Intel's 18A process, a crucial node for its foundry ambitions, has reportedly hit a snag with low yield rates, potentially impacting the division's future and the company's overall strategy.
According to a report from Korean media outlet Chosun, the 18A process is currently achieving less than 10% yield rates. This makes mass production unfeasible and raises concerns about Intel Foundry's competitiveness.
Broadcom, a key Intel Foundry Services (IFS) customer, was reportedly dissatisfied with the 18A process and its low yield, leading them to cancel orders and seek alternative manufacturers. This setback, coupled with underperformance in securing government subsidies, is believed to have contributed to the departure of former CEO Pat Gelsinger, a strong advocate for the foundry business.
Meanwhile, competitor TSMC's N2 (2nm) process, despite a slightly larger node size, is said to outperform Intel's 18A due to higher SRAM density, a crucial factor for performance and efficiency.
With Gelsinger's departure, Intel may shift its focus away from the foundry business, increasing the likelihood of a division sell-off or merger.
