2nm Chip Process Battle Intensifies! Qualcomm (QCOM.US) Returns to Samsung, Shifts from Sole Reliance on TSMC to Dual Foundry Strategy

Stock News
01/07

According to reports citing statements from Cristiano Amon, CEO of Qualcomm Incorporated, Qualcomm (QCOM.US) is highly likely to utilize Samsung Electronics' 2nm-class chip foundry process to manufacture its next-generation mobile application processors. The media report, referencing Amon's interview, indicated that this leading smartphone and PC chipmaker is currently in discussions with several wafer foundries, including Samsung Electronics, regarding contract manufacturing using the cutting-edge 2nm chip fabrication process. Amon stated that the majority of the core chip design work for Qualcomm's next-generation products targeting PCs, smartphones, and even data centers has been completed, paving the way for large-scale foundry production and full commercialization in the near future.

The report further added that this potential deal would mark Qualcomm's return to Samsung Electronics' advanced chip manufacturing nodes, following years of near-total reliance on Taiwan Semiconductor Manufacturing (TSM.US) for its most advanced process needs. On Monday, TSMC's Taiwan-listed shares surged as much as 6.9%, once again hitting a record high and propelling the benchmark Taiwan stock index above the 30,000-point mark. This rally occurred after Wall Street giant Goldman Sachs significantly raised its 12-month price target for TSMC by 35% to NT$2,330, further bolstering market confidence in the long-term robust growth of AI-related computing infrastructure demand.

TSMC's Taiwan-listed shares recently closed at NT$1,675. Meanwhile, in the US market, TSMC's American Depositary Receipts (ADRs) have surged 8% since the start of 2026, bringing its market capitalization close to $2 trillion. The high-demand AI GPUs since 2023 and the recently booming AI ASICs are all heavily dependent on TSMC. As the world's largest contract chip manufacturer, TSMC continues to benefit from the unabated global AI frenzy, with its clients—chip giants like NVIDIA, AMD, and Broadcom—reaping rewards from the surging demand for the core infrastructure of AI: AI chips.

The massive increase in foundry contracts from these chip giants has driven TSMC's consistently stronger-than-expected performance expansion since last year, forming a key rationale behind the repeated new highs for both its Taiwan-listed and US ADR shares in recent years. TSMC, often dubbed the "King of Global Chip Foundries," is also fully focused on the most advanced 2nm and below process capacities. The company explicitly states on its official website's process introduction that 2nm (N2) "commenced volume production as scheduled in the fourth quarter of 2025 (4Q25)," emphasizing that N2 adopts the first-generation nanosheet transistor technology (i.e., GAA/Gate-All-Around direction).

TSMC's website also identifies Fab 20 and Fab 22 as its 2nm advanced process production facilities. Recent media reports further disclosed that TSMC's initial volume production/ramp-up focus leans more towards the Kaohsiung-based Fab 22. Compared to the N3E process (equivalent to 3nm advanced process), N2 offers approximately a 10% to 15% performance increase at the same power consumption, or a 25% to 30% power reduction at the same performance, alongside a 15% to 20% increase in transistor density (depending on specific chip architecture design).

Another chip manufacturing giant, Intel, is skipping the integer 2nm process node to focus on the 1.8nm-class process (i.e., 18A). Intel stated that 18A has progressed from "process node commitment" to "end-product launch and volume production ramp-up," while 14A remains in the R&D and roadmap/customer engagement phase. At CES 2026, Intel significantly unveiled/demonstrated Panther Lake (Core Ultra Series 3) built on the 18A process, confirming it as the "first product platform" for 18A. The key technological features of 18A, namely GAA/RibbonFET + backside power delivery (PowerVia), were highlighted by both media and Intel's official materials as selling points for this generation of products and process, although specific production yield metrics have not yet been disclosed.

What does Qualcomm's return to Samsung Electronics mean for TSMC and Intel? Qualcomm's past "near-total reliance on TSMC" for leading-edge processes means that shifting a portion of its next-generation mobile AP advanced node orders to Samsung's 2nm signifies the initiation of a dual-supply-chain or diversification strategy for its "most core and highest-volume mobile SoCs." This could exert slight pressure on TSMC in terms of market share and pricing power, but it poses no significant disruption to TSMC's fundamental growth prospects.

Qualcomm's choice of Samsung for mobile processor manufacturing appears more like "customer risk hedging and capacity/cost rebalancing," and does not necessarily indicate questioned competitiveness of TSMC's 2nm process. TSMC is likely to retain substantial orders from Qualcomm (for instance, for AI PC core chips and upcoming data center AI inference chips). Furthermore, TSMC's overall high-end capacity faces seemingly insatiable demand from AI/HPC sectors, and with Apple having already secured a portion of TSMC's 2nm capacity, TSMC may need to significantly expand its advanced process chip production in 2026.

For Intel's foundry prospects, however, this could represent a significant setback. At a stage where Intel is striving to attract major external customers with its 18A and more advanced nodes to strengthen its foundry growth narrative, Qualcomm's potential prioritization of a return to Samsung over Intel implies that Intel may still struggle to gain substantive validation from a "flagship mobile SoC customer" in the short term. The 18A and the more advanced 16A technologies are arguably critical nodes for Intel's potential turnaround; therefore, mastering leading-edge chip manufacturing technology that is globally competitive and highly suitable for mass production is crucial for the growth prospects of Intel, which has faced prolonged earnings challenges.

The US-based chip veteran is even investing in more expensive High-NA EUV lithography machines—including the world's first High-NA system delivered by ASML—in its determined effort to develop leading-edge chip manufacturing technology for 2nm and below processes that surpasses both TSMC and Samsung Electronics.

免责声明:投资有风险,本文并非投资建议,以上内容不应被视为任何金融产品的购买或出售要约、建议或邀请,作者或其他用户的任何相关讨论、评论或帖子也不应被视为此类内容。本文仅供一般参考,不考虑您的个人投资目标、财务状况或需求。TTM对信息的准确性和完整性不承担任何责任或保证,投资者应自行研究并在投资前寻求专业建议。

热议股票

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10