If Mac went ARM tomorrow, would you jump?

[cram501]

There isn't one, mainly because infallibility is impossible. Thankfully, us technologists have already solved the problem.

Approved/Reviewed apps are signed by Apple.
Signed apps are procured via the app store.
An issue is found with an application post-release (security/privacy, etc)
The signature on the app is revoked / added to a CRL.

I need a hardened implementation of this exactly as much as I need a completely open and unencumbered ecosystem like Libreboot+GNU.

The only problem that has been solved with the curated apps on the App Store is that you are notified if Apple determines that an app has violated its policy. That may reduce the footprint of hostile applications but it really is a false sense of security. You should eventually get notified if you've been screwed by a bad actor.

The walled garden is more about revenue than application security.

I actually like that and I like the App Store. That doesn't remove the need to install applications I need that are not or will not be available in the App Store (dodging the 30% Apple Tax on apps). There are a lot of applications and utilities that are available for MacOs independent of the app store. It is one of the many strengths that MacOs has over iOS. We'll have to see if that remains true if or when they swap to ARM.

I'm not sure why some are against OTHERS installing apps for their use on THEIR devices. It has no impact on those that wish to stay within the walled garden. I understand why Apple would like to close it down... $$$.

 

[skottichan]

That's the 64,000 dollar question as they say. Depending on who you ask, Apple's Ax processor family will either out perform Intel core i7 CPUs or it will be slower.

The Ax has been great at running one app at a time, but with these multi-threaded apps all running at the same time, who knows how that will perform.

And that'll be my stipulation. I know my useage patterns aren't normal, but I need something to power a Cintiq, run photoshop, and a couple other apps simultaneously.

 

[RecentlyConverted]

There isn't one, mainly because infallibility is impossible. Thankfully, us technologists have already solved the problem.

Approved/Reviewed apps are signed by Apple.
Signed apps are procured via the app store.
An issue is found with an application post-release (security/privacy, etc)
The signature on the app is revoked / added to a CRL.

I need a hardened implementation of this exactly as much as I need a completely open and unencumbered ecosystem like Libreboot+GNU.

Like the Facebook app on iOS?

 

[9081094]

I’m missing the ‘not sure yet’ radio button. Since I’ve made the transitions from Motorola to PowerPC to Intel I experienced what the transition is bringing you. First all your software has to be rewritten to take advantage of the new platform. You can probably run your existing software in some kind of ‘intel emulator’ which translates in slower speed. Second the innovation of MacOS will languish for a couple of years. They’ll work on stability issues and work on bugs. All in all this will take Apple 2-3 years. Than you’ve to wait for third party developers. As the Mac isn’t that important anymore due to the neglect by Apple and developers have to adjust to the new possibilities, it will probably take another few years. Some developers will quit.

The transition will go very slowly and Apple has to support the emulator, working on stability and won’t put much energy in bringing new capabilities to the OS.

I’m not sure I would invest. Because seeing the involvement from Apple to their platform since 2012 and the absence of a stable future roadmap together with excessive pricing... what would you do?

It all depends if the transition is a big jump or not.

As of today I’ll say a big NO. But it’s on Apple to bring back trust and value for money.

 

[russofris]

Like the Facebook app on iOS?

I do not have the Facebook. I can not even comprehend why anyone would.

I would advise that you delete your profile immediately and remove the application from your iPhone.

If you require technical assistance, there are technicians available to help with that.

 

[kwikdeth]

As long as Universal Audio is on board, I'm all set.

LOL. UA is probably the absolute worst example to cite, considering there's little to no MIDI support in the console app, they've never released any kind of supporting app for iPad, let alone a way to control Console from an ipad, despite customers begging for it for years and years and years, and have still to this day refused to make any comment or commitment to doing so.

meanwhile, Console has supported Multi-touch on Windows for years now. And you can already plug in an Apollo Twin to a Surface.

 

[Ledgem]

I agree to a point -- but it's not about today. The full Marzipan rollout will take two years. As software development moves increasingly toward a mobile-centric focus, Apple is trying to avoid future marginalization of Macs as an app development target. There is no Mac NetFlix app and Twitter is dropping their Mac app. Multiply that by 20x over the next few years.

Note your point about just using web apps -- not necessarily needing a native Mac or ported iOS app. This shows how an ARM transition would be less traumatic than past Mac CPU changes.

I'll be honest - I don't use Netflix or Twitter. I thought these were web-based services to begin with. I looked into the Twitter one and there are some interesting theories as to why development was stopped, including a theory that Twitter is porting its iOS application to macOS.

The web apps is something that can't translate well to Mac. The point is that on iOS navigating a website can be cumbersome, so many apps serving as fronts for websites makes sense. On the Mac, programs like Adobe's suite, Final Cut Pro, and so on, aren't served well by such simple designs. Simply put, we're trying to do more on the Mac.

Another advantage to a future convergence of Mac/iOS apps is you'd often only need one device. Many times I take both MacBook Air and iPad Pro because I need apps or functionality available only on one or the other.

I'm not sure that going ARM would lead to this. I'm not opposed to the idea of Apple making a Surface-like device, and I suspect that some day they will, but Tim Cook was pretty adamant that Apple would never do something like that.

There's also a lot more to these devices that battery life. This thread is still in the iMac forum, right? I've done a few touch-ups to presentations using Keynote on my iPad, and I've even thrown together a quick video edit using the iOS version of iMovie, but for serious things I have to use Keynote on the Mac or Final Cut Pro X. Screen real estate is one consideration, and while some day we might have a fold-out 27" iPad, we're not there yet. And it's not just screen real estate but using a mouse/trackpad versus a touch screen. Even with the aid of an Apple Pencil, the touch screen is still a bit fiddly at times.

Any loss in sales from people buying only one device would probably be compensated by increased market share and avoiding marginalization of the Mac app platform.

I like your thinking but I see it a bit differently. If the Mac platform is being marginalized, it's because Apple isn't maintaining it. Hardware-wise, they're doing a poor job. I've said it before, Intel may take some blame for that, but I don't think Intel can shoulder all of the blame. And as was posted before (rather well, I might add), a transition in processor architecture really needs commitment to get developers and others to sign on to it. Apple just isn't showing that commitment, and I'd be worried that many developers are thinking the same thing that I am. Maybe the iOS developers would be thrilled, but those apps are developed with a different focus. If native Mac apps died and were replaced with iOS apps, forget about the Mac being marginalized, it'd be dead.

 

[jinnyman]

Another thing I'm worried is the different situation we are at right now. The world has changed since 2006 when the intel transition happend. Back then, computers were pretty much only means of doing "computing". Windows was really bad back then, and I was happy to go through the transition since I knew I was going to use a great OS with much improved performance.

Nowadays, Macs are not the only computer I'm using. All our gadget does pretty much all the basic stuff that it needs to do, iphone, ipad, etc. The reason I'm still using my personal computer is due to unmatched productivity it offers (multi-tasking, OS being simply better at any other portable at doing hard work at sheer performance).

I had to go through Intel transition because even though occasitional rosetta hiccups, it was best choice for me. I'm not sure about this iteration. The performance is yet to be proven, and what's benefit of going ARM in desktop other than making Apple happy by enabling them to close-control further? Also, Windows are much more improved compared to 2006. I'm already using a custom PC for serious work, and it's perfectly usable. The cost of going through architecture change is pretty significant, and I'm not sure Apple can go through as flawlessly as last time. Heavy users are already leaving Apple. Hopefully, Apple gets their things together and continue support heavy users by introducing a versatile ARM workstation instead of ipad like garbage.

 

[theluggage]

and what's benefit of going ARM in desktop other than making Apple happy by enabling them to close-control further?

Well, desktop probably isn't Apple's #1 priority. We're already at the stage where you have to have a pretty strong aversion to Windows in order to put up with Apple's limited range of desktop options.

There probably won't be any huge benefit to having ARM on mid-range desktops. The advantages are:

Laptops/ultraportables/tablets: power consumption vs. speed, the ability for Apple to custom-tailor A-series CPUs to their designs rather than have to build around whatever CPU/GPU/Thermal permutations Intel decides to release each year. The prospect of a "crossover" machine that can run iOS apps alongside MacOS.

Workstation-class 'pro' desktops: massively powerful chips with more CPU cores + acceleration hardware on-die than is possible with Intel. Probably pitched more at 'appliances' running specific, optimised pro software from Apple and selected partners than as a general-purpose workhorse. With the Mac Pro not having been updated since 2013, the only other option being an all-in-one (who's appeal depends entirely on whether or not a 27" 5k display fits your needs) and Apple stubbornly pushing AMD GPUs in an industry that prefers NVIDIA I'm sure most of the 'general-purpose workhorse' customers will leave the building as soon as their 2012 cheesegraters die.

I still think the best starting point would be MacOS on something like a 2018 iPad Pro - so early adopters would be getting an "iPad Plus" rather than what would inevitably start out as a "Mac Minus".

Overall, compared with past processor switches:
Advantage: the IT world is far more diverse than it was at the 6502-68k, 68k-PPC, PPC-Intel switches, software is written in a more portable fashion, many Apple software houses are already supporting iOS, platform-independent browser-based apps are now a big thing...
Disadvantage: Previous processor switches included a huge performance boost and could promise spectacular performance gains on native apps plus OK performance under emulation - even with PPC to Intel, the PPC (especially the mobile versions) had already fallen a long way behind the curve. The advantages of ARM are going to be a bit more subtle, and will depend heavily on Apple using it to design better all-round systems rather than a raw speed boost.

 

[MandiMac]

My point isn't that Apple can't produce an Arm chip that competes with intel in a laptop or desktop, it is that it will take longer than everyone anticipates and they haven't produced one yet. [...]

So, basically you're saying this about a possible Apple desktop chip? "We've learned and struggled for a few years here figuring out how to make a decent phone. PC guys are not going to just figure this out. They're not going to just walk in." (Ed Colligan, pre-iPhone era)

 

[cram501]

So, basically you're saying this about a possible Apple desktop chip? "We've learned and struggled for a few years here figuring out how to make a decent phone. PC guys are not going to just figure this out. They're not going to just walk in." (Ed Colligan, pre-iPhone era)

Not at all. I think it is possible and they will get there. I don't think it's going to be the fairy tale everyone seems to think it will be. Arm will be competitive. It is a matter of when and how long it takes Apple and what limitations they may have. I doubt money is an issue with Apple. An iPhone/iPad chip is not a desktop chip regardless of how many times people exclaim that artificial benchmarks prove it's better.

I hope they succeed and get competitive chips (as well as ARM as a whole). Competition is good for the market. Intel has been slacking off (mainly because BK's terrible focus when he was CEO).

 

[MandiMac]

An iPhone/iPad chip is not a desktop chip regardless of how many times people exclaim that artificial benchmarks prove it's better.

I'm fully with you except for this quote. You're right, they are artificial benchmarks, but that A12 series (with a power draw of 6 watts) beats Intel chips with much higher wattage. Nobody seems to factor in what could happen if a B13 chip used the same 20/45/95 watts like Intel chips did. Yet.

 

[cram501]

I'm fully with you except for this quote. You're right, they are artificial benchmarks, but that A12 series (with a power draw of 6 watts) beats Intel chips with much higher wattage. Nobody seems to factor in what could happen if a B13 chip used the same 20/45/95 watts like Intel chips did. Yet.

That is kind of my point. Until they can get, display, and benchmark a chip with a real preemptive multitasking operating system like MacOS that can compete with the features of a desktop OS, the comparison isn't valid IMO. The feature set of one of these future chips is also going to be based on Apple priorities so we won't know what capabilities they will bake into it.

I'm not saying they won't or can't . I'm saying we don't know until they do.

 

[RecentlyConverted]

I do not have the Facebook. I can not even comprehend why anyone would.

I would advise that you delete your profile immediately and remove the application from your iPhone.

If you require technical assistance, there are technicians available to help with that.

You obviously completely missed the point. Take off your Apple glasses.

 

[skaertus]

I am giving a second thought and perhaps I even buy an ARM Mac. But it depends.

Lately, my main computer has been a desktop. A Windows-based desktop, far more flexible to configure than a Mac, and still more powerful and cheaper. And yet it has more options of software, and its version of Microsoft Office is far better.

However, a MacBook may be a very good companion to be on the run. A MacBook has a great keyboard, an extraordinary trackpad, and top-notch components overall (such as a great screen and sound). Plus, the MacBook is thin and light and has decent battery life. If an ARM MacBook provides better battery life and LTE connectivity, then why not? It will not be my main computer, but it may be convenient to use on the go. Better than an iPad, that is for sure, but it can be equally fast to turn on.

 

[Ledgem]

I'm not sure why people keep fixating on the battery life. Both the iPad Pro and the current MacBook are rated for about 10 hours of activity. The MacBook is actually rated for 12 hours of iTunes movie playback, according to Apple. Yes, the MacBook's battery is larger (41.4 watt-hours compared to the iPad Pro's 36.71 for the 12.9" model), and I don't think anyone discounts that ARM is the more efficient in power usage... but these aren't Earth-shattering differences, despite the ARM chip being specifically refined for this purpose, while the MacBook represents Intel's relatively early efforts to create a power-efficient, fanless system. And let's face it, Apple would probably use the opportunity not to increase battery life, but to keep that ten-hour rating while shaving a few ounces off of the MacBook by giving it a smaller battery.

I use both devices differently and thus it's hard to make a direct comparison, but I can attest that both the iPad and MacBook have a very impressive battery life. My take is that the battery life bit is overblown.

 

[maflynn]

I had to go through Intel transition because even though occasitional rosetta hiccups, it was best choice for me. I'm not sure about this iteration

Imo, that's the key. As I previously mentioned, I too went through the PPC->Intel transition and for many of us the advantages of embracing Intel were numerous.

What advantage does moving to ARM provide? Will Apple lock down the Macs further and force developers to use the app store (I think this is going to happen)? What will the consumer benefit from this move?

To me, I don't see the benefits, just the negatives and for me, I don't see myself embracing an ARM based Mac.

 

[faraway52]

Jump? no way! But may be moving slowly when Apple shows a significant advantage for me. Another check point will be reached when the current iMac (arrived 5 days ago) should be replaced due to age, slow performance etc.

 

[skaertus]

Imo, that's the key. As I previously mentioned, I too went through the PPC->Intel transition and for many of us the advantages of embracing Intel were numerous.

What advantage does moving to ARM provide? Will Apple lock down the Macs further and force developers to use the app store (I think this is going to happen)? What will the consumer benefit from this move?

To me, I don't see the benefits, just the negatives and for me, I don't see myself embracing an ARM based Mac.

I agree. I would not use a Mac with a locked down app store as my main computer. It could still be a good laptop, but a secondary computer.

The problem is, once Apple decides to move to ARM, what can we do? We can of course stick to older Mac models running Intel, but for how long? Five years? Ten years?

After such transition, Apple will continue supporting Intel Macs for a period of time, and then they will no longer run the newest version of macOS, nor the applications. Everybody will eventually have to move to ARM Macs or to another platform.

So, any choice of not moving to ARM once Apple makes the transition, will be temporary. Not being tied to Apple’s choices is one of the reasons I never made a full transition to the Mac and I do not use Mac-only apps. I want to be free to go to Windows whenever I feel like.

 

[theluggage]

That is kind of my point. Until they can get, display, and benchmark a chip with a real preemptive multitasking operating system like MacOS that can compete with the features of a desktop OS, the comparison isn't valid IMO.

iOS is a real, preemptive multitasking operating system based on a variation of the same unix-like OS kernel as MacOS. The main differences with MacOS are the touch-centric user interface and the 'lockdown'. Also, the iPads have been tested with 'real world' jobs, as well: https://www.tomsguide.com/us/new-ipad-pro-benchmarks,news-28453.html

In any case, the current iPad processors don't have to decisively beat the MacBook Pro or iMac - they just have to outperform the 12" MacBook and maybe the Air which would be the obvious first candidates for replacement by an A12-based system.

The only credible reason why Apple would, in the short term, stick an A12x in a MacBook Pro or iMac, would be as some sort of prototype for developers. If they want to replace Intel in the higher-end Macs they'll need to come up with something new - but that is perfectly feasible and its not only Apple who are looking into server and workstation applications for ARM. Heck, Fujitsu is working on ARM-based supercomputers: https://www.theregister.co.uk/2018/08/22/fujitsu_post_k_a64fx/

Both the iPad Pro and the current MacBook are rated for about 10 hours of activity. The MacBook is actually rated for 12 hours of iTunes movie playback, according to Apple.

Frankly, I trust battery life benchmarks far less than I trust speed benchmarks... Modern Intel CPUs are heavily optimised to power down as much of the CPU as possible for low-demand scenarios such as web browsing and media playback. The 'iTunes Movie' benchmark is basically measuring how much power the screen, the memory and the H264 hardware codec consume - the CPU is barely ticking over feeding data to the codec. Even the lowest-specced Raspberry PIs can happily play back HD video provided that it is in a format supported by one of its hardware codecs.

The $64,000 questions are how much power these chips consume under load (like video editing) and how quickly they need to start thermally throttling the CPU. Bear in mind that battery life and heat generation are joined at the hip (ye canna' change the laws o' physics) and the iPad is thinner (and harder to cool) than even the 12" MacBook.

I agree. I would not use a Mac with a locked down app store as my main computer. It could still be a good laptop, but a secondary computer.

If Apple want to lock down MacOS to the App Store they don't need to switch to ARM to do it: it's already a user option in MacOS for Intel - they just have to take that option away.

What advantage does moving to ARM provide?

Apple get to design their own system-on-a-chip with precisely the permutation of cores, clock speed, cache, GPUs, other accelerators and thermal profile they want for their products - rather than hope that Intel releases what they need.

Part (but maybe not the whole) of Apple's problem with keeping the Macs up-to-date is Intel's tendency to announce Generation X+1 with great fanfare (which Dell immediately shoe-horns into an XPS whether or not it makes sense) before they've actually released the permutation of CPU/GPU/Cores/Thermal that Apple needs for an upgraded Mac.

E.g. the new Mac Mini: Even if Apple wanted to give it a better GPU, Intel don't do a desktop-class i3/i5/i7 chip with an Iris Pro-class iGPU (which would have been ideal) - because desktop PCs either have PCIe slots or are business systems that don't need decent graphics.

 

[cram501]

iOS is a real, preemptive multitasking operating system based on a variation of the same unix-like OS kernel as MacOS. The main differences with MacOS are the touch-centric user interface and the 'lockdown'. Also, the iPads have been tested with 'real world' jobs, as well: https://www.tomsguide.com/us/new-ipad-pro-benchmarks,news-28453.html

In any case, the current iPad processors don't have to decisively beat the MacBook Pro or iMac - they just have to outperform the 12" MacBook and maybe the Air which would be the obvious first candidates for replacement by an A12-based system.

Regardless of the MacOs roots of iOS, the platform has limitations. When benchmarks can run a diverse set of applications on the same operating system, we will see the closest apples to apples comparison. Anything else is disingenuous. The claim that performance is on par with x86 in MacOs has not been proven.

I'm not saying that Apple can't produce competitive ARM processors. It would be cool and interesting if they can. The issue is that Apple will have its own set of goals and objectives. Apple may not want to produce a general purpose processor.

x86 was big because I could use it how I wanted. Build it how I wanted it. Run the OS how I wanted it. Apple grabbed on to that market. Converting to ARM would probably remove the "Pro" users (not that that hasn't already been occurring). It would transform Apple into something different. That could be good or bad.

 

[theluggage]

hen benchmarks can run a diverse set of applications on the same operating system, we will see the closest apples to apples comparison. Anything else is disingenuous. The claim that performance is on par with x86 in MacOs has not been proven.

...they have geekbench results, they have video transcoding results and they have RAW-to-JPEG conversion results, all of which beat x86 devices by a comfortable margin. You have a hand-wavy argument originally based on a false assertion about iOS not being multitasking. Its a UNIX-like OS running a closely-related kernel to MAC OS, supporting a sophisticated graphical UI and all the usual background processes, running the same geekbench code and apparently related application software. You need a better justification to claim that the benchmark results are somehow 'wrong'.

Meanwhile, the x86 machines in that test have more RAM and probably more/faster SSD than the iPad, let alone the likelihood that any actual ARM-based Mac will have something better than the iPad-centric A12X under the hood. So, there are plenty of variables that could skew the tests further to the ARM's benefit.

So, no, nothing has been proven - either way - but so far the weight of the actual quantitative evidence is that the A12X outperforms the x86 chips in laptops and convertibles and can afford quite a significant hit from reality and still be on a par with them.

 

[cram501]

...they have geekbench results, they have video transcoding results and they have RAW-to-JPEG conversion results, all of which beat x86 devices by a comfortable margin. You have a hand-wavy argument originally based on a false assertion about iOS not being multitasking. Its a UNIX-like OS running a closely-related kernel to MAC OS, supporting a sophisticated graphical UI and all the usual background processes, running the same geekbench code and apparently related application software. You need a better justification to claim that the benchmark results are somehow 'wrong'.

Until you get the benchmarks on the same OS with the same priorities, with the same limitations, you are making a "hand-wavy" argument and hoping you are correct. You are making assumptions on a very mobile focused operating system and assume that doesn't impact results.

Until we get similar platforms running the same general purpose operating systems, it can't be accurate.

 

[jinnyman]

...they have geekbench results, they have video transcoding results and they have RAW-to-JPEG conversion results, all of which beat x86 devices by a comfortable margin. You have a hand-wavy argument originally based on a false assertion about iOS not being multitasking. Its a UNIX-like OS running a closely-related kernel to MAC OS, supporting a sophisticated graphical UI and all the usual background processes, running the same geekbench code and apparently related application software. You need a better justification to claim that the benchmark results are somehow 'wrong'.

Meanwhile, the x86 machines in that test have more RAM and probably more/faster SSD than the iPad, let alone the likelihood that any actual ARM-based Mac will have something better than the iPad-centric A12X under the hood. So, there are plenty of variables that could skew the tests further to the ARM's benefit.

So, no, nothing has been proven - either way - but so far the weight of the actual quantitative evidence is that the A12X outperforms the x86 chips in laptops and convertibles and can afford quite a significant hit from reality and still be on a par with them.

I will believe when it comes out and proven to be better. I really don't care iOS test ran on an iPad. The things I can do on iPad and the things that I do on my PC desktop are vastly differ in terms of resource requirements and work flows. The only way to see if ARM really delivers is when I see the same workflows, day to day task tested on MacOS which is not gimped out OS like iOS in multi-tasking. The decision process I went through from 2700x to 9900k is pretty complicated both accounting for IPC, overall multi-core performance, and finding the most efficient chip suitable for the apps I'm using. Several synthetic test results and theoretical assumption are not enough for me to suddenly see ARM as the future.

Of course, the TDP limit will be lifted much more compared to iPad integration, but again, I really have to see it. It's unproven technology no one is yet to integrated in PC really, and win10 for arm never got that much traction or proved to be decent x86 replacement anyway. The chips nowadays are better, but I don't see everybody jumping into Arm chips due to so-called huge performance benefits.

If ARM offer much better performing, more energy efficient chips compared to 9900k, for example, then why should I resist it? If there's considerable performance benefit by going from Intel to ARM transition, then I'd do it. But Apple has to prove to me it's worth it.

 

[jerwin]