Skip to main content

New Samsung and IBM discoveries could one day produce ultra-efficient chips

IBM and Samsung revealed that they are working on a new joint project: The creation of a new semiconductor design.

The goal of the joint efforts of these two companies is to create a new standard of ultra-energy-efficient chips.

The new IBM and Samsung semiconductor design.
Image used with permission by copyright holder

Samsung and IBM made the announcement on the first day of the International Electron Devices Meeting (IEDM) 2021 and followed it up with a press release. The companies report that they have achieved new breakthroughs in their research on manufacturing a brand-new semiconductor design.

Recommended Videos

The innovations that Samsung and IBM are working on involve changes in transistor positioning. In the new design, transistors are stacked vertically on the chip. Current processors and SoCs (system on a chip) have transistors that are positioned flatly on the surface of the silicon. This makes the electric current flow from side to side.

Please enable Javascript to view this content

The new design is referred to as Vertical Transport Field Effect Transistors, or VTFET for short. Stacking the transistors in a perpendicular manner allows for the electric current to flow vertically.

Samsung and IBM discussed the benefits of the new design, highlighting perks both for computing and for the mobility sector. The change in the flow of the electric current should net considerable improvements in terms of energy efficiency. The two tech giants estimate that using VTFET will produce chips that are either twice as fast as their current FinFET counterparts or up to 85% less power-hungry.

Going down the power-saving route in the new chips will be helpful not just for regular day-to-day users, but also for those who utilize their systems to the fullest. Operations such as cryptomining and data encryption may, according to Samsung and IBM, consume much less energy. Aside from all other perks, this might help reduce the enormous carbon footprint left behind by cryptomining.

A 2 nm wafer fabricated at IBM Research's Albany facility. The wafer contains hundreds of individual chips.
A 2 nm wafer fabricated at IBM Research’s Albany facility. The wafer contains hundreds of individual chips. Courtesy of IBM Image used with permission by copyright holder

As our computers grow in power, they also require more and more power. Some of the best processors on the market, especially when paired with equally fantastic graphics cards, demand truly beastly power supplies in order to offer stable performance. It’s no wonder that most companies are looking into energy-saving options. Apple is one such example: The recent M1 chip, although powerful, places a lot of emphasis on energy efficiency. It makes sense for other manufacturers to follow in these footsteps.

Samsung and IBM reference Moore’s Law in their joint press release. Moore’s Law is the principle that states that the number of transistors in an IC chip should double once every two years, improving performance without increasing the size. However, as the two manufacturers point out, engineers are running short on space, so pursuing Moore’s Law is getting trickier. The new design might help extend Moore’s Law into further generations of chips.

Samsung and IBM aren’t the only companies in pursuit of Moore’s Law. Intel made a similar announcement during IEDM 2021 when it revealed new breakthroughs that would fuel the next few years of computing innovations.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
Reviewers agree: Intel’s latest chip is truly ridiculous
Intel's 14900K CPU socketed in a motherboard.

Intel's "Special Edition" KS chips are meant to be over the top. But the latest Core i9-14900KS has just dropped, and it takes things to new heights of insanity.

It's a super-clocked version of the already ludicrous 14900K that sports the same great quantity of cores, but a boost clock that moves even beyond the extremes of the standard 14900K. It can hit an unprecedented 6.2GHz on a couple of cores right out of the box, making it the fastest CPU by clock speed ever unleashed upon the public.

Read more
One of the great rivalries in PCs could collapse
Intel CEO Pat Gelsinger stands in front of a blue and purple background.

Intel always competes against AMD, both make some of the best processors, and the sky is blue -- what else is new? Well, it turns out that one of those statements might not be as true as it was just last week. Intel CEO Pat Gelsinger revealed that Intel's own fabs are open to making chips for some of the biggest companies in tech, and that even includes its long-standing rival AMD.

Gelsinger spoke during the first Intel Foundry Direct Connect in San Jose, California, about the operations of its new fab, which Intel itself refers to as a "more sustainable system foundry business designed for the AI era." Intel has big plans for its foundry, with a goal to beat its biggest rival TSMC in advanced chip manufacturing by 2025. The company reportedly hopes to be making the world's fastest chips already this year with its Intel 18A manufacturing technology. Beyond that, Intel hopes to widen the gap with its new 14A technology in 2026.

Read more
Intel’s new Core Ultra chips needed to be more than this
The Acer Swift Go 14 on a table in front of a window.

A lot has been made of Intel's new Meteor Lake chips. They have a new brand, naming scheme, and even new technology in the form of the NPU (neural processing unit).

And they come at an important time. The stakes have never been higher for Intel, as it faces competition from all sides. They're also an important litmus test for the idea of the NPU, or neural processing unit.

Read more