Intel may discontinue production of nodes 14A and subsequent ones if they fail to secure a significant external client. This decision would effectively transfer leading-edge nodes to TSMC and Samsung.
In the world of semiconductors, Intel's future plans for its 14A process technology, a 1.4nm-class node, are causing a stir. The tech giant is currently evaluating whether to proceed with the development of this advanced process technology, and the decision could have significant implications.
If Intel decides not to move forward with 14A, the company may turn to third-party foundries like TSMC for manufacturing its future designs that require higher transistor density and performance than the current 18A-P. This shift could potentially lead to Intel ceding the leading-edge semiconductor manufacturing market to competitors such as TSMC and Samsung.
During its recent earnings call, Intel's focus was on investing in production capacity, developing next-generation process technologies, improving return on investment, and enhancing profitability. However, the company's strategy for 14A is market-driven, meaning without a significant external customer commitment, Intel may choose to avoid the substantial capital expenditures needed for mass production and tooling of 14A.
The potential halt or abandonment of 14A development could result in several setbacks for Intel. Developing advanced process nodes like 14A costs billions of dollars, involving significant R&D investment. Without an external customer to justify these expenses, Intel may face challenges recouping its investment.
Moreover, failing to commercialize 14A could reduce Intel's foundry competitiveness, especially as rivals dominate the market with smaller process geometries. Intel might lose credibility and business as a major foundry provider for advanced chips.
Internally, the absence of external demand could increase pressure on Intel's internal projects, potentially slowing innovation cycles or causing reliance on external foundries. For instance, TSMC nodes are currently used extensively in Intel's Arrow Lake processors.
In summary, without a major external customer, Intel's 14A process could be discontinued, causing strategic, financial, and market-position risks that may hinder Intel’s ambitions to be a leading-edge semiconductor foundry. This view is supported by Intel CEO Lip-Bu Tan's statements, emphasising the importance of volume commitments from both internal and external customers for continuing 14A investment.
It's important to note that Intel expects to continue building the majority of its products in-house on nodes up to 18A-P through at least 2030. However, if 14A and its successors are ultimately canceled, Intel may continue to rely on third-party foundries for its most advanced products.
Each ASML Twinscan EXE:5000/5200 High-NA EUV tool, essential for 14A's fabrication, costs around $380 million. Procuring two such tools for a fab to enable high volume manufacturing on 14A will cost Intel $760 million. The exact cost of the 14A project is not specified, but it is expected to be in the billions of dollars.
Modern leading-edge process technologies like 14A cost billions to develop, and Intel spent $16.546 billion on research and development in 2024, with a significant portion allocated to the development of 18A, 18A-P, 14A, and other process technologies.
The fabrication process for Intel's 14A process technology must meet crucial milestones to proceed. Intel has at least one internal design planned for 14A, but the company is keeping its options open. The Intel 14A node is planned as the successor to the 18A and 18A-P nodes, and is aimed at both internal products and external clients.
In conclusion, Intel's decision on the 14A process technology will shape its future in the semiconductor industry. The stakes are high, and the company is carefully considering its options to ensure a balanced approach that supports its strategic, financial, and technological goals.
- If Intel determines to forego the development of 14A process technology, they may look towards finance options, such as engaging third-party foundries like TSMC or Samsung, for manufacturing designs with higher transistor density and performance that are beyond the current 18A-P capabilities.
- The potential cost of developing advanced process nodes like 14A, involving massive Research & Development (R&D) investment, could leave Intel vulnerable in the technology sector if they fail to secure a significant external customer to justify these expenses, potentially impacting their foundry competitiveness in the market.
