Will hackintosh be a thing once the ARM transition is complete? Surely Apple will bake in some security into their silicon that will prevent OS11 from running on anything but Genuine™ Apple Bionic.
Doesn't need to be security related at all, where will you get the drivers for your hackintosh ARM GPU, Network, Bluetooth, etc. With Intel Apple was using standard off the shelf components, with ARM it's all proprietary and baked in to their SoC - why would they write support for any other SoC ? Not to mention they will probably include proprietary chip specific stuff and assume it's there which will not be present on other devices.
where will you get the drivers for your hackintosh ARM GPU, Network, Bluetooth, etc.
The community --- providing it still exists --- will write them... or at least some of them. When I was playing around with Hackintosh many years ago (when they were still 32-bit), I took the lack of drivers as a motivation to learn how to write some. Of course, the PC legacy really helped with things like BIOS and related standards, so ARM will definitely have a much steeper curve.
(The diversity of ARM platforms in general put me off writing code for them --- it feels like a losing battle when there are so many platforms, especially when most of then are unlikely to even exist for much time. The fact that it is the same CPU core matters little when everything else is different. Maybe it is the same for many others.)
How long did it take for opensource drivers for stuff like Ras Pi GPU ? On the flipside is there any support for recent nvidia cards in hackintosh ?
I seriously doubt the effort required to write a functional GPU driver is going to come from OSS for something as fringe as Hackintosh on a random ARM SoC, like you said they get replaced constantly and the landscape is fragmented.
I think an important part of the Hackintosh ethos is running on commodity hardware. Even though I'm sure Hackintosh will be ported to Arm I'm not sure it will be very popular because of a lack of commodity hardware. This of course may change in the coming years as ARM moves more into servers and desktops. Apple's move will surely accelerate this.
In the meantime, Intel Macs will be alive and well. There is a 100 million Intel installed base. I doubt that Apple will be treating those customers badly in terms of support and 3rd party software vendors certainly won't. At current sale rates it will take Apple 5 years to draw even with an ARM installed base. Also the question of how popular the move to ARM will be is still open, given the severe backwards step in terms of compatibility. Apple may continue to offer some Intel hardware a bit longer than they intend.
Finally, for me at least, the thought of being stuck at some final version of macOS that supports my Intel Hackintosh is kind of appealing. Each major update seems like a step backward. Hopefully developers will appreciate these Intel diehards and support them longer than Apple does.
Agreed. People need to stop thinking about this as a change in ISA. The important part comes when Apple starts loading their silicon with custom, specialized components aimed at improving OSX.
Apple is unlikely to ever have any major SoC features which are generally useful to end users that are exclusively available in their chips for more than a year or two. Even if Win10 ARM vendors or whatever’s coming for PC are playing catch-up with Apple, we’re unlikely to see major divergence. People still need to get roughly the same things done whether they prefer Apple or otherwise.
It's not enough to have an equivalent capability - if you want a Hackintosh, you'll need to present those capabilities in a way that fools macOS into believing it's running on supported hardware. For now, ARM has been limited to the low-power desktop/mobile and high-end server niches. Apple Silicon will go for the space between as well.
OTOH, Apple is smart enough not to tightly marry its software to a given hardware architecture - they were compiling Rhapsody for Intel from day one and a lot of iOS is directly lifted from macOS. Its lineage started running on 68K, then HP-PA, SPARC, x86, then mostly PPC, then mostly x86 again, and now it'll switch to mostly ARM.
Apple's custom silicon play is not about the users.
Its about the media vendors.
Apple can afford to outright buy most of them. What I think they are doing, is preparing themselves for a time when they can own media, as well as the devices you use to view/experience it.
I don't think the Piracy wars are over, guys. They're gunning for us, and .. unless we've got tunnelling electron microscopes and other tawdry machines, its gonna be harder and harder to crack.
Hypervisor + Emulation? Unless they did some very serious obfuscation of the binaries they distribute it wouldn't be particularly difficult to guess what their new hardware did. Obviously wouldn't be as fast but it seems doable.
You'd need a computer that's substantially more powerful than a Mac in order to emulate one. I believe Apple's premium will not be high enough to make this economically viable.
As long as there's mutable software binaries, DRM can be subverted. If core OS11 software lived on a special R/W chip instead of a disk, that'd be quite a move. Apple hasn't even done that with iPhone though.
Huh, I'd assumed they were doing some type of emulation thing (which is still impressive as heck), are they actually executing native instructions on a non-Apple CPU?
Actually, I'd love to read more about Corellium in general—have they made any technical details public, even high level? They seem to be very secretive about how it all works—which, to be fair, I can totally understand.