Apple Chip Supplier TSMC to Start the Production of 2nm Chips for Future iPhones and Mac Computers

Sophie Webster, Tech Times 22 April 2022, 12:04 pm

Apple is planning to adopt the 2nm process for its Mac and iPhone chips as early as 2025 as the tech giant's leading chip supplier, TSMC, has set in motion a plan to produce that process in the early parts of the said year.

Apple to Adopt 2nm Process in 2025

According to DigiTimes, all of Apple's latest chips feature the 5nm process, including the A15 Bionic in the iPhone 13 series and the entire M1 Apple silicon line.

TSMC will start the mass production of the 3nm chips in late 2022, with 2nm following in 2025, with Apple and Intel being among the first to use the newer technology.

According to MacRumors, TSMC has set a timetable to move its 2nm GAA process to production in 2025 while commercializing its 3nm FInFET process with improved yield rates in the second half of 2022, with Apple and Intel among the first ones to adopt both nodes, further consolidating its dominance in the advanced foundry sector.

Also Read: Dropbox Gets Ready for Apple M1 Chips After Previously Saying They had 'No Plans' to Support It

A report from 2021 claimed that the next iPad Pro, expected to be announced in late 2021, will feature a 3nm process. The current iPad Pro features the M1 chip, and the 2022 version is expected to include Apple's all-new M2 chip.

The 3nm process technology features performance improvements of up to 15% while being at least 25% less battery-hungry than the other processes.

Apple Unveils M1 Ultra

In March, Apple announced M1 Ultra, which is the next leap for Apple silicon and the Mac. The M1 Ultra features UltraFusion, which is Apple's innovative packaging architecture that interconnects the parts of two M1 Max chips to make a system on a chip with unprecedented levels of performance and capabilities.

M1 Ultra also delivers amazing computing power to the new Mac Studio while maintaining industry-leading performance per watt. The new SoC consists of 114 billion transistors, the most ever in a personal computer chip, according to Apple.

M1 Ultra can be configured with up to 128Gb of high-bandwidth, low-latency unified memory that the 20-core CPU, 64-core GPU, can access, and 32-core Neural Engine, giving excellent performance for developers compiling code.

The M1 Ultra also provides great performance for artists working in massive 3D environments that were previously impossible to render and video professionals who can transcode video to ProRes up to 5.6x faster than with a 28-core Mac Pro with Afterburner 1.

Johnny Srouji, Apple's senior vice president of Hardware Technologies, said that M1 Ultra is another game-changer for Apple silicon that will shock the computer industry.

Srouji added that by connecting two M1 Max parts with the UltraFusion packaging architecture, they could scale Apple silicone to unprecedented new heights.

Srouji also said that the M1 Ultra's powerful CPU, massive GPU, ProRes hardware acceleration, great Neural Engine, and the massive amount of unified memory complete the M1 family as the world's most powerful and capable chip for a personal computer.

The foundation for M1 Ultra is the powerful and power-efficient M1 Max. To create the M1 Ultra, the parts of two M1 Max are connected using UltraFusion, Apple's custom-built packaging architecture.

The most common way to scale the performance is to connect two chips through a motherboard, which typically brings trade-offs, including increased latency, reduced bandwidth, and increased power consumption.