The tech powerhouse Apple has recently made public its latest innovations in silicon, introducing the M3, M3 Pro, and M3 Max chips. Heralded as the 3nm chips’ premiere on the personal computer scene, Apple assures that rendering times are now 2.5 times faster compared to those calibrated on the M1 series. It also outlined enhancements in the CPU performance cores, efficiency cores, and the Neural Engine, which respectively achieve speeds that are 30%, 50%, and 60% faster.
With respect to the new M3 chips, they have been unveiled together with new iMac and MacBook versions. An interesting facet of the chips is its GPU, labeled by Apple as a hallmark in the history of its silicon graphics architecture. This GPU features Dynamic Caching, a ground-breaking technology that allows hardware-accelerated ray tracing and mesh shading on Mac devices, providing an enhanced rendering experience.
Johny Srouji, the Senior Vice President of Hardware Technologies at Apple said, “Apple silicon has completely redefined the Mac experience. Every aspect of its architecture is designed for performance and power efficiency”.
The M3 chips also incorporate an element deemed as a unified memory architecture. According to Apple, this enables all technologies within the chips to access an identical set of data from a single memory pool, leading to performance and efficiency enhancement. It also reduces the amount of memory required by a system for tasks due to reduced data copying among multiple memory pools.
Additionally, the M3, M3 Pro, and M3 Max each belong to a distinct performance tier. The basic M3 version showcases 25 billion transistors and provides a graphics performance which is 65% faster than its predecessor, the M1. The M3 Pro sits in the middle, houses 37 billion transistors, and offers single-threaded performance that’s up to 30% faster than M1 Pro. Last but certainly not least, the M3 Max, with its 92 billion transistors and a GPU with 40 cores, is presented as the mightiest among the trio, with CPU speeds that are 80% faster than the M1 Max.
Despite appearing as a mere enhancement of Apple’s desktop and laptop computers’ processing power, these changes also imply a potential shift in the broader landscape of PC CPU chips. Historically dominated by Intel and AMD with x86 technology, the market now witnesses more competition, as a result of innovations from companies like Qualcomm, who, just last week, unveiled the Snapdragon X Elite, their first high-end PC chipset with 2x improved speed against x86 CPUs.
As further proof of this shift, GPU specialist Nvidia and AMD are reportedly planning to design Windows PC CPUs, based on ARM architecture. This disrupts the historical monopoly and may spell an exciting new era for the PC CPU chip sector.