Apple
Announced at its fall Unleashed event in late October, Apple brings us two more high-powered members of the M1 homegrown chip line it launched a year ago, which debut in its shiny new MacBook Pro 14 and MacBook Pro 16 laptops. While they mark a new level of performance available in Apple’s laptops, the Apple M1 Pro and M1 Max also represent another step forward in Apple’s two-year plan to ditch its use of Intel’s competing processors.
The new chips arrived just prior to Intel’s launch this week of its 12th-gen Alder Lake desktop CPUs, which use a similar architecture — incorporating both high-performance cores and low-power cores, a design borrowed from phone chips to better balance performance and battery life.
The real competition between laptop performance will heat up when Intel delivers its mobile versions of the 12th-gen chips in early 2022 and Apple overhauls its iMac 27-inch (and possibly Mac Pro) with top-of-the-line M1s, rumored for mid-2022 as well. It’s not fair to compare Intel’s desktop CPUs to Apple’s mobile chips today.
Watch this:
Apple reveals M1 Pro chip
3:20
Apple’s bound to confuse people with its options, though. The three different lines of M1 chips actually break into multiple different versions you can configure a system with, thanks to variations in the number of CPU and GPU cores that fall under the same model name.
For instance, there are two versions of the M1 Pro, one with eight CPU cores and 14 GPU cores and one with 10 CPU cores and 16 GPU cores. It’s not an atypical practice, but at least companies like Intel and AMD provide clear naming conventions for each, while Apple opts for the friendlier but oversimplified approach that makes comparisons more difficult.
Apple M1 vs. M1 Pro vs. M1 Max
| Apple M1 | Apple M1 Pro | Apple M1 Max | |
|---|---|---|---|
| Total CPU cores | 8 | 8 or 10 | 10 |
| Performance cores | 4 | 6 or 8 | 8 |
| Efficiency cores | 4 | 2 | 2 |
| GPU cores | 7 or 8 | 14 or 16 | 24 or 32 |
| Neural engine cores | 16 | 16 | 16 |
| Maximum memory supported (UMA) | 16GB | 32GB | 64GB |
| Peak memory bandwidth (GBps) | n/a | 200 | 400 |
| ProRes accelerators | None | 1 | 2 |
| Available in | MacBook Air, iPad Pro (5th gen), iMac 24 (2021), Mac Mini | MacBook Pro 14, MacBook Pro 16 (2021) | MacBook Pro 14, MacBook Pro 16 (2021) |
The M1 Max and M1 Pro clearly differ in terms of peak performance, with the M1 Max doubling some important contributors to theoretical performance, notably the number of hardware ProRes accelerators, which is extremely important for pro video editing, and doubled bandwidth for some internal interfaces, like that between memory and the processor. Those won’t necessarily result in double performance delivered in real-world use, though.
One thing all the M1 implementations share thus far is no mention of discrete graphics support. Apple has beefed up the potential integrated GPU performance and boasts that the M1 Max supports the most memory available to a mobile GPU, up to 64GB.
Watch this:
Apple unveils M1 Max chip
4:16
Keep in mind that Apple’s unified memory architecture means that memory is shared by the GPU and CPU; that’s definitely better for performance, since it makes both processing and graphics tasks faster at passing data between each other. In comparison, Intel’s UMA limits the amount of memory accessible by an integrated GPU, and the PCIe-based Resizable BAR used by discrete GPUs only allows the CPU to access video memory, not the other way around.
But if you’ve got applications that are simultaneously demanding of both the CPU and GPU, such as complex 3D simulations, you may still run into some issues.