First of all, I wanted to thank you for taking your time to write down your experience with such detail and kindness.
I appreciate you saying so. It's sometimes easy to forget that we are interacting with real people on these forums, often times distracted by a small minority that have an agenda, making it easy to become jaded, so I'm glad my response was seen as well-intentioned. You are clearly facing a dilemma, one that could impact you long-term, both financially and in practical terms, and you deserve to be taken seriously.
Secondly, I’ve already thought about waiting few months and getting the M2, however, as many of you already know, the M2 offers pretty much the same as the M1 SoC, and the next Mac mini will come with a big price increase because inflation + euro situation.
Unfortunately, the global impact from the plague, war, political instability, recession, chip shortages, and cyclical market forces have impacted us all. A computer purchase today is a substantial monetary outlay, one which can't be boiled down to a soundbite on a forum, or numbers on a benchmark, hence we all need to consider our options quite carefully. As such, I tried to weigh my response to your inquiry accordingly.
Some people on this thread are already mocking the fact that many people are waiting for the next 3nm SoC, but they are probably ignoring the fact that the M2 is a very minimal refresh of the M1.
As I stated previously, I'm currently using a heavily upgraded 2018 Mac mini, featuring a 4-core i3 CPU, but it's satisfactory for my needs, my requirements don't exceed its capabilities, and I won't be replacing it until I need to. I haven't put a timeline on a future purchase for myself; while I'm planning on upgrading roughly once the M3 generation makes it to the Mac mini or Mac Studio, depending upon my use case, the demands of the software I use, and Apple's timeline, it could be an M4. It's always a good idea to keep your options open.
My plans aside, I don't think that the M2 is as minimal an upgrade as some are portraying. The perception problem is that the jump from Intel to Apple Silicon was massive in all respects: efficiency, security, performance per watt, value for money, power consumption, device simplicity, etc. It was impossible for Apple to follow that after such a massive overhaul. Even hardcore Windows PC enthusiasts have been forced to admit the fundamental impact that Apple Silicon has had on the entire industry, even for those who will never buy a Mac. The M2 faced an impossible task, when compared to the uplift from initial Apple Silicon.
That being said, while the M2 was far more evolutionary, rather than revolutionary, it brought its own advances. Unfortunately, many of those details are not getting highlighted because deep dives are no longer being done by the experts at Anandtech, having left the publication to work in the industry, leaving us with little options for unbiased analysis.
The remaining reviews are being done by a limited range of YouTube channels such as Linus Tech Tips (barely competent) to Max Tech (barely incompetent), who are more interested in clickbait headlines for their videos than detailing the enhancements that Apple made to the M2. Most benchmarks are nothing more than synthetic attempts to replicate real-world scenarios that can somehow be run in 30 seconds.
From what I gather, the M2 is a decent upgrade over M1, but nothing earth shattering. Even with TSMC's 3nm process, we're probably not going to see a gigantic upgrade with the M3 compared to the switch from Intel; that was a once in a product lifetime upheaval. M3+ will probably be more impressive than M2, but not to the degree that the hype machine would have us think. Intel's stagnation on 14nm++++++ aside, historically speaking, the majority of CPU upgrades see modest improvements, which is likely to be the case with the M-series going forward. Apple still has significant technological advantages, including the RISC ISA, TBDR GPU, preferred customer status with TSMC, excellent semiconductor design team, and the ability to control the entire product, from SoC to OS.
For a professional opinion on M2, I would suggest consulting Cliff Maier
in the Apple section of that discussion forum, who posts under the handle "Cmaier". He used to visit MR, so many here are familiar with him, but for those who aren't, he's a former Opteron architect and wrote the draft for x86-64. So, literally billions of people are impacted by the technology he helped created, every day, whether we realize it or not. He's quite friendly and willing to answer any questions you may have, so I encourage anyone here that wishes to consult a CPU architect to chat with him, he's approachable and fields any question sent his way. Cliff worked with many engineers within Apple's semiconductor design team over the years, is friends with them, talks with some of them regularly, and has full confidence in their capabilities. He says that most of talented folks are are still working at Apple, despite
claims that there is a brain drain going on over at the fruit company in Cupertino.
While I am no CPU architect, from what I gather from his posts, the following are the main benefits from going to M2 from M1:
• The efficiency cores received a major overhaul in performance and power consumption, which won't show up on 30 second synthetic benchmarks.
• The performance cores were enhanced to increase raw clock speed by about 10%.
• The GPU was improved to increase performance considerably, well beyond what you'd expect by just adding two more cores. As
I pointed out elsewhere, many computer games saw increases reaching 100% gains over M1:
Witcher 3 (running under CrossOver): 55%
Minecraft (native Apple Silicon): 50%
Metro Exodus (Rosetta 2): 100%
Warhammer III (Rosetta 2): 100%
Doubling of GPU performance isn't a small upgrade. There's also the increases in memory bandwidth and cache size, so despite M2 being on the same node as M1, Apple's engineers were hardly sitting still. I'm sure Apple's software engineers are responsible for some of those GPU gains considering how closely they work with the hardware team. Cliff Maier has often talked about how his team at AMD "worked closely" with Microsoft to implement x86-64 within Windows, but that is nothing compared to how closely the macOS and Apple Silicon teams work together.
So, for now, I will keep an eye to the second hand market only looking for 16GB/512GB machines with a good price (what we call a good deal), and in the meantime, I’ll wait to see what Apple unveils in October.
It's always good to keep your options open, and it sounds like you will be doing just that. If a second-hand M1 is capable of doing what you need, then that may be the correct choice, in your use case. I've owned four Mac minis since I switched to the Mac in 2005 after having been a life-long Windows user. Of those minis, the second was a refurb from Apple because, at that time, it was the best value for what I needed it to accomplish.
I've never been a big fan of "future proofing" a computer; buying more than necessary "just in case". At the same time, I knew I was purchasing an under-powered system when I got the base model 2018 Mac mini, with a Core i3 and anemic 128GB internal SSD. Usually, we already know what is going to be a weak spot in the specs of a computer we buy, it's just matter of whether that tradeoff is worth it. I was able to supplement my Mac's deficiencies with various upgrades, such as 64GB of system memory and an eGPU, but I knew those were options before I purchased the computer.
It's more difficult to plan for eventualities, because Apple Silicon Macs are the epitome of Steve Jobs' vision; vertical integration that he could only have dreamed of for the Mac. While that tight integration yields tremendous benefits, it makes it more difficult to plan ahead. I just don't want you,
@Populus, or anyone else reading this, to make the same mistake that I did. A "stopgap" typically makes a lot of sense in the moment, but can bite you in the future, when you realize you need more than it offers. I'm not saying that it is a poor plan, it might be exactly what you would benefit most from, I'm just saying that it's best to examine all of your options and look at the big picture, which thus far, you seem to be doing.
Regardless, I wish you the best in your future computing endeavors. Seeing how you're coming from a 2014 Mac mini, I'm sure you'll be amazed by your next system. I did some benchmarks on my Core i3 Mac mini, and was surprised to see how much better the base M1 is compared to it. My next Mac is either going to be a Mac mini featuring an M(x) Pro Apple Silicon SoC, or a base model Mac Studio with a M(x) Max. So, whenever I do finally upgrade to an M3+ generation from my i3 Mac mini, I'm sure I will be blown away by the performance.
