Apple Exec Explains How M5 Pro and M5 Max Use Two Adjacent Dies [Updated]


UPDATE: Corrections from Heise clarify that the M5 Pro and M5 Max do not use vertically stacked dies. Instead, the chips rely on a new Fusion Architecture that connects two adjacent dies within a single SoC, allowing different functional blocks to work together efficiently without stacking components on top of each other.

Original Article Follows:

Don’t miss the best of The Mac Observer

Set us as a preferred source and our Apple reporting ranks higher in your Google Search results and Discover feed — one tap, no account changes.

Or get it by email

Apple has introduced a major shift in chip design with the M5 Pro and M5 Max, moving beyond the traditional single-die layout to a more advanced structure that uses vertically stacked dies, which allows different parts of the chip to sit on top of each other and communicate faster while using less power.

This change comes through Apple’s new Fusion Architecture, which builds on earlier work from UltraFusion, but instead of combining two identical chips, Apple now splits functions across separate dies, giving each part a specific role and improving efficiency across CPU, GPU, and memory systems.

According to a report by Heise Online, Apple platform architecture employee Anand Shimpi shared key details about this design during a technical discussion.

“In a way, this is a newer version of a similar concept. With the earlier Ultra chips, we connected two identical SoCs together to form a larger SoC. Now, we’ve actually split a number of features into two different dies. They are not two mirror images of each other.”

How stacked dies improve performance

Stacked dies bring the chip closer to a 3D-style layout, where components sit vertically instead of spreading out across a flat surface, and this setup improves how quickly data moves between different parts of the processor.

Because the CPU, GPU, and other blocks sit closer together, the system reduces latency and increases bandwidth, which leads to faster performance in demanding tasks like video editing, AI processing, and software development.

At the same time, Apple focuses on keeping power efficiency in check, as tighter integration reduces the need for long data paths that usually waste energy.

The M5 Pro and M5 Max also introduce a new CPU setup that removes traditional efficiency cores and replaces them with performance cores and what Apple now calls “Super Cores,” which handle high-speed tasks more effectively.

“The super core is the fastest CPU core in the world. It is optimized for single-core performance.”

Apple also added a new type of performance core that sits between previous core types and handles workloads that do not scale well across multiple threads, which helps maintain efficiency without slowing down tasks.

Trade-offs are there as well

Stacking dies can increase heat because multiple components sit closer together, but early testing shows the M5 Max runs cooler than the M4 Max under heavy workloads, which suggests Apple has improved thermal management alongside performance.

At the same time, some details around the stacked die design still need confirmation, as different reports suggest variations in how Apple implements this architecture.

For now, the M5 Pro and M5 Max show how Apple continues to refine chip packaging, pushing performance higher while

Discussion

Join the discussionCommenting as a guest — your email is never published · Log in

Protected by Akismet — be kind, stay on topic.

This site uses Akismet to reduce spam. Learn how your comment data is processed.