Apple once again surprised enthusiasts and analysts with its announcement of the M1 Ultra, a variant of the M1 Max that effectively fuses two chips into one. The result is a dual-chip design viewed by software as a single piece of silicon. Nvidia delivered similar news at the GPU Technology Conference 2022, where CEO Jensen Huang announced that the company will fuse two of the company’s new Grace CPU processors into a single “Superchip.”

These announcements target different markets. Apple has its sights set on the consumer and professional workstation world, while Nvidia intends to compete in high-performance computing. Yet the divergence in purpose only underscores the broad challenges rapidly bringing the era of monolithic chip design to an end.

Moore’s Law, the tenet that the number of transistors on a chip will double every 18-24 months, has driven the electronics industry for decades. Today, there’s no denying that Moore’s Law is showing its age, with some semiconductor industry leaders going so far as to rewrite its definition. In this era of More-than-Moore, chipmakers are turning to new materials, 3D wafer stacking and heterogeneous integration – die with different manufacturing process nodes and technologies integrated within a single package – to keep driving the pace of advancement.

IBM’s $3-Billion Research Project Has Kept Computing Moving Forward
https://spectrum.ieee.org/nanoclast/semiconductors/nanotechnology/ibms-3billion-research-project-has-kept-computing-moving-forward

Back in 2014, under the looming shadow of the end of Moore’s Law, IBM embarked on an ambitious, US $3 billion project dubbed “7-nm and Beyond”. The bold aim of that five-year research project was to see how computing would continue into the future as the physics of decreasing chip dimensions conspired against it.

Six years later, Moore’s Law isn’t much of a law anymore.

Although the chip industry may not feel as constrained by Moore’s Law as it has in the past, the “7-nm and Beyond” project has delivered important innovations even while some chip manufacturers have seemingly thrown up their hands in frustration at various points in recent years.

One example of this frustration was the decision two years ago by GlobalFoundries to suspend its 7-nanometer chip development.

Of course, EUV has enabled 7-nm nodes, but the aim of IBM was to look beyond that. IBM believes that the foundational element of chips to enable the scaling beyond FinFET will be the nanosheet transistor, which some have suggested may even be the last step in Moore’s Law.

The nanosheet looks to be the replacement to the FinFET architecture, and is expected to make possible the transition from the 7-nm and 5-nm nodes to the 3-nm node.

The value of a semiconductor can be difficult to measure because it involves costs and benefits over time. As market segments feel different pressures, maximizing value is going in several directions.

MIT CSAIL researchers say improving computing technology after Moore’s Law will require more efficient software, new algorithms, and specialized hardware.