The Apple Mac Mini is getting its first major update since 2020 this month. The new Mac Mini is up for pre-order now for $599 and up and it will be available starting January 24, 2023.
For the starting price you get a model with an Apple M2 processor, 8GB of unified memory, and a 256GB SSD. But the system can be configured with up to an M2 Pro chip, 32GB of memory, and 8TB of storage (if you’re willing to pay a lot more for some of those upgrades – prices start at $1299 for models with M2 Pro chips).
Apple says even the entry-level models offer significant performance boosts over the previous-gen model, with up to 50 percent faster filter and function performance in Adobe Photoshop or up to 2.4x faster ProRes video transcoding in Final Cut Pro.
You can expect even more of a performance gain from models with the M2 Pro chip, which has more CPU and GPU cores, support for more memory, and twice the memory bandwidth at 200GB/s.
The 2023 Mac Mini measures 197 x 197 x 36mm (7.8″ x 7.8″ x 1.4″) and has an aluminum case (made from 100% recycled aluminum). While there are no ports on the front, if you look at the back of the computer you’ll find HDMI, Thunderbolt 4, Ethernet, USB Type-A, and headphone jacks plus a power jack and power button.
Models with M2 chips feature two Thunderbolt 4 ports, while M2 Pro models have four. Each supports 40 Gb/s data transfer speeds and DisplayPort alt mode for video output. And that Ethernet port? By default it supports Gigabit speeds but you can pay an extra $100 for 10 GbE Ethernet.
Apple says 2023 Mac Mini computers with M2 chips support up to two displays, while M2 Pro models can support up to three displays or a single 8K display, making it the first Mac Mini with that capability. All models support WiFi 6E and Bluetooth 5.3.
Too bad Apple’s gorilla-math for the price of upgrades still makes absolutely no sense.
Upgrading the $599 Mac Mini to 16gb RAM, and 1tb storage doubles the price to $1200.
I use an M1 Macbook, and I’d love an M2 Mac Mini, but I’m not paying those prices.
I just did some comparisons against a similarly priced Mini ITX build, and I just can’t see the value in the Mac Mini, unless you’re totally okay with the specs of the $599 model.
There’s still no reason for Windows users to switch to a Mac Mini above the base-spec, unless they’re already attracted to MacOS or the ecosystem.
The M2 wins easily against similar Intel/AMD price/performance/battery comparisons, but only as long as you only talk about laptops.
As soon as you talk about desktops, there’s no more battery life to talk about, and now we’re comparing desktop-class chips, which are more powerful than laptop chips. The desktop-class Ryzen 5 7000-series chips are significantly more powerful than the M2.
I think I’ll stick with the better value, performance, upgradability, and flexibility of x86, when it comes to desktops. And I’ll stick to my M1 Macbook for my laptop.
What is “Unified memory”?
System RAM is also used as VRAM, as it is with anything that uses integrated graphics.
Not quite “as it is with anything that uses integrated graphics”. It is shared memory, yes, but the memory has a direct connection to the CPU/GPU at a much higher speed than traditional DDR memory on x86_64 machines. This translates to much faster performance in both CPU-heavy and GPU-heavy workloads, and the OS can seamlessly allocate as much or as little RAM to each as necessary to get the performance needed for a particular workload.
This article explains it much better than I can:
https://www.macobserver.com/analysis/understanding-apples-unified-memory-architecture/
Physically it is DDR. I would suspect that the normal CPU/GPU memory controller was modified to allow AXI direct access for any/all AXI peripherals. Most likely the NPU was the most latency sensitive peripheral that would benefit from this approach.