category: Low Power-High Performance

category: Auto, Security & Edge AI

Research Bits: May 26
By Jesse Allen - 26 May, 2026 - Comments: 0

Simultaneous energy generation and emission Researchers from the Institute of Science Tokyo and University of Osaka designed an organic semiconductor device that can both generate electricity from light and emit bright visible light. The researchers used two multiple-resonance thermally activated delayed fluorescence (MR-TADF) molecules, v-DABNA and QAO, in a simple layered structure. Their... » read more

Research Bits: May 19
By Jesse Allen - 19 May, 2026 - Comments: 0

Programmable PIC Researchers from the University of Washington designed a low-power programmable photonic integrated circuit that is electrically reconfigurable and can be mass-produced. “This optical chip could help to accelerate the prototyping cycle while reducing power consumption for applications like AI computing. Our study is also the first time someone has shown that these kinds o... » read more

Why Vision LLMs Force A Rethink Of Edge AI Hardware
By Athish Rao - 14 May, 2026 - Comments: 0

As vision-centric large language models move on-device, performance measured in raw TOPS is no longer enough. Architectures need to be built around real workloads, memory behavior, and sustained utilization, especially at the edge. Vision LLMs are changing the edge AI equation For the last decade, most edge AI silicon has been built to do one job extremely well: run convolutional networks for... » read more

SOCAMM2: Bringing LPDDR5X Benefits To AI Servers
By Piero Bianco - 14 May, 2026 - Comments: 0

The rapid scaling of artificial intelligence is reshaping nearly every dimension of data center design. While much of the focus has been on GPUs, accelerators and advanced packaging, another constraint is emerging as equally critical: power. As AI models grow larger and more complex, power consumption, not raw compute, is increasingly the limiting factor in system scalability. Modern AI work... » read more

Vision-Language-Action Models Arrive
By Mike Leonard - 14 May, 2026 - Comments: 0

The AI model type capturing the most attention across robotics and autonomous vehicles right now is the vision-language-action model, or VLA. At embedded AI conferences this year, particularly the recently held Embedded Vision Summit, VLAs were a main topic of discussion – not as a research curiosity, but as the architecture that teams building autonomous systems are actively targeting. If yo... » read more

Introducing “The Architecture Speaks”
By Jade Alglave - 14 May, 2026 - Comments: 0

What are specifications used for? How do you use them? Are they intelligible? These questions are at the heart of the project that produces a new tool called "The Architecture Speaks". This is an experimental chatbot tool built on generative AI that aims to provide quick answers to complex questions about the Arm architecture. It also provides links to the Arm Architecture Reference Manual. Th... » read more

Chiplets Need A New Workflow
By Ann Mutschler - 14 May, 2026 - Comments: 0

Key Takeaways: Chiplet design turns semiconductor development into a system-level problem, requiring coordinated workflows across design, packaging, verification, test, and reliability. Successful chiplet workflows must handle multi-physics challenges — especially thermal, mechanical, power, and signal integrity — early enough to reduce costly failures before assembly and tape-out. ... » read more

Structured Or Unstructured Meshes: What Works Best For Turbomachinery CFD
By Veena Parthan - 14 May, 2026 - Comments: 0

In computational fluid dynamics (CFD), meshing is a critical step for achieving reliable simulations, especially when combined with a robust solver strategy. As turbomachinery blade geometries become more intricate and design cycles shorten, traditional meshing approaches are often not enough. To keep pace, we must adopt advanced methodologies, and more importantly, quantify their impact on res... » read more

Flash Getting Stacked High-Bandwidth Version
By Bryon Moyer - 14 May, 2026 - Comments: 1

Key takeaways: A new HBF 3D flash stack is similar to HBM for use in AI processing. HBF capacity will be much higher, allowing static storage of AI model weights, with optimized read speed. Samples are due out later this year, with accelerators featuring it coming out next year. AI inference using modern models requires billions of parameters, and moving them to where they c... » read more

Gates Add Functionality, But Wires Create Problems
By Brian Bailey - 14 May, 2026 - Comments: 0

Key takeaways: While transistors see continuous improvement, wires keep getting worse because of the smaller geometries and larger chip sizes. There are limited ways to avoid such problems, but the biggest impact will come from floorplanning. Analysis today is not adequate. New developments, such as backside power and 3D integration, provide temporary relief but new materials are a d... » read more

← Older posts Newer posts →

  • Sponsors

About

Navigation

Connect With Us

Copyright ©2013-2026 SMG   |  Terms of Service  |  Privacy Policy