I'm using a Late 2009 iMac for some things and it's running High Sierra. I have built a Hackintosh but it ran poorly because the hardware was so old. I also run a macOS VM which some consider a Hackintosh. It's just a convenience as I could also just VNC into one of my old MacBook Pros. I expect to upgrade one of my Intel MacBook Pros to an M1X in 2021. I will also keep one of the Intel MacBook Pros in case I need to run 32-bit software or Intel macOS software.
My main work system runs Windows. I do not know if I'll transition over to an M1 system for main work as I have one program which runs far better on Windows 10 than it does on macOS. It kind of runs on Intel macOS but I've heard awful things on it running on M1 macOS. The Hackintosh may still be attractive if Apple continues to charge a fortune for RAM. I like 64 or more GB of RAM in my systems and Apple may not sell such a configuration or may make it extremely expensive.
The M1X should more than well meet my needs. My current work system is 1,280/8,200 Geekbench 5 and I'd expect the M1X to be around 1,700/14,000. I'd want 3x4k support which shouldn't be a problem if they double the GPU cores. I'd be pretty happy if they could put all that in a Mac Mini. Or an iMac. My plan is to get a desktop and a laptop. The desktop would likely sit next to the Windows desktop and take one or two of the monitors. The systems would be tied together via Synergy KV. It's possible that I would add a fourth monitor to my setup too.
So everything is on the table; even Intel if they can deliver Alder Lake before everyone else (I seriously doubt it).
I would think the limitation on newer Intel CPUs is that Apple won't have support for them in the x86-64 version of macOS. That's my main question there. Again, I think there's enough 10th generation support (between the 2020 Intel Air, 4-port 13" MacBook Pro, and 27" iMac) that it probably wouldn't be an issue to build a 10th Generation based Hackintosh. But, I'd imagine that it starts getting more difficult with 11th Gen and beyond (seeing as Apple is extremely unlikely to adopt it in any future Macs).
Not sure about the T1 Macs but by default according to Apple the T2 Macs boot process is similar to Apple Silicon Macs and iOS devices.
The stuff in this article seems to imply the opposite:
https://eclecticlight.co/2021/01/14/m1-macs-radically-change-boot-and-recovery/
This is not surprising really, the T2 is an A10 ARM64 SoC co-processor. The T2 Macs are really hybrids of Intel and Apple Silicon.
That much I'll totally buy. But I think where T2 and M1 Macs still differ is the firmware and booter. M1 Macs don't have UEFI the way all Intel Macs (including T2 Macs) do. But as for them being hybrids of both worlds, that's an absolutely fair assessment.
I agree that MacOS on non-Apple ARM64 is unlikely to work. It's even possible that Apple could drop support for non-T2 Intel Macs before they drop Intel support completely.
I think this is almost guaranteed to happen. The T2 was so substantial and there are definitely likely to be features that it shares in common with Apple Silicon SoCs that Apple will deem essential for continued OS support. Hell, it might be the final dividing line between unsupported and supported Intel Macs for a future macOS release before Intel Mac support is inevitably dropped altogether.
To answer your original question, if I was going to build a Hackintosh, I would stick to a 10th gen CPU. It's not like the 11 gen desktop chips are much of an improvement anyway.
I've seen some pretty good bargains on 9th Gen Intel CPUs and would probably gravitate toward those (it's not like the few 9th Generation Intel based Macs aren't still highly performant beasts as far as Intel Macs are concerned).
My questioning regarding 11th Generation was more of a general "what happens to the ability to make an Intel-CPU-based Hackintosh when using a generation of Intel CPU that isn't used by Apple and is or isn't that going to be similar to the relative struggles of building AMD based Hackintoshes.
Just booting macOS on Non-Apple ARM hardware won't be an issue.
Really? I would think the custom bootloader and firmware would throw a bunch of obstacles in the way of that.
AFAIK someone already has booted M1's Darwin build on a Raspberry PI.
Do you have a link? I'd love to read up on that!
As mentioned in my previous post, the culprit is getting it to work well enough to be a viable option and I really doubt that will ever happen. The M1 consists of so many custom components that macOS is tailored towards that developing all the drivers and patches for third party ARM SoCs would be a huge task. Basically we would need custom drivers for pretty much everything: Graphics, hardware encoders / decoders, ISPs, networking, storage and so on and so forth. And that would be the case for each and every PC ARM SoC that might come in the future. A lot of that hardware will be proprietary, making the job even harder. ARM Hackintoshes won't happen. At least not if you expect something suitable for daily use.
Something that might be a viable option is virtualization thanks to the reverse engineering work of the Linux developers.
Certainly, ARM Hackintoshes would be A LOT more difficult than PowerPC Hackintoshes would've been.