My new AMD linux laptop compiles my projects faster than my old M1, and that's good enough for me. It has 32GB of ram, 4tb of storage, and was half the price of the equivalent Macbook. Apple may make the best laptops right now, but other manufacturers are closing the gap.
I'm personally looking at the middle tier M4 Pro 16" 14/20 with 48gb of ram to replace my 2019 13" i5 with 16gb and 256gb ssd.
It's somewhat frustrating, but although technically I can get by with my current machine, it's starting to feel foolish and needlessly frustrating to hang onto it longer, it's just so sluggish these days. Maybe I'd feel differently if I had got a more higher-end one 5 years ago. Windows remains as my old gaming PCs OS, idk that I'd want to contend with it daily for work or anything unless the software required it, but then I'd probably just upgrade the bits in my PC and stick with my older mac laptop for basic personal use.
Even though I did buy it, I can't say I can get past the cost of upgrading the various bits, but just decided that I should pick a reasonable option and not stress about it too hard, it'll be a nice change, I might even return it within the window if I feel sufficiently robbed. I couldn't wrap my head around the even more comical Max versions.
Maybe they're just trying to capture missing revenue from us never-iPhone people. It's the only part of the Apple ecosystem I participate in aside from an ancient iPad.
I saw some videos from China [1] where people were upgrading the storage in the new M4 Mac Minis. The Mac Mini's SSD is on a removable pcb but it is raw nand flash as opposed to, say, typical m2 2230 SSDs that have a built-in controller. So they had to desolder it. But getting someone in China to desolder and upgrade the SSD to 2 TB is like 90% cheaper than buying it from Apple.
This is unfortunately too late, to justify recouping costs and buying a new M4 Max.
Primarily because I would be using it for large language model inference at higher parameters. What's happened in just one year is that LLM's have gotten muuuch smaller. Llama 3.2 3B only takes 3GB of RAM at 8 bit quantizing. And in addition to that, cost per token in the cloud has subsequently plummuted 99.9% too. Its just not economical.
But yes, I could afford it but no longer a justification. My 64GB RAM M1 Max is going to be future proof for a while.
Can you hear the fans spin in a typical week?
It's not exactly comparable to my MBP, but neither is the cost! The FW cost me about $1,100 all-in and the MBP was $2,900 (both 32GB/1TB).
1: https://knowledgebase.frame.work/optimizing-fedora-battery-l...