How Intel will keep Moore’s Law for years to come

Moore’s law, the observation that the number of transistors on a computer chip doubles every 24 months, has taken a toll when miniature power drops. But chip giant Intel has taken a course to keep the idea alive with a plan to pack 50 times as many transistors on processors as is possible today.

The advancement of Moore’s Law, named after Intel co-founder Gordon Moore, has spread chips from expensive mainframes in the 1960s to personal computers in the 1980s and now to smartphones, watches, cars, TVs, washing machines and almost everything with electric power.

Moore’s Law has worked by shrinking transistors, the elements for processing data on a chip. Intel has plans to shrink them, but also to increase the density by stacking chips in multilayer packages.

“We think there’s going to be a lot more transistor density,” he said. said Intel Chief Architect Raja Koduri, in a speech Monday before the Hot Chips Conference on Innovative Processor Revelations. “The vision will play out over time – maybe a decade or more – but it will play out.”

Koduri’s optimism reflects the excitement of many other companies at Hot Chips, an engineering conference where researchers detail progress. AMD, Nvidia, Google, Microsoft, IBM, and a gaggle of startups showed ways to advance both general-purpose chips and those dedicated to tasks such as artificial intelligence, graphics, and networking.

How Intel expects to deliver chip tracking

Koduri described several steps to cram more transistors into a chip than possible with 10nm chips like its Tiger Lake processor arriving in laptops this fall. First, the most traditional approach will come, reducing transistors and pushing them closer together. That will triple the transistor density, Koduri predicted.

Next up are new transistor designs that continue with the current transformation of transistors of flat circuit elements into 3D structures. These steps, called nanowires and stacked nanowires, would need the triple density.

Then come packaging innovations, with chips stacked in a layer cake of processor elements. That should double again. The total math brings the density up by about a factor of 50.

Years of Intel difficulties

Intel’s optimism contrasts with difficult times in which Moore’s Law clings.

Intel, once the undisputed leader in chip manufacturing, has struggled in recent years. The move from a manufacturing process with transistor functions from 14 nanometers to later 10nm took five years instead of two. A nanometer is a billionth of a meter, and with circuit elements 14nm wide, Intel can pack about 7,000 across the width of a human hair.

Following this, Intel slowed down its move from 10nm to 7nm manufacturing by six months, and Apple dumped Intel chips from its Macs. To help customize, Intel has adopted a more flexible design process that allows it to rely more on other chipmakers like its top rival, Taiwan Semiconductor Manufacturing Corp.

Moore’s Law, but at what cost?

TSMC, which moved to 7nm manufacturing about two years ago and makes Apple’s iPhone chips, declared last year “Moore’s Law is well and live.” But unlike in the past, Moore’s Law steps now create new costs for companies that want to use the most advanced production processes.

Microsoft’s Xbox One in 2013, Xbox One X in 2017, and Xbox Series X coming this year all have chips about the same size, which in the past would mean that the chips cost about the same price. Now, though, “it’s significantly more expensive for the latest,” said Microsoft chip designer Jeff Andrews.

Another challenge besides cost is that new chips often only accelerate specific computer actions. This is handy for tasks such as artificial intelligence and graphics, but it makes life harder for software programmers who have to reckon with processors that work in different ways.

Intel is trying to bridge this chip gap with a new software layer called the oneAPI. It’s a remarkable move: Intel is a hardware specialist, but it includes software as an essential step in making its chips useful.

“More often, hardware architecture teams need to be assembled from software experts,” Koduri said.

New chip ideas

At Hot Chips, retail processors also host a host of innovations. Among the largest:

• Intel’s Tiger Lake processor uses a new incarnation of energy-saving technology called DVFS, as dynamic voltage and frequency scaling. Different parts of the chip can run faster for high priority tasks or slower to save power. Intel now juggles the priorities between its multiple processor cores, the memory system, and the communications that connect it all together.
• AMD’s competing Ryzen 4000 series chips, code named Renoir and now arriving in PCs, are the first chips with eight processing cores for super-thin laptops. AMD had initially planned a six-core design, but realized that a careful design eight could fit for better performance on tasks such as video and photo editing, said architect Sonu Arora. They use half the power for a certain level of performance like their predecessors.
• IBM’s Power10 processors, which have 18 billion transistors and are thanks to massive Unix servers arriving next year, can be bundled into one powerful server with up to 240 processing cores. In addition, a “pod” of interlinked servers can share as much as 2 petabytes of memory. This is useful for massive business computer challenges such as data mining and inventory management.
• Startup Lightmatter has unveiled its Mars chip for accelerating AI work such as image recognition. It powers over a billion conventional transistors with tens of thousands of components that use light instead of electricity to transmit data and perform calculations. The idea behind this photonic technology is to reduce energy consumption.