Taiwan Semiconductor Manufacturing Co. (TSMC) CEO C.C. Wei isn’t worried about Intel’s recently announced product roadmap in the least.

Asked about Intel’s Integrated Device Manufacturer 2.0 roadmap, Wei declined to comment on the rival foundry operator’s technology. Instead, he claimed TSMC’s 2-nanometer process node — slated for release in 2025 — would offer unrivaled transistor density and performance.

“We are confident that our technology leadership will be maintained,” he said.

Intel has moved aggressively to put past missteps and delays surrounding its 10-nanometer and 7-nanometer manufacturing processes behind it since former VMware CEO and 30 year Intel veteran Pat Gelsinger took over as CEO in February.

Less than a month after taking over, Gelsinger announced Intel would compete directly with the likes of TSMC and Samsung Electronics and opened the chipmaker’s foundry services to contract manufacturing.

Intel this summer offered a glimpse at the company’s revised long-term roadmap, which includes a new naming scheme the company claims offers a more accurate comparison of relative performance compared to nodes from competing foundries.

The chipmaker rebranded its 10-nanometer manufacturing process Intel 7, while its ill-fated 7-nanometer process — now slated for release sometime in 2023 — was branded Intel 4. The company also shared details on a novel process using extreme-ultraviolet lithography and gate-all-around transistors the chipmaker is calling Intel 20a.

In addition to powering the chipmaker’s own products, Intel plans to make these process technologies available to contract customers and is providing resources to help those stuck on older nodes modernize their designs.

However, it should be noted that Intel remains a TSMC customer, and its latest generation of GPUs are manufactured in TSMC fabs. Whether Intel will move away from TSMC as its own process nodes improve remains to be seen.

While Intel plans to launch its first 7-nanometer chips — Intel 4 — in 2023, Wei claims TSMC is preparing to ramp production of its 3-nanometer process node — N3 — in the second half of 2022, in line with previous nodes.

“Our N3 technology will use a FinFET transistor structure to deliver the best technology maturity, performance, and cost for our customers,” Wei said. “We expect the revenue will be seen in the first quarter of 2023 because it takes such a long tech cycle time to have all those wafers out.”

TSMC plans to release a revamped version of the node called N3E the following year, which the company claims will benefit from an improved manufacturing window, with better performance, power, and yield.

However, Wei noted the process is expected to cost more than the chipmaker’s current generation 5-nanometer process.

TSMC also announced a new Japanese foundry operation during the company's third quarter of 2021 earnings call. The fab will produce parts based on its 22-nanometer and 28-nanometer manufacturing processes.

"We are expanding our manufacturing footprint to extend and enhance our competitive advantage in providing industry-leading technologies," Wei said.

The fab will be used almost exclusively for specialty technology, he said, dismissing the idea that there could be an oversupply of 28-nanometer capacity in the near future. "Not for TSMC," he added.

The Japanese fab is scheduled to begin construction next year and come online in spring 2024.

Pressed on the potential for a European fab, Wei said it isn't out of the question. "We don't rule out the possibility of building a fab in other areas that include Europe," he said.

Both Intel and TSMC recently met with the European leaders to discuss bringing foundry operations to the region.

TSMC’s revenues grew 12% year over year to $14.88 billion and it realized a net income of $5.57 billion during Q3.

TSMC’s most advanced 5-nanometer manufacturing process accounted for roughly 18% of total wafer revenues during the quarter, while the company’s more mature 7-nanometer process, widely used by chipmakers like AMD, accounted for 34% of sales, the company said.

“Our third quarter business was driven by strong demand across all four growth platforms, which are smartphone, high-performance compute (HPC), IoT, and automotive-related applications,” CFO Wendell Huang said.

Of these, smartphone chips accounted for 44% of revenue and HPC accounted for 37% of revenue. IoT was a particular bright spot for TSMC during the quarter, growing 23% from the previous quarter and accounting for 9% of total revenue.

Huang said he expects the “industry megatrends” of 5G and HPC to drive revenues in the fourth quarter to between $15.4 billion to $15.7 billion.