In the keynote speech, Dr. Siyoung Choi, the head of Samsung’s foundry division, said Said Samsung foundry is expected to use its 2nm manufacturing process to mass produce chips in the second half of 2025. The announcement made at the Samsung Foundry Forum in 2021 indicates that Samsung foundries will continue to develop leading manufacturing technologies and compete with Intel and Intel. TSMC. In fact, in terms of 2 nm, Samsung foundries may have an advantage over its competitors Intel and TSMC.
“2nm is expected to start mass production in the second half of 2025,” Said Dr. Siyoung Choi. “By then, it will represent the process technology applied by the third-generation GAA. Ultimately, due to previous experience in 3nm, we expect the industry’s transition to 2nm will be smooth. To meet the future, Samsung foundries will continue to strengthen their Process technology capabilities.”
At this point, Samsung did not disclose the characteristics of its 2 nm-level manufacturing technology, although it is natural to expect the new node to bring some performance, power and transistor density improvements compared to the previous generation node. The company also did not say whether its 2nm process will originate from its 3nm node, or will become a brand new branch.
Samsung foundry is the first chip manufacturer to announce the transition to full gate field effect transistors (GAAFET), with its 3GAE (early 3nm gate all-around) and 3GAP (3nm all-around gate+) manufacturing processes.At that time, the company stated that its 3GAE node will enter mass production in 2021, but it ultimately proved that the technology will grow slightly Postponed to 2022Samsung revealed at SFF 2021 that it will be ready for mass production in the first half of 2022.Samsung foundries Tape out its first 3GAA test chip, So the technology is preparing for mass production.
In contrast, Intel’s next-generation I7, I4, and I3 nodes will rely on FinFETs, and the company will switch to GAA transistors (RibbonFET in company terminology) as early as the second half of 2024. At the same time, in the second half of 2025, Intel is ready for its 18A manufacturing technology, the second-generation GAA application process. TSMC will also insist on using FinFET’s N4 (2022) and N3 (H2 2022) processes, and is expected to introduce GAA transistors using N2 manufacturing technology later.
For chip manufacturers and chip designers, the transition to the new transistor structure is always a challenge. In addition to the increased variability introduced by all new nodes and resolved by new electronic design automation (EDA) tools, such as new geometries, new layout methods, new layout rules, and routing rules, chip makers also need to learn how to pass new electronic design automation (EDA) tools. Transistors, and chip developers need brand new IP.
But early has its advantages. By the second half of 2025, when Samsung’s 2 nm-level technology enters mass production, Samsung’s third-generation GAA node will rely on the company’s multi-bridge channel field effect transistor (MBCFET). By the second half of 2025, the company will have considerable experience in new transistors, which may translate into higher yields, higher transistor density and/or power/performance advantages.
Regarding Samsung’s 2 nm-class MBCFET manufacturing process, there is an important thing to note. If it enters mass production in the second half of 2025 (starting on July 1) and gets the first batch of chips in October, the actual products based on these chips will not be available until Q1 in 2026 at the earliest if all goes well.
IBM Research earlier this year disclose Its “2nm class” test chip uses GAA transistors and encapsulates 50 billion transistors.
