Intel is considering adopting a double-sided power supply architecture for its 1.4nm process technology to catch up with TSMC and Samsung.

Intel is considering adopting a dual-side power supply architecture (utilizing both front and back sides) for its 1.4-nanometer ultra-fine process to catch up with competitors. Industry sources said Intel originally planned to use PowerDirect, a dedicated backside power supply technology, for its 1.4-nm base process 14A, but is now considering introducing a dual-side architecture that leverages both front and back sides in its subsequent 14A2 process. Intel previously announced plans to achieve 1.3x higher chip density on its 14A process compared to 18A; the 14A process targets an M0 pitch of around 28nm, while the 14A2 process could push the M0 pitch to 21nm via a half-node improvement. Intel will maintain a backside power network as its primary setup, while reallocating some front-side metal interconnects for auxiliary power and clock signals to compensate for insufficient power headroom caused by scaling and lithography limitations. Intel’s 14A process is scheduled to enter risk production in 2028 and mass production in 2029. The chipmaker needs to release the 0.9 version of its 14A process design kit (PDK) to external customers this October, and secure firm orders from large fabless clients within the following 18 months. By comparison, TSMC plans to ship its true 1.4nm A14 products in 2028, while Samsung Electronics aims to commercialize its SF2Z, a modified 2nm process utilizing backside power supply technology, in 2027.