MP 7,1 What does Apple's switch to ARM based processors mean for the future of the Mac Pro

[Blair Paulsen]

Even if ARM Macs have impressive synthetic benchmarks, it won't get us very far if app developers aren't interested enough to devote sufficient coding resources. I could easily see a future where Apple's homegrown apps are blazing fast on new silicon, but not much else. For starters, what about Adobe? Will they cobble together a port just good enough to keep their Mac audience feeding them subscription income, or will they dig deep?

 

[ZombiePhysicist]

You can count on it - just look at the last transition. The software that did transition took about about 4 years 1st version was Intel native, and that was about it - it wasn't until version 2.0 (Intel) that software houses hit their stride.

A good chunk of that was Apple's fault (P.T. Barnum was pushing 64-bit Carbon, since the Cocoa libraries weren't ready, and then did an about face 12 months later 64-bit Carbon is dead - you can trash everything you converted over the last year) - which was why Adobe & Microsoft were so late to the conversion.)

Then look at how long 10.6.8 hung on because of Rosetta. I didn't upgrade until 10.10 came out, and I was far from the only one.

I guess that's one characterization of it. Im not sure carbon was 64 bit or not, but it was basically a way to get old macOS 9 era stuff over. The Cocoa libraries were very much ready. They were the NeXTstep libraries. People didn't want to move to cocoa because it's basically total rewrite of their apps, whereas using the old carbon libraries let them make relatively minimal changes to get things compatible.

I agree with you pointing out that the last migration was moving TOWARDS industry standards and now away from it. But I think it's equally important to point out this time, the apps do not need complete rewrites from carbon to cocoa but are mostly just recompiles of cocoa apps. That makes a huge difference in getting apps over. I'm betting the vast majority of apps are ported quickly and easily because of that. One big exception being virtualization apps will not only be a lot of work but will undoubtedly work way slower than on native x86 chips even if they are ported.

That said, moving away from industry standard x86 to proprietary arm... that is meaningful as well. I'm not sure what it will mean, and I think it's tough to tell what it means until we see the actual chips and how they perform.

If overall performance ends up being a wash over x86 options... You've got to think "was this trip really necessary" and we will forgo great bootcamp and near native virtualization benefits for a wash in performance... That could be a long term turn off. Whereas if they can release machines that are clear performance (both in CPU processing throughput and power consumption) per segment... Well that will leave a very different impression.

We just wont know until the hardware rolls out.

I'm super eager to see what happens. But the range of outcomes is at least somewhere between 'meh, this was a wash at best and unnecessarily proprietary' to 'my god we're doing 2x throughput for 1/2 the power, this was a great decision'.

Time will tell.

 

[Lycestra]

FWIW, it's not like a 7,1 becomes a paperweight overnight. Look how long many people kept getting work done on pimped 5,1 cheese graters. If the delta between how much you can crank out on a MBP vs a 7,1 MP is significant enough to pay for the tower over a couple of years I wouldn't sweat what's next. When the performance delta between the 7,1 Intel MP and an 8,1 ARM MP reaches a certain level then upgrade.

The distance and firmness of that horizon is the issue. I have a 4,1 that has been running the latest os now 3 versions past what Apple officially supports (11+ years). It's mostly because Apple hasn't cut out the CPU I use from being able to boot it. BigSur would never run properly for me if it required AVX. Back on point, the 7,1's will only run the latest os if their CPUs are supported, because the moment they're not, it will be because they're the last to go, and the ARM transition is over. With the ARM transition, the line in the sand will be much more rigid. The PPC->Intel transition from announcement to an Intel-Only mac OS was just over 4 years. After that, you'd get security updates for a year or two, but by that point, nobody will be making new software for it. To keep it running then, you have a machine that only runs heritage software ("legacy" to use the bad word) at best.

I don't think they'll release a proper Mac Pro with ARM until they can arguably compete directly with the old hardware. Otherwise it will be a 2013 Mac Pro, on the order of the Mac Mini that lost half its performance (went from consumer desktop hardware to mobile hardware internally)

All this is a question of how long after Apple's completed the full hardware transition are they willing to maintain the Intel devices? In all likely hood, it will languish on the backend libraries that are being rewritten on ARM, while older devices slowly become excluded from some of the newer features, some bound to hardware (USB5, now with Fiber, aka LightPeak), others bound to software that's a pain to rewrite if they have to (Metal 4, SecurWhozit), or just don't want to.

 

[ZombiePhysicist]

Agreed, but it seems that the 7,1 Mac is very likely to remain on current OSs at least through 2026. That's plenty of time for most if they want to milk the machine and remain current.

I agree the big question is how soon can they get out a processor that can compete with the Xeon segment of the industry? We're all guessing 2 years because that's how long Apple said it might take. But maybe it's sooner. Or maybe it is 2 years but its not super competitive. Maybe it takes them 4 years to get up to parity.

My bet is that in 2 years they will have something competitive with whatever is the then AMD 64core top of the line processor class. But we're all guessing.

 

[Kpjoslee]

My bet is that in 2 years they will have something competitive with whatever is the then AMD 64core top of the line processor class. But we're all guessing.

They don't necessary have to match the top of the line Xeons or EPYCs in terms of core count, since they now have the luxury of integrating specialized cores for specific applications as they see fit.

 

[goMac]

The crux of this issue is time. How long until Apple can put out a legit ARM workstation? How long until apps other than in house programs like FinalCut and Logic are optimized for ARM architecture? Etc.

Apple is vertically integrated, which aids them in times like this. Final Cut and Logic derive their optimizations from Accelerate and Metal. Accelerate and Metal have been optimized for Apple Silicon for a _long_ time because they also ship the same libraries on iOS. Therefore Final Cut and Logic are already well optimized with Apple Silicon. That's why it probably literally took them five minutes to build an optimized version.

Doesn't help companies like Adobe who tend not to use things like Accelerate. Adobe is using Metal though, which means Metal should confer or force them to optimize.

But teams like the Accelerate team are already tuning for each version of Apple Silicon individually. Apple gets to reap their work internally on their own apps.

 

[Stephen.R]

I'm not challenging you bro. Chill. I was asking.

You said you believed it was X, when I described what it is in more detail, part of which is "X", you acted surprised. The level of discourse on this site would improve about tenfold if people actually read and understood things before hitting the reply button.

 

[ZombiePhysicist]

You said you believed it was X, when I described what it is in more detail, part of which is "X", you acted surprised. The level of discourse on this site would improve about tenfold if people actually read and understood things before hitting the reply button.

Yea dude you're cancelled from my feed.
[automerge]1594095995[/automerge]

They don't necessary have to match the top of the line Xeons or EPYCs in terms of core count, since they now have the luxury of integrating specialized cores for specific applications as they see fit.

Agreed. I'm speaking roughly in the sense it has to be in the ballpark in overall performance/throughput. They may well get there with very different core counts and types.

 

[macguru9999]

Odds are the wait is about 2 years for something comparable in power, and likely worse for any intel based virtualization.

Dos anyone know if any large production houses are buying large numbers of 7,1 mac pros ? If music producers and you tube editors are the bulk of the market then Apple will not be making a profit from the current mac pro. Its really just a face saving exercise unless there is an ARM upgrade path for the platform - the mac pro I mean.

 

[Boil]

Dos anyone know if any large production houses are buying large numbers of 7,1 mac pros ? If music producers and you tube editors are the bulk of the market then Apple will not be making a profit from the current mac pro. Its really just a face saving exercise unless there is an ARM upgrade path for the platform - the mac pro I mean.

I want to say Disney / Pixar ordered a whole bunch...?

 

[macguru9999]

I want to say Disney / Pixar ordered a whole bunch...?

yeah ?? you think ?? I would like to know the reality ...

 

[Boil]

yeah ?? you think ?? I would like to know the reality ...

To me, you are coming off as a bit angry about my statement?

I am pretty sure I saw something somewhere at some time about Disney buying a bunch of 2019 Mac Pros.

But I do not know for sure, and Google does not help much.

 

[macguru9999]

To me, you are coming off as a bit angry about my statement?

I am pretty sure I saw something somewhere at some time about Disney buying a bunch of 2019 Mac Pros.

But I do not know for sure, and Google does not help much.

no, not at all i hope its true i thought you were just guessing

 

[Successful Sorcerer]

Even if ARM Macs have impressive synthetic benchmarks, it won't get us very far if app developers aren't interested enough to devote sufficient coding resources. I could easily see a future where Apple's homegrown apps are blazing fast on new silicon, but not much else. For starters, what about Adobe? Will they cobble together a port just good enough to keep their Mac audience feeding them subscription income, or will they dig deep?

I think one of the reasons Adobe creates a complete version of Photoshop for iPad is preparation for ARM macs. It's almost there, the current Photoshop for iPad is already impressive and I can imagine they're also working on the other apps. I use their other iPad apps like Fresco, Lightroom, XD and Aero a lot as well and they're really good. I think they already have a lot of experience with ARM at the moment.

It's too early to say but the new Rosetta also looks promising with efficient emulation. I can imagine a lot of apps will indeed run slower in the early stages of the transition and in the end everything must be native ARM after a few years. Promising but with a few bumps in the road.

 

[BaltimoreMediaBlog]

My PowerMac G5 and my Mac Pro 4,1 are both waving their fingers at me and screaming,

"NO, DON'T GET HOSED AGAIN BY THE REALITY DISTORTION FIELD!"

I'm very skeptical about the whole Arm thing. Just try and use the internet on a PowerPC machine today.
The technology is still functioning and capable, but the inability to even use the internet for everyday things is disturbing.

I'm not an iMac or laptop person either, so I will not be buying a Mac Pro even if it wasn't so outrageously overpriced.

I predict that corporations will be dropping Apple support in droves, even hurting iPhone sales as they like stability, not whatever toy Timmy is playing with next. Steve Jobs was able to pull risky things off, but I can't see bean counter Tim Cook pulling this one off.

 

[leman]

I'm very skeptical about the whole Arm thing. Just try and use the internet on a PowerPC machine today.

Why are you still using a PowerPC? I mean, I understand that people have different hobbies and all that, but deliberately using a machine that has been obsoleted over a decade while expecting good experience ago strikes me as a bit... odd.

 

[BaltimoreMediaBlog]

Why are you still using a PowerPC? I mean, I understand that people have different hobbies and all that, but deliberately using a machine that has been obsoleted over a decade while expecting good experience ago strikes me as a bit... odd.

Well I maintain a PowerPC PowerMac for my elderly parents in their 80s. They do not want to learn new software, just want some level of internet access without losing the apps they've used for 20 years, mostly card playing games. Once you're in your 80s, you do not want to re-learn anything, if that makes any sense. So, I try to keep it working with the help of my trusty G5 as a backup. Sure, they could buy a new machine for a cheap amount of money, but they'd be in a miserable hell trying to figure it out or do what they can do every day now.

I still keep my Mac Pro 4,1 current, being METAL capable and 40GB of memory, several TBs of space. I'm not in my 80s, in fact over 30+ years younger, but I can't imagine myself jumping into Arm very fast without a lot of incentive and less hurt. The hurt in both cases is simple, replacing all of your current apps. With Catalina in such a mess from what I read, I think Mohave will be my last stop with that machine. At least I know Intel apps will not disappear as quickly, at least for very basic things like the internet even if Apple abandons Safari for Intel. There's always going to be people who can code for Intel, not like the PowerPC abandonment.

 

[leman]

Well I maintain a PowerPC PowerMac for my elderly parents in their 80s. They do not want to learn new software, just want some level of internet access without losing the apps they've used for 20 years, mostly card playing games. Once you're in your 80s, you do not want to re-learn anything, if that makes any sense. So, I try to keep it working with the help of my trusty G5 as a backup. Sure, they could buy a new machine for a cheap amount of money, but they'd be in a miserable hell trying to figure it out or do what they can do every day now.

I understand. Yeah, it's a tricky situation. There is such delicate balance between comfort and progress. Apple's way ways always "progressive" (in the sense that they break stuff constantly to bring something new). This approach aligns with how I do things

 

[mode11]

You can surf the web OK on a contemporary Pentium 4 or Pentium M, because they're supported by Windows 7 and so have pretty modern browser support. Whereas G4/5 machines are in a wasteland of obsolescence, so rely on ancient browsers or hacks of newer ones put together my determined individuals. It doesn't help that PPC chips also fell behind in performance compared to Intel.

I would have thought an Intel Mac mini running Snow Leopard would be a better compromise for your parents. The UI is basically the same as Leopard, and it would run their old PPC games with Rosetta. Then they could run a modern-ish x86 browser, which would be vastly more standards compliant, faster and more secure.

Having said all that, the G4 build of Camino runs OK in Tiger on my G4 Pismo. Also, Leopard WebKit runs quite well on my G4 iMac (haven't turned it on in a while though). Some PPC people recommend TenFourFox, but it ran like a pig for me - perhaps it's OK with a G5 and a pile of tweaks.

 

[Unregistered 4U]

They don't necessary have to match the top of the line Xeons or EPYCs in terms of core count, since they now have the luxury of integrating specialized cores for specific applications as they see fit.

This is, and always has been, the key. All these new Mac Pro’s have to do is perform better than the current Mac Pro’s running FCPX and Logic. And, the ones after that just have to do better than the previous Apple Silicon machines.

As a side benefit, if it also happens to run the applications listed on their Mac Pro page a little better, then, again, if you enjoy macOS, you’re getting the most performant macOS experience.

 

[Blair Paulsen]

How much of a performance bump would be needed to justify an upgrade? In some use cases, even a 10% gain would be worth many thousands of dollars - but certainly not all.

For me, and users with similar needs/budget, dropping $10K+ on an ARM MP only makes sense if the performance gains are significant enough that the time saved is worth more than the capital outlay. Will that be true with the first iteration of an ARM MP?

 

[jjhny]

How much of a performance bump would be needed to justify an upgrade? In some use cases, even a 10% gain would be worth many thousands of dollars - but certainly not all.

For me, and users with similar needs/budget, dropping $10K+ on an ARM MP only makes sense if the performance gains are significant enough that the time saved is worth more than the capital outlay. Will that be true with the first iteration of an ARM MP?

All of that is true. But if some software isn't ported over and we lose functionality, then speed makes no difference. If you are only running Apple friendly apps, it makes sense. If you are running some of the other (quite amazing) apps, that developers will just not waste time porting, then you are left worrying about the future. Do we even know what the landscape will look like post-intel. Apple's pleb consumer vision is great for the stock price (lots of consumers of content) but likely bad for those trying to make complicated, high end content. We may be forced out, which I am not happy about.

 

[Lycestra]

All of that is true. But if some software isn't ported over and we lose functionality, then speed makes no difference. If you are only running Apple friendly apps, it makes sense. If you are running some of the other (quite amazing) apps, that developers will just not waste time porting, then you are left worrying about the future. Do we even know what the landscape will look like post-intel. Apple's pleb consumer vision is great for the stock price (lots of consumers of content) but likely bad for those trying to make complicated, high end content. We may be forced out, which I am not happy about.

With Rosetta2, this is less of an issue. The things that aren't support with Rosetta2 are kernel drivers mostly. Anything else that runs in Catalina should run on an ARM machine. One other gap worth mentioning is that the larger rift has already occurred: 32bit apps. Catalina only supports 64 bit apps, and moving forward ARM will be 64bit only. Rosetta2 will also probably only support 64bit apps. So, to reiterate, a majority of apps that run on Catalina should run on ARM without porting thanks to Rosetta2 (expect some slowdown), and the number of incompatible legacy apps probably pale in comparison to the 32bit apps that were abandoned by Catalina. So many Steam apps, for example.

 

[Unregistered 4U]

If you are running some of the other (quite amazing) apps

What amazing apps have made the 64 bit transition where there’s indication that the developer is not supporting Big Sur? The more I think about it, the less I can think of examples of actual apps.

 

[zephonic]

I would have thought an Intel Mac mini running Snow Leopard would be a better compromise for your parents. The UI is basically the same as Leopard, and it would run their old PPC games with Rosetta. Then they could run a modern-ish x86 browser, which would be vastly more standards compliant, faster and more secure.

My '07 Intel iMac running Snow Leopard is good for nothing. I can run an older version of Firefox on it, but nothing that will work with modern security certificates and web standards. I can't even use it as a player for Apple Music. Something messed up Bootcamp on it, otherwise I would have been happy to just use it as a Windows computer.

The hardware still works fine, slow but quiet. It still looks like new. But Apple deems it not worthy of a functional OS and software.

I also have an ASUS laptop from around the same time that huffs and puffs, gets ridiculously hot under moderate loads, but modern browsers work on it, and it still gets used for the occasional game of C&C3.