Free tools Windows power users keep installed
One-click scans. No signup required.
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Rain Neuromorphics taped out a demonstration chip in 2021 that implemented a brain-inspired analog computing architecture using a three-dimensional array of resistive memory devices. The team reported on-chip neural-network weight updates and inference operations, but the tapeout was a feasibility milestone—not proof of the company’s ambitious energy claims or its later commercial shipping forecasts. Rain subsequently shifted its product direction to digital SRAM-based compute-in-memory.
What Rain Neuromorphics taped out
On October 12, 2021, UF Innovate reported that Rain Neuromorphics, a University of Florida startup, had taped out a demonstration chip for analog neuromorphic computing. “Tapeout” means the design was finalized for fabrication; it does not, by itself, mean a product was commercially available. The demo aimed to use a three-dimensional network of resistive memory devices to perform neural-network training and inference at low power. UF Innovate’s announcement described a design that had evolved from randomly deposited resistive nanowires to ReRAM integrated with 3D manufacturing techniques adapted from NAND flash.
How the analog chip was organized
Rain’s design combined conventional CMOS circuitry with resistive memory and patterned connections. EE Times Asia described CMOS layers as neurons, vertical bit-line columns as axons, ReRAM devices at the interfaces, and lithography-defined dendrites connecting the structures. The demo used a 180-nanometer CMOS process and was described as representing 10,000 neurons. EE Times Asia’s 2021 report also said memristor weight updates supported training, while matrix multiplication supported inference.
What memristors contribute
A memristor is a resistive device whose conductance can be changed and used to represent a weight in a neural network. In an analog compute-in-memory approach, the memory devices participate directly in computation instead of sending every value back and forth between separate memory and processing units. Rain’s reported demonstration combined that idea with 3D device connections. The available reporting establishes the intended operations—weight updates and inference—but does not provide a complete independent characterization of precision, endurance, or performance across workloads.
#1 Best Overall
- Temperature measurement range: A version -40℃ to +220℃; B version -40℃ to +500℃. It is recommended to choose a version with a smaller range within the measurement range
- Applicable sensor: PT100/PT1000 3-wire or 2-wire sensor
- Can read temperature and PT100/PT1000 resistance value
- Serial port baud rate: 1200,2400,4800,9600 (default), 19200,38400,57600,115200,N, 8, 1
- Power supply voltage: DC 9V 12V 15V 24V(6-25V) Working current: 18-20MA
Why the connections were sparse and partly random
Rain argued that a large network should not connect every neuron to every other neuron. Sparsity can reduce the number of active connections, while a fixed lattice might impose a particular pattern of information flow before learning occurs. CTO Jack Kendall told EE Times, “The reason randomness is important is if you have a very large neural network, you want to maintain a certain level of sparsity.”
The word “random” needs qualification: the dendrites were defined by a lithography mask, so a fabricated pattern was repeatable from chip to chip rather than changing unpredictably on each device. Rain’s roadmap at the time included testing different sparsity patterns and biological motifs. Kendall described the design as “very brain-like, honestly, and we are quite proud of that.”
Rank #2
- Soil hygrometer humidity:dual output mode, more accurateanalog output,moisture meter plants
- Humidity monitor:adjustable sensitivity (shown in blue digital potentiometer adjustment),humidity gauge
- Moisture probe for plants:using this sensor make an automatic watering system to manage your garden plants without people,hygrometer for plants
- Humidity tester:a fixed bolt hole for easy installation,moisture meters
- Soil hygrometer moisture:with power indicator (red) and digital switching output indicator (green),soil hygrometer detection
What the demonstration showed—and what it did not
The tapeout and reported operations showed that Rain’s architecture could be embodied in silicon and that the demo could perform weight updates and inference. That is a meaningful engineering and scientific milestone. It is distinct from proving that a product could be manufactured economically, run at advertised performance across useful workloads, or ship on a forecast schedule.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteEE Times Asia reported Rain’s comparisons against other systems: more than three times faster training than a SONOS flash array, a ten-fold lower power footprint, and inference latency reduced from hundreds of microseconds to hundreds of nanoseconds. The same report attributed a possible thousand-fold energy reduction versus GPU solutions to Rain. These are company claims reported in 2021, not independent benchmark results established by the report; the comparison conditions and workload details should not be assumed to generalize.
Rank #3
Rain CEO Gordon Wilson acknowledged that “We still have a fair amount of engineering work ahead,” while saying the scientific feasibility question had been addressed. The distinction matters: demonstrating a functional concept does not settle questions such as repeatable manufacturing, usable capacity, write behavior, software support, and commercial readiness.
What happened to Rain’s analog-chip roadmap
Rain’s 2021 plans anticipated first-generation chips with 125 million INT8 parameters and power below 50 watts. EE Times Asia reported that samples were expected in 2024 and commercial silicon in 2025. Those dates and specifications were company expectations published in 2021, not evidence that such chips later shipped.
Rank #4
- Founded in 2010, Chips Gate is a trusted supplier of industrial automation equipment, including PLC modules,motor drives, and control systems for both B2B and B2C needs.
- Wide selection of automation equipment suitable for various industrial and commercial applications.
- Durable packaging keeps your order fully protected in transit.
- Available for single-unit purchases or bulk orders to meet different project needs.
- Dedicated to maintaining consistent quality standards through careful selection and handling of equipment.
In a later public post, Wilson said Rain had taped out two chips and concluded that the materials required for its original analog vision were not mature: “We taped out two chips, and realized that the technology just wasn’t ready.” Rain shifted its product roadmap to digital SRAM-based compute-in-memory while retaining a frontier research effort, including analog projects supported by ARIA. Wilson’s public post describes that change.
Can you buy a Rain AI chip?
The available information does not establish that Rain’s analog demonstration chip or the 2021-planned first-generation silicon is for sale. Rain’s current product page describes a different commercial direction: digital in-memory compute, with IP licensing for a compute tile and a software stack for custom system-on-chip designs. It says the IP targets low-latency, energy-efficient on-device AI and lists hardware as “available soon.” Rain’s product page does not turn the historical analog demo into a retail product or confirm that the listed hardware is currently shipping.
Best Value
- 4-channel PT100 temperature acquisition
- RS485 MODBUS RTU standard communication protocol,Netted with configuration software, PLC or industry touch panel
- Communication circuit designed for thunder protection and interference immunity
- Used for signal collection and control in industrial field
- Temperature measurement range: -40℃ to +300℃
How to read the Rain demo in context
The 2021 chip is best understood as a proof-of-concept for a particular analog, ReRAM-based neuromorphic architecture—not as a preview of the current product’s substrate or a commercially available accelerator. When comparing compute-in-memory designs, useful distinctions include:
Quick Recap
- Compute substrate: analog approaches operate with physical signals and device properties; digital SRAM-based compute-in-memory uses digital representations.
- Memory technology: ReRAM, flash, and SRAM have different device behavior and manufacturing constraints.
- Training support: determine whether a system updates weights on-chip or only performs inference.
- Connectivity: examine whether connections are sparse, fixed, programmable, or learned.
- Evidence level: distinguish a taped-out demonstration and reported operations from independently measured performance and commercial shipment.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
