Newsdesk
Engineering
DevOpsEngineering EditionSecurityExploits, Breaches & AdvisoriesWebBrowsers, Frameworks & the PlatformReleasesWhat Shipped This WeekMobileiOS, Android & Cross-PlatformOpen SourceLicences, Governance & Supply Chain
Data & AI
DataEngines, Pipelines & Query PlansPapersPapers Worth ReadingAIEngineering Edition
Industries
FintechRails, Regulation & Banking TechChipsFabs, Process Nodes & AcceleratorsClimateGrid, Storage & the Transition
Enterprise Systems
Supply ChainFinance, Manufacturing, Warehousing & Enterprise AIHR TechSuccessFactors, payroll and people systems
Go-to-Market
MarketingSearch, Content & DemandSalesB2B Go-to-Market & Retention
Longform
ReadingFrom the Blogs Worth FollowingCraftArchitecture, Practice & Running Teams
DesignIndiaAll stories

Chips & Silicon

11 August 2026

69 new posts across 3 semiconductor sources — 3 worth your time.

One week of Chips & Silicon, 3 stories, as published.

Memory & Storage

Chip Industry Week in Review

SanDisk and SK hynix have released the High Bandwidth Flash (HBF) specification, defining performance expectations, the xPU-HBF host interface, and packaging guidelines for die stacks. Additionally, Samsung demonstrated concept models of zHBM, which utilizes vertical HBM stacking.

Why it matters — Establishes the initial standardization framework for HBF host interfaces and packaging, alongside vertical HBM stacking concepts to address AI memory density bottlenecks.

Semiconductor Engineering · 11 August 2026 · Read the original →

Supply & Shortage

Nvidia reportedly testing lower memory configs of Rubin Ultra as memory shortage bites back — designs tested include as little as 192 GB and step back to HBM4

Nvidia is reportedly testing lower memory configurations for its Rubin Ultra AI accelerators, scaling designs down to as little as 192 GB. The company is also stepping back to HBM4 in response to ongoing high-bandwidth memory shortages.

Why it matters — Demonstrates how severe HBM supply constraints are forcing leading AI chip designers to compromise on memory capacity and architecture targets.

Tom's Hardware · 11 August 2026 · Read the original →

Packaging

Chiplet Architectures as a Practical Path to Scalable Automotive Compute

The automotive industry is transitioning from distributed electronic control units (ECUs) to centralized and zonal compute architectures. This shift relies on powerful central processors built using chiplet architectures to deliver scalable, high-performance automotive computing.

Why it matters — Highlights the automotive sector's migration toward chiplet-based central processors as the primary path to scale compute while replacing legacy distributed ECUs.

EE Times · 11 August 2026 · Read the original →

Nothing like letting the supply chain make your architectural decisions for you.

3 stories, every Tuesday

Published here every week. Follow by RSS to get it as it lands.

Subscribe by RSS All issues
← Previous issue Next issue →