This Review examines the development of neuromorphic hardware systems based on halide perovskites, considering how devices based on these materials can serve as synapses and neurons, and can be used in neuromorphic computing networks.
The big difference is that all the rendering would be handled by chiplets instead of a big compute chip like on its existing GPUs.
The Sydney team exploited stimulated Brillouin scattering, a technique which involves converting electrical fields into pressure waves in certain insulators, such as optical fibers. In 2011, the researchers reported that Brillouin scattering held potential for high-resolution filtering, and developed new manufacturing techniques to combine a chalcogenide Brillouin waveguide on a silicon chip. In 2023, they managed to combine a photonic filter and modulator on the same type of chip. The combination gives the experimental chip a spectral resolution of 37 megahertz and a wider bandwidth than preceding chips, the team reported in a paper published 20 November in Nature Communications.
“The integration of the modulator with this active waveguide is the key breakthrough here,” says nanophotonics researcher David Marpaung of the University Twente in the Netherlands. Marpaung worked with the Sydney group a decade ago and now leads his own research group that is taking a different approach in the quest to achieve wide-band, high-resolution photonic radio sensitivity in a tiny package. Marpaung says that when someone reaches sub-10-MHz spectral resolution across a 100 gigahertz band, they will be able to replace bulkier electronic RF chips in the marketplace. Another advantage of such chips is that they would convert RF signals to optical signals for direct transmission through fiber optic networks. The winners of that race will be able to reach the huge market of telecoms providers and defense manufacturers who need radio receivers capable of reliably navigating complicated radio-frequency (RF) environments.
“Chalcogenide has a very strong Brillouin effect; it’s very good, but there is still a question of whether this is scalable…it’s still perceived as a lab material,” Marpaung says. The Sydney group had to figure out a new way to fit the chalcogenide waveguides in a 5-millimeter-square package into a standard manufactured silicon chip, which was no easy task. In 2017, the group figured out how to combine chalcogenide onto a silicon input/output ring, but it took until this year for anyone to manage the combination with a standard chip.
What just happened? IBM’s concept nanosheet transistor demonstrated nearly double the performance improvement at the boiling point of nitrogen. This achievement is expected to result in several technological advances and could pave the way for nanosheet transistors to replace FinFETs. Even more excitingly, it could lead to the development of a more powerful class of chips.
Liquid nitrogen is widely used throughout the semiconductor manufacturing process to remove heat and create inert environments in critical process areas. However, when brought to the boiling point, which is 77 Kelvin or −196 °C, it can no longer be used in certain applications because the current generation of nanosheet transistors hasn’t been designed to withstand temperatures of this kind.
This limitation is unfortunate because it has been theorized that chips could boost their performance in such an environment. Now, this possibility may be realized, as demonstrated by a concept nanosheet transistor IBM presented at the 2023 IEEE International Electron Device Meeting held earlier this month in San Francisco.
As the company’s first tile-based CPU based on its new Intel 4 process, this is an entirely different CPU from everything that has come before it.
Shirriff’s blog goes into a deep dive with a look inside the HP PHI chip, its construction, and die. He even examines its logic gates, first-in-first-out buffers (FIFOs), and address decoder. Please check out the blog for all these finer details and more.
In conclusion, the computer historian echoes our initial thoughts that this silicon-on-sapphire IC is “interesting as an example of a ‘technology of the future’ that didn’t quite pan out.”
Shirriff also contrasts late 70s era processors built on silicon-on-sapphire vs regular silicon in terms of energy consumption and clock speeds. Would you be surprised to hear that silicon-on-sapphire ICs were far superior using these metrics? Things might have panned out differently if these transparent ICs had been mass-produced at better yields and lower manufacturing costs. A frightening statistic highlighted by Shirriff is that HP’s silicon-on-sapphire yields were a mere 9%.
The researchers embedded this prototype in a biodegradable, chip-like implant that combined energy harvesting and energy storage. When the prototype was attached to a medical implant, power passed through the circuit directly to the device and into the capacitor to ensure a constant power supply.
In rats, the wireless implant worked for up to 10 days and dissolved completely within two months — proving its biodegradability. But it could potentially last longer if the team thickened the protective polymer and wax layers encasing the system, Lan said.
The researchers also tested the wireless charger as a drug-delivery system and delivered anti-inflammatory medicine to rats with a fever. After 12 hours, the rats that had no implants had much higher body temperatures than those with the chips, suggesting the device was successfully delivering the medicine.
Thanks in part to Elon Musk, the field of brain-computer interfaces has captured both public and investor interest, with a cadre of companies now developing implantable devices.
