There's plenty of native linux support for ARM, and there's a native Linux distro for Apple Silicon (https://asahilinux.org/), there's also native Windows for ARM though not running on AS (yet? We'll see if eventually that happens)
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Signs New Deal With Arm to License Chip Designs Beyond 2040
- Thread starter MacRumors
- Start date
- Sort by reaction score
They never will, but if they allowed M2 Ipad Pros to run parallels and windows, I would hands down buy one right this second! I have a buddy that modded his to run it and it was killer. I just want apple to let go and open it up.
The problem with ARM is there’s two licensing schemes. One where you buy the rights to their CPUs and the other you buy the rights to make your own. The latter is a much much cheaper option and doesn’t generate as much for ARM.
Apple has an architecture license (the second one you mentioned).
They’ve been *designing* their own chips for years now, not packaging ARM’s reference designs. Apple’s license is more so that their design can implement the instruction set and not be vulnerable to patent infringement lawsuits.
Now, I wouldn’t expect tech “journalism” to explain the difference to their readers, because they’re still saying Apple’s chips are ARM designed…they aren’t.
?
you said: "cant get the big tech companies to agree on anything"
they literally did with Matter. so you are factually incorrect.
I wouldnt say it's "too different", Apple could definitely move to RISC-V if they had a compelling reason.
I have, in my home office, an array of macs 68k-->PPC-->Intel-->AS/ARM. Apple has been quite competent at switching core CPU architectures.
I *would* say that Apple has invested a truly huge amount into AS and its implementation of the ARM ISA, and for that matter ARM in general (and literally from the beginning, they may have left off it for a while but Apple was the primary financial investor in ARM when it started, in collab with VLSI [now part of NXP née Phillips' Semiconductor group] and Acorn), and it would take a pretty big reason for them to move off the ISA.
Their investment is not small, but much of that investment (most of their chip design work) is portable - applicable to another risc architecture like RISC-V.
Switching would definitely be a significant endeavor, but well within their capabilities.
My sarcasm aside, if you notice in the spec, page 49:
“To support home automation interoperability, this protocol supports the concept of bridging which makes available, through a data model node, devices implementing other home automation tech nologies, transports and link layers.
On page 484, it explicitly allows bridged devices not to be seen.
On page 485, it allows manufacturers to bridge devices, not notify users they are bridged, and thus may not implement the same security protocols that exist in the Matter spec, or any for that matter.
So as I said earlier:
Having worked on a standards committee, the result is usually so broad to include all desired features to get “agreement” which means just because something complies with the standard means it will work with 2 different devices. Look at USB-C and RS -232 for example.
So once again you have a standard that allows for custom implementations and depending on how bridging is implemented may or may not provide 100% feature compatibility across manufacturers. If the all had agreed to settle on Matter no bridging would be allowed. In addition, companies are free to implement only parts of the spec, they just can’t use the Matter logo since it would not be certified; and software only does not require certification.
Belkin has already decided not to support Matter; creating a standard and agreeing to use it are two different things.
A similar situation exists in the USB-C spec, where proprietary implementations can still comply fully with the specification; or in the EU case alongside the defined ones for PD when used for powering devices.
History has shown that a specification does not result in one cable to rule them all, despite how some may wish it did; nor would major tech companies buying into ARM result in one standard cross compatible ARM implementation.
So you could argue they have “agreed” to something, but it is a minimal set of requirements that allow deviations and incompatibilities and still be compliant; which was my point even if I made it sarcastically.
Last edited:
You basically just restated my post without the history lesson detour 🤷♂️, so yup lol
Not entirely. My point is that while their investment in ARM architecture is large, the barrier to switching to another arch is not nearly as large, because most of that investment (or at least, most of the surviving value of that investment, since it depreciates every year) is portable.
I would be shocked if they didn't already have a quality RV implementation, and would be unsurprised to learn that it's already in use as a controller in multiple places in current A and M chips - though I don't know that to be the case.
precisely why I wont be buying any further smart home products from Belkin. or from any company/product not participating.
so only companies that agreed to participate will see my money. it's a pretty simple concept