Not really. It's a convenience that many of us have taken advantage of. In my case, it stops me from having to own a PC to run the handful of Windows-only games I enjoy playing.
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 Plans to Ditch Intel and Use Custom Mac Chips Starting in 2020
- Thread starter MacRumors
- Start date
- Sort by reaction score
Not really. It's a convenience that many of us have taken advantage of. In my case, it stops me from having to own a PC to run the handful of Windows-only games I enjoy playing.
Presumably Apple is aware of this issue but given Apple's preference for controlling their destiny, we can hope they make some effort to mitigate the loss. For the few times I might need Windows, I'm happy to run it in a VM.
Overall this would be a very positive move for Apple. Hopefully they will design, test and implement it carefully.
For developers ( like me ), it potentially introduces some major challenges with existing applications. If Apple plans correctly, it might be as simple as rebuilding in Xcode with a different architecture parameter.
Wow, the negative response is disappointing. Who would have thought the Mac community would attack a decision that promotes competition and innovation so fiercely?
This is excellent news. It's much needed competition in a world dominated by Intel's monopoly on x86.
A few points with regards to some comments here:
1) ARM does not mean low performance. How anyone can suggest that when Apple's own mobile architecture is at least as good as Intel's best architecture already is beyond me. Look outside Apple designs to Qualcomm Centriq and ThunderX2 for further examples of high-performance ARM cores. ARM is just an instruction set, and with ARMV8 and SVE (vector extensions, up to 2048-bit wide(!)) it is possible to build something beyond the capabilities of what Intel offers.
2) You lose Windows support but... price worth paying, IMO. I get that's important to some, but to those people, have you considered the cloud? I get that a local instance is preferable for latency, but it's dirt cheap to spin up a Windows VM for the few occasions you might need to dip into Windows-land.
3) Linux support for ARMV8 is excellent already, you will be able to run native VMs on ARMV8 macOS (xhyve, docker etc.)
4) Moving to ARM means Apple can optimise both the hardware and software to an extreme degree. Think about all the custom logic that makes the iPhone the market leader and imagine the possibilities for the Mac. The CPU cores can be designed to prioritise the needs of desktop specifically, custom accelerators for various system functions can be integrated, the "secure enclave" for TouchID/FaceID moves into the CPU die, extended video decoders, the Neural engine etc. It is an opportunity for Apple to jump to places the generic Wintel competition will not be able to reach for many, many years.
5) Apple have a proven record of chip design after repeated success with A6 (first custom core), A7 (first ARMV8 64-bit), A8, A9 (starting to reach Core level IPC), A10/A10X (arguably, at parity with Core) and A11 (arguably, surpassing Core). Their first GPU design has also been a success.
6) macOS and iOS already share almost everything. Apple have spent many years fine-tuning the low-level code in Darwin, libraries, the compiler etc. to spit out very fast ARM code. While it will take some effort for applications that target x86 specific instructions (e.g. SSE, AVX), it's not as much as you'd think. Adobe (for example) should be able to bring up a native ARM port relatively quickly. Many applications will require zero effort, just changing the target from x86 to ARM in the compiler.
7) Apple will have more options for third parties should they have a mis-step (e.g. new architecture turns out a dud) with companies like Qualcomm, AMD, Cavium, Broadcom, Samsung (yes, Samsung) etc. all capable of delivering high-performance ARM designs. Currently, they have Intel or AMD to choose from...
This is excellent news. It's much needed competition in a world dominated by Intel's monopoly on x86.
A few points with regards to some comments here:
1) ARM does not mean low performance. How anyone can suggest that when Apple's own mobile architecture is at least as good as Intel's best architecture already is beyond me. Look outside Apple designs to Qualcomm Centriq and ThunderX2 for further examples of high-performance ARM cores. ARM is just an instruction set, and with ARMV8 and SVE (vector extensions, up to 2048-bit wide(!)) it is possible to build something beyond the capabilities of what Intel offers.
2) You lose Windows support but... price worth paying, IMO. I get that's important to some, but to those people, have you considered the cloud? I get that a local instance is preferable for latency, but it's dirt cheap to spin up a Windows VM for the few occasions you might need to dip into Windows-land.
3) Linux support for ARMV8 is excellent already, you will be able to run native VMs on ARMV8 macOS (xhyve, docker etc.)
4) Moving to ARM means Apple can optimise both the hardware and software to an extreme degree. Think about all the custom logic that makes the iPhone the market leader and imagine the possibilities for the Mac. The CPU cores can be designed to prioritise the needs of desktop specifically, custom accelerators for various system functions can be integrated, the "secure enclave" for TouchID/FaceID moves into the CPU die, extended video decoders, the Neural engine etc. It is an opportunity for Apple to jump to places the generic Wintel competition will not be able to reach for many, many years.
5) Apple have a proven record of chip design after repeated success with A6 (first custom core), A7 (first ARMV8 64-bit), A8, A9 (starting to reach Core level IPC), A10/A10X (arguably, at parity with Core) and A11 (arguably, surpassing Core). Their first GPU design has also been a success.
6) macOS and iOS already share almost everything. Apple have spent many years fine-tuning the low-level code in Darwin, libraries, the compiler etc. to spit out very fast ARM code. While it will take some effort for applications that target x86 specific instructions (e.g. SSE, AVX), it's not as much as you'd think. Adobe (for example) should be able to bring up a native ARM port relatively quickly. Many applications will require zero effort, just changing the target from x86 to ARM in the compiler.
7) Apple will have more options for third parties should they have a mis-step (e.g. new architecture turns out a dud) with companies like Qualcomm, AMD, Cavium, Broadcom, Samsung (yes, Samsung) etc. all capable of delivering high-performance ARM designs. Currently, they have Intel or AMD to choose from...
I don't need a walled garden with higher walls. I'm not sure how that promotes competition.
I need to do my work. Any decision by Apple that negatively impacts my work is frowned upon.
Very amazing and seamless.
This would actually be the 4th of 5th CPU architecture transition and previous times it was quite smooth.
Today you can choose a high-performance PC with an Intel CPU or an AMD CPU... that's it.
If Apple move to custom designs, you will have Intel, AMD and Apple (yay competition).
The walled garden thing, I guess you mean the software? I'm not convinced they would fundamentally change what macOS is (e.g. I don't think you'll lose the console access, ability to install apps from anywhere etc.). One of macOS's biggest advantages is being a *nix desktop that doesn't suck, that alone explains why the Mac is so popular with developers and engineers. They're not going to alienate engineers/developers by blocking their favourite tools/open-source software/package managers/virtual machines etc.
Such a regressive move. Running on x86 makes porting Windows code much simpler, not to mention actually running Windows itself.
Where is the gain for the user? Intel's chips are by far the fastest and there is already a huge software library which runs on them. It's an illogical, irrational decision.
As for making it easy to pretty iOS apps to macOS, really, who cares. Mobile apps are designed for touch and are always horribly dumbed down. Microsoft already tried this with Windows 8 and we all know how that worked for them.
I'm glad I ditched my Mac and and back to Windows a year ago. I'm safe on Intel for the foreseeable, plus I get to run with high end Nvidia GPUs.
Where is the gain for the user? Intel's chips are by far the fastest and there is already a huge software library which runs on them. It's an illogical, irrational decision.
As for making it easy to pretty iOS apps to macOS, really, who cares. Mobile apps are designed for touch and are always horribly dumbed down. Microsoft already tried this with Windows 8 and we all know how that worked for them.
I'm glad I ditched my Mac and and back to Windows a year ago. I'm safe on Intel for the foreseeable, plus I get to run with high end Nvidia GPUs.
1. This is not accurate. Compare an A11 to an i9. Not even close.
Also the A11 for all intents is an embedded processor that doesn't support the amount of memory, PCIe lanes, etc that a true desktop processor must.
2. Sorry, but a VM won't work well in emulation. Current VMs use virtualization.
3. Android is Linux just like iOS is BSD so no argument about Linux support. Cavium and Qualcomm have Linux server level ports.
4. Apple can do no more optimization than they already do. They do not control the ARM instruction set. Yes they can modify the number of pipeline stages but that is different.
5. The A6-A11 are not laptop CPUs, desktop CPUs or server CPUs. I'm not saying Apple cannot do it. But they are going to spend a lot of the cash they have sitting around. It's a decision of ROI. Is there ROI if they do this?
6. iOS is a dumbed down version of MacOS with a mobile GUI. It'a BSD. It may or may not be simple that is yet to be seen.
7. They won't have more options,
Broadcom left the server market for ARM CPUs in 2016 in a layoff that happened in Oct. of that year.
Samsung left the ARM server CPU market before Broadcom.
Cavium and Qualcomm make server class silicon that cost like Xeon processors.
Currently they don't ship much with Xeon because they made poor decisions on thermal and physical enclosure design.
A box is better than a can. MacPro......
Translating CISC code to RISC on the fly will result in colossal performance loss.
Ok, so, since Microsoft has created the "Weaksauce" version of Windows, and Microsoft "understands" that Enterprise IT is "what matters", then where does that leave your argument?
Plus, I don't know about you; but *I* can't create ANY Processor, ARM or not; so I'm not sure who and what THAT part of your fake scenario that is even supposed to apply-to.
You make no sense. You had some arguments until I started to cite actual FACTS, then you just kind of fell-apart.
Precisely. I read his reply and wondered if he bothered reading my follow up posts. I can turn on CPU throttling in my Windows BIOS (workstation at home) and it throttles down when:
a) I'm doing very little
b) It gets too hot
You can turn off CPU throttling and the processor will throttle down and up by the second depending on what you're doing. This may seem dumb, but it actually extends the life of the processor, even though it seems as if it wouldn't.
If you recall, processors of past would run at full speed prior to AMD and Intel introducing these measures.
Apple designing it's own proprietary chips and chipsets... I don't see that as any more competitive than it is today. I doubt they'll put them on the direct market to compete with other chip makers. I would still have the same number of choices I do today. Apple just has the potential to make more profit and require me to buy more PC's to do my job (assuming I stuck with Apple).
I don't care what processor it uses as long as I can do what I need to do. Intel may make less money but I also will be able to do less with an ARM Mac. Fewer capabilities = bad.
1) You don't think Apple can throw a bunch of PCI-E lanes, more memory channels etc. in? They can build something as good/better than "Core i9".
2) I never mentioned emulated x86, only native Linux ARMV8 virtual machines. I don't think x86 Windows virtualisation would be a realistic expectation.
3) I don't understand this point?
4) They already do way more than "modify the number of pipeline stages". Apple's cores are custom from the ground up they, just implement ARMV8 ISA. They can stick to the ARM spec and still do things like encryption acceleration (like AES-NI in x86), wide SIMD/vector instructions (like AVX) etc. to be at parity with x86. When I say custom accelerators, I mean beyond the CPU cores, custom logic outside the CPU core.
5) They can do it. Apple's core is already in the same league as Intel's, while it's not trivial to build up from that to a desktop/server class part it's not beyond Apple's engineering capability. By far the hardest part is designing the core and they've already nailed that.
6) iOS is macOS. Same kernel, same graphics, same basis for the APIs etc.
7) Broadcom sold their IP to Cavium who've used it to build ThunderX2 (I should have left Broadcom off the list).
Samsung never joined the server race, but they are actively developing high-performance ARM cores.
My point regarding third-party options wasn't that there are off the shelf designs available, but there are companies Apple can work with (Qualcomm, Cavium, AMD, ARM etc.) should they falter. It would be a case of integrating the third-party IP into Apple's design, rather than farming out the whole SoC.
It's worth remembering that the core in your Intel-powered MacBook is a "server" core. AMDs Zen is a "server" core. The logic around the core changes depending on the application (e.g. the number of PCI-E lands, memory channels, cache) but the core remains the same e.g. between Core m and Xeon. (well, the core differs with Skylake EP but that's unusual!)
[doublepost=1522791739][/doublepost]
Why does ARM only have to mean lower power consumption? It's more than that, it can mean better performance (as we've seen with Apple's SoCs already)
Ultimately, the focus on a single architecture is a good thing for users though. It's a shame that AMD and Intel are the only competitors when it comes to x86, but having mostly homogenous architectures in personal computing is something that helped developers and thus users in the past few decades a lot. It made things very simple. It's vaguely like having one standard for power plugs.
Also, I suppose Apple producing CPUs isn't even actual competition. They will only ship those in their own devices and the rest of the market will be left unchanged.
If Apple wants a large market share they need to:
1) Drop the price significantly.
2) Create a better file browser than Finder. Check out Path Finder or Windows Explorer.
3) Include a supported VM by default, so you can easily run Linux and Windows (embrace and extend). Even if few uses Linux, power users are the ones who recommend computers to others.
4) Support NVIDIA and work for better compatibility with (Windows) games and GPU acceleration (AI etc)
5) Switch to a robuster / faster Web browser by default, or make Safari on par with Firefox,Chrome or Edge.
Switching to a new CPU architecture will not help, and it's not what customers are asking for.
Last edited:
Because MS does lots of stuff just to see what people want. Some of it works and some of it doesn't.
The existence of Windows for ARM does not mean that those who use Windows on a mac now will be able to use Windows on a mac with ARM processors. It's one more nail in the enterprise coffin and one more way to drive away users. Stupid.
I didn't imply that you made the processor -- just that you bought it.
To listen to Jobs in old interviews like this is to realize that his philosophies regarding corporate structure are still spot on. I disagree with another poster that the Apple of today can't be analyzed in context of Jobs' pronouncements of decades ago. The current "success" of Apple has become only skin deep, primarily market driven - style trumping substance. The "regular users" of Apple products these days aren't concerned with Macs; they are only concerned with luxury mobile devices powered by iOS. Being primarily a market/style driven corporation under Cook, Apple follows the money trends. It is much less concerned with Macs these days. The very titles of websites like this one - "MacRumors" or "9-to-5 Mac" - are anachronisms. If a new cpu platform for the Mac line actually improved it as a general computing product, as opposed to moving the Mac closer to iOS integration, then we'd see that the engineers are again being given some control over Apple's direction. I don't hold much hope over that happening.
Well, while I'm one of the users that would definitely leave Mac, I really doubt the user exodus would significantly affect the sales. The pros are no longer the core of the Mac owner base as the products had developed from "tools for professional work and production" into overhyped lifestyle accessory for rich-ass kids that need a screen to browse facebook on. Like you cannot do that on a $400 laptop, a $1500 Mac just seems to be better fit for the purpose, right? No wonder pros are leaving the platform when Apple charges much-more-than-premium prices for products, that are based on tech that's outdated before the product even hit the stores.
But given this, it would be actually a smart move. Apple is really good at dictating people what they shall do; doesn't your old software work anymore with the new platform? Oh, nevermind - stop using it and buy a new version! People would stop comparing Macs to Wintel PC's because noone would have a clue what's inside as it is with iPhones and iPads.
Honestly I'm getting fed up with the way everything about Apple that used to be handy to use becomes iOS-ified, so all these iOS to Mac halo switchers could feel welcome... God, how I miss Snow Leopard times!
In short: No need for performance as long as the new MacBooks will have animojis </sarcasm>
I wish I could remember the name of the chip, but Intel Macs produced in late 2007/early 2008 included a hardware chip which prevented installation of non-blessed OS software (translate "Linux"). Blessed Windows would install. They produced this hardware for less than a year, then decided that it was more trouble than it was worth, Linux having so little economic affect on their product. I found out several years ago, after I had successfully installed Linux on my MBP (1,1), the first Intel Mac generation, circa 2006. A friend tried my same procedure on an MBP(3,1) of 2007, using ReFind to modify the UEFI, etc., but the Linux system would not be bootable. It turns out that it was prevented by that mystery chip. Wish I could remember the infamous chip's name.
Addendum: The chip was called, back ten years ago, a "T-Chip". Well, fancy that, as apparently the same company behind the original T-Chip may be brought back into the proported new Mac cpu design process as the "T2-Chip". Though it serves various functions (see article below), it is again intimately involved with the EFI boot process, in effect providing the function Apple ditched a decade ago. Apparently, the new Apple finds proprietary walled garden enhancement profitable these days. We'll see. The article:
https://www.macworld.com/article/32...e-imac-pro-the-start-of-a-mac-revolution.html
Last edited:
I get where you're coming from, but I do think it'll up the pressure to innovate/compete.
Wild hypothetical scenario time. Let's say Apple continues its momentum and improves its core beyond where Intel will be with Cannon Lake. Apple builds an 7nm SoC for the MacBook that supports 10-cores, a large custom GPU and some high-bandwidth memory. Intel, meanwhile, is shipping 6C/12T Cannon Lake (10nm shrink of Coffee Lake) leaving PC OEMs at a performance disadvantage. Intel would no doubt respond with further enhancements to keep pace.
(disclaimer: I think early ARM Macs will start out with something less exotic, but the hypothetical 10-core SoC is totally doable)
Also, I get the desire to standardise on one ISA, I personally wouldn't care if that ISA was x86, but we're forced to look to other ISAs given Intel's control of x86.
Yea I typically do, but I am not set in any present or past way of computing. I am all for Apple moving forward in this area. I already use an iPad Pro as my main work and personal computer so I have no issue adapting.