Hacker News Reader

Comment by Stranger43 | Hacker News Reader

Have anyone checked with adobe and autodesk if this essentially kills the professional mac workstation as a viable product going forward?

I have no doubt they can make it fast and usable enough for the naive consumer living inside the bubble of what relatively inexpensive(consumer grade) app-store apps can do and that microsoft will release something pretending to be MS office for a new ARM macbook line, but photoshop, autocad, mathematica(ok that one is fairly portable), AfterEffects and whatever it is the pro's use for video editing those days are a different ballgame.

We saw microsoft stumble down that path with their successfull port of windows to arm where the OS itself worked but where no major 3rd party business essential app ever got ported to windows for arm, with a lot of IT departments choosing to keep windows 7 around our of fear of what 10 would morph into until MS begun to dial back their ambitions for windows everywhere.

Im guessing that the short run result of Apple going though with OSX on arm is that a lot of the CAD and video editing heavyweights just drop support for whatever OSX morphs into to support running on IOS hardware.

View on HN

If you remember back to the mid 2000's when Apple switched from PowerPC to (Intel) x86, the higher end machines still ran with PowerPC chips in them while the mid-low end products had x86 chips. All of the old applications that weren't yet ported to x86 had to use "Rosetta." It wasn't until a little while later, presumably once most of the professional software companies had good working x86 versions of their software - that the high end mac pro systems started seeing Intel chips in them.

Microsoft started down that path last year with Windows 10 on ARM, and Intel had some legal issues with that - https://arstechnica.com/information-technology/2017/06/intel... It will be interesting to see how Apple handles this.

There's two problems with that story: first, the famous switch from PowerPC to Intel happened when the desktop was the king, and app developers had strong incentive to recompile programs to the new platform. Right now? Not so much.

Second, while Apple's mobile processors are currently the best in ARM world, for "pro" desktop computers Apple needs desktop-grade processor. Making something comparable to i7 isn't such an easy and cheap job.

> the famous switch from PowerPC to Intel happened when the desktop was the king,

I though one of the major reasons for the switch to Intel was the power consumption and performance on laptops.

I'm thinking that in this case "desktop" is cast in opposition to tablets and cell phones and laptop computers can be considered "desktop" devices.

> Second, while Apple's mobile processors are currently the best in ARM world, for "pro" desktop computers Apple needs desktop-grade processor. Making something comparable to i7 isn't such an easy and cheap job.

Qualcomm, NVIDIA, and Cavium have created very powerful ARM chips for non-mobile application. Apple definitely has the chip architects, the money, and the experience to do such a thing.

I have serious doubts that the marketplace can support 3 desktop/laptop workstation CPU manufacturers.

(Amd, Intel, and now Apple)

I don't think it matters to Apple since they will be using their own chips. Intel will simply sell less, and AMD will not change at all.

By the way, don't forget Broadcom, Marvell, Cavium, IBM etc. as they are doing big bucks with CPU's and SoC's as well.

It matters to Apple, CPUs are a very big expense for them, and the performance of their computers are very important to them.

I'm specifically worried about the performance workstation/gaming cpu market, which Broadcom Marvell, Cavium, and IBM are not major players in. It's currently AMD and Intel. Apple would need to sell a lot of macs with the new CPUs to make the R&D cost worth it and the new CPUs would have to be very fast or they won't sell macs.

I meant the impact on global CPU markets won't matter to Apple, they don't care if Intel makes less money.

Adobe for one is making a boatload of money from CC subscriptions, record numbers, so I wouldn't worry too much about that (I imagine MS is equally happy with Office 365 on the desktop).

Besides, there's no Carbon to Cocoa or CodeWarrior to Xcode port, so aside from a very small percentage of optimized assembly code, it shouldn't be nearly as hard.

>Second, while Apple's mobile processors are currently the best in ARM world, for "pro" desktop computers Apple needs desktop-grade processor. Making something comparable to i7 isn't such an easy and cheap job.

Good point, I guess Apple never though of that.

It's quite impressive how many people in the comments can so easily come up with big show-stoppers that Apple never thought about. Must be why Apple is struggling financially and can't get anyone to buy their products.

While being sarcastic you might have missed the sarcasm of the parent :)

I both understood and was adding to the sarcasm.

Also worth noting is Windows and x86 desktop dominance at the time likely made porting Mac software to x86 easier

I remember all the way back to the early 90s and the original x86 emulator running on ARM. Admittedly that was before optimizations like JIT, and a lot of research has gone into binary translation since then, but it was a painful experience at the time. I wonder how much performance would be lost these days.

If they’re making their own CPUs, they can presumably add whatever additional functionality and/or instructions are necessary to make emulation or JIT translation work effectively.

But even if flat-out performance isn’t as good, what about if it proves cheap and easy to scale to more cores? Like, what if you get twice as many cores as the Intel equivalent has threads, albeit at 2/3 the speed (maybe worse or better for some workloads) but no hyperthreading-style contention? This wouldn’t be useless.

And GPU-accelerated functionality will be mostly unaffected.

Couldn't they achieve that today by running legacy apps in a hypervisor?

The issue is implementing the features for the hypervisor’s emulation system without Intel-licensed hardware or software.

lol wut? VMware Workstation, Connectix/MS VirtualPC, Oracle VirtualBox, qemu, bochs...what Intel-licensed hardware or software?

How about FX!32 or it’s latest AArch64 cousin seen on Windows 10?

Of the above, qemu can target x86 and run on anything, Virtual PC ran originally on PowerPC, and FX!32 ran on the Alpha

https://arstechnica.com/information-technology/2017/06/intel...

You know, the article refers to Windows ARM emulation, but I now see the Intel post as a warning to Apple, as patent licensing discussions may have started about that time last year.

It usually takes 4-5 years to design a chip, and if Apple planned on releasing the Mac CPU in 2020, well...

I see an article that is not congruent with actual tech facts. They can warn all they want, because the precedent says otherwise. Even if you wanted to make the argument that Intel is against emulation on non-x86 systems, that’s not true - see Virtual PC for Mac, FX!32, qemu, bochs and countless 8086 emulators to run VGA ROMs on various legacy free systems (either in OS or firmware). If you wanted to make the case that efficient emulation on non-x86 is is disallowed, well, FX!32, qemu and Virtual PC all used and use JIT.

Intel’s position amounts to cage rattling and vague insinuations that efficient execution of x86 code is impossible without hardware support. That’s nice, and their ham-fisted approach already hurt customer choice in the past (Transmeta and nVidia Denver, before the later implemented the Arm ISA) but the Windows on Arm solution is completely software and not coupled to any specific Arm chip implementation.

There's not really any real precedent other than people not being sued. That's more the complete lack of precedent rather than precedent that it's legal.

the x86/x64 applications in the hypervisor would be super slow, as the x86 ones are with Windows 10 for ARM (Windows 10 for Arm doesn't support x64 emulation)

> It wasn't until a little while later, presumably once most of the professional software companies had good working x86 versions of their software - that the high end mac pro systems started seeing Intel chips in them.

The switch to Intel started six months [correction: seven months] from the announcement, and took seven months to complete.

June 5th, 2005: Intel switch announced at WWDC

January 10th, 2006: First Intel Macs (MacBook Pro and iMac) released

August 7th, 2006: Mac Pro and Intel Xserve released, last PowerPC Macs discontinued

https://everymac.com/systems/by_timeline/index-macs-by-timel...

Apple has done this twice already. 680x0 to PowerPC, then PowerPC to Intel. Adobe's come along on both of these.

Those changes were all to more powerful processors such that emulating the old architecture wouldn't be that bad. In this instance, Apple would likely be going to a less powerful processors and thus emulation at any reasonable level of performance wouldn't be that viable.

The processors wouldn't necessarily be less powerful. Case in point is the A11 outpacing the CoreM processors.

The greater threat is the ios-ofcation of macOS. We've seen many companies pull their apps from the Mac App store due to the limitations of the sandbox.

That's not "likely" at all. Apple's CPUs are already faster in single-core benchmarks than the Intel competition, even though the Apple processors in question are designed for mobile devices and have much less power to work with.

Apple would not be making this switch if they didn't think they could improve performance.

Faster, probably, but 2x+ Faster like in PowerPC > Intel very unlikely. Which is why there’s much less likely to be enough performance budget for effective emulation.

Presumably their "desktop-grade" A-whatever chip will have a higher TDP and clock speed than their mobile counterparts.

If they could just turn up the TDP and clock speed on their chips and get 2x the performance of Intel's best chips that easily- they would already own the entire desktop market.

And some of us, with actual CPU design backgrounds, have been saying that for a while now.

We've been asking why Apple doesn't already do it.

As someone without a CPU design backround, would something like this scale pretty linearly with added power and thermal headroom? I assume there are limit that would have to be overcome, but what would an ARM chip in the conditions of an i7 look like?

It's more of a logarithmic scale.

The same architecture can generally scale to 10x over a few process generations.

This does not hold at 10nm and below.

TDP typically scales as somewhere between the cube and fourth power of the clock speed if you're pushing the envelope (in the sense of running at frequencies where further frequency increase also needs a voltage increase). So having 10x the thermal envelope means you can probably clock about twice as fast, all else being equal.

Seems that it’s more of a logistical and network-effect issue in getting everyone to support it, rather than a technical issue.

Possibly a patent issue also

My only question is if they actually use ARM for their next-generation architecture, or something completely new...

Computer architectures routinely see 3x performance jumps across different power budgets. This rule has held over decades.

Clock speeds alone can probably increase by 30%. Caches and internal datapaths can double or more. Then you can start to add in more execution units or more expensive branch-prediciton or even new more power-hungry instructions.

A 4 Watt Intel Pentium 4410Y Kaby Lake for mobile devices gets about 1800 on Geekbench, while a 115 Watt Intel Core i7-7700K Kaby Lake for desktops gets 5600.

I'm just going to say it: the Apple laptop CPU is going to get Geekbench score... above 9000!

And, yes, I do have a CPU design background.

So artificial benchmarks already do a very poor job of capturing performance. The apple laptop cpu does not exist. If it did exist it would likely suffer a very substantial performance hit if forced to emulate x86 software. So you are going to speculate on the meaningless benchmark numbers of an imaginary cpu that will take a wholly unknown hit if everyone does not rewrite everything why?

Artificial benchmarks do a great job of capturing performance, since they're more controlled and eliminate unnecessary variables.

Once you understand this, then you can understand how CPU designers work to predict future performance. CPU designers use artificial testbenches.

You making up numbers doesn't appear to be a useful endeavor.

I suspect if Apple designs a desktop CPU, performant x86 emulation will be a key design criteria. I know very little about CPU design, but I imagine it would be possible to have hardware optimisations for x86 emulation just like we have today for video codecs.

Or even further they could bake a "rosetta" into the chip's microcode and have their CPU natively support the x86 instruction set along with ARM or whatever they come up with.

If someone disagrees they should state why

Which is the previous gen and was sandbagged as the 50% increase in cores for coffee lake show.

Do you think they will use the same chip? In 2020?

That's not remotely true. There is some crossover, which is notable. But that means that the fastest iPhone is faster than some of the slower desktops. Intel's current single core leader is a coffee lake running at 4.7 GHz with like 12MB of core-speed cache. Be real.

The current A11 chip is faster single-core than the current MacBook Pro.

The Intel chip you mention isn't in any Apple machine at all. (Except maybe the iMac Pro? Is it even in that?) So you be real. I don't think you've actually looked at the benchmarks. You should do so. I think you'll be shocked at just how much overlap there is, just like I was.

that's because apple is fixed on choosing the intel parts with the least tdp power they can get away with. it's their choice to make the mbp slow, up and including the numerous thermal issues that force the cpu out of the nominal turbo mode for the longer workloads

Those are valid points, but it's still shocking that Apple's own CPUs can outperform Intel laptop CPUs with far higher TDP.

The 8700k which the previous poster mentioned is a hex core with 12 Threads!

>I don't think you've actually looked at the benchmarks.

What benchmarks? Geekbench? That's the only one I've ever seen where the fight is close, and Geekbench (like everything that boils performance down to one number) is nearly completely useless.

If you see a two-watt phone CPU beating a 45-watt actively-cooled laptop CPU, you can either conclude that the phone is alien technology decades ahead of anything else on the market or that the benchmark is broken. Which is more likely?

https://www.realworldtech.com/forum/?threadid=136526&curpost...

Does Linus not realize that Intel processors also have hardware assisted SHA?

Geekbench may not be the best tool for determining the power of a processor, but at the least it gives you a general idea of power across devices in a consistent set of tests, which can be extrapolated.

They didn't at the time he wrote that.

That post is pretty outdated. It's about old hardware and an old version of Geekbench. But people still use Geekbench 3 and old x86 hardware, so it's not totally irrelevant.

We are talking about the MacBook, IE passively cooled Intel Core m3 running at 1.2 GHz on something like 8 Watts TDP.

Benchmarks aside, because they can and are gamed, the only real way to see this, is loading a big workload such as Photoshop, Kernel compilation, ect., that professionals do every single day. I very much doubt that A11 is even remotely close to a mid range quadcore Intel cpu, and let's not forget that we're in the middle of moving to 6-8 core cpus.

Gotta remember that all these previous ARM chips have been designed in the context of the phone. If you suddenly tell the chip designers they have 10x the power and thermal headroom, theres a lot you could do.

Sure, but "could do" isn't remotely the same thing as "already faster" now, is it?

Apple's A CPUs are already the best in their category, by far.

Knowing Apple had Intel Macs running in the background for years before they announced the jump to x86, I wouldn't be surprised if they already have the chips and know how well it performs.

And I also wouldn't mind if they did another demo like they used to: https://youtu.be/oxwmF0OJ0vg?t=24m45s

> Apple's CPUs are already faster in single-core benchmarks than the Intel competition

Extraordinary claims require extraordinary evidence. "But Geekbench..." is not extraordinary evidence. Give me a board where I can do an Apple to apples comparison, booting both under the same revision of the same OS, and running industry standard benchmarks (read: plural), and then we can come to that conclusion.

No one is stopping you from doing that experiment.

> Extraordinary claims require extraordinary evidence.

"Extraordinary" is in the eye of the beholder. We have some evidence; you're welcome to believe it or not, but unless you supply alternative evidence you're not really making an argument. You're just saying "it might not be true," which is something anyone can say at any time about anything.

Nice, that was a good explanation for your stance. I am not supporting either, just liked the turnout. Obviously this isn't my usual kind of hn comment.

"booting both under the same revision of the same OS"

It should be irrelevant in Apple's case anyway since they run their own OS. If their OS is faster for a certain task than some other OS then that's part of what you would define as "faster" for a user of the device.

Right? Am I missing something?

There's no reason to think that the ARM chips they use won't be as powerful as their intel equivalents in a few years.

This is true. The rate of progress in CPU performance has slowed to a crawl and will only get slower in the future. Because the pace of innovation of fabs has slowed Intel has also lost their fab lead and GF/Samsung/TSMC should all be very competitive with Intel over the next decade.

Apple is probably looking at all these trends and concluding that they don't really need Intel.

After the Meltdown patch and Intel's exposed security flaws, Apple might still be better off with less powerful processors.

When Apple switches, their processors are likely going to be 2x faster than Intels. It will be a huge jump.

> In this instance, Apple would likely be going to a less powerful processors…

Arguably, the A11 Bionic is already faster than mainstream desktop CPUs.

"The iPhone 8 even edged out the score from the 13-inch Apple MacBook Pro with a 7th-generation Core i5 processor. That notebook notched 9,213. Is Geekbench 4 really comparable from phone to desktop? According to the founder of Geekbench, John Poole, 'the short is answer is yes that the scores are comparable across platforms, so if an iPhone 8 scores higher than an i5, then the iPhone 8 is faster than the i5.'"[1]

[1] https://www.tomsguide.com/us/iphone-8-benchmarks-fastest-pho...

Geekbench is a poor tool. Maybe for simple mathematical operations, but x86-64 has a larger instruction set than what the A11 supports

I dont think there is strong connection between a larger instruction set, and performance.

You are 100% correct.

Using a bad example, let's say x86-64 and A11 can both add numbers up to 3 digits. The a11 can do that faster than x86-64. But I want to add a 5 digit number. The x86-64 can do that with one instruction, but the A11 has to break that 5 digit number into multiple 3 digit numbers in order to process it.

From my understanding, which might be wrong, Geekbench only tests what both can do. Comparing a pickup truck and a sports car you need to look at more than just how fast they can do. You can both move furniture, but the sports car will need to do a lot more to be able to what the pickup can.

This doesn't make any sense.

Yes, Geekbench is a terrible CPU benchmark, but both of them are Turing complete, so there is no functionality difference. And complex instructions are broken into uOps in x86 too.

On the other hand, Apple hasn't been shy about putting fixed function hardware on the Ax chips to eke out the maximum performance for a given power budget. They were much faster at adding h265 decode/encode than Intel, for instance. I would imagine they'd spend their (larger) transistor budget on a lot of this type of dark silicon that would assist in speeding up common operations.

Geekbench doesn't benchmark single instructions.

Faster is not necessarily more powerful.

Faster != Faster. Adobe and Autodesk likely have many architecture specific optimizations, and take advantage of the large x86-64 instruction set.

A 6-core CPU beat a 2-core one in a perfectly scaling multithreaded benchmark? That's just shocking! Truly, that is definitely, unquestionably, beyond any doubt representative of what a battle at the workstation level would be.

But more seriously and less snarky there's a reason they used the 13" macbook pro there specifically and that's because the 15" with the quad-core i7 destroys the A11, 15k vs. 10k. Of course that's with wildly different power & thermal budgets, and I haven't seen much reason to think these geekbench results are at all representative of anything, so grains of salt and all that.

Don’t forget that the current gen A processors are designed to run in something with a tiny battery compared to a laptop. It’s likely that any MacBook bound Apple CPU won’t have such a limitation and will be able to run much faster as a result.

Also if you look at the performance trajectory of A-series in the last five years, 2020 sounds about right for where the crossover will come even between iOS based A-series processors and the fastest of what Intel has to offer, provided they can keep up the pace of performance improvement.

Why would Intel stand still while arm improves dramatically?

Apple has been gaining on Intel while both have been improving. He’s projecting where those lines cross.

> Apple has been gaining on Intel while both have been improving.

Have they really, though? What lines are actually crossing? Intel's mobile CPUs have continued to shed power, so how much of this is actually Apple "catching up" to Intel vs. Intel just optimizing for power at the cost of performance? As in, is there any performance gap that's actually shrinking, or are phone SoCs just getting more power hungry while laptop ones are getting less power hungry?

Because phone SoCs have gotten rather monsterously power hungry compared to years past. They are actually 5w under real world load parts now, with devices just letting them thermal throttle rapidly to achieve higher burst rates. Laptops, by contrast, are ~10w TDP, vastly less than they were a decade ago.

This would probably make a fascinating in-depth analysis, but the singular data point of a 5w SoC being within spitting distance of a 10w SoC is hardly a revolutionary story. It's pretty much what you expect.

but the question would be:

will apple up the battery requirements for their own cpu/gpu, or would they simply drop battery while striving for a thinner laptop?

this from someone who's been buying/using Macs for a long time - I'd love to see some very fast arm competition, but is that what it'd really turn out to be?

> It’s likely that any MacBook bound Apple CPU won’t have such a limitation and will be able to run much faster as a result.

Geekbench is based on the unthrottled peak performance of the CPU. Increasing the TDP headroom eliminates throttling but Geekbench doesn't measure throttling.

You can perhaps increase performance by overclocking but this requires the CPU to be designed for higher frequencies. AMD's Ryzen CPUs usually top out at around 4 GHz. Intel's CPU can be overclocked up to 5 GHz. The maximum clock speed is limited by the slowest component of the CPU. If the slowest operation takes 0.25 nanoseconds to complete this limits your frequency to 4GHz. If Apple had enough foresight to design their chips with this in mind then maybe but in reality they probably optimized the chip entirely for mobile TDPs.

...in one benchmark, from a heavily Apple-favoring company.

I'm sorry, we need better evidence.

ARM less powerful? Wait and see what Apple can do with their own ARM-design if they target Desktops / Laptops. We might get to see 32 or even 64 cores in a regular MBP. Now tell me how THAT would be less powerful than the current Intel offering.

Apple has always gone down the path of fewer, more powerful cores instead of packing many, weaker ones. Apple's chips were dual core until A10.

And it has let them eke out performance that absolutely leaves almost all Android phones in the dust, and all that with a battery that's 30-50% smaller than the rest (although that does bring its own problems, as we now know.)

Imagine if they're not held back by the battery at all, or even just have to match the power consumption of the power-hungry x86 chips.

Let's be realistic. 64 A72 or A11 cores is a pipe dream. It's not going to happen without a massive nm process advantage.

If we had a MBP with 64 wimpy a53 cores it would still only have 50% of the performance of a Intel i7 8700k or AMD Ryzen 1800X on paper. Of course that is faster than most notebooks but if you truly care about CPU performance then you usually wouldn't use a notebook in the first place.

There are also fat binaries - a recompile for apps that don't do anything too esoteric and you've got both formats in one bundle.

Microsoft is running 32-bit x86 applications on ARM today:

https://channel9.msdn.com/Events/Build/2017/P4171

I'm envisioning a Mac with multiple Ann chips in it. Why stop at 1 when you could put 8 or 14 Ann chips in it?

As have Microsoft and other major companies. Unless you're writing large portions of the code in assembly language, porting to a new operating system can be greatly eased by the developer tools. In the past, Apple has provided a transitional emulation layer that meant that even architecture-specific code ran on their new machines.

Don't forget Apple's NeXT heritage. NeXT supported and migrated between even more CPU platforms: 68k, x86, PA-RISC, and SPARC

Whenever there is a big change by Apple, I always see a lot of relatively niche comments like this, and they're really important, IMHO.

These sorts of "great, now I can't use Mac for XYZ" problems might, at first glance, be dismissed as "well, looks like Apple out-grew your niche market... sorry."

Upon further investigation, however, it appears that Apple might letting go of too many niche customers whose needs will be met by niche products; niche products that will grow concentrically, and the Apple we know now will die (has already been dying) a death of 1000 cuts, and the shell that remains will the consumer tech toy company we see taking shape now.

I don't feel that the parent comment was "niche". This argument comes up every time anyone discusses a architecture switch, and the invariable response is that Apple has already done this twice before with an emulation layer.

And, if anything, it's easier now than in the past. Most code has been abstracted away from the bare metal, so it's more a matter of porting runtime environments over (V8, JRE, etc) than it is rewriting apps.

Yep, it's the slow death of the general purpose computer (Cory Doctorow reference).

The niche market of "software developer" is in an interesting position if you include developers who create the apps that Apple needs to sell phones. Clearly "iOS app developer" is only a subset of "software developer" but I think there is only so far down the "consumer tech toy" road they can go without doing longer term damage to their ecosystem

I would think that non-iOS software developers are still a big money maker for Apple. It seems like half of professional developers use Macbooks.

Even though I am a non-iOSdeveloper who likes using a macbook, I can see how it would make sense for Apple to push out a small group of power users if their requirements conflict with features that enhance the experience for the majority of their users, for example something like sandboxing the entire OS away from the user. This would (ostensibly) be good for security and would only be a dealbreaker for a small minority.

As long as iOS developers aren't impacted, I'm not sure what the incentive is for Apple to allow that kind of access.

They could make a Linux distro, a "yellow box" for Linux. I know it sounds totally ridiculous, but if Apple really doesn't want to make computers for software engineers, it doesn't stop them making an OS for software engineers.

:-)

Apple did just this, twenty years ago.

https://en.m.wikipedia.org/wiki/MkLinux

http://www.mklinux.org

Interestingly, they claim to have incorporated some of the technology from MkLinux into OSX, according to this page:

https://developer.apple.com/library/content/documentation/Da...

Make of that what you will.

...and they passed that point years ago, to boot.

I'm not really sure how you reach the conclusions you do considering you don't know:

- Adobe's plans

- Apple's transition plans

- Really, anything other than the vague statement that Apple will start using their own chips in two years

Adobe has a lot of stuff that works well on iOS. ARM has neon instructions for SIMD. I think they will be okay. Changing over from one SIMD architecture to another is not that hard.

AutoCAD isn't ultra optimized for Mac anyhow, it will be okay.

This transition may screw up or significantly slow down VMs of Intel-based OSes like Ubuntu.

A lot of progress has been made recently in ARM virtualization (see its push into datacenter units). An ARM VM should run reasonably quick, and Ubuntu works fine on ARM.

Are there extensions with a stronger memory model by at very low cost (matching the one of x86)? If not, we might not get excellent perf with x86 emulated by arm in the multithread & multicore era we are.

> An ARM VM should run reasonably quick, and Ubuntu works fine on ARM.

Sure, but if we're deploying to x86 servers running Linux, then running Ubuntu on virtual ARM doesn't seem terribly wise[0] — it won't tickle the right bugs.

[0] For my money, running Ubuntu in general is a poor idea. It's not a great server OS, so it shouldn't be on the server — and if it's not on the server, it shouldn't be on the development systems either. Just run Debian everywhere.

Your last point may be a bit of an understatement. It's going to absolutely annihilate x86 virtualization on the Mac. Like, I don't think parallels or VMware will even attempt to write an x86 emulator for ARM, they'll just pack up and leave.

But the Mac thrived for a very long time without x86 compatibility, and it can do the same again, virtualization was always kind of a happy accident for us anyway.

Parallels and VMWare will just abandon their primary market?

No they won't.

Maybe not Parallels, but VMWare Fusion always seemed like a skunkworks side-project that almost got killed off at least once. Their cash cow is enterprise hypervisors, not $60 consumer windows emulation, unfortunately. (Fusion has been serving me very well for windows and .NET development purposes so far)

They can try, but you can't get blood from a stone. Emulating x86 on ARM is going to give you grotesquely low performance, it's arguably not even the same category of problem that they currently solve (virtualization, not emulation).

One thing they could run with is ARM-on-ARM virtualization, which is a more direct (heh) parallel to what they do on Intel now, but ARM hypervisors are less mature, and their practical applications, less readily apparent.

Didn't Parallels start off with an Intel emulation before Apple switched to x86.

There was "Microsoft Virtual PC for Mac" which was x86 emulation for PowerPC but performance was abysmal.

VMWare's primary market is not Fusion.

VMWare's primary market is vSphere -- and you don't run vSphere on a MacBook.

There's Windows for ARM, so important Windows applications will be recompiled to ARM and those who stay behind probably can take that performance hit.

Except, like, all the games.

IIRC Windows games already sucked on Mac hardware, even booting natively.

The hardware isn't top-end by any stretch, but you still have the back catalogue of pretty much the entire history of PC gaming, plus less demanding (but still un-emulatable) indie games — to say nothing of eGPU use for modern gaming. (And even without an eGPU, I can still play Overwatch and CS:GO on my 2013 MBP at 60fps on low settings.)

If my Steam library couldn't come with me, there's no way I'd get an ARM Macbook.

What's so special about games? Sure, old games might be abandoned, but supported and new games probably will be compiled for ARM as well. Of course it depends on whether ARM will be a commercial success and there will be sizeable number of ARM laptops.

But old games are the classics of the genre! Imagine if switching platforms meant you couldn't watch movies made from 1960 to 1990 anymore. Also, game companies lose source code all the time, and they often aren't interested in recompiling the code that they still have. (This comes up whenever compatibility-breaking iOS updates happen.)

Frankly, I'd be surprised if even a game like Overwatch got an Arm port.

But adobe's arm work is probably related a completely different series of products directed at the not quite a demanding pro segment of the market, and you still need to port autoCAD something they might not think is worth it.

The problem is that you are not just porting from a low TDP atom(by any other name) chip with a intel gfx card but a class of computers that include dual Xeon systems with 2 or more high end ati cards, to something that don't really scale upwards from a macbook air in terms of raw power.

And when you remember how badly outperformed even the highest spec mac pro is compared to their competitors, who is likely still going to be around a decade from now the question is to what extend a maker of high margin niche product targeted at demanding professionals(with corporate procurement budgets) to even consider supporting the mac with anything but a simplified html5 based consumer version that lacks most of the advanced features going forward.

I think developers flocked to Intel Macs because it gave them one machine that could run Mac, Linux, Windows in virtual or bootcamp environments. If x86 compatibility goes away, developers may jump ship to Linux or Windows machines.

UNIX developers you mean.

I used to have a PPC G5 iMac running Linux. It was fine; hitched me onto the Linux-on-Mac train for more than a decade once I finally ditched Mac OS X. And Linux already works nicely on ARM - it's the most common deployment.

This may be a terrible outcome for multi-OS users, but it's not guaranteed.

Server versions of major Linux distributions already run on ARM. Gnome and KDE too, although I hear that there can be issues. Part of the problem with ARM support is the lack of hardware for Linux developers to test on. MacBooks with ARM would actually help this process.

There are several aspect to the perf hypotheses you make, and I don't see why Apple could not design its own CPU with excellent perfs.

Current Apple best CPUs are already performing very well compared to some Intel CPU. Not the fastest Intel CPU, obviously, but a) Apple Arm implementation is already above all the Arm competitors b) we are talking about an Arm implementation optimized for low energy consumption and other components and functions essential to a high end mobile phone -- and even then it starts to catch up chips that are dissipating more and integrate far less features c) the hypothetical Apple latop/desktop cpu won't be coming today.

From a theoretical pov I see nothing that prevent Apple to achieve what they want. Especially with the kind of money they have, and the kind of team they have to already design chips. Oh btw, the A11 has more transistors than a 4core Skylake. That does not means much, except that we are dealing with the similar kind of projects, and that in that kind of project, the perf and thermal budget you have is not something that suddenly makes the people working on it go: oh my god, but we don't know how to make that, we are only good for mobile phone SoC. I suspect some of Apple employees working on their chips even worked at Intel before...

Another problem with what you say is that the workloads you consider are especially well suited for accelerations, and it is even simpler to compete in this area compared to old school CPU power. You just need bulk compute power, that can come in the form of, at least: vector instructions; GPU; DSP; fixed function acceleration units. Those areas are not especially dominated by Intel.

Some last points are that: top CPU speed are (mostly) not growing anymore. That lets others catch up Intel in the few area that they did not yet -- others also are catching up on the process side.

The result of that equation is that I really don't see why Apple could not manage to create high performance CPU to the point of not needing Intel anymore. Maybe at first they would limit to their laptops, and switch the Mac Pro last, but I'm not even sure of that.

>Some last points are that: top CPU speed are (mostly) not growing anymore. That lets others catch up Intel in the few area that they did not yet -- others also are catching up on the process side.

Meanwhile in reality the top end non apple ARM SoCs only have the performance of a single highend intel core.

http://browser.geekbench.com/android-benchmarks/ vs http://browser.geekbench.com/v4/cpu/7779910

Yes but Apple has an edge compared to other ARM (and that's what we've been discussing about in the first place) and comparing the CPU speed of an energy optimized phone SoC to an high end desktop core is absolutely not representative of what you can expect from a team tasked with creating a CPU core dedicated to performance in the first place, with less power constraints. (And that's also why the work Apple has already done on their own CPU is so impressive, btw.)

Otherwise I could as well take the perf of an IBM z14 mainframe and compare it to an existing smartphone SoC, and declare that Arm is doomed because too slow. Note than even increasing the clock alone of those ARM cores would increase the perf greatly -- it might not be possible without modifying them in various degree right now (if you don't target high freq to begin with, you can afford to work less on the length of some critical data path), but given today processes it is very probably not extremely hard to tune the design if needed.

Apple has more experience with architecture changes than many other companies. In general, transitioning to a new processor architecture is very straightforward, and way easier than switching to a new OS.

The reason why windows on ARM never caught on was simply that windows was still available for x86. With apple it's different though, because once (Or if) they switch, then the only way to be on OS X is through the use of the new architecture.

There are a variety of features of OS X that are going to keep the design community there for the foreseeable future.

Well, except it's not that simple at all. Even if they immediately discontinue all previous models and only introduce new architecture models in one go, you still have several years worth of computers sitting on x86/x64. They won't be able to get rid of it in the OS until at least the mid-2020s.

As an example with the PPC->x86 move, they released Rosetta in 2005, stopped pre-installing it in 2009, and only in 2011 did then completely remove the ability to install it. I foresee a similar experience here.

The technology to do this is already there. Apple will likely extend some piece of open-source work or acquire a small company which specializes in emulating x86 on whatever architecture they transition to. Emulators are complicated pieces of software, but they are well understood ones. We know how to make them reasonably performant, and Apple has had around four major architecture changes under its belt from which it can draw experience.

The existence of Rosetta does not prove the transition to be 'complicated'. It just shows the transition was a thing

I'm not a chip designer, but how practical would it be for Apple to try to make a CPU that has good hardware support for both ARM and x86/x86-64 binaries?

It's a radical direction, but they are designing the chip by themselves and for themselves, they have deep pockets, and if it worked well it would reduce an important risk.

There are already several chips capable of running multiple instruction sets. Most desktop machines have chips that can run either x86 or x84-64. Some(?) ARM chips have a Thumb mode. So that seems very practical, and obviously there is a bigger "impedance mismatch" when trying to do both x86 and ARM on the same chip, how much worse is it?

If you could do some kind of hardware real-time instruction translation without wasting much silicon and without reducing performance (for x86) that much, it seems like it would take a lot of the pain out of the transition.

I'm not a chip designer either, but IIRC, the current architectures are actually something similar, whereas the CISC is broken down into custom RISC-ish instructions by a HW layer, which is then executed by the RISC-ish cores.

And while it might be doable (x64 and ARM on the same chip), I think it might not be a smart decision, since some things will not function correctly or will have to be redone (e.g. speculative execution).

I don't know either architectures on a deep understanding level (x64 or ARM), but I have a feeling that there should be some big architectural changes that would prevent x64 code to run on ARM natively.

See: Transmeta.

AFAIK the biggest impediment might be that Intel would need to license Apple to do so, not to talk about patents etc.

How do you think the announcement that OSX will be supporting external graphics cards fit into Adobe/Autodesks plans?

I don’t really do much video, photo or 3d editing these days but isn’t that all GPU driven?

If that’s the case then couldn’t someone like Nvidia pick up the slack for Apple with a really nice external modular offering?

It is also quite likely Apple will try to add some kind of SIMD chip to their ARM efforts. Things just got a hole lot more interesting if we can wrestle Intel from their current leader position.

I use both Adobe and Autodesk, and neither yet support eGPU well. Essentially, if they wrote their own graphics engines (both adobe and autodesk did) then eGPU doesn't work that well. Lightroom has started to come over, and some features of Fusion360 seem to work OK, too, but for the most part the only eGPU accelerated parts of the apps seem to be the user interfaces.

Given Apple's recent obsession with AR/VR, I wouldn't be surprised if the new apple architecture was really GPU first, CPU second. Apple has always loved graphics, and in all likelihood they recognize that a more parallel OS model might actually suit people's day to day needs better.

Sure nvidia could build a traditional co-processor card(just like in the old SBUS days) but what are the chance of apple releasing hardware with that kind of extendability after going all in on the IOSification of the mac?, even thunderbolt(which is a much lower level bus then USB3) did not support the same kind of performance as a proper PCIe slot.

And even if you could most of the software in question is composed of countless components some of whom predates the mac itself and porting that to support anthing but the traditional big core CPU platform is not a trival thing to do, which is why the Xeon powered workstations and servers still exists and why manycore designs only really work in the supercomputer space.

You could argue that the workstation itself is a dead concept and that the future is "cloud" with html5 frontends even for video and CAD workloads* which is something autocad and adobe is likely working on as a long term strategy, but in the short term they need the workstation that post Jobs apple don't appear to to have any interest in building.

Apple goes where the money is. I'm sure they have ran the numbers. If this makes financial sense I'm sure they are willing to drop the professional graphics market. Still waiting on that new Mac Pro?

If Apple manages to lose both designers and developers, I think the whole ecosystem (not just macOS) will suffer more than Apple's management realizes. I also don't think it's possible to run the numbers on this scenario.

Without the insistence of so many designers on using a Mac, would macOS really have won a foothold in corporate environments? If enough designers and developers switch to Windows, who's to say that macOS won't slowly turn into a hobbyist OS that no IT department wants to support (like Linux, or Macs ~12 years ago)?

Apple doesn't care about the corporate market. They care about individual users. And as long as the CEO wants their shiny Mac (also, see iPhone, iPad) then the IT dept will support it. Your argument might have been valid a decade or so ago, but since then, every home has a laptop or three, and BYOD happened.

In my limited experience, management often uses a shiny iPhone, a shiny iPad, but then a Windows laptop (or none at all). I can't find any statistics to back it up, though.

>this essentially kills the professional mac workstation as a viable product going forward?

Some would argue it's already been dead for a long time. I'm seeing less and less directors and artists using Macs for their workstations when it was close to 100% about 10 years ago.

They just didn't supply a competitive workstation machine for so long that many were forced to move to windows if they wanted power. I don't know a 3D artist anymore who isn't doing all their rendering through multiple Nvidia GPUs in a tower PC.

The iMac Pro has only been out for a few weeks after years of neglect and I'd struggle to recommend it to anyone at that price point.

(Ironically many I know still use a Mac laptop but it's relegated purely to office/email/slack/web tasks that back in the day we'd associate with Windows machines, not the actual creative work)

> is going to require significant work by third parties outside Apple who haven't even heard of the possibility of this happening until today

I don't see recompiling their next release with a different version of Xcode qualifying as "significant work".

> whatever it is the pro's use for video editing those days are a different ballgame.

That would be Final Cut Pro X, which Apple itself makes. (Also, Motion, Compressor, and honestly, iMovie does a lot of stuff you wouldn't think it could do for a free app.)

There are very very few pros using Final Cut Pro X (lots of wedding videographers though). Every single edit house I've been in recently is on Avid or Premiere (usually both) with a FCP7 computer somewhere for legacy projects.

I've never encountered anyone using Motion professionally. It's all Nuke and After Effects with Fusion growing.

If Apple needs to, they could buy either Autodesk or Adobe. To be honest, I'd prefer the buy and port Solidworks.

FTA: The Pro units will remain on Intel until Apple's chips could compete for this reason.

My concern is that they are going to completely lock down and cripple MacOS and turn it into iOS, making it worthless as a professional machine. The CPU doesn't matter but the versatility and user-centric-ness of the OS does.

If they're talking about making MacOS more like iOS I think it might be time to start looking for the exit. A laptop/desktop is not just a portable dumb terminal for accessing siloed cloud services.

I agree and it's unlikely my next laptop will be an Apple device...

I was thinking the same thing, but Apple doesn't care. Folks like us do not represent the majority of the market.

exactly. It's a shame... written on old macbook air

I see no reason to suspect that changing CPUs would be related in any way to "crippling" macOS. Those are two separate issues.

They are separate issues, but history has shown that changes in architecture (or form factor) is often used as a pretext to also make other changes.

Mobile transformed the computer into a portable surveillance and addictive media device aimed at the user rather than a personal computing device built for the user. Those are two separate issues of course -- there is nothing in mobile that mandates that it be designed to commoditize its users -- but the shift to mobile was used to also smuggle in a total inversion of the user/machine relationship.

Apple is kind of a strange company. On one hand they have stuck up for the privacy and security of their users, but on the other hand they were a factor in this inversion of relationship. They weren't the only or even the primary factor, but the locked down nature of their iOS platform helped other actors such as Facebook and Google implement surveillance-capitalist and gamified attention-capitalist user experiences there.

Whether or not I stick with Apple depends on whether or not they keep my computer mine even as they swap out the CPU. I could care less about the architecture as long as it performs well, but I do care about the nature of my relationship with the technology that I use. I care about this for both personal and professional/pragmatic reasons.

Why assume this is an arch switch to ARM? What's stopping Apple from making an x86-64 compatible chip? Or even a chip with dual instruction sets? I'm sure they have the in-house talent at this point to make it happen.

Interesting question. I think the advantage Apple has here over MS, is that partners like Adobe already have experience making apps run on Apple's ARM chips. Lightroom on iOS is surprisingly full featured and fast.

Whats the big deal? We all write code that is cross platform, don't we? :)

No one wants to be the next QuarkXPress.

Why are you so sure that it will be ARM processors tho?

Reply on news.ycombinator.com