Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
You say something that is contradictory.
"Because Apple can modify the processor to add Apple specific private instructions (their license allows it as long as it adheres to the entirety of the ARM instruction set otherwise)"
That just means that the processor, provided with commands that should execute on all ARM processors, executes those successfully and completely, Apple can add custom instructions. This is currently implemented in the A13 processors, but I’m looking for the article from last year that referenced it :) It was around October when they announced that ARM is now allowing custom instructions.

Question: How do you add custom instructions and still maintain the ecosystem?

Segars: We’re trying to enable the best of all worlds. One key thing is that the custom instructions that can be added are purely additive to the core instruction set. The operating system that might sit on top of this processor isn’t going to rely on custom instructions. It might call a routine where, if the custom instructions are there, it runs very efficiently, and if they’re not there, it goes and does something else. But the core operating system that you can take off the shelf and run, that will still run on every ARM processor, because all the instructions that are defined today are still there.
[automerge]1591749335[/automerge]
Interesting. All Apple computer designs with CPUs from any company but Intel were commercial failures. We'll see if they can change this pattern.
Yeah, when you think about it, the Apple II WAS a commercial failure.
Glad I didn't shell out $4000+ AUD for the 16 inch MBP. It's been a fun ride with Macs. without my Legacy gaming Windows XP VM via Parallels, I just can't go down the ARM path.
Actually, if you had bought the 16 inch MBP, that would be the very definition of not going down the ARM path. Because Apple’s not shipping anything with an ARM processor in it yet.
 
Last edited:
  • Like
Reactions: chabig
RISC-V might have something to say about ARM ISA.
Not much.
[automerge]1591749589[/automerge]
Actually Xeons have 40-48 lanes.
Skylake desktop are 28/44.

You don't run at 1x until demand dictates.
PCIe negotiates link width and speed as the link comes up.
You need to do emphasis, pre-emphasis and equalization.
To do what you say, the link would need to go down and renegotiate speed and width.
That's not what devices do. I worked on and designed PCIe Gen3 switches.

Moden CPUs burn a significant amount of power driving I/O.
That is one reason why LPDDR is popular. Standard DDR consumes more power.
[automerge]1591748096[/automerge]


Once again.
If you add an instruction; the processor will no longer be compliant.
Arm only allows custom instruction in the Cortex-M embedded processor.
Just because they design the processor from the ground up means nothing about instruction set.
It allows them to chose implementation.
If you want to call a custom processor ARM (A12); it must pass the compatibility suite for the class of processor A7x, etc.

You can argue if you like, but I design chips for a living and some of them in the ARM ecosystem.
Apple doesn’t need to “call it ARM” :)
 
Not really what I’m saying. When I buy a Mac Pro which I did, personally I want to still be using it in a few years. But if arm comes I won’t want it anymore. And that’s a bummer!
I am feeling much the same way and I'm torn. I am within the return window for my 7,1 MP, but I am not sure if it is worth returning.

This thread is causing a lot of hype and panic that Apple is going to suddenly obsolete machines they just released within a year or two. Would they really do that with this all new 2019 MP design? It seems to me that they put a whole lot of effort into developing it. It is quite a polished machine and built phenomenally.

I would think that Apple has had a roadmap laid out long before they released it. So if ARM was right around the corner why wouldn't they just save it for that?

The 7,1 sure does feel like the modular design all us MP users were begging for. It seems like a machine that should still be current 10 years from now...?
 
Damn it, I just bought an Intel Mac mini, too (my first Mac). They'd better not make it "obsolete" soon!

mheh . . . every piece of hardware from Apple I've ever purchased was 'obsolete' by the time it felt my fingers ;)

Just enjoy it, while you can . . . .

Regards, splifingate
 
  • Like
Reactions: chabig
SJ goes on to say that Mac is set for the next twenty years with this transition. Using that timetable we were expecting the next transition to start in 2026. So if we start in 2021 and finish in 2022 that would shave off 5 years from SJ vision. Then the next transition away from the ARM in will happen in 2036 to the next architecture.
The original timetable was probably assuming Intel being consistent with their tick-tock cycle. However, recently it has become tick-tock-tock-tock cycle. And Imo even Steve would be pissed this time around.
 
Last edited:
  • Like
Reactions: JosephAW
Imagine how bulky, slow, hot and with terrible battery life our 2018 iPad Pro would be if Apple had to use Intel chips? Yikes!
Today’s iPad Pro would be imposible if Apple had just stuck with Intel.
 
My question is how graphics will be handled. ARM with a bigger chip size and more processing units plus a fan could run circles around current until chipsets I’m sure. Just needs proper coding and or emulation
Imo this is one issue with intel. Their integrated graphics is just sad, especially compared to the current Ax chip.

In PowerPC days, Apple was big on thermals and performance per watt. In the last few years, they added GPU in their priorities (as seen with how they focused on GPU on the Ax chips), and that part just don't match with intel's priorities.
[automerge]1591753835[/automerge]
What does this message now mean for an upcoming new purchase? That i would be an idiot to buy an "Intel" Macbook now?
Not really. Today is different compared to the PPC days. Buying an intel mac today is completely fine.

Besides, remember the first wave of intel Macs? They were 32bit only first Core Duo/Solo, and quickly got "obsoleted" by the second wave of Core 2 Macs. So I wouldn't sweat too much. Just enjoy the marketing and the hype.

2020-2021 will be just a warming up, a tease. 2021-2022 onwards will probably when the game is actually on. And we will see intel pumping up muscles as well to convince Apple otherwise (I'm sure Apple's contract is significant for intel, and they wouldn't just let that big pile of money go).
 
Last edited:
It’s going to be fun to see how powerful the first Apple powered MacBooks are. I bet Windows users will suddenly care a lot less about CPU performance, just like Android users have for the past few years.
 
Assuming that this rumor is finally true, I think it's logical to see a MBA replacement that will utilize a slightly beefed-up version of the A12z in the iPad Pro. If it's fast enough it will run emulation on "native" MacOS software fast enough that it will not be a noticeable performance hit, and native iPadOS apps would run just like on iPadOS. For a large segment of Apple's laptop market, this would be a perfect device. Meanwhile, they could leave the Intel chips in the iMacs and the "Pro" line of MacBooks for now. As applications transition to full native capabilities on the ARM version of MacOS, they could transition to higher-performance ARM chips in their higher-end machines. More likely they won't though, as their install base on those machines is pretty tiny so engineering a chip for that makes little sense, when Intel (or maybe AMD some day) is already putting the work in to build those chips for a vastly larger install base. The added plus of having the ARM in the low end is it also turns up some heat on Intel, even if its minor.

I don't see a huge plus for Apple in trying to move their entire computer line to ARM vs. the costs to do so. They are profit-motivated and the sales in those segments seem unlikely to recoup the costs required. At the low end, however, they would likely make *more* profit as they can leverage what they're already doing for their iPads vs. paying Intel for what amounts to pretty weak cheap with poor thermals.
 
Interesting. All Apple computer designs with CPUs from any company but Intel were commercial failures. We'll see if they can change this pattern.
Hmm, the PPC Macs were not a failure I believe. They just met a thermal wall so they couldn't push the performance further. Thus the Intel switch, performance per watt. A similar thing is happening again with Intel stagnating on their cycle.
 
RISC-V might have something to say about ARM ISA.
Not directly of course. But a ton of university is going into RISC-V, security as well as performance and efficiency. Some of that research could possibly give it the type of advantage that Apple would like to use, even though their ARM architecture license fee is just a tiny portion of Apple's petty cash, and might let them use it anyway.
 
It’s going to be fun to see how powerful the first Apple powered MacBooks are. I bet Windows users will suddenly care a lot less about CPU performance, just like Android users have for the past few years.
I'm sure number of ports will be thrown out a lot as a performance metric. :D
 
One of the main problems is that most Steam games on PC and Mac don't support ARM processors and there really isn't a way to refactor or emulate those games. So Apple needs to get their game straight to figure this issue out.

I may go Razr thin laptop next round after my MacBook Pro 16 is my main workstation at home.
 
Last edited:
  • Like
Reactions: Lozza013
One of the main problems is that most Steam games on PC and Mac don't support ARM processors and there really isn't a way to refactor or emulate those games. So Apple needs to get their game straight to figure this issue out.

I may go Razr thin laptop next round after my MacBook Pro 16 is my main workstation at home.

Apple has never been particularly concerned with your ability to run games on Mac.
 
There will always be something new. At some point there will be 128-bit processors, and that alone will be a new architecture jump. The only reason arm is so popular now is due to them eclipsing intel in efficiency. Someone will invent a more efficient architecture, and it will eclipse arm the exact same way without a doubt. That is the historical trend for nearly 80 years, and until proven wrong, it will be inevitable.

Definitely! Because macOS is a 64 bit operating system running on a 64 bit chip, I am limited to 16 exabytes of memory. (about 17179869184 GB). I need more room! And that's where a 128 bit cpu would come in handy.
 
I would think that Apple has had a roadmap laid out long before they released it. So if ARM was right around the corner why wouldn't they just save it for that?

The 7,1 sure does feel like the modular design all us MP users were begging for. It seems like a machine that should still be current 10 years from now...?

I’m going to wait and see what they actually say at WWDC. Maybe Apple can actually support multiple architectures. Unless they say something like that, it seems obvious x86_64 will become second class or worse well before 10 years, but who knows. Maybe Apple will say nothing at WWDC.
 
Definitely! Because macOS is a 64 bit operating system running on a 64 bit chip, I am limited to 16 exabytes of memory. (about 17179869184 GB). I need more room! And that's where a 128 bit cpu would come in handy.
You only need 128 bit address bus. Can still be a 64-bit processor :)
 
It’s going to be fun to see how powerful the first Apple powered MacBooks are. I bet Windows users will suddenly care a lot less about CPU performance, just like Android users have for the past few years.
Did you read the article? It's about the battery life and thinness. Expecting ARM-based Macs (at least the first ones) to be more powerful is probably naive. Macs have been under-powered for how long? Always? That's probably not going to change.
 
Eh I don't think that's case. Or else why the Arcade shop? Or allowing Steam to be installed? There has to be an audience for gaming on a Mac if those concepts are in reality?
Arcade is really an ios thing, and is hardly AAA games. Mac is just along for the ride.

And “allowing steam to be installed?” Apple doesn’t decide what you can install on your mac.

Apple is notorious for poo pooing the gaming market.
 
Hi All,


Assuming that the ARM rumors come to fruition in the near future, it certainly will have different effects on users. This of course, depends on what a person typically does with his/her computer. If a person's workflow is already quite smooth on an iPad, then I don't think the ARM transition to Apple laptops/desktops will be a big deal. However, if a person cannot "get stuff done" solely on an iPad, then the ARM transition will indeed be disruptive (until all their major software is ported over to ARM). Of course, there will be some software which will never be ported over.

For myself, I use Mathematica extensively in my workflow. As it stands now, I would not be able to use Mathematica on an ARM platform. See picture below.


richmlow
 

Attachments

  • Screen Shot 2020-06-09 at 9.58.01 PM.png
    Screen Shot 2020-06-09 at 9.58.01 PM.png
    145.9 KB · Views: 121
You only need 128 bit address bus. Can still be a 64-bit processor :)

The way I understand it is that a 128 bit processor can manipulate 128 bit pointers in a single cycle. Using bank switching or, worse, multiple 64 bit registers would slow things down.

wikipedia notes

Larger virtual address spaceThe AMD64 architecture defines a 64-bit virtual address format, of which the low-order 48 bits are used in current implementations.[11](p120) This allows up to 256 TiB (248 bytes) of virtual address space. The architecture definition allows this limit to be raised in future implementations to the full 64 bits,[11](p2)(p3)(p13)(p117)(p120) extending the virtual address space to 16 EiB (2^64 bytes).[16] This is compared to just 4 GiB (2^32 bytes) for the x86.[17]This means that very large files can be operated on by mapping the entire file into the process' address space (which is often much faster than working with file read/write calls), rather than having to map regions of the file into and out of the address space.Larger physical address spaceThe original implementation of the AMD64 architecture implemented 40-bit physical addresses and so could address up to 1 TiB (240 bytes) of RAM.[11](p24) Current implementations of the AMD64 architecture (starting from AMD 10h microarchitecture) extend this to 48-bit physical addresses[18] and therefore can address up to 256 TiB of RAM. The architecture permits extending this to 52 bits in the future[11](p24)[19] (limited by the page table entry format);[11](p131) this would allow addressing of up to 4 PiB of RAM. For comparison, 32-bit x86 processors are limited to 64 GiB of RAM in Physical Address Extension (PAE) mode,[20] or 4 GiB of RAM without PAE mode.[11](p4)
 
For myself, I use Mathematica extensively in my workflow. As it stands now, I would not be able to use Mathematica on an ARM platform.

Does your workflow correlate with the benchmark? They might test more things than you use. In any case, wait for benchmarks on actual hardware, and start pressing developers to optimize their software as soon as Apple announces development kit for ARM. These chips are capable of much more.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.