The Curious Engineer

All blogs / Tech news / industry

Chipstrat

newsletter.chipstrat.com

Independent research on semiconductors, AI infrastructure and business strategy by Austin Lyons.

Tech news / industryCPU / microarchitectureAI / ML

’23’24’25’26
Posts
168 from 20 Dec 2023 to 27 Oct 2026
Posting
weekly (72 in the last year)
Kind
personal blog
Feed
https://newsletter.chipstrat.com/feed

2026 62

  1. Right-Sizing Memory
  2. My Conversation with Dylan Patel and Stacy Rasgon
  3. Optics Primer, Part 4: Coherent Optics
  4. Dell's Tailwinds
  5. 🎙️ The Compute 100: Startups vs. Incumbents in Semis, Intel Foundry, Physical AI, and more
  6. Is EDA Dead?
  7. AI With No Decode (Jev), and What It Means for AI Infrastructure
  8. 🎙️The Race to Build the Next Trillion-Dollar AI Chip Company
  9. A Billion Agents Need Somewhere to Run
  10. Oligopoly Equilibrium
  11. Open Models & American Diffusion; Nvidia & Poolside
  12. NPO State of the Union
  13. Build Your Own DRAM Supply Model
  14. AMD on Agentic CPUs, and Why There's No One Size Fits All
  15. The Hardest Computing Problem Ever, According to Arm's Physical AI Chief
  16. Why Did AMD Buy Taalas?
  17. Will ASML Finally Charge for Its Value?
  18. TSMC could spend $10B more on N2 capacity. Why won’t they?
  19. 🎙️ PicoJool’s Al Yuen: The Case for GaAs VCSELs in Scale-Up Interconnects
  20. The Next Trillion-Dollar Chip Company
  21. High Bandwidth Flash: The Full Report
  22. Micron's Blowout and the Case for a Memory Supercycle
  23. The ISA Doesn't Matter Where It Counts
  24. Are Agentic CPUs a Commodity? It’s Complicated.
  25. Power Moves Into the Package. Empower, PowerLattice, and the IVR Socket
  26. An Interview with Nvidia's Deepu Talla About Physical AI and Robotics
  27. Inside the 800G → 1.6T → 3.2T Race
  28. An Interview with the Gimlet Labs Team About Heterogeneous Inference for AI Agents
  29. Advanced Packaging: Intel's EMIB vs TSMC's CoWoS
  30. VCSELs + 200G Wall In AI Datacenters?
  31. Rethinking Amazon + Not All SaaS Is Dead
  32. TSMC's Margins in Uncharted Territory
  33. An Interview with Meta VP Matt Steiner About Ads Infrastructure
  34. Substrate
  35. An Interview with MatX CEO Reiner Pope About LLM Chips
  36. The Agentic Computer: New S-Curve or Another iPad?
  37. Coherent's Vertical Integration Strategy
  38. Agentic AI Needs CPUs. Whose CPUs?
  39. The Multi-Silicon Era Is Here
  40. GTC 2026 Keynote Debrief
  41. Meta's MTIA Roadmap
  42. Optics Primer, Part 3: Co-Packaged Optics (CPO)
  43. An Interview with Rivian’s Mukund Chavan About RAP1
  44. Broadcom Makes Lasers?
  45. Lumentum and the Laser Bottleneck
  46. xLight. Photons as a Service.
  47. Arista's Ethernet Bet
  48. The Nvidia Business Hiding in Plain Sight
  49. The $200 Billion Bet
  50. Meta’s ROIC Strategy: GEM Now, LLMs Later
  51. The MI450 Waiting Game
  52. This Month in Review (Jan '26)
  53. Intel Q425: Back to Reality
  54. An Interview with Microsoft's Saurabh Dighe About Maia 200
  55. Optics Primer, Part 2: LRO & LPO
  56. Why LIDAR Is Consolidating Now
  57. Right Systems for Agentic Workloads
  58. LiDAR, Explained: How It Works and Why It Matters
  59. Credo’s Reliability Thesis
  60. Intel's Product Marketing at CES: B-
  61. When Frontier AI Goes From Cloud to Desk in Five Years
  62. Lithography Economics

2025 51

  1. Right Tool for the Job
  2. Optics Primer, Part 1: Traditional Pluggable Optics
  3. GPU Networking, Part 4: Year End Wrap
  4. Nvidia, AMD, Amkor, Arista @ UBS Tech Conference
  5. Rivian's Silicon & Physical AI
  6. Nvidia Q3 Earnings
  7. Intel is "All In On 14A"
  8. Understanding SkyWater
  9. AMD & Apple Earnings
  10. The Four Horsemen: Google, Microsoft, Amazon, Meta Earnings
  11. When Broadcom’s Customers Start Buying Chip Startups
  12. Intel Q3 2025 Earnings: Doing Fine
  13. SerDes Matters!
  14. TSMC Earnings and Intel Foundry
  15. An Interview With Mohamed Awad About Chiplets
  16. Right-Sized AI Infrastructure. Marvell XPUs
  17. Credo AECs
  18. GPU Networking Basics Part 3: Scale-Out
  19. Building the Chiplet Ecosystem
  20. Eliyan: The Ultimate Chiplet Interconnect
  21. AWS Is Missing the AI Infra Moment
  22. Why GenAI is Good for Google
  23. Nvidia's Climb to $4T and the Peaks Ahead
  24. Intel’s 18A: Misstep or First Step?
  25. Thinking Bots + Nvidia’s Robotics Play
  26. AMD's Strategic Bets
  27. WWDC25: Apple’s Playing a Different Game
  28. Enterprise AI: Nvidia, Dell, and Implications for AMD
  29. Chiplets and the Future of System Design
  30. Anticipatory Intelligence and the Return of Windows
  31. Qualcomm Automotive Primer
  32. The Long Arc of Semiconductor Scaling
  33. Intel Foundry’s Cultural Reboot
  34. Robots That See, Part 2: Vision Transformers
  35. Robots That See, Part 1: The Evolution of Computer Vision
  36. GPU Networking Basics, Part 2
  37. Jensen, We’re With You. But We’re Not There Yet.
  38. GPU Networking Basics, Part 1
  39. Dispelling DeepSeek Myths, Studying V3
  40. Nvidia 4Q25 Earnings
  41. DeepSeek Mixture of Experts & V2
  42. AMD's AI Accelerator Business: Puts and Takes
  43. Reasoning Models and DeepSeek, Semis Implications
  44. DeepSeek Pulls Forward Copilot+ PC adoption
  45. Personal Update
  46. Apple's AI Strategy
  47. CPU Server Growth, Enterprise AI Agent Adoption Challenges
  48. Chipstrat Chat with Jared O’Leary, CEO of SirenOpt
  49. Edge AI at CES 2025
  50. What Happens If TSMC Controls It All?
  51. The Power of Nvidia's Systems Thinking Approach

2024 53

  1. AMD's AI Woes & Trust Crisis
  2. Jensen The Great and His 50% Edge AI Price Cut
  3. Semiconductor Metrology
  4. Arm's Edge AI Strategy: Is It Enough?
  5. Meta & Contextual AI at the Edge
  6. Gelsinger Retires; Intel's Future
  7. Why America Must Invest in Intel Foundry
  8. MatX Prediction
  9. Tokens Per Second Per Watt: A Useful Metric for Edge AI
  10. High-Bandwidth Memory
  11. GenAI for Gaming, Netflix, CapEx
  12. X86 Commoditization and Competition
  13. Etched's Oasis: Creating a Market For Sohu
  14. Android, AI, and Qualcomm Phones
  15. AI Phones: Qualcomm Gets It
  16. Monopsony and TSMC
  17. Groq's Business Model, Part 3: Competing with Google, AWS, Microsoft
  18. Groq's Business Model Part 2
  19. Groq’s Business Model, Part 1: Inference API
  20. Follow Ups: Intel AI PC, Qualcomm, Apple Intelligence, Microsoft Recall, Nuclear Power
  21. Chipstrat Chat with Groq's Mark Heaps
  22. Data Center Energy, National Security, Nuclear Power
  23. Qualcomm + Intel
  24. Intel's AI PC Mirage
  25. IDM Death Spiral
  26. Intel Skips 20A
  27. ZML's Hardware-Agnostic Distributed Inference
  28. Understanding Intel, Part 2
  29. Understanding Intel
  30. Semis Funding Heats Up, Groq, Block and China
  31. AMD's Earnings: Playbook Strength and New Strategic Update
  32. Jack Dorsey's Bitcoin Push: A Noble Effort, But Will It Matter?
  33. Hardware as a Differentiator: Block's Unique Approach to Financial Services
  34. Memory: From Caches to Commodity Cycles
  35. Arm's Journey to Value-Based Pricing
  36. Etched: Silicon Valley's Speedrun
  37. Intel CPU follow-up, Nvidia Superchips, AMD MI300A
  38. Is Intel's CPU Role Diminishing In The Age of AI?
  39. Lisa Su - Steady As She Goes
  40. The Rise of Groq: Slow, then Fast
  41. 🎧 Chipstrat + The Circuit - Episode 70 - NPUs
  42. Qualcomm’s Opportunity
  43. Rethinking AMD
  44. The Future of Edge Inference Hardware
  45. FPGAs - Speed of Hardware, Flexibility of Software
  46. Can AMD Bridge Nvidia’s Software Moat?
  47. 🎧 Chipstrat + The Circuit Podcast
  48. Nvidia and Microsoft - The Impact of LLM-Powered Copilots on Microsoft's Hardware Strategy
  49. Amazon's Nvidia Entanglement
  50. Nvidia’s Risk
  51. GPU Bloat Stifles AI
  52. ASICs: Custom Chips to Improve Performance
  53. GPUs: General-Purpose Parallel Computing

2023 2

  1. CPUs: Hardware, Yet Infinitely Adaptable Through Software
  2. Welcome to Chipstrat