byte if you don’t want to define a custom struct to represent the texture’s format.When accessing individual pixels, it’s a good idea to define a custom struct with some utility methods for ease of use. For example, an R5G5B5A1 format struct could be made up out of a ushort data member and a few get/set methods to access the individual channels as bytes.The above code is an example from an implementation of an object representing a pixel in the R5G5B5A5A1 format; the corresponding property setters are omitted for brevity.SetPixelData can be used to copy a full mip level of data into the target texture. GetPixelData will return a NativeArray that actually points to one mip level of Unity’s internal CPU texture data. This allows you to directly read/write that data without the need for any copy operations. The catch is that the NativeArray returned by GetPixelData is only guaranteed to be valid until the user code calling GetPixelData returns control to Unity, such as when MonoBehaviour.Update returns. Instead of storing the result of GetPixelData between frames, you have to get the correct NativeArray from GetPixelData for every frame you want to access this data from.The Apply method returns after the CPU data has been uploaded to the GPU. The makeNoLongerReadable parameter should be set to “true” where possible to free up the memory of the CPU data after the upload.The RequestIntoNativeArray and RequestIntoNativeSlice methods asynchronously download GPU data from the specified Texture into (a slice of) a NativeArray provided by the user.Calling the methods will return a request handle that can indicate if the requested data is done downloading. Support is limited to only a handful of formats, so use SystemInfo.IsFormatSupported with FormatUsage.ReadPixels to check format support. The AsyncGPUReadback class also has a Request method, which allocates a NativeArray for you. If you need to repeat this operation, you will get better performance if you allocate a NativeArray that you reuse instead.There are a number of methods that should be used with caution due to potentially significant performance impacts. Let’s take a look at them in more detail.These methods perform pixel format conversions of varying complexity. The Pixels32 variants are the most performant of the bunch, but even they can still perform format conversions if the underlying format of the texture doesn’t perfectly match the Color32 struct. When using the following methods, it’s best to keep in mind that their performance impact significantly increases by varying degrees as the number of pixels grows:GetPixelGetPixelBilinearSetPixelGetPixelsSetPixelsGetPixels32SetPixels32GetRawTextureData and LoadRawTextureData are Texture2D-only methods that work with arrays containing the raw pixel data of all mip levels, one after another. The layout goes from largest to smallest mip, with each mip being “height” amount of “width” pixel values. These functions are quick to give CPU data access. GetRawTextureData does have a “gotcha” where the non-templated variant returns a copy of the data. This is a bit slower, and does not allow direct manipulation of the underlying buffer managed by Unity. GetPixelData does not have this quirk and can only return a NativeArray pointing to the underlying buffer that remains valid until user code returns control to Unity.ConvertTexture is a way to transfer the GPU data from one texture to another, where the source and destination textures don’t have the same size or format. This conversion process is as efficient as it gets under the circumstances, but it’s not cheap. This is the internal process:Allocate a temporary RenderTexture matching the destination texture.Perform a Blit from the source texture to the temporary RenderTexture.Copy the Blit result from the temporary RenderTexture to the destination texture.Answer the following questions to help determine if this method is suited to your use case:Do I need to perform this conversion?Can I make sure the source texture is created in the desired size/format for the target platform at import time?Can I change my processes to use the same formats, allowing the result of one process to be directly used as an input for another process?Can I create and use a RenderTexture as the destination instead? Doing so would reduce the conversion process to a single Blit to the destination RenderTexture.The ReadPixels method synchronously downloads GPU data from the active RenderTexture (RenderTexture.active) into a Texture2D’s CPU data. This enables you to store or process the output from a rendering operation. Support is limited to only a handful of formats, so use SystemInfo.IsFormatSupported with FormatUsage.ReadPixels to check format support.Downloading data back from the GPU is a slow process. Before it can begin, ReadPixels has to wait for the GPU to complete all preceding work. It’s best to avoid this method as it will not return until the requested data is available, which will slow down performance. Usability is also a concern because you need GPU data to be in a RenderTexture, which has to be configured as the currently active one. Both usability and performance are better when using the AsyncGPUReadback methods discussed earlier.The ImageConversion class has methods to convert between Texture2D and several image file formats. LoadImage is able to load JPG, PNG, or EXR (since 2023.1) data into a Texture2D and upload this to the GPU for you. The loaded pixel data can be compressed on the fly depending on Texture2D’s original format. Other methods can convert a Texture2D or pixel data array to an array of JPG, PNG, TGA, or EXR data.These methods are not particularly fast, but can be useful if your project needs to pass pixel data around through common image file formats. Typical use cases include loading a user’s avatar from disk and sharing it with other players over a network.There are many resources available to learn more about graphics optimization, related topics, and best practices in Unity. The graphics performance and profiling section of the documentation is a good starting point.You can also check out several technical e-books for advanced users, including Ultimate guide to profiling Unity games, Optimize your mobile game performance, and Optimize your console and PC game performance.You’ll find many more advanced best practices on the Unity how-to hub.Here’s a summary of the key points to remember:When manipulating textures, the first step is to assess which operations can be performed on the GPU for optimal performance. The existing CPU/GPU workload and size of the input/output data are key factors to consider.Using low level functions like GetRawTextureData to implement a specific conversion path where necessary can offer improved performance over the more convenient methods that perform (often redundant) copies and conversions.More complex operations, such as large readbacks and pixel calculations, are only viable on the CPU when performed asynchronously or in parallel. The combination of Burst and the job system allows C# to perform certain operations that would otherwise only be performant on a GPU.Profile frequently: There are many pitfalls you can encounter during development, from unexpected and unnecessary conversions to stalls from waiting on another process. Some performance issues will only start surfacing as the game scales up and certain parts of your code see heavier usage. The example project demonstrates how seemingly small increases in texture resolution can cause certain APIs to become a performance issue.Share your feedback on texture data with us in the Scripting or General Graphics forums. Be sure to watch for new technical blogs from other Unity developers as part of the ongoing Tech from the Trenches series. Source: https://unity.com/blog/engine-platform/accessing-texture-data-efficiently #accessing #texture #data #efficiently