MACs Are Not Enough: Why “Offload” Fails


For the past half-decade, countless chip designers have approached the challenges of on-device machine learning inference with the simple idea of building a “MAC accelerator” – an array of high-performance multiply-accumulate circuits – paired with a legacy programmable core to tackle the ML inference compute problem. There are literally dozens of lookalike architectures in the market t... » read more

2025: So Many Possibilities


The stage is set for a year of innovation in the chip industry, unlike anything seen for decades, but what makes this period of advancement truly unique is the need to focus on physics and real design skills. Planar scaling of SoCs enabled design and verification tools and methodologies to mature on a relatively linear path, but the last few years have created an environment for more radical... » read more

What’s The Best Way To Sell An Inference Engine?


The burgeoning AI market has seen innumerable startups funded on the strength of their ideas about building faster, lower-power, and/or lower-cost AI inference engines. Part of the go-to-market dynamic has involved deciding whether to offer a chip or IP — with some newcomers pivoting between chip and IP implementations of their ideas. The fact that some companies choose to sell chips while... » read more

AI Won’t Replace Subject Matter Experts


Experts at The Table: The emergence of LLMs and other forms of AI has sent ripples through a number of industries, raising fears that many jobs could be on the chopping block, to be replaced by automation. Whether that’s the case in semiconductors, where machine learning has become an integral part of the design process, remains to be seen. Semiconductor Engineering sat down with a panel of e... » read more

Using Test And Metrology Data For Dynamic Process Control


Advanced packaging is transforming semiconductor manufacturing into a multi-dimensional challenge, blending 2D front-end wafer fabrication with 2.5D/3D assemblies, high-frequency device characterization, and complex yield optimization strategies. These combinations are essential to improving performance and functionality, but they create some thorny issues for which there are no easy fixes. ... » read more

Semiconductor Manufacturing’s Transformational Challenges


Semiconductor manufacturing is going through massive transformational challenges driven by strong demand for advanced computing, fueled by AI, cloud, the electrification of the economy, and the need for compute power in data centers to support these applications. With the slowdown of Moore’s Law, more compute power will not be achieved by just increasing transistor density. Not only is Moo... » read more

Do More With Less In Semiconductor Manufacturing


The recent resolution of labor disputes sheds light on a universal concern: the balance between automation and workforce dynamics. These situations mirror a challenge faced in semiconductor manufacturing—embracing AI without displacing the people driving the industry. Moving beyond automation fears US port workers expressed concerns about automation technologies, such as autonomous cranes a... » read more

Accelerating The AI Economy Through Heterogeneous Integration


The world is rapidly transitioning from an internet economy to an AI economy. In the internet economy, we stayed constantly connected to the internet 24/7 through our smartphones, PCs, and IoT devices. However, in the AI economy, every aspect of our lives is interwoven with artificial intelligence. You may already be familiar with AI tools such as ChatGPT or Google Gemini, which answer question... » read more

Integrating Ethernet, PCIe, And UCIe For Enhanced Bandwidth And Scalability For AI/HPC Chips


By Madhumita Sanyal and Aparna Tarde Multi-die architectures are becoming a pivotal solution for boosting performance, scalability, and adaptability in contemporary data centers. By breaking down traditional monolithic designs into smaller, either heterogeneous or homogeneous dies (also known as chiplets), engineers can fine-tune each component for specific functions, resulting in notable im... » read more

Why PCIe And CXL Are Essential Interconnects For The AI Era


As the demand for AI and machine learning accelerates, the need for faster and more flexible data interconnects has never been more critical. Traditional data center architectures face several challenges in enabling efficient and scalable infrastructure to meet the needs of emerging AI use cases. The wide variety of AI use cases translate into different types of workloads. Some require high ... » read more

← Older posts Newer posts →