TSMC Will Begin Process Technology Research at 1.4nm
Taiwan Semiconductor Manufacturing Co. has set a roadmap for its N2 (2 nm-class) fabrication process, which will enter high-volume manufacturing (HVM) in 2025, it is time for the company to begin considering a further node. According to a recent report, TSMC will reveal its 1.4 nm-class technology formally in June.
Business Korea says that TSMC aims to move the team that developed its N3 (3 nm-class) node to the development of its 1.4 nm-class fabrication process in June. Foundries and chip designers rarely openly announce R&D milestones, therefore it is unlikely that TSMC will issue a press release stating that the development of its 1.4 nm technology has begun. TSMC is scheduled to host its Technology Symposium in mid-June, where the company may provide brief details regarding the node that will replace its N2 production process.
The phases of standard process technology design flow are pathfinding and research and development. Pathfinding includes fundamental materials and physics research and is frequently conducted concurrently for several nodes. Pathfinding for TSMC's N2 is likely to complete at this point, thus teams focusing on fundamental physics and chemistry are working on a successor for N2, which may be designated as 1.4 nm or 14 angstroms. TSMC's N2 is based on gate-all-around field-effect transistors (GAAFETs), although it will utilize existing extreme ultraviolet (EUV) lithography with a 0.33 numerical aperture (0.33 NA). Given the current information regarding TSMC's N2, it is probable that its successor will keep GAA transistors, but the major question is whether it will transition to EUV tools with a 0.55 NA (or High NA).
Considering that TSMC's N2 reaches HVM in late 2025 (thus expect the company's first 2 nm chips to be delivered around 2026) and TSMC's two-and-a-half to three-year node introduction cadence, we can expect TSMC's 1.4 nm (or 14 angstroms) process to be used for commercial products beginning in 2028. Given the timeline, it will be advantageous for the node to implement High NA lithography, which Intel aims to implement in 2025.
Regarding Intel, it is unknown which of Intel's nodes will compete with TSMC's 1.4 nm. Intel plans to launch its 18A (18 angstroms) technology in 2025, therefore by 2028, the corporation will have introduced at least one new fabrication technique. It will be interesting to see if it will be referred to as 16A or 14A, given that Intel appears conservative with node improvements at now.
Senior Member
Posts: 3766
Joined: 2013-02-05
There was a time when the number of a process node related to the gate length of a transistor.
For a decade now, that the number in these process nodes have no correlation to real physical dimensions.
Senior Member
Posts: 1300
Joined: 2014-11-19
Are we getting near the end of the line?
Senior Member
Posts: 3766
Joined: 2013-02-05
No. Like I said, these numbers are marketing.
But producing smaller chips is getting exponentially harder and more expensive.
A silicon atom is 0.2nm. 0.45 if we consider the atom's lattice.
For example, on TSMC N7, Contacted Gate Pitch is 57nm and Minimum Metal Pitch is 40nm.
Senior Member
Posts: 144
Joined: 2018-05-04
If you mean near the end of the line for misleading process names, then, yes


Senior Member
Posts: 1134
Joined: 2008-10-31
That is just insanity how small that is. A DNA strand is 2nm. I wonder what is the smallest they can go. then once they hit that, whats next?