Apple's M6 chip proves TSMC's 2nm node is ready for next-gen PC processors

Reviews of Apple's latest M6 SoC reveal the chip is performing exceptionally well, and the significance extends beyond Mac users to the broader PC gaming market. The M6 uses TSMC's latest N2 silicon node, a development process that historically signals major performance and efficiency improvements across the industry.

The M6 demonstrates substantial gains in real-world benchmarks. According to Notebookcheck, the M6 scores 4,698 single-core points in Geekbench 6.7, compared to 2,872 points for AMD's Ryzen AI 9 HX 470, one of the fastest laptop CPUs available. The M6 also runs approximately 10 percent faster than Apple's previous M5 chip in the same benchmark. In multi-threaded workloads, the advantage grows significantly. The M6 features 12 CPU cores versus the M5's 10, resulting in roughly 50 percent more overall CPU performance in a single generation.

Power efficiency improvements are equally notable. The M6 achieves nearly 20 percent better performance per watt than the M5, and operates twice as efficiently as the Ryzen AI 9 HX 470. These gains demonstrate TSMC's N2 node is performing as expected after years of uncertainty around Moore's Law.

Intel remained stuck on 14nm for roughly a decade, raising serious questions about whether each new silicon process would deliver meaningful benefits. The M6's success with N2 suggests that concern may ease as the technology matures.

The timeline for N2 adoption in PC chips remains uncertain. AMD has already begun producing server variants of its next-gen Zen 6 CPUs on N2, potentially skipping TSMC's N3 node entirely for some products. Whether consumer PC variants of Zen 6 will use N2 remains unclear. Intel has begun using TSMC's N3 node for Lunar Lake and may adopt N2 for some variants of its upcoming Nova Lake family, likely launching in early 2025.

Nvidia's approach appears more conservative. Current Blackwell graphics chips use the N4 process, while next-gen Rubin GPUs are being built on N3. Rumors suggest Rubin may not arrive before 2028, potentially delaying an N2 Nvidia GPU until 2030 or later.

Despite the promising performance gains demonstrated by the M6, PC gamers should expect a lengthy wait before N2 becomes widespread in consumer hardware. Competing demand from AI chip production is currently limiting access to TSMC's advanced nodes. However, the M6's success confirms the process works reliably when available.