Eviden’s Post

This is a super useful perspective for those that don't know what a silicon chip designer works with!!! NYSE Wired - AI Factories TSMC - AMD - NVIDIA - Cerebras Systems - Groq - Celestial AI - Ayar Labs - Lightmatter - Bill Tai - John Furrier - David Vellante #Silicon #ChipDesign #AtomicScale #NanoEngineering #SubMicro #Science #SmallerThanYouThink

NextSilicon reveals new processor chip in challenge to Intel, AMD: https://lnkd.in/eVCEe4YA (Reuters) - NextSilicon, an Israeli startup whose computing chips are being evaluated by U.S. national labs, on Wednesday said it is developing a central processor that it hopes will rival Intel Corporation and AMD and help it compete with NVIDIA's systems. NextSilicon's has raised $300 million in funding and its flagship "Maverick-2" chip is designed to speed up precision scientific computing tasks such as modeling nuclear weapons. That field was once dominated by Nvidia, but as Nvidia has focused its attention on lower-precision computing tasks such as artificial intelligence, startups like NextSilicon have tried to take advantage of the AI giant's shift. #semiconductor #manufacturing #technology #innovation #chips  #semiconductormanufacturing #advancedtechnology #engineering #lithography #nanometer #research #development #AI #EUV #DUV #mobileprocessors #semiconductorindustry #semiconductormarket

🧠 Adaptive Architecture, Proven Performance: Maverick-2 Technical Disclosure For decades, computing has forced you to adapt to the hardware. What if hardware adapted to you instead? Today: Maverick-2's dataflow architecture makes this real - hardware that adapts and optimizes to your applications in real-time, not the other way around. The official performance results prove it: Built on our revolutionary Intelligent Compute Architecture #ICA, Maverick-2 delivers up to 10× higher performance than leading GPUs while using less than half the power, and it does so out-of-the-box, with no code changes required. Already powering breakthroughs at Sandia National Laboratories and other leading institutions, Maverick-2 is transforming what’s possible in #AI, #HPC, and scientific research. Plus: We’ve unveiled #Arbel, our enterprise-grade #RISCV core built on TSMC's 5nm process—surpassing current RISC-V competition and the latest generation cores from industry leaders. Proof that NextSilicon innovates at every level of the stack. 🎥 Watch the full presentation, hosted by Elad Raz, Eyal Nagar & Ilan Tayari, and witness how these innovations are redefining performance and efficiency. 👉 https://lnkd.in/dbGMKKm4 #NextSilicon #Maverick #Innovation #Milestone #IAHPC #TechNews #TechTrends #Israel #StartUpNation

Last week, I shared that every computing breakthrough unlocks human potential we didn't know was waiting. Today, as we reveal more details about Maverick-2 and a few other fun surprises, that potential becomes real. I'll let the technical details speak for themselves—they are pretty impressive. 😉 Instead, I want to focus on what this really means. NextSilicon's Intelligent Compute Architecture delivers more than fast, efficient computing at scale and solutions to the biggest pain points for software porting—though we deliver that too. It's really about accelerating time to discovery and insight. It's about enabling breakthroughs that lift everyone: fusion energy that makes renewable power abundant, precision medicine that makes treatments accessible and affordable to all, discoveries that prove we can build the hopeful, innovative society we believe humanity is capable of becoming. The future of innovation shouldn't be limited by computing architecture. It should be empowered by it. That's the mission. See our full launch below. 👇 Read additional thoughts on today's news by NextSilicon Founder and CEO, Elad Raz here 👉 https://bit.ly/47gpDUP #NextSilicon #Innovation #Dataflow #HPC #AI #ICA #RISCV

🚀 The alternative path for compute Lately I’ve been thinking a lot about what "next-gen compute" really means. It must be a smarter architecture that adapts to how we actually work. NextSilicon just dropped Maverick-2—the masterpiece of eight years work to challenge the fundamental assumptions and tackle the limits of today's computing architectures. 💡 My takeaways: Problem: Organizations building AI and HPC applications face a choice between three fundamentally limited options. Whether it's architectural inefficiency, programming complexity, or prohibitive costs and inflexibility—CPUs, GPUs, and ASICs each force organizations to compromise. Solution: Maverick-2 addresses the fundamental shortcomings of all three traditional approaches. The secret sauce? Dataflow + NextSilicon's unique software layer. Performance: It delivers up to 10x the performance of leading GPUs while consuming as much as 60% less power. 🤔 Details: Traditional processors spend most of their silicon managing this sequential execution like me following a cooking recipe. Dataflow eliminates all of that overhead because there are no instructions to fetch, decode, or schedule like my mom's way of making dumplings. NextSilicon's software layer identifies the most computationally intensive parts of your code and dynamically optimizes the hardware to accelerate them in real-time. If the results hold up, Maverick-2 could mark a major leap in modern compute design. 👉 Elad Raz’s blog is always a delight to read—worth checking out. #AI #HPC #Compute #Architecture #Dataflow #NextSilicon https://lnkd.in/gUbH9sfE

Exciting to see NextSilicon sharing more details on Maverick-2. The company’s Intelligent Compute Architecture -- a non-Von Neumann dataflow architecture -- addresses fundamental shortcomings of traditional CPUs/GPUs/ASICs. The performance data released today speaks volumes: up to 10x the performance of leading GPUs while consuming as much as 60% less power. Achieved with unmodified, out-of-the-box code. Congratulations team!

I’m thrilled to announce the launch of #Maverick-2, NextSilicon’s latest #HPC and #AI accelerator. For the first time, we’re opening the curtain on the technology that was previously under NDA — revealing how NextSilicon's ICA (Intelligent Compute Architecture) based Maverick-2 delivers massive compute acceleration with less power and less effort than leading CPU and GPU solutions. This 1-hour YouTube deep dive gives you a closer look at how Maverick-2 works, what makes it different, and the performance results we’re already achieving. Spoiler Alert: There is a bonus reveal at the end... #RISCV If you’d like to learn more, feel free to reach out directly. And if you’re attending #SC25, I’d be happy to arrange a closed-door meeting with our technical experts and leadership team to explore how NextSilicon can help you unlock your next level of performance.

Intel Corporation is rewriting the rulebook on silicon! 🚀 Their groundbreaking 18A process technology isn't just an incremental step—it's a massive leap forward, set to power the next generation of computing. What's so exciting about it? It all comes down to two revolutionary innovations: RibbonFET™ Transistors: A complete redesign of the transistor (the first in over a decade!) that wraps the gate all the way around the channel for faster switching and incredible power efficiency. PowerVia™ Backside Power: A genius move to put the power lines on the back of the chip, decluttering the data pathways on the front. Think of it as creating a dedicated superhighway for power, leaving more room for data to fly. The result? A huge boost in performance-per-watt that will fuel everything from data centers to the upcoming "Panther Lake" processors for AI PCs. This move is a cornerstone of Intel's strategy to reclaim process leadership and a major win for the U.S. domestic tech industry. We're on the cusp of a new era of on-device AI and efficiency. What are your thoughts on this comeback? #Intel #18A #Semiconductors #Innovation #AI #FutureOfTech #Engineering #Technology

Intel Corporation shows its new "Panther Lake" processor, powered by its 18A semiconductor tech This processor is part of the Intel Core Ultra series 3 and is the first to use the 18A process, marking a significant advancement in semiconductor technology. The Panther Lake processors will be produced at Intel's Fab 52 facility in Chandler, Arizona, and are expected to start shipping later this year. Intel CEO Lip-Bu Tan said the new technology represents a major leap forward in computing and will shape the future for decades. Read more: https://lnkd.in/euGbiVTW 📰 Subscribe to the Daily AI Brief: https://lnkd.in/eq56P3bj #ai #artificialintelligence #ainews #aistartup #aichips