Over the past four years, Intel has been on a mission to transform itself into a company that manufactures chips worldwide, having invested over $ 90 billion in its foundry aspirations. The new 18A process adopted by the company, and based on its powerful 1.8nm technology, is at the heart of this change. The method integrates new technologies such as RibbonFET gate-on-all-side transistors and PowerVia backside power delivery, created to increase performance and energy efficiency. These developments are particularly significant with the continued use of high-powered chips in artificial intelligence, mobile phones, and servers.
Notwithstanding these endeavors, the road has been rocky at Intel. Its foundry segment recorded close to $13 billion of losses in the past year and the source stock has suffered a loss of nearly half of its value when compared to its 2024 high. In the interim, TSMC has been able to maintain its market lead by having more than two-thirds of the foundry capacity and aggressively venturing into mass production of its own 2nm chips. TSMC also has its next generation in the 2nm node, which will also be based on gate-all-around technology and should be 15% faster and use 30% less power than the previous 5nm node. The company has already reported some great yields of 60% on its 2nm chips, topping the reported 20% to 30% yields of Intel on 18A, and 40% by Samsung.
This is another obvious advantage of TSMC which has a powerful customer base and includes such tech giants as Apple and AMD. Even Intel is playing it safe as it expects some of its future processors to be based on TSMC technology. Although Intel says 18A will be better in other ways than TSMC 2nm chips, TSMC is liable to be ahead with connectivity and cost. Intel is not spared of delays and lost early customers which makes their path even more difficult. Going forward, TSMC has a good chance of leading based on its size, loyal customers, and its track record, whereas Intel will need all the capacity to shorten the gap.
