TSMC will be the Leading Foundry for the Next 5 Yrs; 3D Packaging to Determine Future

Taiwanese chipmaker TSMC is expected to lead the foundry business for at least another five years, as per analysts. With the company’s 5nm EUV process already in mass production and the 3nm node slated for volume production in 2022, the future is looking bright for Taiwanese pure-play foundry.

Reports indicate that the 2nm process is also nearly complete and will debut in 2024. As per the Research Director of the Industrial Technology Research Institute, Yang Rui, TSMC will dominate the foundry industry for another five years, after which 3D Packaging will be the primary challenge.

Intel at its Q2 earnings announced the delay of its 7nm process which caused its stocks to drop by more than 20%. Speculation that Team Blue will also become a client of TSMC in the further worsened the situation. More recently, Qualcomm also reported that 5nm that it would be transferring its upcoming SD 875 SoCs from Samsung’s foundries to TSMC.

TSMC’s 3nm process will continue to use the fin-field effect transistor technology (FinFET) and will be the most advanced node of 2022. The following 2nm process will adopt the GAA technology in 2024, up from FinFET, and will continue to lead the industry.

That’s the limit, however. After the 2nm node, it’ll mostly be 3D stacking and other advanced packaging techniques that will determine who rules the foundry business. As of now, TSMC hasn’t made any significant headway in this department.

Although the foundry introduced the InFO packaging technology in 2016, with several improvements in recent years, we’re yet to see them in any major processor.

Yang Ruilin believes that TSMC will continue to be ahead of Samsung even in the field of advanced packaging. In the coming years, advanced packaging will be the key to ruling the foundry business, as the cost barriers increase, widening the gap between competitors.



Computer Engineering dropout (3 years), writer, journalist, and amateur poet. I started my first technology blog, Techquila while in college to address my hardware passion. Although largely successful, it was a classic example of too many people trying out multiple different things but getting nothing done. Left in late 2019 and been working on Hardware Times ever since.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button