dorsaVi (ASX:DVL) starts fabrication of its first RRAM-CMOS validation chip, a 180nm run that de-risks the 22nm program with NTU and ITRI for physical AI applications.
dorsaVi (ASX: DVL) has begun physical manufacturing of its first resistive random-access memory (RRAM) validation chip, moving the program from design into an active silicon wafer run. The company confirmed that its manufacturing data package passed foundry intake and the wafer lot is now in fabrication on the front-end CMOS base.
The current run uses a 180nm node. That base layer will support the RRAM memory array, peripheral circuits, and evaluation features before the RRAM-specific layers are added in a later stage. dorsaVi is taking a staged approach, generating process knowledge at each step before moving to the next. The goal is to produce silicon-level electrical testing results and performance benchmarks that can de-risk a subsequent 22nm program developed with NTU and ITRI.
Foundry acceptance means the design files, mask data, process documentation, and wafer-lot instructions all met the intake requirements. The CMOS base now being produced will provide the foundation for back-end-of-line preparation, RRAM stack integration, and formation of the integrated memory cells and arrays. Once the base is complete, the wafers are expected to move into partner-led processing and integration before electrical testing and characterisation assess how the completed architecture performs.
DorsaVi is building the RRAM platform for physical AI systems that process information on-device rather than relying on cloud connectivity. The company has identified target applications including smart exoskeletons, electromyography sensor nodes, robotic joint controllers, drones, self-guided vehicles, and always-on Internet of Things edge sensors. The 22nm program is designed to bring memory and compute closer together, reducing latency and energy associated with moving data between memory and processing.
Group chief executive officer Mathew Regan said the intention is to prove the memory architecture through dorsaVi's own sensing platform and then, subject to successful validation, commercialise it for the wider physical AI market. Manufacturing the validation chip gives the company physical silicon that can ultimately be tested and demonstrated to prospective partners and customers. Successful validation would support future supply, partnering, and licensing discussions.
The 180nm validation run will generate silicon-level learning that can inform the broader 22nm RRAM program. Electrical testing and characterisation will follow once the integrated wafers are returned from partner processing.
Drafted by a large language model from the source reporting linked above, then screened by automated publishing checks. It is not read by a journalist before publication. Some articles cite our Alpha Score. Verify prices and figures against the original source. Educational coverage, not personalized advice.