In the foundry (semiconductor contract manufacturing) 2.0 market in the third quarter of this year, TSMC ranked No. 1 with a 39% market share, according to estimates. "Foundry 2.0" refers to an expanded market that includes not only pure-play foundry firms but also integrated device manufacturers (IDMs) in non-memory, outsourced semiconductor assembly and test (OSAT) corporations, and photomask suppliers.
According to market research firm Counterpoint Research on the 23rd, global foundry 2.0 market revenue reached $84.8 billion in the third quarter of this year, up 17% from a year earlier. It said pure-play foundry firms such as TSMC drove growth as steady demand for artificial intelligence (AI) graphics processing units (GPUs) continued across front-end manufacturing and back-end advanced packaging.
TSMC is estimated to have posted a 41% year-over-year revenue increase in the third quarter of this year. Counterpoint Research said ramp-up of the 3-nanometer process for Apple's flagship smartphone solutions and high utilization of 4- and 5-nanometer processes for AI accelerator customers such as Nvidia, AMD, and Broadcom boosted revenue. The quarterly revenue growth rate of foundry firms excluding TSMC came to 6%. Chinese foundries grew 12% year over year, aided by local policy support.
In the third quarter of this year, the foundry 2.0 market shares were TSMC 39%, ASE 6%, Texas Instruments 6%, Intel Foundry 5%, Infineon 5%, and Samsung Electronics 4%.
Jake Lai, research director at Counterpoint Research, said, "The annual revenue growth rate of the foundry 2.0 market is expected to be around 15%," and added, "As the main revenue growth driver, the 4- and 5-nanometer processes, run at full capacity, reach their limits, and constraints emerge in TSMC's CoWoS advanced packaging facilities, it will be difficult for TSMC to post another large quarter-over-quarter revenue gain in the fourth quarter of this year."
He added, "The pure-play foundry market is projected to grow 26% year over year," and said, "Sustained shipments of AI GPUs and AI application-specific integrated circuits (ASICs) over the next few quarters will support this and be a key contributor to overall market expansion."