Intel envisions MESO logic devices superseding CMOS tech in ten years

Cal Jeffrey

Posts: 4,482   +1,609
Staff member
Why it matters: Faster and more energy-efficient computing is becoming harder to develop. Fortunately, chip researchers are looking well beyond current technologies and have prototypes that will replace our current CMOS transistors ten years from now.

As physical limitations in chip architecture are reached, it becomes increasingly more important for manufacturers to develop new ways to overcome those limits. Quantum computing is one step toward that goal. Intel hopes it has the next step in its MESO technology.

Researchers from Intel, the University of California at Berkeley, and Lawrence Berkeley National Laboratory (LBNL) have published a paper in Nature titled “Scalable energy-efficient magnetoelectric spin-orbit (MESO) logic.” In it, the scientists explain how MESO-based devices have the potential to lower voltage by a factor of 5 versus complementary metal-oxide-semiconductor (CMOS) technology and increase energy efficiency 10-30 fold “when combined with ultralow sleep state power [sic].”

Currently, Intel is still working on CMOS scaling, but the researchers see MESO logic driving growth in the post-CMOS era. They expect their technology to emerge in the next decade.

“We are looking for revolutionary, not evolutionary, approaches for computing in the beyond-CMOS era. MESO is built around low-voltage interconnects and low-voltage magneto-electrics. It brings together quantum materials innovation with computing. We are excited about the progress we have made and are looking forward to future demonstrations of reducing the switching voltage even further toward its potential.”

Intel has already developed an experimental MESO-device prototype using magneto-electric materials developed by Ramamoorthy Ramesh at UC Berkeley and the LBNL. The technology uses quantum materials at room temperature to produce “spin-orbit transduction” effects.

“MESO is a device built with room temperature quantum materials,” said Senior Staff Scientist Sasikanth Manipatruni. “It is an example of what is possible, and hopefully triggers innovation across industry, academia, and the national labs. A number of critical materials and techniques are yet to be developed to allow the new type of computing devices and architectures.”

While the experimental prototype shows promising results, the technology is in its pre-infancy. Much more research is needed and practical devices are still quite a way off — at least a decade, according to Ramesh.

“There’s still a lot of work to be done,” he told Berkeley Lab. “Your computer today operates at 3 volts. This device in the Nature paper proposes something at 100 millivolts. We need to understand the physics a lot better.”

Permalink to story.

 
I am all for cutting-edge research. However, I cannot help be reminded of research into nuclear fusion when reading this article. Back in the 1970's, fusion power was 20-years away. Well, those 20-years came and went and we still don't have fusion power today.

I did not read the original paper that this article refers to, however, the statements in this article make me think they have no clue if or when it will be ready.

I once read somewhere that if you ask someone when something will be ready and they tell you five years or more, what the person you asked is really saying is that they have no clue when it will be ready.
 
"In it, the scientists explain how MESO-based devices have the potential to lower voltage 400 percent"

In my mind dropping voltage 100% would result in 0 voltage. And then dropping the voltage another 300% would triple the voltage on the negative side. What exactly is the baseline here?

"versus complementary metal-oxide-semiconductor (CMOS) technology"

That tells me nothing. There are no numbers in that quote.

"and increase energy efficiency 10-30 fold “when combined with ultralow sleep state power [sic].” "

And here the power saving could all be in the sleep state.
 
It gets even worse. The use this phrase twice:

"“MESO is a device built with room temperature quantum materials,” said Senior Staff Scientist Sasikanth Manipatruni."

The phrase "quantum materials" means absolutely nothing. All materials of every size are made of quanta, which are molecules and atoms, baryons, leptons, and the fundamental quanta, the photon itself. We've detected nothing smaller. So what is a quantum material? It's a catch-phrase, just like "quantum computing", which is still stuck on the electron and not "quantum" at all. Quantizing the wrong particle doesn't make it "quantum computing". It's a sales pitch and that's it.

On top of that, these guys don't even know what magnetism is to begin with, or else they wouldn't be chasing the wrong quanta in everything they've ever done. This project is and will be a failure in this form, since there's simply no physics behind it.
 
"In it, the scientists explain how MESO-based devices have the potential to lower voltage 400 percent"

In my mind dropping voltage 100% would result in 0 voltage. And then dropping the voltage another 300% would triple the voltage on the negative side. What exactly is the baseline here?
Yeah, I'm surprised Techspot didn't pick up on that. 1.0v - 400% = -3v. Impressive indeed! If the article meant "used only 1/5th of the voltage", that's -80% not -400%. The rest sounds like another "breakthrough battery article"...
 
I thought that computers today ran, not at 3 volts, but at voltages like 1.15 volt or 1.4 volt.
 
Back