Research Bits: Apr. 15


Shape-morphing OLED panel with built-in speaker Researchers from Pohang University of Science and Technology (POSTECH) developed a flexible OLED panel that can freely transform its shape while simultaneously functioning as a speaker. The design is based on a based on a specialized ultra-thin piezoelectric polymer actuator that when integrated into a flexible OLED panel enables electrically ... » read more

Board-Level Packaging Method For Device Encapsulation To Enable Water Immersion Cooling


A new technical paper titled "Thermally Conductive Electrically Insulating Electronics Packaging for Water Immersion Cooling" was published by researchers at University of Illinois, Urbana, University of Arkansas and UC Berkeley. Abstract "Power densification is making thermal design a key step in the development of future electrical devices. Systems such as data centers and electric vehicl... » read more

ReRAM-Based, In-Memory Implementation Of Stochastic Computing


A new technical paper titled "All-in-Memory Stochastic Computing using ReRAM" was published by researchers at TU Dresden, Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI), Case Western Reserve University, University of Louisiana at Lafayette and Barkhausen Institut. Abstract "As the demand for efficient, low-power computing in embedded and edge devices grows, tradit... » read more

Advanced Packaging Fundamentals for Semiconductor Engineers


Advanced packaging is inevitable. Large systems companies and processing vendors already are working with various types of highly engineered packaging. The rest of the semiconductor industry will follow at some point, whether they're designing their own packages, developing the tools, processes, materials, and methodologies to create them, or developing components that will be used inside of th... » read more

Chip Industry Week In Review


Don't have time to read this? Check out Semiconductor Engineering's Inside Chips podcast.  The U.S. Department of Commerce is investigating TSMC for potential export control violations involving Huawei chips, reports Reuters. The probe follows TechInsights' teardown of a Huawei AI accelerator chip last year. The foundry, meanwhile, maintains it has not shipped any chips to Huawei since 2020... » read more

Are You Ready For AI?


The semiconductor industry sits at the heart of the artificial intelligence revolution, providing the essential computational foundation that powers AI's rapid evolution. However, new research suggests that while AI adoption is accelerating globally, strategic implementation remains a significant challenge across industries. According to the recently released Arm AI Readiness Index Report, 8... » read more

Lego-Style Software For Automotive And Industrial Chiplet Systems?


Chiplets are a key topic in the semiconductor industry today, as they offer the potential to greatly increase the performance and flexibility of chips. The current focus is primarily on implementation, in particular on the architecture and the development of die-to-die interfaces that enable efficient communication between the individual chips. These technologies hold out the promise of meeting... » read more

Delivering Breakthrough Performance And Power Efficiency With PCIe 6


PCI Express (PCIe) has been the backbone of high-speed data transfer in computing systems for nearly two decades. With each iteration, PCIe has not only increased data bandwidth but also introduced innovations aimed at enhancing power efficiency—a critical factor in today's energy-conscious computing environment. PCIe 6 introduced significant technical improvements that optimize power consump... » read more

Chiplet Tradeoffs And Limitations


The semiconductor industry is buzzing with the benefits of chiplets, including faster time to market, better performance, and lower power, but finding the correct balance between customization and standardization is proving to be more difficult than initially thought. For a commercial chiplet marketplace to really take off, it requires a much deeper understanding of how chiplets behave indiv... » read more

Implementing AI Activation Functions


Activation functions play a critical role in AI inference, helping to ferret out nonlinear behaviors in AI models. This makes them an integral part of any neural network, but nonlinear functions can be fussy to build in silicon. Is it better to have a CPU calculate them? Should hardware function units be laid down to execute them? Or would a lookup table (LUT) suffice? Most architectures inc... » read more

← Older posts Newer posts →