Angus Pacala

Why 28nm is a Half Node of 32nm: Incremental Scaling: The transition from 32nm to 28nm represents a modest reduction in feature size, typically achieved through an optical shrink or process optimizations rather than a complete overhaul of the manufacturing process. This is characteristic of a half node, where the improvements are incremental rather than a full leap to a new node (like 32nm to 22nm). Process Characteristics: The 28nm node, as developed by foundries like TSMC, often used similar manufacturing techniques as 32nm (e.g., high-k metal gate technology and 193nm immersion lithography) but with refinements to improve transistor density, power efficiency, or performance. For example, TSMC’s 28nm process was a refinement of the 32nm process, offering ~10-15% better power efficiency and slightly higher transistor density without major changes to the underlying technology. Industry Context: In the semiconductor industry, half nodes like 28nm are used to extend the life of a process technology, providing chip designers with better performance or efficiency while the foundry prepares for the next full node (e.g., 22nm or 20nm). The 28nm node was particularly significant because it was widely adopted and remained a workhorse node for many years due to its cost-effectiveness and maturity. Not a Full Node Jump: A full node transition (e.g., 32nm to 22nm) typically involves a significant reduction in feature size (roughly 0.7x scaling), new materials, or transistor architectures (e.g., moving from planar to FinFET). In contrast, 28nm was more of an evolutionary step from 32nm, with smaller tweaks to lithography, design rules, or process optimizations, fitting the definition of a half node. Example: TSMC’s 32nm and 28nm processes both used planar transistor technology and high-k metal gate (HKMG) processes. The 28nm node improved on 32nm by tightening design rules and optimizing the process, resulting in better power efficiency and performance without a major architectural shift. This contrasts with the jump from 28nm to 20nm or 16nm, which involved more significant changes (e.g., introduction of FinFETs in 16nm). Summary: Yes, 28nm is considered a half node of 32nm because it represents a refined, incremental improvement over the 32nm process, focusing on better density, power, or performance without the major technological leap of a full node.

57

10 Comments

Python and pytest for hardware testing offer a fast, modern, and free approach that any test engineer can adopt. pytest-f3ts streamlines test execution to keep projects consistent from early test development to mass production!

5

I work closely with several leading autonomy and robotics companies in the Bay Area 🚗🤖 and regularly connect with engineers across planning, perception, simulation, and robotics systems. One trend I’m seeing across many teams: strong systems engineers are increasingly in demand. The stack (often C++ / Python) matters, but what matters more is deep domain experience in robotics or autonomous systems, particularly around safety, evaluation, validation, and fault handling. If you’re working in this space or curious about what’s happening across AV and robotics, feel free to reach out. #robotics #autonomoussystems #physicalAI

28

4 Comments

One of the most common objections I hear is: “We can’t use AMRs, our gangways are too narrow.” That’s exactly why we built our LightWeight AMR. Designed for fast, high-frequency movement and tight aisles, it can navigate in ~1.0 to 1.2 meter gangways, where traditional movement becomes messy: ● people + trolleys + forklifts ● blind corners ● constant micro-stoppages Where it fits well: ● Electronics manufacturing (totes/trays, frequent cell changes) ● Automotive sub-assemblies (small parts replenishment) ● FMCG / eCommerce / warehousing (cartons/totes in tight pick aisles) Our philosophy is: The robot should adapt to the factory layout & not the factory being forced to adapt to the robot. If your aisles are tight: What’s the narrowest gangway you regularly move material through today?

19

1 Comment

– Always providing the latest updates on tech innovations and trends. #TechNews – Waymo discusses the significance of their robotaxi expansion in San Francisco. #Waymo #Transportation – Recent jury verdict reveals Apple owes Masimo $634 million for patent infringement. #Apple #PatentInfringement – Disney and YouTube TV have reached an agreement to end service blackout issues. #Disney #YouTubeTV – Leaked documents uncover payments OpenAI makes to Microsoft. #OpenAI #Microsoft – Tesla shares a detailed safety report following requests for more data on their autonomous driving systems. #Tesla #AutonomousDriving – Comprehensive list released discussing tech layoffs of 2025. #TechLayoffs #Startups – WhatsApp plans to enable third-party chat integration in Europe shortly. #WhatsApp #Apps – Apple adjusts its app review guidelines, restricting apps that share user data with third-party AI. #Apple #Privacy – ChatGPT rolls out pilot group chats in several countries including Japan and New Zealand. #ChatGPT #AI

1

5 Comments

🚗 No Room for Error: How OCXOs Guide Autonomous Vehicles When a vehicle drives itself, timing isn’t just critical—it’s everything. Behind every safe lane change, smooth stop, and real-time decision lies an unsung hero: the Oven-Controlled Crystal Oscillator (OCXO). Here’s how OCXOs enable autonomy: 📍 Precise Sensor Sync – Coordinates LiDAR, cameras & radar with microsecond accuracy 🌡️ Temperature Resilient – Stable performance from freezing mornings to hot asphalt 📡 GNSS Reliability – Enhances GPS accuracy for exact positioning 📊 Clean Signal Integrity – Low phase noise ensures trustworthy navigation data Where OCXOs Drive Impact: • Sensor fusion for 360° awareness • Real-time path planning & obstacle avoidance • Vehicle-to-everything (V2X) communication At Dynamic Engineers, we engineer the timing solutions that help autonomous vehicles navigate our world with confidence—because on the road to autonomy, every nanosecond matters. #AutonomousVehicles #OCXO #AdvancedDriving #SensorFusion #SmartMobility #EmbeddedSystems #DynamicEngineers #Innovation #EverythingRF

11

11 Comments

While for Robot taxis, it is almost always ok to rely on the cloud, for real level 4 systems, you are going to be able to deal with the all O.D.D locally in the car, that means, having a local map always ready to take over, a lot of regulation of traffic understanding on board that Robot taxis don't actually need, as they use data telling them the most commonly driven path for places they know very very well. Level 4 is not Robotaxi, by far, and if you understand that a Level 4 car needs to deal with unknown roads, then you understand that you are going to pile multiple systems, dedicated to solve multiple different problems that robottaxi does not meet. The multiple systems on Chip of Athos Silicon was design to solve Level 4, and many other problems that requires multiple systems to work together without steeling the performance of each other, like most SDV systems are doing with VMs. The #mSoC is the solution if you have analyzed properly the problem you got to solve for really safe ADAS. https://lnkd.in/gSnFCpbM

12