TSMC’s N4X Technology Targets HPC
New process technology unveiled by Taiwan Semiconductor Manufacturing Co. is geared toward high-performance computing (HPC) workloads and devices. The foundry giant said its N4X process technology is scheduled for trial production during the first half of 2023.
The new processing node is TSMC’s first foray into the growing HPC market. NX4 targets designs with transistors as small as 5 nm, delivering a claimed 15-percent performance boost compared to its N5 processes. TSMC also claims the N4X offers a 4-percent performance improvement over its N4P at 1.2 volts.
The company previously released HPC process technology based on its N5A and N6RF technologies. The N4X design is an enhancement of its N5A, improving upon its performance and maximum clock frequencies.
The “X” denotes TSMC’s growing push into the HPC chip design sector, which the company said is among its fastest growing business segments.
TSMC’s third-quarter revenue report showed a profit of NT$156.26 billion (U.S.$5.56 billion), up 13.8 percent compared to last year. The company attributed the revenue boost to increased demand in its four growth platforms, including smartphones, IoT, automotive applications and HPC. It expects process technologies aimed at HPC to continue providing growth into 2022.
“The demands of the HPC segment are unrelenting, and TSMC has not only tailored our ‘X’ semiconductor technologies to unleash ultimate performance but has also combined it with our 3DFabric,” said Kevin Zhang, TSMC’s senior vice president of business development.
Yujun Li, TSMC’s director of HPC business development, added that N4X can address an increase in computing demand as HPC chips reach their maximum reticle size.
“Our customers can expand the number of leading-edge comput[ing] chips for maximum compute power. Or, chips can be partitioned into multiple chiplets with each adopting the optimum technology of choice – logic optimized, IO and analog optimized or memory optimized.”
The post TSMC’s N4X Technology Targets HPC appeared first on EETimes.
