中文 English

Can Analog Make A Comeback?


We live in an analog world dominated by digital processing, but that could change. Domain specificity, and the desire for greater levels of optimization, may provide analog compute with some significant advantages — and the possibility of a comeback. For the last four decades, the advantages of digital scaling and flexibility have pushed the dividing line between analog and digital closer ... » read more

Zero Dark Silicon


Planning for AI requires an understanding of how much data needs to be processed and how quickly that needs to happen. Nick Ni, senior director of data center AI and compute markets at AMD, talks with Semiconductor Engineering about data bubbles and domain-specific designs, why dark silicon is no longer as useful as in the past, and how to optimize power and performance in both the data center ... » read more

Architecting Faster Computers


To create faster computers, the industry must take a major step back and re-examine choices that were made half a century ago. One of the most likely approaches involves dropping demands for determinism, and this is being attempted in several different forms. Since the establishment of the von Neumann architecture for computers, small, incremental improvements have been made to architectures... » read more

Better Video Compression


Video data is increasing, but bandwidth is not increasing quickly enough. One solution is to compress that data, but the challenge is to do that without impacting resolution. Rob Green, AMD’s senior manager for Pro AV, Broadcast & Consumer, talks about what’s changing in video compression, from new standards to better analytics, virtualization, what impact AR/VR will have, and a compari... » read more

Power Now First-Order Concern In More Markets


Concerns about energy and power efficiency are becoming as important as performance in markets where traditionally there has been a significant gap, setting the stage for significant shifts in both chip architectures and in how those ICs are designed in the first place. This shift can be seen in a growing number of applications and vertical segments. It includes mobile devices, where batteri... » read more

Improving Performance And Simplifying Coding With XY Memory’s Implicit Parallelism


Instruction-level Parallelism (ILP) refers to design techniques that enable more than one RISC instruction to be executed simultaneously in the same instruction, which boosts processor performance by increasing the amount of work done in a given time interval, thereby increasing the throughput. This parallelism can be explicit, where each additional instruction is explicitly part of the instruc... » read more

Design Technology Co-Optimization


Rising complexity is making it increasingly difficult to optimize chips for yield and reliability. David Fried, vice president of computational products at Lam Research, examines the benefits of automated rules to manage the relationship between layout and design requirements on one side, and process flows and rules/checks on the other. Benefits include reduced margin, shortened time to market,... » read more

SOT-MRAM To Challenge SRAM


In an era of new non-volatile memory (NVM) technologies, yet another variation is poised to join the competition — a new version of MRAM called spin-orbit torque, or SOT-MRAM. What makes this one particularly interesting is the possibility that someday it could supplant SRAM arrays in systems-on-chip (SoCs) and other integrated circuits. The key advantages of SOT-MRAM technology are the pr... » read more

Amdahl Limits On AI


Software and hardware both place limits on how fast an application can run, but finding and eliminating the limitations is becoming more important in this age of multicore heterogeneous processing. The problem is certainly not new. Gene Amdahl (1922-2015) recognized the issue and published a paper about it in 1967. It provided the theoretical speedup for a defined task that could be expected... » read more

HBM3: Big Impact On Chip Design


An insatiable demand for bandwidth in everything from high-performance computing to AI training, gaming, and automotive applications is fueling the development of the next generation of high-bandwidth memory. HBM3 will bring a 2X bump in bandwidth and capacity per stack, as well as some other benefits. What was once considered a "slow and wide" memory technology to reduce signal traffic dela... » read more

← Older posts