SCROLL NEWS / DISCOVERY
Search headlines
Find the stories shaping the conversation.
Results for " developer " 1,405 found 🔀 AI-powered Shuffle
Land sale for 324-unit downtown Sarasota apartment project called off
The cancellation of the sale of the 1800 block of Fruitville Road to an apartment developer isn't stopping plans for the Breakfast House-led move to a property 1 mile east of the current bungalow village location.
Forza Horizon 6's Community Is Fed Up With Playground Games, and the Studio Has Finally Responded
Forza Horizon 6 developer Playground Games has finally addressed its player community after months of constructive feedback about the game has been met with nothing but radio silence from the renowned British studio.
IceCure Expands ProSense® ChoICE Study for Low-Risk Breast Cancer adding Legacy Health in Portland, OR as U.S. Clinical Site
/PRNewswire/ -- IceCure Medical Ltd. (NASDAQ: ICCM) ("IceCure," "IceCure Medical" or the "Company"), developer of minimally-invasive cryoablation technology...
Supermassive Games suffers layoffs as more than 70 developers seem to have left the company this week
According to a LinkedIn post by Supermassive's IT and facilities director, over 70 developers have been laid off as of this week.
Apple invites developers to submit iPhone Duo-ready apps to the App Store
With the iPhone Duo launch fast approaching, Apple is now accepting submissions of iPhone Duo-optimized apps and games built with Xcode 27.1.
GTA 6 devs tried "not to jump on the bandwagon" of passing trends and instead created their "own slanted, overblown versions"
Grand Theft Auto 6 developers knew development would take "years and years," so avoiding trends was important in making the game "feel vital" come launch.
Northwestern Stadium Developer: Seats Empty Because Fans Watched From Concourse
The 35,000-seat stadium was officially sold out in its first game.
Apple releases beta 3 for iPadOS 27.2, watchOS 27.2, and more
Apple is now rolling out new beta software for developers, including beta 3 for iPadOS 27.2, tvOS 27.2, and more.
Apple Releases Third iOS 27.2 Beta
Apple today seeded the third beta of an upcoming iOS 27.2 update to developers for testing purposes, with the update coming two weeks after Apple released the second iOS 27.2 beta. Apple also seeded a new iPadOS 27.2 beta. Registered developers can download the beta updates by opting in under Settings > General > Software Update and selecting the iOS 27 developer beta. iOS 27.2 includes Apple's redesigned Health app with new Insight and Longevity tabs.
Apple releases third iOS 27.2 developer beta for iPhone
Apple has released the third iOS 27.2 developer beta for iPhone. The update follows the release of iOS 27 to...
Claims that genAI speeds up PC ports of classic games are either "false" or missing the point, says Banjo Kazooie: Recompiled developer
A Banjo Kazooie: Recompiled developer has spoken out against the idea that generative AI allows for faster PC ports and, therefore, better preservation.
Ninja Theory developers announce they are facing redundancy, as the studio starts laying off staff following widespread Xbox cuts
Ninja Theory staff on LinkedIn have shared either farewell messages or sombre posts on the ongoing redundancy process happening at Ninja Theory.
Jagex responds to use of generative AI in RuneScape trailer made by "trusted external partner"
Developer removes promotional material as it investigates situation, says it highlights "an issue with our own processes which we must address"
The science behind targeted oral peptides
A new type of medication is tackling a problem that has long frustrated drug developers: how to deliver peptide therapies orally and hit their target more precisely.
Google admits not every Android app runs great on Intel Googlebooks
‘A small handful’ of apps apparently need some ‘developer tuning.’
ASUS Announces New 64GB Configuration of the Ascent GX10
ASUS today announced a new 64 GB configuration of ASUS Ascent GX10, expanding the lineup alongside the existing 128 GB model. This compact AI supercomputer is designed to bring local agentic AI development within reach of more developers, researchers and data scientists. Based on the NVIDIA DGX...
The New Denpa Men to end service on December 6
Free-to-play RPG The New Denpa Men will end service on December 6 at 6:00 p.m. PT / 9:00 p.m. ET / December 7 at 11:00 JST, developer Genius Sonority announced.
Psikyo Memories launches February 18, 2027
Psikyo shoot ’em up collection Psikyo Memories will launch for PlayStation 5 and Switch 2 on February 18, 2027, developer Edia announced. In its “Edia Game Catalog” distributed at…
Grand Theft Auto VI developers literally forced Sony to develop QSSR technology
According to insider PainkillerQ8, the Rockstar Games team literally forced Sony developers to dig up the technology...
Editorial: Public input should come before approval
A recent Reno town hall revealed tensions between developer Jacobs Entertainment and residents regarding urban design and public engagement.
What is in common between Super Mario and top WA developer?
The family-owned firm behind plans for a controversial aggregate mine in South King County has become one of the region's most prominent landowners and developers.
Fourth RuneScape MMO Built With Unreal Engine in the Works, Jagex Announces RuneScape Animation With MoistCr1TiKaL
Developer Jagex has announced a fourth MMO in the long-running RuneScape franchise, as well as an animated adaptation.
Gears of War: E-Day Devs 'Proud to Be One of Xbox's Main Titles' Amid Newfound Focus on Exclusivity
The Gears of War: E-Day developers at The Coalition say they are "proud to be one of Xbox's main titles" as Microsoft refocuses on exclusivity.
Capcom Announces Plans to Transform the RE Engine Into an AI-Generation Game Engine — ‘Our Goal Is a Future Where We Create Games Together With AI’
Resident Evil and Street Fighter developer Capcom revealed some big changes coming to its RE Engine, which they plan to gradually improve and transform into an "AI-generation game engine."
Manufacturing Sector Highlighted through Day of Recognition
For some, pumpkins and the spectrum of spookiness represent October, but for economic developers, the month — more specifically its first Friday — holds National Manufacturing Day.
Bishop’s computer science chair says AI is eliminating ‘grunt work’ and reshaping digital jobs
Dr. Russell Butler, chair of Bishop’s University’s Department of Computer Science, says artificial intelligence is increasingly taking over routine digital “grunt work,” changing the skills students and workers may need in the years ahead. Photo courtesy of Bishop’s University. Click here to read our platform! Get local news like this every morning. Subscribe free below. Subscribe Pulse Newsroom Artificial intelligence is already taking over some of the routine digital work once performed by entry-level employees, according to Bishop’s University Computer Science chair Dr. Russell Butler — a shift he believes will make expertise, judgement and the ability to evaluate AI-generated work increasingly important. Butler described the change as automation reaching what he calls the “bottom of the cognitive pyramid” during an interview with The Pulse about AI, university education and the future of digital work. His argument is not that humans are disappearing from the workplace. Instead, he sees their role changing as AI becomes capable of handling more of the repetitive tasks involved in computer-based work. “Everything’s converging on just AI basically doing all the grunt work,” Butler said. The bottom of the cognitive pyramid Butler is from Lennoxville, completed his undergraduate studies at Bishop’s and later earned his PhD at the Université de Sherbrooke. He joined Bishop’s in 2019 and now chairs its Department of Computer Science. He uses AI in research, teaching and other creative work, and said the technology is increasingly capable of handling routine digital tasks that once required substantial human time. “You’ve got the grunt work, like moving things around in Excel, doing basic programming tasks,” Butler said. “That’s like the bottom grunt work that the entry-level worker would have done two or three years ago.” Butler went further, saying in his view that much of this type of work has already been absorbed by AI tools. He connected that change to a shift he sees taking place in computer science, where entry-level roles have declined while experienced workers can use AI to increase what they can produce. “A senior developer now has access to essentially an unlimited amount of junior developers on demand,” Butler said. He was describing AI systems metaphorically, not literal employees. His point was that one experienced developer can now ask AI tools to produce code or other technical work that might previously have been assigned to more junior workers. That raises a broader question for universities and employers. If some of the simpler work traditionally given to newcomers can now be done by machines, how do people gain the experience needed to become senior workers themselves? Students doing work that once required years of programming Butler said he is already seeing the change inside his classroom. He created a Bishop’s course called CS208, Coding in Plain English, to show students — including those without traditional programming backgrounds — what they can accomplish using AI. Some come from disciplines such as psychology or neuroscience with no previous programming experience. “What I’m finding is that students who have zero programming experience, they come in from another discipline like psychology or neuroscience or something, they are now able to get up and running with real data sets,” Butler said. Butler’s own doctoral work involved neuroimaging, and he said he has assigned students work he considers comparable to the kinds of analysis he performed during his PhD. He described giving students a real dataset, an expected result and instructions to use AI to complete the analysis. Butler said those students have been able to carry out work at a level he associates with tasks he once performed after years of programming experience. That does not mean a student using AI instantly acquires the same knowledge as an experienced computer scientist. But Butler sees it as evidence that AI can dramatically lower the technical barrier between having a problem to solve and producing a working result. The role of the person using the system then changes. They need to know what to ask for, understand enough about the subject to judge the answer and take responsibility for what is ultimately produced. What about AI hallucinations? One persistent concern surrounding generative AI is hallucination — when a system produces information that sounds convincing but is false or unsupported. Butler said that problem has improved substantially in newer systems and is not as severe as it was with earlier models. He pointed in particular to newer AI systems’ ability to search the web, consult scientific databases and return source links rather than simply generating a citation from their internal model. Butler nevertheless said users should verify what AI produces. “You should verify,” he said while discussing citations generated with AI. His position is that hallucinations have become less common in the systems he uses, not that incorrect AI output is impossible. Human checking remains part of the process. Hours of work compressed Butler offered another example from his administrative work at Bishop’s. The university has been comparing CEGEP courses with first-year Bishop’s courses to determine where students may qualify for advanced standing. Traditionally, Butler said, that means reviewing two sets of course descriptions and manually matching one institution’s courses with another’s. As department chair, that could require hours of reading and comparison. He said AI can now be given both sets of material and asked to generate an initial mapping automatically. “It saves me hours and hours of work,” Butler said. Butler said the AI-generated mapping can also produce a better result than he would produce himself. He sees similar opportunities in businesses. A company may receive specifications, drawings, emails or other digital material from customers and then have employees convert that information into an internal format before anything can be manufactured or delivered. AI can increasingly perform parts of that intermediate processing. In Butler’s view, the employee does not necessarily disappear. Instead, the job becomes more about checking the result. “The human is still there,” he said. Judgement could become more valuable That change is central to Butler’s view of where digital work may be headed. If AI can produce a first draft, write basic code, compare documents or process large amounts of information, he argues that knowledge and judgement become increasingly important. “The real scarce resource going forward is going to be domain knowledge, expertise, and… credentials, authority, the ability to work with other people,” Butler said. He is therefore encouraging students not only to learn AI tools but also to broaden their education. He mentioned acting, fine art and photography as examples of courses that can help develop what he described as taste and intuition — the ability to recognize what is good, what is not and what should be changed. “And that’s the skill that they need now because the grunt work has been automated and now it’s about evaluating what the AI produces and telling it how to improve or expand the output,” Butler said. That does not remove the need for technical knowledge. Butler’s argument is that the emphasis may shift away from manually completing every step of a task and toward directing, checking and improving work produced with AI assistance. He also stressed that humans remain responsible for final decisions. “At the end of the day, the human has to make the final call,” Butler said, pointing to the question of accountability. Physical work may be different Butler draws a distinction between digital work and jobs requiring people to manipulate the physical world. He expects AI to continue advancing quickly in tasks performed almost entirely on computers. He is more cautious about claims that humanoid robots will soon be able to replace workers such as plumbers or electricians. Butler said those trades would not be replaced “anytime soon” in his view and described some current robotics demonstrations as overhyped. He described the situation as counterintuitive. “It’s almost like we solved the highest level first,” he said, contrasting advances in digital AI with the harder problem of building robots capable of operating reliably in the physical world. Butler’s argument is that some digital occupations could face automation sooner than trades requiring complex physical work. The wider AI shift in the Townships Butler’s comments come as local institutions are increasingly grappling with how AI should be used. The federal government recently announced up to $162 million over five years for 10,000 AI-related placements through Mitacs. The Université de Sherbrooke told The Pulse that the number of its collaborative research and innovation projects supported through Mitacs has approximately doubled since 2018-19, while the number of placements has approximately tripled. UdeS also said it already has projects involving both AI adoption and the application of AI to other fields. At Bishop’s, Butler said part of his role involves building the Computer Science department’s local presence, connecting students with regional partners and strengthening relationships with the Sherbrooke technology ecosystem. Those developments put universities in a difficult position. They have to prepare students to use the tools employers increasingly have available while also making sure graduates understand enough to recognize when those tools are wrong, incomplete or being applied to the wrong problem. The risk of losing human skills Butler does not see AI as risk-free. Later in the interview, he raised a separate concern: as AI becomes better at reading, writing, programming and other cognitive tasks, people could lose some of the underlying skills that historically helped produce new ideas. “The other danger is that we’re going to lose the human skills,” Butler said. He specifically mentioned reading and writing and argued that people could lose some of the lower-level understanding that contributes to breakthroughs. Butler said AI systems are based on existing human knowledge and argued that they become much more powerful when paired with a capable person. That tension runs through his view of the technology. AI can remove hours of routine work. It can allow beginners to perform tasks that once required substantial technical training. It can give experienced workers significant leverage. But the more capable the technology becomes, the more important it may be for people to develop the judgement needed to decide what should be done in the first place. For Butler, that is where education may have to change. Students still need to understand their field. They still need to know when something is wrong. And they still need to be able to make the final decision. But if AI increasingly handles the bottom of the cognitive pyramid, the value of a worker may depend less on how quickly they can perform the grunt work and more on whether they understand what the work is supposed to accomplish. Don’t miss the next story that affects your community. Join thousands of readers getting The Pulse every morning—plus breaking news alerts when it matters most. If you value this kind of local journalism, consider supporting it. Subscribe, share and like! Subscribe Share
IQHQ Lands $245M for Boston Life Science Campus
The first part of funding enabled the developer to pay off an existing bridge loan.
Apartment developer Newstreet joins conversion boom with deal for Gold Coast office
Though the deal will be the firm’s first office-to-residential conversion, it joins a string of similarly sized deals by other developers capitalizing on rental demand.
Google’s new Android sideloading rules are live — and here are all the ways to already bypass them
Google’s developer verification rules are now live, but users can bypass new APK sideloading blocks using ADB, Shizuku, and Advanced Flow.
Invincible VS DLC character Angstrom Levy launches October 13
Publisher Skybound Games and developer Quarter Up will release three-versus-three tag fighting game Invincible VS downloadable content character Angstrom Levy on October 13 as part of the game̵…
Metro Redux coming to PS5, Xbox Series on October 29
Publisher Deep Silver and developer 4A Games will release PlayStation 5 and Xbox Series versions of Metro Redux—which includes Metro 2033 Redux and Metro: Last Light Redux—on October 29…
Elta: Defy All Gods launches February 2, 2027, demo now available
“Adrenaline-fueled action platformer” Elta: Defy All Gods will launch for PlayStation 5, Xbox Series, and PC via Steam on February 2, 2027, publisher Focus Entertainment and developer A…
Ace Combat 8 developer apologises for Belkan War pre-order backlash, explains why it happened and vows to explore possible solutions
The team behind Ace Combat 8 has issued a statement in response to the controversy surrounding the game's pre-order bonus game, The Belkan War.
Nerial shuts down after 13 years
Reigns series and Card Shark developer Nerial will shut down after 13 years of operation, the studio announced.
Kena: Scars of Kosmora delayed to 2027
Publisher Sony Interactive Entertainment and developer Ember Lab have delayed Kena: Bridge of Spirits sequel Kena: Scars of Kosmora from its previously planned 2026 release window to 2027. It will …
Report: Gears Of War: E-Day Developers Fear For Their Jobs And Bonuses
With the game due out next week, many at The Coalition appear to fear Microsoft's axe will fall on them next
Gears of War: E-Day Missing Classic Features From Original Games
Classic features such as taking enemies hostage as a "meat shield" and duelling with the chainsaw attachments on Lancer assault rifles won't be in Gears of War: E-Day at launch. According to a developer blog from The Coalition, it wasn't possible to build these features out for launch because this prequel was built from scratch, rather than on the bones of previous Gears of War titles.
As Xbox swings the axe, Gears of War: E-Day developers fear they'll be laid off straight after launch - and right before they'd be due their bonuses
Eurogamer speaks to anonymous sources from The Coalition on the change in morale over the past few months, linked to layoffs and Metacritic bonuses.
September Monthly Review: Health MCP Guy?
For all the time I spend talking about APIs and data exchange, this newsletter has had embarrassingly little interoperability of its own. It’s old-school SaaS, baby - UI or bust. Not all of that is my fault! Substack, for all its popularity, is (as mentioned before) really frustrating as my system of record. The primary user it optimizes towards is the reader, whose attention feeds the app, the Notes feed, and the recommendation network that Substack is actually selling. Health API Guy is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe They’re falling into the same trap that Medium once sprinted into, albeit at a much more glacial speed. Younger generations may not remember (or never knew) it, but Medium once lured publications like The Ringer and Signal v. Noise onto its platform, only to lose them as it rebuilt itself around a paywall and an algorithmic feed. Substack’s original pitch circa 2020 was the antidote (you own your list), but every new feed feature chips away at it. I get it: writers are generic supply in the eyes of the Substack product team. Generic supply follows demand, so the consistent pressure on the PMs is to generate demand, which leads to every product cycle going toward discovery (Notes, recommendations, the app, live video). The end result lackluster author tools (laggy and broken analytics are truly crippling) all despite the premium price point. The archive is where it crosses from authorial gripes to actual reader impact, though. Across Substack and LinkedIn, I’ve published 1,218 posts and nearly 900,000 words since 2018. Substack’s native search can see only the 464 posts that live on Substack, and even there it’s woefully inadequate at surfacing relevant content (even for the guy who wrote it)! Readers have the same problem, writ large. Ask the Archive Fixing search (for both my readers and my own use) turned out to be extremely non-trivial, unfortunately: There’s no public API for readers to access content RSS exists for new content notification, but it only carries the 20 most recent posts and cuts paid ones off at the preview The only Substack MCP is an admin tool for analytics The export exists for authors, but it’s painfully awkward and slow. So I built something to work around that: the Health API Guy MCP. It’s a private MCP server holding the full text of everything I’ve written across Substack and LinkedIn. Add it to Claude or ChatGPT as a connector and you can ask things like: It pairs keyword search (for acronyms and case names, of which we have many) with semantic search. Every post is also labeled by topic, form, and court case, so you can pull a lawsuit’s (or multiple cases’) entire docket across Substack and LinkedIn in one query: Hail, Healthapiguyhydra This is also a bit of me eating my own cooking. I’ve spent the year arguing that headless healthcare is arriving and that developers want primitives over abstractions. Last week in Hail, Benihydra, I watched Salesforce plug its platform into Claude and bet that the records underneath outlive whatever interface sits on top: Salesforce must believe that the Hydra’s immortal head is the records, permissions, and business logic (in that they will survive whatever happens to the interface). It felt a bit hypocritical to keep my own archive behind a Substack search bar. The archive is the body. Substack is one head, and now whatever tool you might use is another. Will the EHRs follow my lead? Probably not imminently, but one can hope! Why Founding Members Only The MCP is available to Founding Member subscribers. A few reasons: It costs money and time to run. Hosting, the database, Clerk usage and AI compute for search and labeling are significant operational costs (that increase with usage). Substack gives me no way to automate access. There’s no API to check whether someone is a paid subscriber, so every user is manual operations for now. It’s a beta. We’ll need to improve things and parts will break. I’d rather they break for a small group of dedicated readers who will tell me about it. The Founding tier deserved a real benefit. Until now it was mostly a generous tip. Now it comes with something. If you’re already a Founding Member, reply to this email with the address you’d like provisioned. If you’re not and want in, upgrade here. Larger teams (such as companies) can get access too: group subscriptions at the Founding tier come with the group discount. The admin for the subscription can reply with the list of addresses to provision. Get 20% off a group subscription Pricing Founding membership is now a flat $200 a year, replacing the old suggested $150 that readers could adjust. Existing Founding Members will have access grandfathered in at their previous rate. Month in Review Articles Published Make Networks Dumb Again (Sep 11): AI is making old analog networks cheaper to use and flooding them with machine traffic in the process. A plea for agent-native rails with the intelligence at the endpoints, and for Washington to drag them to ubiquity. Interoperability, Unbuttoned (Sep 29): Interoperability is a series of workflows in a trenchcoat, and this post finally takes the trenchcoat off. A color-coded map shows which provider workflows went digital, which are fragmented, and which still run on fax. Video Content Antitrust’s System of Record Problem (Sep 3): The FTC’s investigation into Epic prompted a ninety-minute webinar, now posted with chapters and a smoothed transcript. We get into why market definition, missing B2B share data, and the SSNIP test leave antitrust poorly suited to systems of record, plus why information blocking may be doing more than any pending lawsuit. The Information Exchange: Cerner Milking Edition (Sep 18): This was so fucking fun. Ryan Tucker, Ryan Brickner, Brad Thorson, and I try an Around the Horn format, complete with points, penalties, and a mute button. We debate whether Oracle Health is dead, who would actually buy Cerner (IBM enters the chat), p(Doom), and why the front desk still hands you a clipboard. Regulatory Ask Not What Your Network Can Do (Sep 23): CDC’s RFI on public health data intermediaries lists use cases already served, unevenly, by eCR, lab reporting, IZ Gateway, and syndromic surveillance. The real opportunity sits in one question about AI: a network that can handle the next emergency without another interface project. Court cases Particle v. Epic: The Naughty Schoolteacher (Sep 4): After a summer-long vigil, the answer on Epic’s early summary judgment bid is two paragraphs and a remedial assignment. Particle advances to the next grade, and market definition gets another year of discovery. The Epic Discovery Files (Sep 8): Texas’s motions to compel arrive with a 251-page appendix of requests, objections, and meet-and-confer letters with Cravath. Inside: Epic using generative AI for document review, denied interoperability requests in scope, and Noerr-Pennington raised against lobbying discovery. Epic v. Health Gorilla: “Working as Designed” (Sep 9): Epic wants nothing to do with the data breach MDL, and explaining why puts fresh discovery on the public docket. The spiciest allegation: Health Gorilla told GuardDog to scrub its law firm business from its website instead of cutting off access. Veeva v. Epic: Legal Bypass Surgery (Sep 15): Veeva asks the Wisconsin Supreme Court to skip the Court of Appeals, while law professors and three policy groups line up as amici. Meanwhile, Epic’s lawyers apologize for another round of citation errors, including a quote that never existed. Texas v. Epic: Behind the Black Bars (Sep 17): The court portal served up Epic’s unredacted filings, revealing a $5 million document review budget and the AI workflow behind it. Also inside: the 200-plus company names Texas wants searched, and one very expensive Zoom invitation. OpenEvidence v. Doximity: Consequences Will Follow (Sep 25): OpenEvidence tells the court it will not comply with a discovery order, and its own lawyers at Quinn Emanuel say they won’t defend the choice. Add years of Slack reported unrecoverable, and Rule 37 sanctions come into view. Health Tech Keeps Choosing Violence (Sep 28): Doximity moves to end OpenEvidence’s case, Epic asks Texas to show its work, and a sleepy Audacious Inquiry patent fight sprouts antitrust and information blocking counterclaims. The main event is a long-sealed Cognizant v. Infosys ruling that hints where system of record antitrust claims can survive. EHRs One Verb for Oracle Health (Sep 14): Oracle’s earnings call quantifies every data center to the decimal point, while Oracle Health gets a single verb: “accelerate.” Panning the few healthcare mentions turns up an agentic care management system, a research-to-care ledger play, and the question of whether healthcare gets Larry’s patience. Epic’s Forretress Under Siege (Sep 30): Judy Faulkner tells a panel Epic paused hundreds of projects to harden its software and calls it a “shame,” while a spokesperson insists the roadmap hasn’t changed. Project Glasswing, a Pokémon-branded QAN sprint, and a trove of Jodel grievances reconcile the two Epics and preview the Jevons paradox coming for all software. Industry Analysis Scanning for a Network (Sep 1): Scan.com raises $220M to build the imaging network labs never got, speedrunning Zocdoc’s consumer-to-B2B pivot. Its wedge to network density may be the most American one available: personal injury litigation. The Front Desk Wants the Damn Whole Building (Sep 16): Hello Patient buys Converse Health, pushing its AI front desk into referrals, chart work, and prior auth. The copilot categories keep blurring, and your distribution partner is still your final boss. Cross-industry Comparisons A Different Patchwork Quilt (Sep 10): Healthcare loves to envy open banking, but the US version has no mandate, a toll booth at Chase, and data quality problems of its own. The grass isn’t greener; it’s another American patchwork quilt, and perhaps a worse one. Hail, Benihydra (Sep 21): Salesforce brings its platform into Claude, Slack, and its own agent, inviting the interfaces that might replace it. Cut off one head and more grow back, so long as the records underneath stay put. Living on A Trade Secret Prayer (Sep 22): Six credit unions are suing Fiserv, the Epic of core banking, by claiming their own member records as trade secrets. Banking has no Cures Act, so its lawyers reach for legal alchemy that healthcare has already tried. Other News None this month External Media Nabla Accelerate NYC: I kicked off fall conference season at Nabla’s Accelerate event in New York, where Delphine Groll opened by asking the room to picture the company it wants to become and the one it doesn’t. I liked her framework of bureaucracy, fear, and ego as the blockers. I also love vendor conferences/events in general - different vibe than industry biggies. Healthcare 101 with Nikhil Krishnan: I briefly guest lectured Out-Of-Pocket’s Healthcare 101 class. Appreciated Nikhil’s takeaways cover whether AI doctors can query HIEs, why healthcare’s data-liberation rules beat most industries’, and the hi-res graphic is in the comments. HCN: Epic Probed by the FTC: I joined the HCN guys to talk through the FTC’s investigation into Epic. Cedric’s new hairdo was an unexpected plot twist. The episode also covers MFN drug pricing, healthcare M&A and Eli Lilly’s acquisition streak. Fall Conferences: At eHealth Exchange’s Annual Meeting in Austin on Oct 27, I’ll moderate a panel on the tensions between AI, HIEs, and timely patient access with Jean Ross (Primary Record), Therasa Bell (Kno2), Byron Crowe (Doctronic), and Michael Marchant (Freenome). Here’s the rest of my schedule if you want to meet up: CommonWell (Redwood Shores, CA): Oct 13-14 Will be participating in a fun Jeopardy event Open@Epic (Madison, WI): Oct 20-23 Will be enjoying Verona, Willy Street, and other local adventures eHealth Exchange and Sequoia Project (Austin, TX): Oct 27-29 Yeehaw HLTH (Vegas): Nov 15-18 Groundhog Day, Venetian Edition RSNA (Chicago): Nov 30-Dec 2 See me at “Governing AI in Breast Imaging: What Health Systems Are Actually Doing” Digital Health Counsel 2026 AI Summit (Seattle): December 2-3 Will likely be speaking to a bunch of lawyers about information blocking Posts I Liked ChatGPT’s Epic integration: game changer or nothing burger?: Joshua Liu, MD offers five thoughts on OpenAI’s Epic launch, starting from the point that it rides the same open APIs available to anyone. He argues read-only access leaves ordering to Art and the ambient startups, and asks why HCA is piloting it alongside Commure. He’s sorta a must-follow at this point - consistently excellent, thought-provoking takes. Epic, Lilly, and the Discovery pass: Seth Chaney reads Epic’s first major Discovery deal (with Eli Lilly), as well as its revised developer terms on commercial displays, as the pipe finally opening between EHRs and pharma. Life sciences and health continue to collide. Scott Rossignol on EHR payer platforms and ePA fees: Scott argues the top ambulatory EHR vendors are turning the prior auth mandate into a per-member-per-year toll on payers, for an API whose operating cost has nothing to do with covered lives. His fix is two lines of regulatory text: put the CRD/DTR/PAS certification criteria in the Base EHR definition, and bring payer-facing ePA connectivity under § 170.404’s cost-based fee rules. It’s a good solution. Health API Guy is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe
Unfinished business: Stalled hotel construction in Solvang remains inactive amid yearlong litigation
More than a year after Solvang red-tagged his unfinished hotel, Ed St. George said he has no intention of raising a white flag over the dormant development. “We’re already 90 percent done. If we were 40 percent done, I would be willing to tear it down. But we are literally 90 percent done,” the developer […]
Alexandria's business community gathers as AEDP tallies first results of ALX Forward
The economic development group announced a $200,000 state grant at its annual meeting, where the developer of the Old Town North power plant site said it expects to break ground next year
Over 85 Percent Of Japanese Game Developers Are Using AI
A survey by a Japanese industry body suggests developers primarily use it for debugging, but that its usage extends to in-game assets as well
Iowa's opportunities to expand digital infrastructure
The Technology Association of Iowa brought developers, energy companies, and business leaders together to discuss digital infrastructure in Iowa, highlighting an opportunity for the state to lead in the industry. Brian Waller, president of TAI, delivered the keynote speech during the Iowa Digital Infrastructure Summit on Sept. 29. He argued that data centers support a…
City seeks housing developers for vacant North Corktown lots
Detroit is offering the opportunity on 16 lots for 20 homes in the neighborhood near Michigan Central plus $2.5 million through its Small Buildings Program.
Grasshopper Manufacture Leaves NetEase Games to Operate as Independent Studio
Suda51's studio, maker of No More Heroes and Killer7, has split from NetEase Games to work as an independent developer again.
Grasshopper Manufacture splits from NetEase Games
Grasshopper Manufacture has split from now-former parent company NetEase Games, the company announced. NetEase Games acquired the developer from previous parent company GungHo Online Entertainment …
Developer eyes Wise for pumped storage hydropower
Rye Development plans two pumped storage hydropower projects in Wise County, creating jobs and supporting Virginia's clean energy goals.
Star Fox version 1.2.0 update now available
Publisher Nintendo and developer Velan Studios have released the version 1.2.0 update for Star Fox, which adds new stages to Battle Mode, split-screen support for up to four players on a single Swi…
Optimi Health Appoints Melanie Pump to Board of Directors
Seasoned finance executive brings more than 25 years of leadership experience across technology, digital infrastructure, software, fintech and emerging growth companiesVANCOUVER, British Columbia, Sept. 29, 2026 (GLOBE NEWSWIRE) -- Optimi Health Corp. (NASDAQ: OPTH) (CSE: OPTI) (FSE: 8BN) ("Optimi" or the “Company"), a commercial-stage manufacturer and clinical-stage developer of regulated psychedelic drug products, today announced the appointment of Melanie Pump to its Board of Directors (the "
Naughty Dog Will 'Fully Reveal' Intergalactic in 2027, but No Release Window Yet
Naughty Dog studio head Neil Druckmann concluded this year's "The Last of Us Day" by giving updates on the developer's current projects, teasing two new The Last of Us projects that "will expand the canon of The Last of Us beyond Part I and Part II" without confirming whether or not either one of them is a game, as well as saying that its next game, Intergalactic: The Heretic Project, will be "fully reveal[ed]" next year.
Intergalactic: The Heretic Prophet to be fully revealed in 2027
Publisher Sony Interactive Entertainment and developer Naughty Dog will lift the veil on science-fiction adventure game Intergalactic: The Heretic Prophet in 2027, Naughty Dog studio head Neil Druc…
Mat3ra and the University of Tsukuba’s Center for Computational Sciences Sign MoU to Expand Access to Cloud-Based Materials Modeling in Japan
Mat3ra Japan K.K., the Japan entity of Exabyte Inc., the developers of Mat3ra.com, and the Center for Computational Sciences (CCS) of the University of Tsukuba today announced the signing of a Memorandum of Understanding (MoU) that establishes a framework for the use and dissemination of the Mat3ra.com computational materials platform within CCS and associated academic institutions in Japan. Under the MoU, Mat3ra provides CCS with free access to the platform for research and educational p...
MediaDailyNews: Meta Forms Business-Facing AI Product Platform
Meta will begin selling its AI tools to businesses and developers with its newly founded "Meta Enterprise Platform."
Tampa Drone Company Lands $10 Million Investment as Defense Business Grows
Tampa drone developer Draganfly is raising $10 million as it expands its defense business following a Canadian military contract.
God of War Laufey pre-orders open September 29, editions detailed
Physical and digital pre-orders for God of War Laufey will begin on September 29 at 10:00 a.m. local time, publisher Sony Interactive Entertainment and developer Santa Monica Studio announced.
Experimental Emulator Brings Nintendo Switch Games to Jailbroken PS5 Consoles
An unofficial port of the Eden Nintendo Switch emulator is now running natively on jailbroken PlayStation 5 consoles. Developer BlackBearReloaded released ProsperoEden, which adapts Eden, itself a fork of the now-defunct Yuzu project, to run directly on Sony's hardware as homebrew rather than...
Modder Squeezes 74% More Performance Out of DLSS 5 on AMD Radeon in One Day
Modders working on bringing NVIDIA's DLSS 5 Neural Rendering to AMD Radeon GPUs just had their busiest day yet. The developer behind the DLSS-NR-on-AMD tool shipped two new versions within 24 hours, Alpha 0.3.3 and Alpha 0.4.0, together adding roughly 74% performance on Radeon hardware, according...
Meta Launches $1 Million Competition For "Hands-First" Apps Ahead Of VR Glasses
Meta launched a $1 million developer competition for new and updated "hands-first" Quest and VR Glasses apps.
Intel Panther Lake Teardown, 18A, BSPD, GAAFET, SemiAnalysis STEEL
Panther Lake debuts the first commercial implementation of backside power delivery (BSPDN), introduces Intel’s first iteration of gate-all-around (GAA) transistors, and showcases their advanced packaging capabilities with its Foveros-S assembly. With Panther Lake, Intel’s manufacturing arc has shifted from nebulous roadmaps to shipped silicon, a significant milestone on their long road back to competitive semiconductor manufacturing. To evaluate the extent of Intel’s comeback, we tore down Panther Lake. The SemiAnalysis STEEL teardown lab breaks down advanced datacenter and AI hardware. To learn more about our pipeline or to commission a teardown, contact sales@semianalysis.com. WE’RE HIRING: Architecture, floorplan, packaging, manufacturing, and labs experts. Opportunities from system to transistor and everywhere in between. Check out our Careers page. Our teardown traces 18A from its four-sheet RibbonFETs (Intel’s marketing name for GAAFETs) and gate stacks through contacts, frontside and backside wiring, and the bonded carrier. We explain how these material and integration choices improve gate control and reduce resistance, while adding capacitance, thermal resistance, and process complexity. Our measurements put Panther Lake’s 18A compute logic and TSMC N3E GPU logic at similar logic density. However, 18A does not lead TSMC N3P, N2 or Samsung SF2 in peak density. Panther Lake’s CPU cores are incremental updates, and the high-end GPU still uses TSMC N3E. Panther Lake assembles one compute tile, one GPU tile, and one I/O tile atop a passive base tile using Intel’s Foveros-S advanced packaging. Both compute tile variants use Intel 18A. The Xe3 GPU options are a 4-core GT1 tile on Intel 3 and a larger 12-core GT2 tile on TSMC N3E. Both I/O tile variants use TSMC N6. [1], [2] Our analysis centers on the PTL-U compute tile, both the 4-core and 12-core GPU tiles, as well as the 12-lane I/O tile. In conventional chips, power and signal are routed through the same frontside metal stack towards the device frontend. Power rails consume scarce routing resources near the transistors, while tall via stacks carry VDD and VSS from the coarse upper wires to local rails. Backside power delivery (BSPD) moves the main power network behind the transistor layer, to the backside, separating it from frontside signal routing. We covered BSPD and its impacts in 2024. [3], [4], [5] Intel’s BSPD implementation, branded as “PowerVia”, routes power through dedicated backside metals to nano-TSVs, which connect those rails to local source/drain (S/D) contacts. Implementing that separation requires Intel to build the interconnect stacks from both sides of the wafer. The frontside comprises the M0-M14 signal stack, while the backside comprises the BM0-BM5 power stack. M0 and BM0 are closest to the transistors. The nano-TSVs connect the two sides, but Intel patterns and etches each via from the front after forming the contacts. A narrow via runs from the side of the contact deep into the silicon substrate. Intel then completes the frontside signal metal stack, bonds the wafer to a carrier, flips it and removes the original substrate until the buried via tips are exposed. The backside metal stack is then deposited directly on the revealed vias. The nano-TSV and backside-via profiles taper in opposite directions because Intel forms them from opposite sides of the wafer. The transistor structures form the FEOL. Local contacts and nano-TSVs connect them to the wiring. M0 begins the frontside interconnect stack. The silicon carrier remains attached above the frontside interconnects. It supports the device wafer during substrate removal and backside processing and remains part of the finished chip’s thermal path. PowerVia removes the main power distribution from the congested frontside metals, routing supply through shorter and wider backside wires. Its lateral landing still occupies area in the standard cell, so it recovers less cell area than a direct backside contact. [3] Nano-TSVs beside the logic devices carry VDD or VSS from the backside power network, while signal connections continue upward through the frontside metals. Backside Interconnects Samsung SF2 data is included for comparison to Panther Lake’s within this article. SF2 is the incumbent GAA foundry node but lacks BSPD, serving as a useful reference to evaluate 18A. A full teardown of Samsung’s S26 products, processed on SF2, will be shared soon.Nanosheet-cut EDS comparison. The PowerVia supply path runs from the backside Cu rails through Mo-lined W nano-TSVs to the local transistor contacts. In this cross section, the tapered connection spans roughly 150 nm from the contact level to BM0. The Ta liner confines Cu and promotes adhesion to the surrounding stack; the AlOₓ etch stop controls the next dielectric etch above the rail. Dielectric beneath the ribbons electrically separates the devices from the backside wiring and removes the conducting silicon body below the channel. [6] AlOₓ serves as an etchstop (ES), enabling endpointing and protecting the underlying layers. Low-volatility aluminum fluoride reaction products resist the fluorinated plasma, allowing a thin AlOₓ film to protect the metal while the surrounding low-k dielectric is removed. [6], [7]. While the BM0 and layers above the M1 lines show double AlOx layers, Our SMIC N+3 teardown showed single AlOₓ layers. SMIC uses a simpler local AlOₓ substack, while the remaining cap and etch sequence provide the required landing protection. So why double layers? The closely spaced AlOₓ doublets provide two protected endpoints in the etch sequence. Intel documents an AlOₓ/SiN/AlOₓ stack that explains the benefit. The main dielectric plasma etch stops on the first AlOₓ film; a selective wet clear opens that film; a second plasma etch removes the intermediate SiN and stops on the second AlOₓ film. The final wet clear exposes the metal landing surface. SiN is the intermediate dielectric in Intel’s published example. [8] The second stop protects the metal through a cap breakthrough. Wide openings can etch faster than narrow ones, and etch depth varies across the wafer. Metal under an early-clearing opening would otherwise be exposed while other openings still need more etching. Staged protection widens the process window and reduces metal erosion, corrosion and void formation. [8] TSMC documents AlN/AlOₓ/SiOC/AlOₓ above Cu, with AlN blocking Cu diffusion, and a simpler AlN/SiOC/AlOₓ variant that omits one AlOx film. [9] Levels with different opening sizes, aspect ratios, pattern densities and cap materials need different etch margins. A double AlOx stop is useful where another protected endpoint justifies the added processing. The extra film adds formation, selective opening and cleaning steps, plus another set of interfaces to control adhesion, moisture, and stress. These blanket films are opened through the existing via pattern, so each film does not require another lithography mask. AlOₓ adds parasitic capacitance when it replaces lower-k dielectric; two thin AlOₓ films can nevertheless contain less AlOₓ than one thick film. Total thickness, placement, and theintermediate dielectric determine the electrical cost. Deposition chemistry also changes AlOx permittivity and residual hydroxyl content, which can oxidize the underlying metal. [7], [8], [10], [11] The backside stack separates into relatively fine BM0-BM2 wiring near the devices and coarser BM3-BM5 power distribution. The largest pitch increase occurs between BM2 and BM3. BM0’s pitch closely matches the logic-row height, fitting local power delivery to the cell rows. Higher levels aggregate current through larger conductors: routing density becomes less important than low resistance and current capacity as the network approaches the package. This hierarchy provides wide power wiring for the power delivery network without consuming scarce frontside signal-routing resources. [3] The SemiAnalysis STEEL teardown lab breaks down advanced datacenter and AI hardware. To learn more about our pipeline or to commission a teardown, contact sales@semianalysis.com. WE’RE HIRING: Architecture, floorplan, packaging, manufacturing, and labs experts. Opportunities from system to transistor and everywhere in between. Check out our Careers page. Frontside Interconnects Intel 18A combines Mo-lined W contacts and nano-TSVs with a separate backside Cu power network. Samsung SF2 keeps power on the frontside, using Ti-based contact interfaces and Ta-based barriers and Co liners around Cu wiring. In 18A standard-cell rows, backside power rails supply the devices through nano-TSVs within the cells, freeing frontside routing resources. Samsung’s M0 accommodates both power and signal connections. From the device toward M0, the connection runs through a Ti-based S/D interface, W contact fill, a Mo-lined W via, and the Cu M0 wire. Mo supplies a conductive nucleation and adhesion layer for W, replacing the resistive TiN liner used in conventional W integration. This increases the effective conduction volume within the feature while retaining W fill and its established polishing, cleaning and etching processes. Intel’s Mo/W patent describes this integration tradeoff. The nano-TSV uses the same Mo-lined W construction in the backside supply path. [12] The move from TiN to Mo is an incremental change. While a full Co or Mo fill can also reduce the volume lost to liners in very small features, it requires new integration schemes that increase complexity and risk. Cu remains attractive for wider wires due to its low resistance. As wires and vias shrink, the diffusion barrier consumes an increasing fraction of their cross-section. [12], [12], [14] Intel uses Co/Ru liners at M0-M1, Co at M2-M4, and Nb at M5-M9. The lower-level liners help Cu adhere and reduce void formation during trench fills. Applied Materials’ Endura has new thermal control that facilitate wetting process, so the thin film continuity is good enough that good capillary pressure will drive Cu atoms to the via bottom without voiding. Intel’s choice to use Nb is particularly interesting. Intel’s Nb patent describes a conductive diffusion barrier intended to reduce the barrier’s contribution to resistance relative to conventional Ta-based barriers, particularly at via bottoms where all current crosses the barrier. The patent pairs Nb in coarser levels with the option of lower-cost PVD processing. [15], [16] The upper metal layers support thicker barriers formed through physical vapor deposition (PVD) despite its worse coverage and uniformity. Meanwhile, the lower metal layers require thinner barriers deposited through conformal atomic layer deposition (ALD). Co/Ru adds another material interface and requires controlled deposition and Cu fill. Changing liners and barriers by metal layer allows Intel to optimize interconnect resistance, process complexity, and reliability. [15, 16] RibbonFET, Intel’s name for its gate-all-around FETs (GAAFETs), replaces the FinFET’s vertical fins with four stacked horizontal silicon nanosheets, allowing the gate to surround the channel on every side. The path to GAAFET begins with the planar transistor. A planar MOSFET places the gate above the channel between its source and drain. Pairing an NMOS with a PMOS transistor creates a CMOS inverter, in which the NMOS pulls the output low for a high input, and the PMOS pulls it high for a low input. The gate must retain electrostatic control of the channel to ensure clean switching. As gate lengths shrank, the drain began to compete with the gate for that control, increasing off-state leakage. Electrostatic control was restored through an architectural evolution that raised the channel into a vertical fin and wrapping the gate around three sides. Called “FinFET”, this new architecture packed more effective channel width into a smaller footprint. Further scaling made it harder to maintain both drive current and leakage within smaller cells, and reintroduced the same problems planar MOSFETs faced. Nanosheet GAAFETs close the fourth side by replacing the vertical fin with a stack of horizontal nanosheets, each surrounded by the gate. The tighter electrostatic control suppresses leakage at shorter gate lengths while stacking adds effective channel width within the cell footprint. In a FinFET process, channel width changes in discrete steps as designers must add or remove whole fins. Nanosheet width can instead be adjusted continuously within the process’s design rules. Wider sheets increase drive current, while narrower sheets reduce capacitance at the cost of drive current. Intel 18A uses stacks of four nanosheets each and varies their widths across logic and SRAM. At the process level, adding more sheets to each stack increases effective channel width and drive current, but complicates fabrication. RibbonFET vs MBCFET Samsung began GAAFET production in 2022 with SF3E, following with SF3 and now SF2. Its ‘MBCFET’ provides a useful structural comparison with Intel’s first RibbonFET implementation. [17] STEEL is digging deeper into SF2, used in the Exynos 2600, and TSMC’s GAAFET N2, used in Apple’s A20 Pro, in upcoming newsletter articles. We’re throwing some teasers on X. Let’s compare Samsung SF2’s MBCFET with Intel 18A’s RibbonFET. Subscribe Even to the untrained eye, Intel’s extra nanosheet is obvious. Intel stacks four ribbons to Samsung’s three. Samsung’s sheets are much wider in these fields, so both sheet count and width matter to the available channel perimeter. Sheet width also changes which silicon surfaces carry current. On conventional (001) silicon, wide nanosheets emphasize the broad top and bottom surfaces, favoring electron transport; the larger sidewall contribution in a narrow sheet favors hole transport. Thinner sheets improve gate control but increase confinement and scattering. This makes width and thickness part of the NMOS/PMOS balance, alongside strain and threshold voltage. [18], [19] GAAFET designs like 18A use different work-function-metal (WFM) stacks for NMOS and PMOS. Around each ribbon, a thin SiOx interfacial layer separates the silicon channel from the HfOx high-k dielectric, with La providing dipole tuning and the WFM wrapping the dielectric. NMOS uses a TiAl-based stack, while PMOS uses TiN WFM. W fills the remaining gate trench, providing a lower-resistivity path where the work-function layers are no longer needed. In this field, the PMOS stacks leave room for W between ribbons, while the NMOS stacks occupy more of those gaps. A silicon-based dielectric marks the P/N boundary, allowing the PMOS and NMOS gates, sharing the same gate trench, to be processed sequentially. Fast logic paths, retention circuits, and SRAM need a family of threshold options. Changing threshold without substantially changing device dimensions, capacitance or fabrication complexity is valuable. FinFET processes typically use different work-function-metal stacks. In a four-ribbon GAA stack, the narrow sheet-to-sheet gap limits how much WFM can fit around each channel. La in the gate dielectric creates interfacial dipoles at the SiOx/HfOx boundary, shifting effective work function and tuning threshold voltage. This gives Intel another control alongside its NMOS and PMOS WFM stacks. Low-threshold devices improve critical-path drive; higher thresholds reduce leakage elsewhere. Dipole tuning is especially useful in GAA because it changes threshold without consuming the narrow intersheet gap with thicker WFM. Precise control of La incorporation, diffusion and interface quality has long been a challenge, limiting viability in high volume production but is now seen from every leading-edge foundry. Intel’s patent describes depositing a dipole-forming oxide above HfOx and annealing it toward the interfacial oxide before completing the work-function and fill metals. This separates threshold tuning from the space available for metal. Newer research addresses the thermal cost: imec’s 2026 dipole-middle research inserts the shifter between two HfOx depositions, shortening the diffusion path while protecting SiOx during patterning. [20], [21]Matched-cut EDS, Intel 18A (left) vs. Samsung SF2 (right). Intel retains raised source/drain epi beneath its contacts, while Samsung recesses W deep into the epi to form a V-shaped Ti-lined interface. The deeper contact increases metal-to-semiconductor area and shortens the current path from the lower sheets, reducing contact and spreading resistance. It also removes epi volume and brings the contact etch closer to the channel ends. Retaining more epi preserves the material available for strain transfer, especially from SiGe into PMOS. These geometries balance contact access against stress engineering and etch margin. [22], [23] Samsung stacks three sheets to Intel’s four ribbons, and both processes use sheet width to tune drive strength. In our Samsung cross-sections, widths range roughly from 19 to 30 nm in the NPU rows and 37 to 50 nm in the CU cell. The Samsung nanosheets taper, with the widest sheet at the bottom and the narrowest at the top. Both processes use HfOx gate dielectric and Ti-based work-function stacks, with Al in the NMOS stack. In the Samsung devices shown here, the dielectric and WFM occupy the intersheet gaps, leaving W above the top sheet. Intel’s PMOS stack leaves more room between ribbons, and W fills those gaps while the thicker NMOS stack leaves W mainly in the upper trench. Gate-stack EDS maps. The W between Intel’s PMOS ribbons provides a conductive path close to the lower gates. Where WFM fills the entire gap, the gate still surrounds the channel, but voltage reaches it through the more resistive work-function films. Thinner WFM and dipole tuning preserve room for low-resistivity fill; Mo and Ru are alternative fill metals being developed for further scaling. [24] A masked, sequential WFM flow explains the different gate heights and inter-nanosheet fill. The proposed sequence below shows how separate NMOS and PMOS work-function steps produce that geometry. Enabled by the BSPDN process, Intel replaces the dense-logic silicon subfin with dielectric, removing the parasitic conduction path below the ribbons and reducing substrate-related capacitance. A retained silicon body as in classical, non-SOI, planar and FinFET designs needs junction and punchthrough-stop engineering to suppress leakage. Dielectric isolation makes that leakage less sensitive to the subfin doping profile but adds removal and fill steps. It also weakens the direct thermal path through silicon, making the contacts, metal stacks and package more important for heat extraction. [24], [26] Fluorine is concentrated around selected Intel device structures in the maps. WF6 is a standard precursor for W nucleation and fill, while barrier films protect adjacent dielectrics from fluorine attack. Low-fluorine W processes reduce the residual-F burden. Chloride-based precursors avoid introducing F during W deposition, but require control of chlorine attack, nucleation and fill quality. The integration target is a continuous, low-resistance W path with a thin protective liner and minimal chemical damage to the surrounding stack. [20], [27], [28] The SemiAnalysis STEEL teardown lab breaks down advanced datacenter and AI hardware. To learn more about our pipeline or to commission a teardown, contact sales@semianalysis.com. WE’RE HIRING: Architecture, floorplan, packaging, manufacturing, and labs experts. Opportunities from system to transistor and everywhere in between. Check out our Careers page. We measured cell height, gate pitch, metal geometry, and ribbon dimensions at the XTEM sites shown below. The tables group these dimensions by site and device polarity. Our “sheet cuts” cross the silicon channel and show the ribbons end-on. “gate cuts” run along the channel through successive gates. The 18A logic cell dimensions point to a five-track logic library while the N3E and Intel 3 cell dimensions evidence a seven-track logic library. The DDR-PHY uses wider M0 wires and much larger spacing than core logic. That trades routing density for lower wire resistance and weaker coupling between neighboring nets. The geometry suits the current delivery and coupling requirements of analog, clock, and I/O circuitry. PowerVia lets 18A combine a compact cell height with wider M0 geometry by moving the main power rails off the signal-routing tracks. That relaxes local wire scaling while preserving a small cell footprint. Cell height and gate pitch set the geometric density; pin access and routability determine how much of it a real block can use. [29] The biggest takeaway from our gate-pitch measurements is that Intel 18A compute logic and TSMC N3E GPU logic have similar density in the Bohr representative-cell model. The 18A example is 18.6% denser than the Intel 3 GPU example. Gate pitches are nearly identical across the three sites, so cell height drives most of the difference. The Bohr model combines a four-transistor NAND2 spanning three gate pitches and a 32-transistor scan flip-flop (SFF) spanning nineteen pitches, weighting their densities 60:40. The sensitivity column shows how independently changing cell height and gate pitch by ±1 nm changes the result. This compares representative cell geometries; whole-die density also depends on cell mix and placement. The 18A P-core gives M0 substantially more metal cross section than the N3E vector engine. Treating each profile as a trapezoid gives 2.63 times the area per line and 1.84 times the area after normalization by routing pitch. The larger section reduces the geometric contribution to line resistance and lowers current density for a given current. Taller and wider wires also add capacitance, so circuit delay depends on the balance of resistance and capacitance. The DDR-PHY has less metal area per routing width than the 18A core fields, while remaining above N3E. [30] Area = height × (top CD + bottom CD) / 2, including liners. Area/pitch normalizes by routing width. Taper is the symmetric sidewall angle from vertical, with the largest angle belonging to the DDR-PHY. Compute tile The measurements show how ribbon dimensions and gate-stack geometry vary across the compute tile and between NMOS and PMOS to balance channel drive, gate load and the space needed for the dielectric/WFM stack across logic, SRAM and the DDR-PHY. Width mainly changes available channel perimeter; thickness also changes electrostatic control and carrier confinement. Gate-stack thickness then determines the space left for low-resistivity fill P-core and LP E-core logic Both the P-core and LP E-core use multiple nanosheet widths. Widths are measured on high-magnification XTEMs while wider-field images demonstrate additional width choices within the LP E-core. Multiple widths are expected even within an LP E-core. Timing-critical paths, buffers and cells with different fanout need different drive strengths. The lower-magnification fields show this width diversity beyond the sites quantified in the table. L2 and L3 SRAM GAA gives SRAM designers another way to balance the pull-up (PU), pass-gate (PG), and pull-down (PD) transistors. FinFET bitcells set device strength through fin count while GAA adds nanosheet width as a sizing knob. In a 6T SRAM cell, a strong pull-down relative to the pass-gate limits read disturbance, while a strong pass-gate relative to the pull-up improves writability. During a write, the pass-gate and write driver pull the node storing “1” below the inverter trip point. During a read, the pull-down holds the node storing “0” low. Bias, threshold voltage, mismatch and assist circuitry set the remaining margin. FinFET high-current cells commonly use a PU:PG:PD fin-count pattern of 1:2:2, a device-sizing ratio rather than a current ratio. Ribbon width lets Intel balance SRAM strengths without adding whole fins. The L2 cell uses its narrowest ribbons for PU and widest for PD, improving writability and read stability respectively. Intel’s disclosed HCC operates without assist; its denser HDC uses negative-bitline write assist. Pulling the selected bitline briefly below ground increases pass-gate overdrive so it can overpower the pull-up at lower supply voltage. That buys density and low voltage writability at the cost of boosting circuitry, switching energy, and additional voltage stress that must be controlled. [31], [32] Four rectangular ribbons give the perimeter = 8 × (width + thickness), before corner rounding. PG/PU is 1.49 and PD/PG is 1.16. The L3 structures closely resemble L2 in layout and cell height. Fewer L3 nanosheet widths are tabulated because fewer high-magnification images were available. DDR PHY The DDR-PHY trades density for controlled analog behavior and reliable off-chip signaling. It contains drivers, receivers, delay circuits, and calibration logic that set drive strength, sampling time, and voltage margin. Repeated four-sheet devices with similar widths fit the use of regular transistor units for matching and programmable drive. Its wider local wiring provides room for current delivery and separation of sensitive signals, while consuming more area than a dense core-logic grid. The layout serves the memory channel’s electrical requirements as well as digital logic density. [33] The SemiAnalysis STEEL teardown lab breaks down advanced datacenter and AI hardware. To learn more about our pipeline or to commission a teardown, contact sales@semianalysis.com. WE’RE HIRING: Architecture, floorplan, packaging, manufacturing, and labs experts. Opportunities from system to transistor and everywhere in between. Check out our Careers page. Intel 3 GPU devices Vector engine logic Intel 3’s XVE logic uses two-fin PMOS and NMOS devices with power rails in M0. Its cell height and M0 pitch give a seven-track geometry, two tracks more than the 18A logic. One-fin groups also appear among the two-fin devices. Intel 3 L2 SRAM The Intel 3 L2 SRAM uses the familiar HCC sizing pattern: one PU fin, two PG fins, and two PD fins. N3E GPU devices Vector engine logic The N3E XVE field contains repeated two-fin devices with seven-track cell geometry. N3E remains a FinFET process, giving Panther Lake a direct FinFET-to-RibbonFET comparison. N3E L2 SRAM The N3E L2 SRAM uses the same PU:PG:PD fin-count pattern of 1:2:2. Panther Lake-U follows Lunar Lake’s floorplan quite closely. Both pair 4 P-cores with 4 LP E-cores and NPU, media and display engines in similar locations. Lunar Lake also uses Xe2, the direct predecessor to Panther Lake’s Xe3 GPU. This makes Lunar Lake the most direct basis for our comparisons. Arrow Lake differs in core count and uses the older Xe-LPG GPU architecture, so we only use it where it offers a more direct component-level comparison. Compute tile Panther Lake compute-tile floorplans remain sparse even months after launch. Intel 18A’s backside metal and dielectric stack must be removed without damaging the underlying structures before a clean transistor-level floorplan can be imaged. Most published die shots hide or heavily process the background, but we are quite proud of the die shot we achieved and are excited to show the work we have done. We measured the areas of the key components on the compute tile and compared them with their Lunar Lake predecessors on TSMC N3B. These help us to capture changes in block area and compare the two chips across process nodes and designs. Our total tile areas exclude the scribe-line area. The compute-plus-GPU subtotal below uses the PTL-U compute tile and GT1 GPU; it excludes the I/O tile and passive base. Individual block areas use the boundaries marked on the floorplans The compute-plus-GPU row is recomputed from the displayed PTL-U and GT1 areas. Component rows use their stated per-region counts and are not an additive partition of the whole tile. The P-core area remains almost unchanged between Lunar Lake and Panther Lake, despite L2 capacity increasing from 2.5 MiB to 3 MiB. Arrow Lake uses the same Lion Cove core as Lunar Lake but also has a 3 MiB L2. Cougar Cove fits 20% more L2 into the same P-core area. The larger private cache keeps more of each core’s working set close to its execution units, reducing access to shared L3 and DRAM. Extra capacity adds storage leakage and lookup energy, so designers balance it against avoided lower-level accesses. The shared P-core L3 cache also shrank by 14.8%. [2] Cougar Cove combines a similar footprint with Intel’s reported power-efficiency improvements. RibbonFET’s tighter channel control reduces leakage, while PowerVia reduces supply droop and allows tighter voltage guardbands. [1] Darkmont’s four-core LP E-core cluster is 5.0% smaller than Skymont’s on Lunar Lake, with most of the reduction in its L2 regions. The 1 MiB region shrank by 8.4% and the 1.5 MiB region by 14.9%. The tag arrays also use one fewer visible row. Tags identify which memory addresses the data array holds, so rearranging them changes the cache’s layout and wiring without requiring less data capacity. [2] The LP E-cores share one L2. This pools capacity and avoids duplicating all the cache machinery, but the four cores contend for its banks and bandwidth. Their separate cluster also keeps light work away from the performance cluster and its L3, allowing that larger domain to sleep. [1], [2] Cache area includes more than the storage cells. Tags identify each line, decoders select rows, sense amplifiers read the small bitline signal, and wires connect to the banks. Splitting an array into smaller sections shortens wordlines and bitlines, improving access speed, but duplicates peripheral circuits. Panther Lake’s smaller cache regions therefore reflect the complete memory implementation, including how much of each region is devoted to storage. [34] Unlike Meteor Lake and Arrow Lake, Panther Lake has no separate SoC tile. The NPU, LP E-cores, memory controllers, PHYs, media and display engines now share the compute tile. This removes an active die and keeps CPU memory traffic on one die. The cost is moving PHY and I/O-related circuitry onto 18A: drivers, receivers and analog circuits must still meet external voltage, loading and signal-integrity requirements, so their area does not shrink like dense digital logic. [1], [2] The biggest shrink comes from the NPU, which occupies 36.9% less area. NPU 5 consolidates the same total INT8 MAC count into half as many neural compute engines. Each of the three NCEs has a larger MAC array to make the complete NCE envelope 22.6% larger than an NPU 4 engine. Consolidation also halves the number of scratchpads and SHAVE DSPs, from 12 to 6. The MAC array handles matrix multiplication and convolution, while SHAVE executes vector and custom operations that fit the array poorly. [1], [2], [35] The paired floorplans identify each NCE envelope and its scratchpad, MAC, and SHAVE regions. Each measured MAC polygon is counted once per NCE in the area accounting below. The scratchpads store weights, activations, and intermediate results near the MAC arrays, allowing repeated use without fetching them again from DRAM. Halving their number delivers the largest measured area saving but leaves less local storage for the same total MAC count. Layers that no longer fit locally require smaller working tiles or more transfers of intermediate data. The benefit depends on keeping the enlarged arrays busy while managing that tighter storage budget. [36] NPU 5 also adds native FP8. Using half the operand width of FP16 reduces storage and transfer demand, helping workloads fit the smaller local memory budget. Lower precision and format-dependent range make scaling and model validation part of deployment. Hardware activation functions further reduce work that would otherwise occupy the programmable DSPs. [1], [2] Microsoft requires an NPU to deliver at least 40 TOPS for Copilot+ PCs. Both Lunar Lake and Panther Lake meet this threshold, but Panther Lake uses significantly less silicon. GPU tiles Panther Lake is Intel’s first product with Xe3, its latest GPU architecture. It offers two different GPU tiles: a smaller GT1 tile with 4 Xe3 cores on Intel 3 and a larger GT2 tile with 12 Xe3 cores on TSMC N3E. Panther Lake allows us to compare the same GPU architecture across both Intel 3 and TSMC N3E. Wildcat Lake adds a third Xe3 implementation on Intel 18A. A future newsletter will detail Xe3 and its implementation differences across all three process nodes. GT2 scales Xe3 to a different physical layout, with render slices arranged vertically instead of GT1’s horizontal arrangement. Slice placement sets the distances to shared cache banks and the D2D interface. Those wires consume area and add delay, so scaling the number of Xe cores also requires a new balance of cache placement, routing and timing. [1] What’s immediately obvious is that the GT2 tile on TSMC N3E has much smaller Xe cores than GT1. These block areas include logic, caches, and routing. An Xe core on the GT1 tile is ~69% larger than one on Lunar Lake, and ~55% larger than one on GT2. Intel 3 therefore uses substantially more area per Xe core. The block-area gap exceeds the measured logic and SRAM density gaps, bringing routing, timing targets, cell mix, and floorplan allocation into the comparison. The measured vector/matrix engine region is almost unchanged between Lunar Lake and Panther Lake’s GT2 tile. Xe3 retains eight 512-bit vector engines and eight 2048-bit XMX engines per core. Its gains also come from feeding those engines more effectively: more resident threads hide stalls, and variable register allocation lets shaders trade registers per thread against the number of threads kept active. [1] The shared L1/SLM capacity increased by 33% from 192 KiB to 256 KiB, while its area increased only 5%, raising effective density by 27%. L1 retains reused cache lines, while software-managed SLM lets a thread group share data locally. Both reduce traffic to more distant memory. Allocating more SLM per group can also limit how many groups reside on a core at once. [1], [37] The GT1 tile carries 4 MiB of L2 against 16 MiB on the GT2 tile. GT1 divides its L2 cache into four 1 MiB banks, while GT2 uses eight 2 MiB banks. Each bank contains 128 macros, but each N3E macro stores 16 KiB, twice the Intel 3 macro’s 8 KiB capacity. The N3E macro is only 54% larger while holding twice as many bits, giving it 30% higher density: ~23.7 Mbit/mm² versus 18.3 Mbit/mm². Including bank-level circuitry, the gap widens to ~16.9 Mbit/mm² on GT2 versus ~10.4 Mbit/mm² on GT1. GT2 gains density with its macros storing more bits per unit area, and those macros occupy more of each cache bank. Larger macros spread decoder and sense-amplifier overhead across more storage, while a more compact bank layout reduces the share spent on control and routing. The compromise is longer wordlines and bitlines that carry more capacitance. [34] I/O tile Panther Lake uses two I/O tile variants, both fabricated on TSMC N6. The smaller one provides 4 PCIe 5.0 and 8 PCIe 4.0 lanes and serves lower-tier systems as well as those without a discrete GPU, while the larger one adds 8 PCIe 5.0 lanes, bringing the total to 20 lanes, for discrete-GPU connectivity. Panther Lake SKUs with the larger 10- or 12-Xe GPUs use the smaller I/O tile. [38] The smaller I/O tile adds a PCIe 4.0 block and a Thunderbolt block to Lunar Lake’s I/O layout, providing four additional PCIe 4.0 lanes and another Thunderbolt 4 port. Its repeated N6 blocks retain nearly identical areas and layouts. Reusing these proven PHYs and controllers avoids porting and requalifying external interfaces on 18A, where faster digital logic offers less benefit to circuits constrained by the off-chip link. [38] SemiAnalysis’s teardown lab (STEEL) dives deep into the world’s advanced datacenter and AI hardware. To learn more about our pipeline or to commission a teardown, contact sales@semianalysis.com. We’re hiring technical experts from system to transistor and everywhere in between. Check out our Careers page. Panther Lake offers scalability and modularity through its disaggregated packaging that partition compute, GPU, and I/O silicon into separate tiles allowing for a suite of tile configurations. This partitioning makes the package part of Intel’s node economics as it determines how much leading-edge wafer area each product consumes, which functions can remain on other processes, and how much configuration freedom Intel can offer from a shared set of tiles. Furthermore, fabricating the compute and GPU tiles separately confines the new 18A process to the compute tile and allows graphics and I/O to use other, more established, and more cost-effective processes. For Panther Lake, the GPU and I/O tiles are assembled alongside the compute tile on a passive silicon base using Foveros-S. Intel’s current technology brief lists a nominal 36 µm pitch for Foveros-S. Through-silicon vias (TSVs) in the base connect the fine wiring above to the larger package connections below. The functional tiles sit side by side on that passive base in a 2.5D configuration. [39] Our cross-section through the compute and GPU tiles shows the package’s wiring hierarchy. Microbumps connect each active tile to the passive silicon base; its fine redistribution layer (RDL) carries the short, dense tile-to-tile links. TSVs carry connections through the base to the package substrate, which fans them out to the much coarser motherboard solder joints. The base supplies interconnect, while computation remains in the active tiles above it. [39] At the compute-tile edge, the higher-magnification inset shows a local microbump spacing of approximately 25.24 µm and a feature width of 12.33 µm. These local spacings are finer than Intel’s nominal Foveros-S value. The X-ray fields further confirm tighter neighboring bumps, consistent across every die-to-die area found on each tile. Additional X-ray analysis is offered after the paywall. Putting the memory controller beside the CPU removes the D2D transfer that CPU memory requests required in Meteor Lake and Arrow Lake. This avoids the extra transmitter, receiver, and link traversal, saving interface energy and latency. Panther Lake’s separate GPU still crosses a D2D link to reach DRAM, so its larger local caches also help contain package traffic. [1], [40] Smaller dies are less likely to contain a random fatal defect, and screening them before assembly prevents one bad tile from consuming a complete package of good silicon. Reuse also spreads design and qualification work across more products. Against those gains, Intel pays for the passive base, D2D circuits, extra bonding and test steps, and losses during assembly. Cost per working product across the portfolio captures the combined effect of wafer yield, reuse, test, and assembly. [29] Wildcat Lake packaging Intel launched Core Series 3, formerly Wildcat Lake, on 16 April 2026 for value mobile and edge systems. Wildcat Lake keeps 18A but removes the passive base and combines more functions on one die to simplify the package. The two products therefore reveal two distinct ways to commercialize the same leading-edge process. [41] Wildcat Lake’s 18A die combines up to two Cougar Cove P-cores, four Darkmont LP E-cores, two Xe3 cores and a smaller NPU. A separate platform-controller die supplies I/O, connected through UCIe, Intel’s first processor implementation of the standard. Consolidating graphics remove a tile boundary and the passive base, reducing assembly complexity for a modest-bandwidth value product. It also ties CPU and graphics scaling to the same die, giving up Panther Lake’s ability to swap in a much larger GPU. [42], [43] In July 2021, Intel CEO Pat Gelsinger set out an ambitious process roadmap aimed at regaining performance leadership by 2025, later described as five nodes in four years. Five years and one CEO later, Intel’s comeback story is not as unambiguously positive as Pat may have hoped. [44], [45] Intel once set the pace for process technology, bringing high-k metal gate technology and FinFETs into volume production years ahead of the rest of the industry. Its 22 nm FinFET process reached consumers with Ivy Bridge in 2012. [46] Intel’s integrated device manufacturing (IDM) model allowed its architects and process engineers to co-optimize products and processes. Starting with Sandy Bridge, Intel dominated x86, while AMD struggled with Bulldozer. That lead faltered at 14 nm and broke at 10 nm. Intel targeted a massive 2.7× density increase, but the node arrived years late and required several revisions before it could support Intel’s full lineup. This delay forced Intel to stretch 14 nm across six generations, while TSMC moved ahead in process technology and AMD recovered in x86. By 2019, Intel was still shipping 14 nm across most of its product stack, with its 10 nm client ramp focused on Ice Lake mobile processors. Meanwhile, TSMC was shipping N7 and N7+, and AMD’s Zen 2 compute chiplets used N7 to raise core counts and improve efficiency. Intel’s process failures were central to its decline, but unsound business decisions furthered their downward slide. Product delays compounded product mistakes, pushing client, server, and FPGA roadmaps off schedule. Several attempts to enter AI (Nervana and Gaudi) and networking (Tofino) also failed to establish lasting businesses. Intel’s recovery has focused on consumer CPUs and advanced packaging. Tiger Lake, Alder Lake, Lunar Lake and now Panther Lake have restored Intel’s consumer roadmap. On the process side, Intel 4 shipped with Meteor Lake, Intel 3 with Granite Rapids and Sierra Forest, and Intel 18A with Panther Lake. Intel has also made advanced packaging part of its foundry offering. However, Intel is still playing catch-up in servers. Several Xeon generations arrived years late and trailed contemporary AMD and Arm server CPUs in performance, efficiency, and core count. The process roadmap is back, but Intel does not hold the same process-technology leadership position it held prior to 10 nm. The introduction of gate-all-around nanosheets and backside power delivery are two of the biggest changes to transistor integration in a decade. Intel took on both changes at once: 18A paired its first RibbonFET with PowerVia in Panther Lake. Panther Lake is a substantial manufacturing milestone. Our cross-sections show how RibbonFET and PowerVia reshape local contacts and wiring, while the floorplans show where architectural consolidation and process choices save area. A sustained competitive lead depends on product performance, cost, yield, and the next implementation. The SemiAnalysis STEEL teardown lab breaks down advanced datacenter and AI hardware. To learn more about our pipeline or to commission a teardown, contact sales@semianalysis.com. WE’RE HIRING: Architecture, floorplan, packaging, manufacturing, and labs experts. Opportunities from system to transistor and everywhere in between. Check out our Careers page.. [1] Intel, “Intel Tech Tour 2025 Panther Lake recap,” Doc. 866361, 2025. Accessed: Sep. 14, 2026. [Online]. Available: https://cdrdv2-public.intel.com/866361/ITT_2025_Panther_Lake_Recap1.pdf [2] Intel, “Core Ultra processors Series 3 datasheet, volume 1,” Doc. 872188-001, Jan. 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://cdrdv2-public.intel.com/872188/872188-001.pdf [3] N. Horiguchi and E. Beyne, “Backside power delivery: How to power chips from the backside,” imec Reading Room, Nov. 25, 2022. Accessed: Sep. 14, 2026. [Online]. Available: https://www.imec-int.com/en/articles/how-power-chips-backside [4] W. Hafez et al., “Intel PowerVia technology: Backside power delivery for high density and high-performance computing,” in Proc. IEEE Symp. VLSI Technol. Circuits, Kyoto, Japan, Jun. 2023, pp. 1–2, doi: 10.23919/vlsitechnologyandcir57934.2023.10185208. [5] K. Fischer et al., “Intel 18A platform technology featuring RibbonFET (GAA) and PowerVia for advanced high-performance computing,” in Proc. Symp. VLSI Technol. Circuits, Kyoto, Japan, Jun. 2025, pp. 1–3, doi: 10.23919/vlsitechnologyandcir65189.2025.11075006. [6] K.-F. Cheng, C.-L. Teng, H.-Y. Huang, and H.-C. Chen, “Metal oxide composite as etch stop layer,” U.S. Patent 12 176 247 B2, Dec. 24, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US12176247B2/en [7] S. W. King, “Dielectric barrier, etch stop, and metal capping materials for state of the art and beyond metal interconnects,” ECS J. Solid State Sci. Technol., vol. 4, no. 1, pp. N3029–N3047, 2015, doi: 10.1149/2.0051501jss. [8] A. V. Mule’, D. J. Towner, D. Seghete, C. R. Ryder, and A. A. Gonzalez, “Multi-layer etch stop layers for advanced integrated circuit structure fabrication,” U.S. Patent Appl. 20220102343 A1, Mar. 31, 2022. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US20220102343A1/en [9] C.-C. Wang and J. H. Wang, “Interconnect structure with dielectric cap layer and etch stop layer stack,” U.S. Patent Appl. 20230253247 A1, Aug. 10, 2023. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US20230253247A1/en [10] Y. J. Wu, K.-F. Cheng, C.-C. Lee, H.-K. Chang, and H.-Y. Huang, “Etch stop layer for interconnect structures,” U.S. Patent Appl. 20240332070 A1, Oct. 3, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US20240332070A1/en [11] M. G. Rainville, N. Shankar, K. S. Reddy, and D. M. Hausmann, “Deposition of aluminum oxide etch stop layers,” U.S. Patent Appl. 20180197770 A1, Jul. 12, 2018. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.justia.com/patent/20180197770 [12] J. S. Leib et al., “Conductive lines having molybdenum liner and tungsten fill for advanced integrated circuit structure fabrication,” U.S. Patent Appl. 20240429126 A1, Dec. 26, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.justia.com/patent/20240429126 [13] M. Naik, “Cobalt enables power and performance scaling at single-digit logic nodes,” Applied Materials, Dec. 17, 2018. Accessed: Sep. 14, 2026. [Online]. Available: https://www.appliedmaterials.com/us/en/blog/blog-posts/cobalt-enables-power-and-performance-scaling-at-single-digit-logic-nodes.html [14] Lam Research, “Breaking through AI’s invisible barrier with molybdenum,” Lam Research Newsroom, May 6, 2025. Accessed: Sep. 14, 2026. [Online]. Available: https://newsroom.lamresearch.com/molybdenum-metallization-ai-revolution [15] P. Yashar, G. Malyavanatham, and H. Vijwani, “Integrated circuit interconnect structures with niobium barrier materials,” U.S. Patent Appl. 20240112951 A1, Apr. 4, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US20240112951A1/en [16] Applied Materials, “Applied Materials unveils chip wiring innovations for more energy-efficient computing,” Applied Materials Investor Relations, Jul. 8, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://ir.appliedmaterials.com/news-releases/news-release-details/applied-materials-unveils-chip-wiring-innovations-more-energy [17] Samsung, “Samsung showcases AI-era vision and latest foundry technologies at SFF 2024,” Samsung Semiconductor, Jun. 13, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://semiconductor.samsung.com/news-events/news/samsung-showcases-ai-era-vision-and-latest-foundry-technologies-at-sff-2024/ [18] C. W. Yeung et al., “Channel geometry impact and narrow sheet effect of stacked nanosheet,” in Proc. IEEE Int. Electron Devices Meeting (IEDM), Dec. 2018, pp. 28.6.1–28.6.4, doi: 10.1109/iedm.2018.8614608. [19] S. Mochizuki et al., “Evaluation of (110) versus (001) channel orientation for improved nFET/pFET device performance trade-off in gate-all-around nanosheet technology,” in Proc. Int. Electron Devices Meeting (IEDM), Dec. 2023, pp. 1–4, doi: 10.1109/iedm45741.2023.10413854. [20] D. G. Ouellette et al., “Fabrication of gate-all-around integrated circuit structures having molybdenum nitride metal gates and gate dielectrics with a dipole layer,” U.S. Patent 12 051 698 B2, Jul. 30, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US12051698B2/en [21] imec, “Advancing the CFET-based device roadmap with novel integration modules and standard-cell architectures,” imec, 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://www.imec-int.com/en/articles/advancing-cfet-based-device-roadmap-novel-integration-modules-and-standard-cell [22] A. Reznicek, X. Miao, C. Lee, and J. Zhang, “Wrap around contact for nanosheet source drain epitaxy,” U.S. Patent 11 302 813 B2, Apr. 12, 2022. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US11302813B2/en [23] J. A. Kittl, J. G. Hong, D. R. Palle, and M. S. Rodder, “Methods for varied strain on nano-scale field effect transistor devices,” U.S. Patent 9 978 833 B2, May 22, 2018. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US9978833B2/en [24] S. Gandikota et al., “Method of reducing metal gate resistance for next generation NMOS device application,” Int. Patent Appl. WO 2024/137272 A1, Jun. 27, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/WO2024137272A1/en [25] J. Zhang et al., “Full bottom dielectric isolation to enable stacked nanosheet transistor for low power and high performance applications,” in Proc. IEEE Int. Electron Devices Meeting (IEDM), Dec. 2019, pp. 11.6.1–11.6.4, doi: 10.1109/iedm19573.2019.8993490. [26] C. Yoo, J. Chang, Y. Seon, H. Kim, and J. Jeon, “Analysis of self-heating effects in multi-nanosheet FET considering bottom isolation and package options,” IEEE Trans. Electron Devices, vol. 69, no. 3, pp. 1524–1531, Mar. 2022, doi: 10.1109/ted.2022.3141327. [27] S. S. Pradhan, D. B. Bergstrom, J.-S. Chun, and J. Chiu, “Tungsten gates for non-planar transistors,” European Patent Appl. EP 3 506 367 A1, Jul. 3, 2019. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/EP3506367A1/en [28] L. Schloss and X. Ba, “Tungsten films having low fluorine content,” U.S. Patent 9 754 824 B2, Sep. 5, 2017. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US9754824B2/en [29] Intel Foundry, “Accelerating AI and HPC with advanced process and packaging technologies,” Doc. 367370-001US, Jun. 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intel.com/content/dam/www/central-libraries/us/en/documents/2025-11/intel-foundry-hpc-ai-brief.pdf [30] M. Lofrano, X. Chang, H. Oprins, and Z. Tokei, “Mitigating the thermal bottleneck in advanced interconnects,” imec, Sep. 28, 2023. Accessed: Sep. 14, 2026. [Online]. Available: https://www.imec-int.com/en/articles/mitigating-thermal-bottleneck-advanced-interconnects [31] S. Bair, “Intel at ISSCC 2025: Navid Shahriari invited talk, eight papers, forums, panelist & product details,” Intel Community, Feb. 19, 2025. Accessed: Sep. 14, 2026. [Online]. Available: https://community.intel.com/t5/Blogs/Tech-Innovation/Edge-5G/Intel-at-ISSCC-2025-Navid-Shahriari-Invited-Talk-Eight-Papers/post/1667592 [32] D. Chandra, E. Potladhurthi, D. R. S. Reddy, and K. S. Rengarajan, “Tunable negative bitline write assist and boost attenuation circuit,” U.S. Patent Appl. 20160203857 A1, Jul. 14, 2016. Accessed: Sep. 14, 2026. [Online]. Available: https://patents.google.com/patent/US20160203857A1/en [33] Synopsys, “Advantages of firmware-based training in high-speed DDR IP,” Synopsys IP Technical Bulletin. Accessed: Sep. 14, 2026. [Online]. Available: https://www.synopsys.com/articles/firmware-based-training-ddr-ip.html [34] N. Muralimanohar, R. Balasubramonian, and N. P. Jouppi, “CACTI 6.0: A tool to understand large caches,” Hewlett-Packard Laboratories, 2009. Accessed: Sep. 14, 2026. [Online]. Available: https://users.cs.utah.edu/~rajeev/cacti6/cacti6-tr.pdf [35] Intel, “Intel Tech Tour 2024 Lunar Lake AI hardware accelerators,” Doc. 824436, 2024. Accessed: Sep. 14, 2026. [Online]. Available: https://cdrdv2-public.intel.com/824436/2024_Intel_Tech%20Tour%20TW_Lunar%20Lake%20AI%20Hardware%20Accelerators.pdf [36] Y.-H. Chen, J. Emer, and V. Sze, “Eyeriss: A spatial architecture for energy-efficient dataflow for convolutional neural networks,” in Proc. 43rd ACM/IEEE Annu. Int. Symp. Comput. Archit. (ISCA), Jun. 2016, pp. 367–379, doi: 10.1109/isca.2016.40. [37] Intel, “Introduction to the Xe-HPG architecture,” Doc. 758306, Nov. 4, 2022. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intel.com/content/www/us/en/developer/articles/technical/introduction-to-the-xe-hpg-architecture.html [38] Intel, “Core Ultra processors Series 3 for the edge,” Doc. 855291. Accessed: Sep. 14, 2026. [Online]. Available: https://cdrdv2-public.intel.com/855291/Intel%C2%AE%20Core%E2%84%A2%20Ultra%20Processors%20Series%203%20for%20Edge%20Overview_2.pdf [39] Intel Foundry, “Foveros technology brief,” Doc. 366411-001US, Feb. 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intel.com/content/dam/www/central-libraries/us/en/documents/2025-07/foveros-25d-product-brief.pdf [40] D. Das Sharma, “UCIe: Building an open chiplet ecosystem,” UCIe Consortium, 2022. Accessed: Sep. 14, 2026. [Online]. Available: https://www.uciexpress.org/_files/ugd/0c1418_c5970a68ab214ffc97fab16d11581449.pdf [41] Intel, “Intel launches Intel Core Series 3 processors,” Intel Newsroom, Apr. 16, 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intel.com/content/www/us/en/newsroom/news/client-computing/intel-launches-intel-core-series-3-processors-changing-the-game-for-everyday-computing.html [42] Intel, “Core Series 3 launch press deck,” Apr. 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://download.intel.com/newsroom/2026/Intel-Core-Series-3/Intel-Core-Series-3-Launch-Press-Deck.pdf [43] Intel, “Intel outlines architectures for agentic AI at Hot Chips 2026,” Intel Newsroom, Aug. 24, 2026. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intel.com/content/www/us/en/newsroom/news/client-computing/intel-outlines-architectures-for-agentic-ai-at-hot-chips-2026.html [44] Intel, “Intel accelerates process and packaging innovations,” Intel Investor Relations, Jul. 26, 2021. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intc.com/filings-reports/all-sec-filings/content/0001193125-21-224438/d199788dex991.htm [45] Intel, “Intel reports third-quarter 2021 financial results,” Intel Investor Relations, Oct. 21, 2021. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intc.com/news-events/press-releases/detail/1505/intel-reports-third-quarter-2021-financial-results [46] Intel, “3rd generation Intel Core processors bring exciting new experiences and fun to the PC,” Intel Investor Relations, Apr. 23, 2012. Accessed: Sep. 14, 2026. [Online]. Available: https://www.intc.com/news-events/press-releases/detail/577/3rd-generation-intel-core-processors-bring-exciting
Illinois Just Passed a Tough New AI Law. Even Small Businesses Should Pay Attention
Illinois’s new AI law may target frontier developers, but it offers every business a preview of where AI regulation is headed.
Remedy has confirmed performance issues with AMD graphics cards in CONTROL Resonant.
Developers from Remedy Studio have officially confirmed technical issues with CONTROL Resonant for AMD graphics card users. Players...
MindsEye: Ex-GTA developer's studio faces closure
Build A Rocket Boy, founded by Leslie Benzies, has gone into administration after its debut game flopped.
Ace Combat 8: Wings of Theve ‘101 Overview’ trailer
Publisher Bandai Namco and developer Bandai Namco Aces have released a five-minute overview trailer for flight action game Ace Combat 8: Wings of Theve.
Cellid and Megane Top Partner to Commercialize and Expand the AR Glasses Business
/PRNewswire/ -- Cellid Inc., a developer of displays and spatial recognition engines for next-generation augmented reality (AR) glasses, announced today that...
Sony reportedly surveying developers about PlayStation disc-ending decision - is it having second thoughts?
Sony's decision to end PlayStation disc production has been hugely controversial, seemingly even among developers. A rumoured survey suggests Sony has noticed.
"So Long, Sweet Prince" - Ubisoft's Rogue Prince Of Persia Game Rolls Out Its Final Update
Developer Evil Empire says it's been "one hell of a run"
Unicorn Overlord director Takafumi Noma announces fantasy IP IGNISPHERE and action RPG DAWNGAZER for PS5, Switch 2, and PC
Publisher Good Smile Company and developer Monochrome Corporation have announced new fantasy intellectual property IGNISPHERE, and the first game set in that world, DAWNGAZER, led by Unicorn Overlo…
PlayStation Reportedly Polling Game Developers After Decision to Kill Discs Sparks Backlash
PlayStation is reportedly sending a lengthy survey to game developers with a question about how they sell physical games after their decision to stop producing physical game discs prompted intense backlash.
Housing geared toward low-income families opens in Detroit
A group of developers and city officials on Thursday celebrated the opening of Preston Townhomes, a $14 million affordable housing project in the Islandview neighborhood.
Meta Launches $1M Developer Competition to Stock 'VR Glasses' with More Hand-tracked Apps
At Meta’s annual developer conference today, the company announced its next VR Start Developer Competition, this time slated to award a prize pool of $1 million to spark the community to get cracking on Meta VR Glasses-compatible content. Meta is hoping to appeal to developers of both new and existing projects across entertainment, gaming, and […]
Sony Allegedly Surveyed Game Developers on its Disc-Killing Plan for 2028, Hinting at a Possible Reversal amid Fan Boycotts
According to leaker MLID, Sony has surveyed game developers on its disc-killing plan, suggesting they may consider a shocking reversal.
Placeholder domain used in dev docs now serves ClickFix attacks
The "third-party.com" domain, commonly used as a placeholder in developer documentation and code examples, is serving a fake Cloudflare verification page that attempts to trick Windows users into executing PowerShell commands.
Regent Bank Amphitheater Opens Its Doors This December, Marking VENU's Entry into the Oklahoma Live Entertainment Market
BROKEN ARROW, Okla., September 23, 2026--Venu Holding Corporation ("VENU" or the "Company") (NYSE AMERICAN: VENU), an owner, operator, and developer of premium live entertainment destinations, today announced that Regent Bank Amphitheater will open to the public on December 4 and 5, 2026 with its Winter Preview, featuring A Magical Cirque Christmas. Fans can sign up at regentbankamphitheater.com to be the first to know when tickets go on sale.
Xbox's Latest Merger Means 'Forza Motorsport' Is Almost Certainly Done
Forza Horizon developer Playground and Forza Motorsport creators Turn 10 are merging into one studio focused on the Forza and Fable franchises.
Florida developer splashes into Raleigh with 264-unit project near Downtown South
Hillpointe, a Florida developer, plans a 264-unit apartment community near Raleigh's Downtown South development on Rush Street.
What's ahead for the $250 million funding tool for Belton's Uptown Entertainment District?
At Tuesday night's meeting, there was plenty of discussion to better understand what Belton and developer, PETRA is getting into.
State of Decay Developer Undead Labs Confirms 'Significant' Layoffs Following Independence from Xbox
State of Decay 3 developer Undead Labs has confirmed that it has had to undergo "significant" layoffs following its independence from Xbox.
Apple Releases First Public Betas of iOS 27.2 and macOS Golden Gate 27.2
Apple today seeded the first betas of iOS 27.2, iPadOS 27.2, macOS Golden Gate 27.2, tvOS 27.2, and watchOS 27.2. The software comes a day after Apple provided the second betas to developers. To get the new betas, you can sign up on Apple's beta testing website. After you're signed up, the betas can be installed through the Software Update section in the Settings app on each device. iOS 27.2 includes the redesigned Health app that Apple introduced at its September iPhone event.
Rayman Legends Retold ‘Overview’ trailer
Publisher Ubisoft and developers Ubisoft Montpellier and Ubisoft Milan have released a six-minute game overview trailer for Rayman Legends Retold.
‘Halo’ to move under Call of Duty publisher as Xbox announces layoffs
The cuts affect Halo Studios, other first-party developers, and management and central functions within Xbox, according to an announcement posted this morning by Matt Booty, Xbox’s chief content officer. The company is now about 75 percent finished with a restructuring announced earlier this year.
Hellblade developer Ninja Theory to be shut down in new wave of Xbox cuts, as two separate agreements collapse
Ninja Theory - developer of the Hellblade series - is shutting down as part of today's Xbox cuts. Its proposed buyers pulled out, resulting in the closure.
XBOX Moves Rare Under Activision & Obsidian Entertainment to Bethesda
XBOX's ongoing changes have seen developer Rare moved under Activision's wing, whereas Obsidian is now under Bethesda.
Xbox is merging its Forza developers into a single studio
Playground Games and Turn 10 Studios staff will now focus on Forza Horizon and Fable together.
Florida Chamber gives JB Pritzker 'runner-up' for 'Economic Developer of the Year' award on Chicago billboards
Florida Chamber of Commerce billboards along Chicago's Stevenson Expressway dub Gov. JB Pritzker the runner-up "Florida Economic Developer of the Year."
AI connectors: How the App Store-big internet keys work
Meta is opening Muse connectors to developers. Here's how they could make AI agents more useful — and reshape online commerce.
Meta's Muse could become Zuckerberg's best platform opportunity yet
Meta's Muse AI agent could give Mark Zuckerberg another chance to build a powerful internet platform, and developers are paying attention.
How Epic turns America's health data into answers
Software developer Caleb Cox and chief medical officer Dr. Jackie Gerhart discussed how Cosmos, Epic Research’s database, spurred tools like health alerts.
Apple Seeds Second macOS Golden Gate 27.2 Beta to Developers
Apple is testing the next macOS Golden Gate update, and developers can download the second macOS Golden Gate 27.2 beta. The second beta comes a week after Apple released the first beta. Registered developers can download macOS Golden Gate 27.2 by opening up the System Settings app, clicking on General, selecting Software Update, and opting into macOS Golden Gate betas. macOS Golden Gate 27.2 introduces Siri AI in new languages. macOS Golden Gate 27.2 introduces Siri AI in new languages.
Apple Releases Second iOS 27.2 Beta
Apple today seeded the second beta of an upcoming iOS 27.2 update to developers for testing purposes, with the update coming a week after Apple released the first iOS 27.2 beta. Apple also seeded an iPadOS 27.2 beta. Registered developers can download the beta updates by opting in under Settings > General > Software Update and selecting the iOS 27 developer beta. iOS 27.2 includes Apple's redesigned Health app with new Insight and Longevity tabs.
Seattle startups DexCare and Mila Health combine to expand AI patient scheduling
Seattle health-tech startup DexCare has acquired fellow local startup Mila Health, an AI agent developer that automates patient calls and texts for appointment scheduling. The deal, whose terms were undisclosed, integrates Mila's rapidly deployable AI communication tools into DexCare's patient-access platform serving 57 million patients nationwide.
'Project Hail Mary: Journey Among the Stars' lets players do jazz hands with Rocky in mixed reality
We played the upcoming mixed-reality and VR adventure game based on the hit sci-fi movie at Gamescom, and talked with one of its developers.
KCC Entertainment Design | Mega City shortlisted for award
The large-format social entertainment venue, delivered for developer Wellstone, is one of three KCC-designed projects on the shortlist
Welltower Pays $230M for Warner Center Site
Senior housing developer Harbert South Bay sold its 339-unit Warner Center assisted living community for $230.3 million, or about $539 per square foot.
BALL x PIT expansion ‘Risen Ballbylon’ announced
Publisher Devolver Digital and developer Kenny Sun have announced BALL x PIT expansion “Risen Ballbylon.” It will launch on November 6 for $6.99.
Blizzard developer says World of Warcraft: Forever's beta numbers are the "craziest" he's seen in 17 years
World of Warcraft: Forever's beta was overwhelmed by players this week, with a Blizzard developer calling the numbers the biggest he's seen in 17 years.
Wolverine developer Insomniac denies using generative AI after fans spot strange restroom sign
Insomniac denies using generative AI on Marvel's Wolverine after fans spotted a strange bathroom sign, blaming a "UV map bug" instead.
Latest batch of Nintendo Switch backwards compatibility fixes for Switch 2
Nintendo has been striving to get all third party Nintendo Switch games up and running and compatible with the Nintendo Switch 2. The company has been working with third party developers to ensure any…
Engrams Embedding Entendre: Codesign for Efficient DRAM/SSD Offloading
Engram extends standard token embeddings with learned multi-token lookups. Recurring local patterns retrieve vectors directly, reducing the need to reconstruct them through attention and feed-forward layers. With Engram model architecture optimization, it allows for lower HBM capacity to be needed for models at the same quality. This does not mean there won’t be an insane demand for HBM but it just means that model architecture will continue to innovate around constraints. This model architecture design is naturally codesigned for parameter offloading: each token accesses a few embedding rows whose addresses depend on token IDs, not hidden states. The runtime can prefetch those rows from host DRAM while earlier layers compute, keeping the table outside HBM without transferring entire weight matrices. Our Memory model contains our latest estimates of quarter by quarter HBM, DRAM, & NAND supply and demand. Offloading frees HBM for model weights and KV cache, potentially supporting larger batches or more concurrent sessions. When DRAM becomes the next constraint, NVMe offers another tier. Recommendation systems already cache frequently or recently accessed embedding rows in faster memory while backing colder rows with SSDs. After NVIDIA roadmap had to change due to massively despec’ing Rubin Ultra from 1024GB to now ~200GB of HBM per chip, model architecture optimizations like emgram maybe helpful. Our DeepSeek-V4.1-Flash configuration uses roughly 189 GiB of memory for Engram. We replace it with a memory-mapped (mmap) file and measure serving performance with offload to SSD. Later on in our report, we will show our Engram offloading experiments along with the official InferenceX agentic inference serving results on Engram models like DeepSeekv4.1 Flash across all 6 NVIDIA GPU SKUs along with MI355X. Unsurprisingly, the CUDA Moat is still mogging MI355X on the ultra popular DeepSeekV4.1 Flash model. We also show how even on high capacity HBM SKUs, offloading emgrams to DRAM could result in even better performance for most of the pareto than keeping the emgram in HBM. Our benchmark has been widely reproduced, validated and/or supported by almost every major buyer of compute from Google Cloud to Microsoft Azure to Oracle, to Meta and many more. Furthermore, it has the support of the ML community including from vLLM, LMCache, SGLang, PyTorch, Huggingface and the support of major labs like OpenAI, MiniMax, ZAI, Qwen, Moonshot Kimi, etc. Star the InferenceX GitHub repository if you find the open-source benchmark and data useful!. InferenceX is the only inference benchmark in the world to have TPUv7, Jalapeño, Nvidia Rubin NVL72, AMD, and soon, SambaNova and Trainium. Due to how realistic AgentX scenario is to real world agentic inference workloads, AMD has committed to collaborating on MI455X UALoE72 too. DeepSeek did not release the original paper’s two trained Engram models. We replicated its setup on fineweb-edu using the released code and training hyperparameters, at an estimated 6E18 FLOPs per run. We observed the same U-shape scaling: Engram improved performance over pure MoE baselines. We also reproduced DeepSeek’s results, where earlier-layer representations with Engram resembled those of later layers. Like the original Engram paper, it is possible to probe Engram’s gate scores to see what n-grams DeepSeek-V4.1-Flash makes the most use of. Our gate scan finds names, code fragments, relational phrasing, and boilerplate. These examples prioritize interesting-ness over gate strength. One unexpected result was Wright : Ace Attorney. These examples suggest learned memory optimizes the training objective, not a judgment of which facts deserve storage. Licenses, bibliography fragments, API scaffolding, and website furniture can provide prediction shortcuts, so the value of additional Engram capacity may depend on what survives data preparation. This does not show that table capacity is “wasted”: the evaluation-corpus scan establishes neither training exposure nor the capacity occupied by each category. For offloading, strong gates do not identify cache-hot rows. Low gates do not automatically save reads either: computing the gate requires the retrieved key and undoes the performance gain of a fused kernel. Skipping reads would require a separate usefulness predictor before retrieval. In the original paper’s inference-time ablation, factual-knowledge benchmarks retained just 29–44% of their original performance, while reading comprehension retained 81–93%. This is due to the training–inference mismatch. The resulting degradation therefore measures this trained model’s dependence on Engram, not the performance difference between models trained with and without it. In our ablations, suppressing Engram worsens token likelihood across all evaluated domains, especially encyclopedia text and several code corpora. Surprisingly, GSM8K accuracy stays within measured run-to-run variation and removing Engram has no effect. Engram is not a detachable dictionary beside an unchanged MoE. Removing it changes downstream features and expert selection. We tested whether rerouting hurts or compensates by holding tokens fixed in a teacher-forced experiment on CRUXEval, a code-reasoning benchmark of small Python functions where the model predicts a function’s output from its code and an input, and scoring the reference answer. Removing Engram raised answer loss from 0.2848 to 0.3093 bits/token. Forcing the ablated model to use the original Engram-on expert choices made it worse still, at 0.3375 bits/token. Rerouting partially compensates for the missing memory. Memory features and expert selection work together, rather than following a clean “memory stores facts; experts reason” division. On the same CRUXeval, removing Engram during either phase reduced accuracy and increased generated tokens; removing it throughout produced the largest changes. Keeping Engram for prefill leads to more correct answers than keeping for decode likely due to semantically richer KV cache transferred to decode workers, allowing it to mitigate some of the performance loss. Engram’s table is large, but each lookup is small. DeepSeek-V4.1-Flash requests 24 rows at each of two Engram layers, about 12.4 KiB per processed token position across the model, or 3.1 KiB per GPU when split across four GPUs. Currently as of Day 7 since Model Release, MI355X is still 2-4x worse performance per dollar compared to B200 even when normalized by Mi355X’s lower TCO. Our full total cost of ownership breakdown comes from our AI Cloud TCO Model along with monthly market surveys of over 100+ gpu clouds & gpu cloud customers. On the Day 0 release of DeepSeekv4.1 Flash, NVIDIA vLLM works out of the box with zero issues across all 6 SKUs: H100, H200, B200, B300, GB200, GB300! This was thanks to the amazing work by the NVIDIA & Interact teams! In comparison, AMD vLLM did not work on day 0 for DeepSeekv4.1 Flash.Source: SemiAnalysis InferenceX AMD’s vLLM documentation points to using vllm/vllm-openai-rocm:deepseekv41-flash-0909, but from hour 0 of the model release to hour 23, AMD has not publicly released the image. AMD claims, “SPEED IS THE MOAT,” yet it has still not released it by the 23rd hour. We wish that, going forward, the AMD team has a better process for hour 0 model releases. Eventually when they did publicly release for “day 0” image support, performance-wise, it is currently up to 14.8x worse perf per dollar than H200 and up to 42x worse perf per dollar than B200/B300. The power of the CUDA MOAT is NVIDIA’s collaboration with its massive 6 million-developer community ecosystem including most of the vLLM & SGLang & Tokenspeed maintainers which means that CUDA is optimized on day 0. Overall, AMD did make significant improvements but the performance per dollar is still currently 2-4x worse than B200. AgentX Engram DRAM Offloading Improving Performance The HBM and DRAM offload use the same GPU kernel to select and dequantize rows. With HBM, it reads GPU memory; with Unified Virtual Addressing (UVA), it reads pinned host memory directly. Both support full decode graphs. Moving the table into HBM accelerates only the sparse lookup, leaving decoder computation and communication unchanged, resulting in little overall benefit while consuming memory otherwise available to KV cache. Another benefit of Engram offloaded to DRAM which means you can reduce the communication overhead by using less HBM GPUs per replica. For example, when enabling Engram offloading on B300, we are able to switch from TP4 to now TP2 which improves the pareto curve by up to 1.6x. When iso-model quality, less HBM is required as the engrams could be offloaded to host DRAM. Thus HBM bandwidth matters way more than HBM capacity. For inference workloads where memory bandwidth matters the most, 4-hi HBM provides the best $/bandwidth and therefore lowest cost per token. If China continues to make more and more revolutionary model architecture innovations, soon it could potentially 0Hi HBM stacks. [ ](https://newsletter.semianalysis.com/p/long-live-the-short-king-why-4-hi) [ Long Live the Short King: Why 4-hi HBM Wins ](https://newsletter.semianalysis.com/p/long-live-the-short-king-why-4-hi) Myron Xie, Bryan Shan, and 3 others · Sep 13 [ Read full story ](https://newsletter.semianalysis.com/p/long-live-the-short-king-why-4-hi) Moreover, on B300 and week-0 stack, moving the Engram table back to HBM did not improve results and stay within run-to-run variance. This is a result of the work optimizing DRAM offload, such as async, and overlap. SSD Offloading On B200, we created a unoptimized vLLM fork and stored the Engram tables in memory-mapped files on local SSDs. File backing lets the OS reclaim table pages when other applications need RAM. Pages already cached in memory can be served without reading the SSD again. Furthermore, note that we were unable to turn on GDS The unoptimized SSD implementation changes how rows reach the GPU. It copies row IDs to the CPU, deduplicates them, gathers the requested rows into pinned buffers, copies those rows back to the GPU and dequantizes them. This work runs between segments of the GPU execution graph. Native UVA performs row selection and dequantization directly on the GPU, avoiding the CPU round trip. A warm filesystem cache removes physical SSD reads, but leaves the coordination, row gathering and transfers. This is why a file already cached in RAM can still perform worse than a pinned DRAM table. The comparison measures the whole serving path; it does not separate the time spent on each of these operations. B200 DRAM dominates both measured SSD serving curves in total tokens per dollar and P90 interactivity. Near 125 tokens/s/user, DRAM delivers 121 million total tokens per dollar versus 52 million for SSD. For production serving, SSD offloading is likely not worth the tradeoff. On the B200 configurations we measured, SSD offloading loses on both measures: every observed SSD point has a DRAM alternative that delivers higher P90 interactivity and more total tokens per dollar. The only points where Cheaper storage does not automatically produce a cheaper inference service. Moving Engram to SSD leaves the same four expensive GPUs and the rest of the server in place. Reclaiming RAM only creates an economic benefit if it enables a cheaper server configuration or additional useful capacity. The current unoptimized path provides neither benefit, and the filesystem cache still consumes RAM when table pages are resident. Next we will look at the mechanisms and specific implementation of ngrams from DeepSeekv4.1 Flash, LongCat, Qwen3.8 Flash Next.
Another new TOD project takes shape in Hicksville
Developers and construction crews mark topping off of $58 million Fieldstone at North Broadway, a transit-oriented mixed-use project in Hicksville.