Apple has introduced its new M5 Pro and M5 Max chips for the latest MacBook Pro, and this time the company is highlighting a new Fusion Architecture that changes how its high end silicon comes together. Instead of scaling performance by simply grouping multiple base dies as in earlier generations, Apple now connects two third generation 3 nanometer dies into a single system on a chip, which allows the company to increase CPU, GPU, and AI performance while keeping power efficiency under control.
In earlier Apple Silicon generations from M1 to M4, the Pro version effectively combined two base chips while the Max scaled further. With M5 Pro and M5 Max, Apple uses advanced packaging to fuse two dies into one unified SoC that integrates the CPU, scalable GPU, Media Engine, unified memory controller, Neural Engine, and Thunderbolt 5 controllers on the same package. As a result, the design reduces latency and increases bandwidth between components, which directly benefits pro workflows.
Apple detailed these changes in its newsroom announcement, where it described M5 Pro and M5 Max as its most advanced chips for pro laptops and confirmed that they power the new MacBook Pro models now available for pre order.
Fusion Architecture and “Super Cores”
At the center of this update sits the new Fusion Architecture, which connects two dies with high bandwidth and low latency links inside a single SoC. Apple also renamed its high performance CPU cores as “super cores,” a term it now uses across M5 based products.
M5 Pro and M5 Max feature a new 18 core CPU that includes:
- 6 super cores designed for the highest single threaded performance
- 12 new performance cores optimized for power efficient, multithreaded workloads
Apple says this CPU architecture boosts performance by up to 30 percent for pro workloads compared to the previous generation. The company also claims up to 2.5 times higher multithreaded performance than M1 Pro and M1 Max, which positions the new chips for complex simulations, data analysis, and demanding compilation tasks.
Johny Srouji, Apple’s senior vice president of Hardware Technologies, said:
“M5 Pro and M5 Max are a monumental leap forward for Apple silicon, leveraging our new Fusion Architecture to scale the capabilities of Apple silicon while preserving its core tenets of performance, power efficiency, and unified memory architecture.”
He added:
“Both chips underscore our relentless pace of innovation, integrating the world’s fastest CPU cores, a next-generation GPU with Neural Accelerators, a faster Neural Engine, and high-bandwidth, high-capacity memory.”
GPU, AI, and Memory Gains
On the graphics side, Apple scales the next generation GPU architecture up to 40 cores on M5 Max, with a Neural Accelerator inside each GPU core. Apple says the new chips deliver over 4 times the peak GPU compute for AI compared to the previous generation. For apps that use ray tracing, the company claims up to 35 percent higher graphics performance compared to M4 Pro and M4 Max.
Unified memory also scales significantly:
- M5 Pro supports up to 64GB of unified memory with up to 307GB per second bandwidth
- M5 Max supports up to 128GB of unified memory with up to 614GB per second bandwidth
This increase in memory capacity and bandwidth benefits large language models, 3D rendering, complex simulations, and high resolution video workflows.
Both chips also integrate:
- A faster 16 core Neural Engine with higher memory bandwidth for on device AI
- An updated Media Engine with hardware acceleration for H.264, HEVC, AV1 decode, and ProRes encode and decode
- Memory Integrity Enforcement, which Apple calls an always on memory safety protection without performance tradeoffs
- Thunderbolt 5 controllers directly on the chip, with each port powered by its own custom controller
Apple positions M5 Pro for data modelers, sound designers, and STEM students who need strong CPU and GPU performance with large unified memory. Meanwhile, M5 Max targets 3D animators, app developers, and AI researchers who require maximum GPU compute and the highest memory bandwidth.
With Fusion Architecture and the introduction of super cores, Apple is clearly pushing its pro silicon strategy further, focusing on scaling performance within a tightly integrated SoC design that emphasizes CPU speed, GPU compute, and on device AI.
