Apple M5 Chip's Dual-Use Design Will Power Future Macs and AI Data Center Servers

Written on 07/04/2024
Taylor Bell

Apple will reportedly use more advanced SoIC packaging technology for the M5 chips. This is part of a two-pronged strategy to meet growing demand for silicon to power consumer Macs and improve performance in data centers and future cloud-dependent AI tools.

System-on-integrated chip (SoIC) technology, developed by TSMC and introduced in 2018, allows chips to be stacked in a three-dimensional structure, resulting in better electrical performance and thermal management compared to traditional two-dimensional chip designs.

According to the Economic Daily, Apple has expanded its collaboration with TSMC on a next-generation hybrid SoIC package that further combines thermoplastic carbon fiber composite molding technology. The package is reportedly in small-scale test production. The chips are expected to be mass-produced in 2025-2026 for new Macs and AI cloud servers.

References to what appears to be Apple's M5 chip have already been spotted in official Apple code. Apple is developing its own AI server processors using TSMC's 3nm process, aiming for mass production in the second half of 2025. However, according to Haitong analyst Jeff Poo, Apple plans to assemble AI servers with M4 chips at the end of 2025.

Currently, Apple's AI cloud servers are thought to run on multiple connected M2 Ultra chips, originally designed exclusively for desktop Macs. Whenever the M5 is introduced, its advanced dual-use design will be seen as a sign that Apple is realizing its plan to vertically integrate its supply chain of AI capabilities across computers, cloud servers, and software into the future.