Intel CEO says Intel 18A is better than TSMC N2 process
Intel has almost entirely staked its future on rapidly advancing its manufacturing process nodes. According to its announced technology roadmap, the plan involves completing “five process nodes in four years,” a strategy that will directly influence the expansion of Intel Foundry Services (IFS). Intel is poised to launch Intel 18/20A into the market, aiming to reclaim its leadership in semiconductor manufacturing technology.
Recently, Intel CEO Pat Gelsinger, in an interview with the media, stated that “Intel 18A is somewhat superior to TSMC’s N2 process.” This assertion is based on Intel 18A’s use of RibbonFET, an all-around gate transistor technology, and PowerVia, a backside power delivery technique. These innovations, purportedly years ahead of competitors, offer enhanced area efficiency for chips, translating to lower costs, improved power delivery, and heightened performance. Moreover, Gelsinger hinted at the higher cost of the N2 process, suggesting that Intel 18/20A could attract orders from clients seeking greater cost-effectiveness.
According to Intel’s recent statements, chips manufactured with the Intel 18A process are expected to debut in the first quarter of 2024, with the first mass-produced products hitting the market in the latter half of that year. In comparison, TSMC’s N2 process is not anticipated to reach mass production until the second half of 2025, theoretically placing Intel a year ahead. Although TSMC has introduced GAA architecture transistors in its N2 process, it still relies on traditional power delivery technology, which, in Intel’s view, does not match the technical prowess of the Intel 18A process.
However, TSMC disagrees with this perspective. Next year, TSMC plans to introduce the N3P process, which is expected to offer energy efficiency, performance, and transistor density comparable to Intel 18A. Furthermore, the next-generation N2 process is projected to surpass both N3P and Intel 18A in all these aspects.