Apple to debut multiple ARM MacBook, desktop models in 2021(appleinsider.com) |
Apple to debut multiple ARM MacBook, desktop models in 2021(appleinsider.com) |
On the other side, such a transition always is painfull. A lot of software needs to be rebuilt, and even more legacy software gets killed, as it is not ported to the new platform. Though Catalina already did a lot of that, and that might not be a coincidence, if you think about it. Catalina might be intentionally restrictive to ease the transition to another architecture.
Also, a lot of Mac users are running VMs on their machines. Running Windows and Linux VMs on your Mac was one big benefit of the switch to Intel. I myself use a Linux VM for my professional work on a MB Pro. And that VM needs to be able to run x86 software. So it remains to be seen, how Apple deals with the MB Pro (and of course the Mac Pro).
I wonder whether they’ve got any special killer apps in mind for the desktop and MacBook ARM machines. They could theoretically customize them greatly since they’re designing the architecture. Also wonder whether they view keeping the unix developer pipeline.
But of course, Apple does often surprise us all, maybe this could be something cool after all.
People asked the same question during both previous transitions.
ARM macbooks will be for youtube influencers. Sure they’ll be able to edit 8K (and when uploaded to youtube they can’t even view it themselves). Devs need AMD.
My big worry is that actually this is just acting as cover for deep underlying issues in a lack of real vision for how the iPad can become a real replacement for laptops and no vision at all for the desktop line. How crazy is it that Apple is pumping out products that re-implement Microsoft features from 5 years ago (after spending 5 years mocking Microsoft).
- Keyboard with more travel - ESC key
Now people are calling the 16" the greatest laptop ever. It's like you've been abused for 4 years, then the abuser does something nice, and you think he/she is the nicest person in the world.
Macbook Pros are still bad:
- Useless touchbar that makes basic functions harder - Resting your palm over an oversized touchpad - Dongle-hell - No battery life improvement in 8 years - Overpriced due to focus on thinness and useless touch bar
They should have taken the 2015 Macbook Pro and gave it a 16" screen, touch ID, newer ports, newer CPUs/GPUs and it would have been 10x better than the overpriced junk we have today.
The A12X was so fast when it came out, reviewers were like "holy shit, this is nearly as fast or faster as comparable x86 laptop," and that was what... two years ago?
Apple's ARM processors are likely _so fucking fast_ that they'd absolutely be showing their hand if they put one in a tablet right now. Reviewers would probably be like "uhh this is faster than a 13" macbook pro and it's a fucking tablet?!" Not to mention, that'd be the nail in the coffin for a good (for Apple) relationship with Intel until they're totally ready to transition.
My guess is Apple is probably just not "quite" there yet for it's highest-end products or we'd see them this year. It's pretty lucky for them their custom silicon to-date has been so far ahead of everyone else or (appearing to be) sitting on their asses for a year or so might hurt.
TSMC is already capable of taking a shit on Intel (Ryzen 3000) and Apple is obviously capable of making very fast custom CPUs. It's a pretty dangerous combo if you're Intel right now and already getting hammered by AMD.
For me, the real question, and pipe dream, will Apple partner with AMD during the transition and perhaps make some some hybrid AMD x86/Apple ARM machines since they're both being built by TSMC?
There is an interesting thought. Apple could embed one AMD-chiplet with up to 8 cores into their designs, where Apple provides the ARM-CPUs as well as the io-hardware the chiplets require. This would retain the ability to run x86-code, while the system would use the ARM-hardware for most task.
I remain deeply sceptical that Apple will move the Mac to ARM, and given their bungling of the MacBook lineup over the last 5 years I have doubts they understand what their users need from a laptop. I don't think they could pull it off without a great deal of pain for their users and developers. I don't think their OS engineers need the burden.
* signed software restrictions
* no kernel extensions
* SIP
If absolutely nothing else, native developers: you can't have more than one version of a MAS app installed. Nearly every iOS developer I know has at least two versions of Xcode at any given time.
They may provide sandboxes for development, some kind of WSL (Windows Subsystem for Linux) or Docker on Mac, and a "Classic" emulator for apps that haven't been recompiled like they did for the transition between Mac os 9 and Mac os X, and from PowerPC to x86.
Meanwhile they will continue to limit web apps.
There will be some cross platform compatibility issues. Interpreted languages hide the CPU (think Python, Ruby, Node) but their extensions in C could need some extra work that not all of them require now.
On the other side there will be the rise of the ARM servers. Apple is notoriously not interested in running server farms for their customers but I expect that the big cloud providers will start offering ARM servers.
It could be the x86 developers to have to fire up an ARM VM on the cloud to be able to work in ARM Macs heavy teams.
Apple already has an x86 build chain that cross compiled to ARM.
Virtualization extensions have been available since the Cortex A15 or thereabouts.
Linux supports normal KVM on both ARM32 and ARM64 although there are mutterings about dropping support for ARM32.
In practice, the hypervisor is usually already grabbed by the platform (e.g. on Android/Snapdragon it's taken by the TrustZone driver and can't be used yourself, even if you rebuild the kernel with CONFIG_KVM).
Also, the move to Catalina helps to check how fast app developers and users switch to new hardware.
I'm pretty sure Apple have the ability to put 32-core ARM systems into the MacBook, and push the power/processing ratio into the stratosphere. The only reason I'm not using ARM right now, personally, is because such a thing doesn't really exist .. but I look at my old Touchbook up on the shelf and wonder what could have been...
"Apple will no longer be held to the whims of Intel"
Apple switched to Intel because IBM couldn't ship a good enough G5 for notebooks and for scale issues. Given that Intel is not on track with 7nm and rumours of their CEO being seen in Taiwan the other day it seems history is repeating itself.
However, breaking compatibility with Windows and Linux would be a terrible idea. People would flee to other platforms leaving Mac for those who are already locked in, like iOS developers.
People care most about running their software to do their work, than what os they use.
Will this allow for greater freedom with ARM devices not running new OSX?
It seems like Apple lately has very much internalized 'if you build it, they will come' even though it seems like it is heavily alienating their professional user base.
Maybe I’ll get an Arm Air in ‘22. My 2017 Pro should still be viable until then for what I do.
Each of them have made the transition from 68K -> PPC -> OS X -> x86.
I don't think transitioning to ARM is a bad choice given the benefits, but a lot of serious (and casual) users are going to be very unhappy their programs aren't going to work after dropping $1k+ on a new computer.
That means next-day onsite repairs. Not 3-5 days at a distant location, leaving the user without their computer.
Or I was, now that I work from home all the time my need for a laptop is much reduced.
For every person who complains about the laptops there’s someone like me who is happy with the direction (doesn’t make either of us wrong). I’m a software developer and miss the portability of my 12” MacBook. I enable telemetry so they know I almost never plug anything into my machine. An iPad just doesn’t cut it.
If this ARM speculation is even substantive it will likely mean a light super-portable machine for me and continued investment in the 16” MBP for you.
Wouldn't this stand as a reason to not be skeptical of the big move? Switching to ARM (badly) and breaking everything seems like it'd be exactly in line with all of this bungling.
I ordered a new MacBook Air just over a week ago, right when it was announced. If it turns out to be the last Intel-based MBA before years of pain and suffering with the ARM transition then I expect to be hanging on to it for a long time.
Having said all that, Apple is well known for having successfully navigated multiple CPU architecture transitions over the life of their operating system (s). If anyone can do it again successfully, I would expect Apple to do so.
IIRC, It took a while for all the "pro" tools I used to get refreshed for x86.
They did this before with the move to Intel from PPC. And back then, Apple was a great deal smaller.
but the transition will likely be more gradual (for instance, replacement times for computers are longer now), so x86 macs should stay around for a long while i think.
We're talking about Apple here. I expect the new machines to have less perf but be as expensive as before.
Better battery life and less heat, no? Isn't that the point of ARM?
They've done processor transitions multiple times before.
And whatever "bungling" you're talking about, I don't see what that has to do with it. It's not like people have stopped buying MacBooks and Apple has serious remedial stuff to work on first. Besides, whether or not your MacBook has a touchbar or whatever its key travel is has literally nothing to do with what it's doing with processors.
Citation needed. We’re not talking about Year of the Linux Desktop here.
I’ve been hearing rumblings and speculations about ARM Macs since at least five years ago, but IFAICR never heard a concrete timeline attached to it, let alone “a year away”. And 2021 would be fairly consistent with speculations.
https://9to5mac.com/2019/02/21/mac-marzipan-arm-next-year-ma...
Intel is one of two suppliers for nearly every laptop CPU sold today. If Apple was just disappointed with Intel they could switch to AMD instead. Either way they wouldn’t fall behind the rest of the personal computer market as they were at the end of the PPC era.
Instead, they’re making basically the same bet they made with PowerPC—that a different CPU architecture can significantly outperform x86. Last time, it turned out that Intel and AMD, by focusing on the PC market, could keep up with and ultimately outperform Motorola, since Motorola was more focused on embedded systems. This time, it’s Apple producing the chips, except Apple themselves are also more focused on mobile devices than they are on personal computers. So it will be interesting to see if this bet pays off this time around.
(notwithstanding their core "supply chain tech" as per Ben Thompson's point on Stratechery).
The other thought...is it yet another loss of American technical expertise that they cannot chip fab w/ the best of them anymore?
Apple: People who are really serious about hardware make their own processors.
1. We don't know what impact this will have for Bootcamp and Windows, since Windows 10 on ARM right now is customized for specific ARM CPUs like Snapdragon.
2. Same for virtualization, we don't know the performance hit it is going to have. A lot of people still need to use Windows for specific software that is not available on Windows or macOS where 32-bit support can be retained with older macOS releases.
3. Going the other direction, ARM means we could also see easier porting of iOS apps to macOS via Catalyst with more consistent APIs. But that could also mean less focus on macOS overall and everyone switching to iOS to port to macOS rather than working on two separate versions. This has both pros and cons and we won't know the full extent until a few years later.
It makes me sad to say that given the Apple fanboy I once was but reality is reality.
Controlling the stack w/ better vertical integration as well will probably yield more coherent product design. You can tell re how "put-together" a product an iPad or iPhone is vs. a Mac these days.
While maybe not yielding direct customer, user-facing benefits, the under-the-hood benefits would definitely improve customer experience.
If this would happen, most definitely the improvement on margins. Why would they lower prices?
Even the low end $329 iPad can hold its own performance wise to low and midrange laptops. The high end iPad Pros are cheaper and faster than lowend and midrange laptops.
They did the same thing with PowerPC and x86-32. No reason to expect them to support x86-64 in 5-10 years if they switch all Macs to ARM.
Apple has done this three times if you include the Apple //e transition. Each time they brought customer’s along.
The first transition from 68K to PPC, almost all of their revenue came from Macs. The second time about half (the other half was iPods). Now it’s less than 10%. It’s much less risky this time.
How is moving Macs to Arm going to improve this?
Apple pays much more for a Intel chips than its own ARM chips.
Again, you can look at the price/performance of iPads compared to laptop Macs.
But during the PPC transition, you could buy a PPC daughter card for 68040 Macs to run PPC software but you couldn’t use them simultaneously.
So you don't have to kill all of the software currently running on your customers' machines.
I bought a PPC Mac 9 months after they first made the switch. Most of my software was already native. Within the next year, all of the important software was native.
With recent changes to the OS, a decade of instability with every update every year, the recent hard deprecation of 32 bit binaries, and now moving to an unproven ISA for those particular workloads - all while Windows and Linux have closed the gap, mostly... the situation is a bit different.
A lot of the plugins those users acquire over time are produced by individual devs or small Indy shops, who may not be around any more or inclined to develop Arm versions of older products. You'll find threads on gearslutz.com where some people aren't upgrading to Catalina because 32-bit plugins aren't supported any more.
However, macOS is still better IMO because I still have a lot of issues on Windows; all stemming from MS shipping bad updates. Two months ago, a Windows update broke my bluetooth completely and I had to wait a month for another Windows update/driver to fix it. Last year, a bad update forced me to reinstall Windows because it was freezing all the time and I couldn't restore a system point either.
As for macOS, I never had to do a reinstall (except one time that was entirely my fault) in more than a decade. W7/W10, I had to do it 5 or so times in last few years. Catalina however lost my respect for macOS, that was the worst update of all time.
Most development tools have Windows versions: Node, NPM, Git, Docker. If I'd encounter one which doesn't run ob Windows, there's WSL.
I pretty much disliked Windows XP, Vista and preferred Linux, but since Windows 7,the OS got a lot better from my point of view.
I have the largest possible software library and the added possibility to run *NIX tools when needed.
I don't need to fiddle with Linux desktop or pay 3x the price for Apple branded hardawe which you can't even maintain or extend yourself. I can use cheap chargers, perriferals and I can connect almost any device to my desktops or laptops.
Windows is slowly becoming a very developer friendly platform. Will it work out for Microsoft? Who knows.
I'm more excited for Microsoft than Apple at this point but I am looking forward to seeing iPadOS/macOS later this year.
Of course this is anecdotal, but my macbook (made solely from Apple-supplied parts) seems to crash 3x as often as my cobbled-together-from-spare-parts Windows 10 desktop.
> Otherwise, they simply won't release it at all.
Do you think post-Jobs Apple retains that discipline? I'm not sure.
Apple, of all companies, is not a monolith. There are parts of the company who've been just killing it from a software perspective (Swift, SwiftUI, XCode preview for SwiftUI, etc.) Where they've had severe issues has effectively been quality management across the broader OS.
Also, I'll be very interested to watch what happens over the next couple of major release cycles. I think Apple got a major wake-up call with the shitstorm that was the iOS 13 / Catalina release cycle. I'm hoping that they'll be putting in place an outright culture shift to fix that long-term, vs. a one-off "Snow Leopard" tech debt paydown release.
Time will tell, but right now I think Swift isn't the languages that I once hoped for. And objective-C, despite all of its problem, is still doing well.
Apple didn't bother with 32-bit compatibility for Catalina which would have been much easier to implement. You're being way too optimistic. Apple doesn't care about backwards compatibility.
If the ARM macs can run Intel 64 bit software, but not 32 bit, then removing it in Catalina makes perfect sense
Nah. It won't be seamless at all. It will be like touchbar macs - you buy it because last 3 generations were stuck in place with little improvement and you are forced artificialy to upgrade.
Intel CPUs are not customized by Apple for their own APIs, they're for general purpose use. yes, they have ISA extensions that Apple could use like QuickSync but it's not enough for Apple.
Apple customize their A series with the same APIs they use, such as Metal, CoreFoundation, Javascript Core (they have hardware-based JS acceleration support), etc.
It's why they added T2 chips to their Macs to help accelerate a lot of tasks like disk encryption, more locked down security with TouchID and so on.
So, for example, my computer can encode 4 4K videos in real time simultaneously. Why don't I use this feature? Because the encode quality is subpar. So unless you give me a benchmark of the iPad Pro on ffmpeg or any other open source, non-hardware accelerated video encode software, comparison is completely moot. And the benchmarks that have been done on ffmpeg on x265 or x264 show that the iPad Pro is multiple times slower than an Intel laptop. Now obviously x265 is optimized for x86, but not on the order of multiple times slower. Unless it is the case, and that means that your benchmarks don't apply either.
Herein lies the issue, for many many workloads hardware acceleration while faster offers results that are not comparable to the CPU in terms of quality. So the only way to compare is by disabling hardware acceleration, and x86 processors tend to win.
Metal is a trash API that no one uses in the real world. It offers literally nothing better than Vulkan. And the fact that the chip is "customized for the API" is vacuous. All GPUs are optimized for OpenGL, DirectX or Vulkan. Same for a lot of the "Core" APIs. They will not succeed outside of the mobile market.
The T2 chip is simply a glorified security processor. There is absolutely nothing the T2 chip does that a traditional security processor in say a Zen chip can't do. x86 CPUs can already do AES at speeds so high you would need something like 4 RAID-0 NVMe SSDs to have a performance bottleneck, and even then the limitation isn't the CPU but RAM speed. There is no real world scenario where you would need to "accelerate" disk encryption or other kinds of cryptography beyond what an x86 CPU can do. Cryptography isn't some kind of magic you can't implement without some specialized chip, literally everything the T2 chip does can be done using a trusty old x86 processor and TPM. The only use of the T2 chip is for Apple to have more control over your hardware, and literally nothing else.
I'd only be impressed if both used the exact same high-quality software encoder. Most likely the iPad uses the fast but less quality dedicated hardware encoder of the A-Series SoC and the MacBook uses a high quality but slow software-only one, which is what you typically use in any non-real-time encoding scenario due to way better bitrate-to-quality ratios.
> Javascript Core (they have hardware-based JS acceleration support)
Do you have a credible source for this? AFAIK JS VMs have gotten to the same place that Java VMs (for which some people also envisioned dedicated silicon a long time ago, but it was a dud) reached: so frickin fast on standard x86 ISA that putting any special instructions for them into the ISA isn't worth it, because it's more important to stay flexible to be able to adapt future extensions of ECMAScript.
> It's why they added T2 chips to their Macs to help accelerate a lot of tasks like disk encryption, more locked down security with TouchID
That has more to do with having a secure element under Apple's control in the T2 chip and nothing with performance. Any modern x86 CPU can do accelerated AES just as fast as any ARM with hardware crypto support.
For JS: https://twitter.com/codinghorror/status/1049082262854094848
It looks like it's not exclusive to Apple's CPU, it's the specific instruction features in ARM 8.3 ISA that makes JS faster.
Added here: https://bugs.webkit.org/show_bug.cgi?id=184023
> Any modern x86 CPU can do accelerated AES just as fast as any ARM with hardware crypto support.
Right but back then, Intel mobile chips weren't that fast. I had MBP with Filevault that took a massive hit and I had to turn it off to get back disk performance. I can't prove that T2 is the reason the encryption doesn't take any hit on T2 Macs, all I can see from my trial of rMBP 16, there was zero performance hit with it on or off.
Other synthetic benchmarks of memory bandwidth and so on use all acceleration features of the plafortm on ARM but don't support AVX2 or AVX512, although some other workloads in the benchmark do. And of course you would have to choose exactly which instruction is used in which scenario in which processor(Intel vs Zen 1 vs Zen 2) in order to have the same kind of optimization as for ARM processors. Then comes the issue that vector operations are "hand-tuned", which is not realistic and depending on the skill of the programmer and their affinity with a given uArch can yield vastly different results. Which is why they should either use the fastest library for each processor, or leave all the optimization to the compiler.
The only way to do a proper comparison between two uArch is with an open source benchmark compiled specifically for the processor.
The perf hit with Filevault became practically zero when Intel added the AES encryption hardware into its chips, which was quite a while ago (and definitely long before the T2 was a thing). I don't remember exactly when that was, but I remember noticing a considerable difference, because I've used FDE since it became available in Filevault. It wasn't even on Macs only, my Windows machines also showed the same improvements using (IIRC) TrueCrypt back in the days.
The dedicated AES hardware extensions in ARM and x86 cores are probably the same logic anyway, so it shouldn't matter too much who decrypts the data. Maybe it's a tiny bit faster with the T2 though, because then the CPU doesn't necessarily have to pipe all the data through it's buses for decryption then. But that is more or less a feature of having a dedicated separate chip for it, and is thus not tied to the question of whether that chip uses ARM or x86 or any other ISA.
I don't really see how that would help, CPU i/o has to handle the exact same amount of incoming data encryption or not. Maybe a bit less RAM impact though.