For the very first time, at its ‘Scary Fast’ event on October 30, Apple unveiled not one, not two, but a total of three chipsets – the M3, M3 Pro, and M3 Max. These three chips have been mass-produced using cutting-edge 3nm technology, making them the first set of chips to be manufactured on this next-generation process. With this enhanced technology, Apple has equipped each member of the M3 family with more powerful CPU and GPU cores, along with advanced technologies.
While the M3, M3 Pro, and M3 Max do exhibit differences in their specifications, they also share common features, including hardware-accelerated ray tracing and hardware-accelerated mesh shading. Another significant change introduced by Apple in these chips is the dynamic allocation of local memory for their GPUs, allocating only the precise amount required for a specific task, as opposed to the traditional method where software determines GPU memory allocation. Additionally, Apple’s M3, M3 Pro, and M3 Max offer support for a variety of codecs, such as H.264, HEVC, ProRes, ProRes RAW, and AV1.
Commencing with the M3, Apple’s foundational chip boasts 25 billion transistors and provides support for up to 24GB of unified memory. Apple has adhered to an 8-core CPU configuration, consisting of four performance cores and four efficiency cores, similar to the M1. However, the company asserts that this setup delivers up to a 35 percent increase in performance compared to the M1 and up to 20 percent more speed compared to the M2. The 10-core GPU incorporates Dynamic Caching and offers performance gains of up to 65 percent compared to the M1 and up to 20 percent compared to the M2.
Talking about the M3 Pro. Apple has increased the transistor count to 37 billion, providing the capability to accommodate up to 36GB of unified RAM. The 12-core CPU is composed of six performance cores and six efficiency cores, and the entire setup is claimed to be 20 percent swifter than the M1 Pro. Similar to the M3 GPU, the 18-core GPU of the M3 Pro incorporates technologies like Dynamic Caching and delivers performance improvements of up to 40 percent in comparison to the M1 Pro, albeit it is only 10 percent faster than the M2 Pro.
Finally, let’s not forget the M3 Max, which boasts a staggering 97 billion transistors and offers support for up to 128GB of unified RAM. The 16-core CPU is comprised of 12 performance cores and four efficiency cores, marking a remarkable leap in performance as it achieves an 80 percent speed improvement over the M2 Max and a remarkable 50 percent gain over the M2 Max. As for the 40-core GPU, it surges ahead by being 50 percent faster than the M1 Max and up to 20 percent faster than the M2 Max.
Apple’s M3 series also introduces a swifter and more efficient neural engine, with the tech giant asserting that it outpaces the M1 family by up to 60 percent. The adoption of the 3nm manufacturing process across all three chips contributes to enhanced performance-per-watt. To illustrate this, Apple conducted a comparison with a 12-core PC chip, revealing that the M3, M3 Pro, and M3 Max achieve the same CPU performance while utilizing only 25 percent of the power. Furthermore, on the GPU side, these System-on-Chips (SoCs) consume merely 20 percent of that power for equivalent performance.
The trio of custom silicon will be featured in the upcoming 14-inch and 16-inch MacBook Pro models, which will introduce a fresh black finish and have a starting price of $1,599. In summary, Apple’s event was notably brief, but the introduction of the M3, M3 Pro, and M3 Max chips in the current year sets the stage for their anticipated launch in 2024, and we appreciate this proactive move by the company.