New Flexible RISC-V Semiconductor Has Great Potential

"For the first time, scientists have created a flexible programmable chip that is not made of silicon..." reports IEEE Spectrum — opening new possibilities for implantable devices, on-skin computers, brain-machine interfaces, and soft robotics.

U.K.-based Pragmatic Semiconductor produced an "ultralow-power" 32-bit microprocessor, according to the article, and "The microchip's open-source RISC-V architecture suggests it might cost less than a dollar..." This shows potential for inexpensive applications like wearable healthcare electronics and smart package labels, according to the chip's inventors: For example, "we can develop an ECG patch that has flexible electrodes attached to the chest and a flexible microprocessor connected to flexible electrodes to classify arrhythmia conditions by processing the ECG data from a patient," says Emre Ozer, senior director of processor development at Pragmatic, a flexible chip manufacturer in Cambridge, England. Detecting normal heart rhythms versus an arrhythmia "is a machine learning task that can run in software in the flexible microprocessor," he says...

Pragmatic sought to create a flexible microchip that cost significantly less to make than a silicon processor. The new device, named Flex-RV, is a 32-bit microprocessor based on the metal-oxide semiconductor indium gallium zinc oxide (IGZO). Attempts to create flexible devices from silicon require special packaging for the brittle microchips to protect them from the mechanical stresses of bending and stretching. In contrast, pliable thin-film transistors made from IGZO can be made directly at low temperatures onto flexible plastics, leading to lower costs...

"Our end goal is to democratize computing by developing a license-free microprocessor," Ozer says... Other processors have been built using flexible semiconductors, such as Pragmatic's 32-bit PlasticARM and an ultracheap microcontroller designed by engineers in Illinois. Unlike these earlier devices, Flex-RV is programmable and can run compiled programs written in high-level languages such as C. In addition, the open-source nature of RISC-V also let the researchers equip Flex-RV with a programmable machine learning hardware accelerator, enabling artificial intelligence applications.

Each Flex-RV microprocessor has a 17.5 square millimeter core and roughly 12,600 logic gates. The research team found Flex-RV could run as fast as 60 kilohertz while consuming less than 6 milliwatts of power... The Pragmatic team found that Flex-RV could still execute programs correctly when bent to a curve with a radius of 3 millimeters. Performance varied between a 4.3 percent slowdown to a 2.3 percent speedup depending on the way it was bent.

Sew it into clothing

[jfdavis668]

Control your integrated clothing lights during your next rave!

Re:

[backslashdot]

That can be done with non-flexible chips .. even the smallest CPU will do (millimeter size). The biggest issue has always been power supply needing a fat battery (unless you have a tiny display, which is lame). Second biggest issue is non-flexible WS2812b LEDs or expensive flexible OLED.

Re: Sew it into clothing

[dotslashdot]

Stop fat shaming batteries.

Re: First demo of Doom on a pacemaker in 3..2..1..

[LindleyF]

Dibs on the Beat Forcing Gun.

Cool, but silicon is still king

[serviscope_minor]

A Cortex M0 running at 60kHz will take 0.3 to 3 microwatts of power and takes 0.03mm^2 at 90nm lithography. On a patch that size, the bending radius will not be large.

But this is still super cool!

Re:

[gweihir]

It is a nice demo. It does how viability of a new approach and that is always cool. It is not a product to be used now, many people do not get that on these types of announcements.

For actual application, the micropower Cortex M0 will be preferable, better power, better known, easier to program, established toolchain.

Re:

[algaeman]

Right. It's a demo, so why talk at all about cost? The advantage here is the flexibility for healthcare applications. Saying that IGZO is going to be significantly cheaper than silicon is some straight up BS.

Eh?

[fuzzyfuzzyfungus]

"Our end goal is to democratize computing by developing a license-free microprocessor," seems like an interesting claim to have above the fold when the project involved grabbing someone else's RISC-V implementation, tacking on a custom "ML" acceleration unit because of course it has to have AI(discussed in the paper only at a fairly high level), and noting merely that the implementation of the design on a specific vendor's flexible substrate fab process was performed "using commercial Electronic Design Automation (EDA) tools" which really clears things right up.

This isn't to say that I think they are trying to pull some sort of "oh, the 'community edition' is under a FOSS license; but you need the enterprise edition features..." type fast one; just that it seems a bit hyperbolic to talk about "democratizing computing" when you are talking about a situation where table stakes is either having a punchy FPGA that needs a CPU core(and doesn't already avoid using the expensive and low performance programmable cells by just slapping some basic 'hard' ARM cores on for housekeeping; and the more realistic low end is at least small scale fabless semiconductor operations that have engineering capacity and need architectural flexibility; or fairly large scale ones that are looking to get price concessions out of one of the proprietary ISAs or ISA implementations.

Given the comparatively enormous costs of taking something from ISA to chip(along with the fact that ISAs pretty much have to be disclosed if you want software to run on them; but cannot be economically modified once fabricated); it just seems weird to talk about ISA licensing as though it's a major barrier to 'democratized' computing.

The flexible semiconductors themselves seem neat, if niche; but also seem like something that the common man is going to be buying from a commercial provider under more or less the same terms whether that provider used RISC-V or licensed something from ARM. That's certainly how it has been with RISC-V on the silicon side: ISA is open but if you actually want RISC-V parts you can either be reasonably serious fabless semiconductor engineers yourself; or you can buy from someone licensing from SiFive rather than someone licensing from ARM.

Indium availability

[manu0601]

How can it be cheap? Indium is as scarce as silver [wikimedia.org].

Re: Indium availability

[OrangeTide]

You don't need a lot of it to dope a little fleck of silicon.

Just like I don't need a lot of silver to solder the pipes in my house. The rest of the house isn't made of silver, thankfully.

Re:

[mistergrumpy]

This is a thin film technology - not a bulk crystal technology like Si. It's already used in some displays and TVs.

Flexible ECG

[Rick Schumann]

They already have very small wearable ECG patches [irhythmtech.com] that can collect data for up to 2 weeks. For scale, consider that it must be powered off a garden-variety CR2032 coin cell; it's not much bigger in diameter than a CR2032, and maybe twice as thick overall as a CR2032. You can't get much better than this, really, because regardless of how tiny the electronics are (imagine it's totally custom silicon, maybe even chip-scale, no package, bonded directly to the PCB) you still need to power whatever it is. Also being flexible may not be the 'flex' they think it is in this instance, if the patient intentionally or unintentionally removes it, and it's damaged in the process because it's flexible, you won't be able to download the collected data from it.