GoKawiilRobert Triggs / Android Authority
TL;DR TSMC has reportedly developed a 1.6nm chip manufacturing process.
The process is said to deliver an 8% to 10% increase in computing speed, and a 15% to 20% decrease in energy consumption.
TSMC is expected to start mass production in Q4 2026.
While everyone’s excited about 2nm chips, IBM recently blew past that limit, showing off its sub-nm architecture. As a result, TSMC and other chip makers are now left playing a game of catch-up. However, IBM estimates that it will take five years before this design is ready for production. As the company works towards that goal, TSMC is hot on IBM’s heels with a sub-2nm manufacturing process that could be ready for mass production soon.
A report from Liberty Times (via Chosun) claims that TSMC has finished the development and verification of its angstrom-class A16 (1.6nm) process. TSMC will reportedly enter the mass production phase in the fourth quarter of 2026.
The process is said to use the company’s backside power delivery network, known as Super Power Rail (SPR). This new architecture separates the signal wiring and power wiring, placing the power on the backside of the wafer. Previously, both were placed together on the front of the wafer.
By separating the signal and power wiring, the company managed to reduce bottlenecks during data transmission. As a result, computing speed and energy efficiency are improved. Compared to 2nm architecture, this process is said to deliver an 8% to 10% increase in computing speed, and a 15% to 20% decrease in energy consumption. The chip density is also reported to be 8% to 10% higher.
It appears that TSMC expects to use the 1.6nm process as a stepping stone toward its 1.4nm process. The report states that the company’s goal is to start mass production of its 1.4nm process by 2028. However, TSMC still has a lot of work to do if it wants to match IBM’s 0.7nm node.
Follow