Apple's M6 chip, introduced with the Mac mini in August, represents a significant architectural shift to the 2-nanometer fabrication process. This move allows for denser chip design, resulting in notable performance improvements over previous generations. Die shots shared by High Yield and SemiAnalysis reveal the physical changes, showing the M6 is slightly smaller than the M5 despite using a more advanced process node.
The M5, built on TSMC's N3P process, occupies 154 square millimeters, while the M6, using the N2 node, measures 141.6 square millimeters. This size reduction enables Apple to fit more transistors into a smaller area. The memory interface remains at the top of the die, with the 16GB model retaining the same 9,600 MT/s DRAM bandwidth as the M5. Higher configuration models feature increased bandwidth via faster DRAM.
A key design change in the M6 involves the CPU configuration, moving from four super and six efficiency cores in the M5 to two super, four performance, and six efficiency cores. The efficiency core cluster remains separate, while the performance and super cores share large L2 caches. The M6's GPU cores are approximately 10% smaller than the M5, allowing for two additional cores within the same die area.
Source: AppleInsider · Summarized by HeadlinesBriefing