Apple's Workforce Growth in Oregon Could Point to Desktop-Class CPU Ambitions

[bluecoast]

I remember when the PPC transition was rumoured and there was much angst amongst the Apple faithful.

PPC Macs couldn’t even run Windows at the time (one of the big worries this time round is not being able to run Windows).

In fact, much of the angst was that Intel were seen as part of the enemy (Wintel).

However, very quickly the Apple faithful were ok with it all as the machines were fast and Rosetta worked really well.

I think that that will be the case this time too. Especially when laptops with double (maybe even triple) the battery life are unveiled.

This is going to be great.

 

[Yvan256]

Keep dreaming. Much more likely that lower component costs will mean fatter margins on each unit sold. New Apple strategic thrusts are usually not revolving around lowering prices for consumers.

Even though Tim says otherwise, with the higher prices of the latest Macs combined with the currency exchange rates world-wide, they are now a luxury reserved for the rich in a lot of countries.

 

[MacLawyer]

Personally I wouldn't touch a Mac that couldn't do x86 virtualisation; but it could be very interesting for low-end, long battery life Macs.

I hope it's not a sign of Apple going the Nintendo route. ("Ignore performance, if we go our own route, the user won't care about performance").

Yes. I would be forced to abandon the Mac if I could not run Windows speedily either in virtualization or in Bootcamp. I shudder when I recall SoftPC.

 

[whooleytoo]

What I’m curious about is where does iPad fit into all of this?

I love my iPad, but for consuming... (casual browsing, music-listening, news-reading, gaming), not for work.

In order for an iPad to be a work replacement, it'd need to have a bigger screen, physical keyboard, keyboard shortcuts, a better mechanism for text selection, better multi-tasking support etc. You'd basically need to make it into an ARM based laptop.

And even then it'd still suffer from no virtualisation.

I'm interested to see what direction Apple are going with this, but I doubt it'll be for me personally.

 

[bluecoast]

Yes. I would be forced to abandon the Mac if I could not run Windows speedily either in virtualization or in Bootcamp. I shudder when I recall SoftPC.

The thing is, is that Windows is moving to ARM to - for the battery life.

In fact it has moved though I think that it only supports 32 but apps right now. Almost certainly they’ll remedy that and move to support 64 bit apps.

So by the time these new Mac ARM machines come out, this shouldn’t be an issue.

 

[DNichter]

What I’m curious about is where does iPad fit into all of this?

iPad's with some serious power potentially. Maybe the ability to run macOS apps as well. Could be huge.

 

[CWallace]

Considering Hillsboro is a design and validation center for Intel CPUs, I was wondering if the announcement might herald a closer collaboration between the two companies.

Apple has also evidently head-hunted a number of Intel CPU engineers from Hillsboro so this might be in support of the rumored x86 to ARM transition or perhaps a hybrid CPU that could run both x86 and ARM code.

 

[ipponrg]

iPad's with some serious power potentially. Maybe the ability to run macOS apps as well. Could be huge.

I question your definition of "serious power" here.

 

[DNichter]

I question your definition of "serious power" here.

That's fine, you come on here just to hate anything positive about Apple. I wouldn't expect you to agree.

 

[macduke]

The resident tech writer has spoken. Will people in the forums finally start believing now that Apple could rival Intel’s desktop chips? It’s not a matter of if they can, but when they will.

 

[deconstruct60]

...
Digging deeper into these job positions reveals keywords indicating performance validation in non-iOS workloads, as well as a heavy focus on memory concepts such as memory controllers, memory hierarchy, and cache coherency protocols. The focus on the memory subsystem is significant because this is one area where mobile device and PC form factor usage models differ based on their power consumption profiles, along with PCs featuring tools that can stress a memory system in ways not typically seen in mobile device workloads.

The above has nothing to with anything that is "desktop" processor specific at all. Cache Coherency protocols are important when you have multiple computational cores all working in parallel. The current A-series SoC have that in spades already. As the core count increases so does the importance of doing design validation increase in this space. As long as the A-Series iOS keeps upping cores (not just ARM but GPU and "AI" cores that happen to drawn on the same pool of memory) cache coherency complexity/verification is going to go up. Absolutely nothing specifically to do with desktop at all.

As the iPad Pro goes to a dock and running multiple applications at the same time. I could see a split between iPad Pro and iPhone processors. It is already there now in a relatively mild form. Apple could fork those a bit more. But that would have nothing necessarily to do with macOS. [ e.g. Apple brings back the iBook as an iOS based system. Something to meet the Chromebook onslaught with more directly. ] iPad expands out its functionality , etc. etc. ( to lessor extent iPhones go to 8-10 GB of RAM and there is more cached/paging activity on phones (instead of pushing stuff to NAND Flash most of time and single app focus. ), for example larger background "AI" that working in parallel with apps all the time. ]. Hence have more memory hierarchy issues.

System boards in iPads could expand into non-stacked on SoC RAM chips that sat on independent memory busses ( so could get to wider memory ). [ more integration of the board components frees up space, without necessarily eating into battery space by expanding overall logic board size. ]

PC systems also tend to feature much higher memory bandwidths, due to both wider memory busses and higher memory clocks. Apple is often rumored to be working on an ARM-based MacBook, but such a device would likely feature some variant of LPDDR memory seen in Apple's mobile devices, as well as Apple's entire line of MacBooks.

The iOS devices already have LPDDR memory. For the last 5-6 years Apple has boasted the memory bandwidth and caching complexity of the A-series chips. Why would they stop now ?

This means that Apple already has the necessary memory controller designs to interface with LPDDR memory. Apple is also no stranger to the 128-bit memory busses seen in MacBook Pros with LPDDR3 RAM, along with previous iPad iterations featuring a 128-bit memory bus.

Using a memory controller that Intel gave you in the package you bought is significantly different than implementing one. That is not familiarity. Practical familiarity would be having done it.

Perhaps the most demanding component of a memory subsystem in custom SoC destined for macOS devices is the GPU. On the iOS side, Apple recently shifted from licensing Imagination Technologies' family of PowerVR GPUs to designing its own custom GPUs. Additionally, Apple's Orlando-based GPU design center has gone through several hiring spurts, and is also hiring now, including a position for a platform architect aimed at creating a family of GPUs across multiple Apple platforms.

Right! Orlando went through hiring spurts and it all got plowed into the A-series solutions for iOS devices.
So why is another hiring spurt all about desktops and non-iOS ????????????????? There is no rational connection there.
Orlando is not likely suddenly building a laptop/desktop GPU. As the transistor budget get bigger then incrementally build bigger, more complex GPUs but that doesn't take them out of a primary focus on iOS space.


Great click bait fodder to fan the "macOS on ARM" flames but very little substantive there.


The run rate of the Macs ( which use 3 different baseline die designs ) is substantially much lower than the iPhone and has none of the uniformity of the iPad ( which either borrows an used iPhone processor or has one and only one model (that is slight derivation of iPhone SoC.). The notion that Apple is going off to do a giant fork of the iPhone SoC is completely lacking in how Apple is going to actually pay for that. (raiding the money pit is not a legitimate answer. ). Folks don't build large, custom processors just to run at a 3-4 million (or fewer ) rate for relatively modest prices points.

 

[ignatius345]

Could have been worse? I was extremely proud of Apple for how well and seamlessly they did the transition for the end user.

Yeah, that's what "could have been worse" means. It means that it was better than it could have been. You might be thinking of the phrase "couldn't have been worse"?

Totally agreed, though. I was around for that PPC to Intel transition and using a Mac for my design job, and I don't remember ever being locked out of anything I needed to do. I would hope for a similar experience with this rumored transition, but let's also remember that at the time Apple was moving toward a mainstream chip architecture that everyone else was already using. I just hope this wouldn't be us all moving into some iOS-style walled garden.

 

[ipponrg]

That's fine, you come on here just to hate anything positive about Apple. I wouldn't expect you to agree.

No, I come on here with rationale and without rosy glasses unlike yourself, which is why I am asking you what do you mean by "serious power"?

To me, serious power implies I can use terminal and do things that isn't just a simple click/run behavior. It means I can multi-task and able to see more than 1 window at a time. It means I can simply substitute the iPad for my MBP and not need to still launch the MBP to finish.

You seem to be bent on that anyone who **questions** you or your opinions implies that they hate Apple.

In another thread, a poster seemed to be convinced that the translation layer for apps run natively instead of emulated. I'm skeptical, but if Apple were able to accomplish that, then it'd be a mighty feat. How's that?

 

[ignatius345]

If there is no Rosetta 2, there will be migration pain that could last years for those that lean on Mac software NOT created in house at Apple. Adobe? Microsoft? Etc.

My fears exactly. And all this is even assuming an Apple computer with its A-series processor would even allow non-App Store apps to be used at all. Pre-iOS Apple had to keep the Mac a viable platform through that Intel transition because their only other product was the iPod. Now? We have all seen over the past 5 years just what kind of priority the Mac is for Apple.

Is it so hard to imagine they won't just flip this switch at some point?

I personally don't look forward to the day I have to jailbreak my Mac.

 

[MichaelDroste]

I believe these will initially be for the MacBook air - transitioned to ARM

(Windows people will be able to run windows on arm also)

What is interesting - Can it scale to 10 cores 20 cores etc? Then the whole line would switch!
(That is years off - first touch screen arm computers on the low end)

 

[AxiomaticRubric]

Perhaps Intel will begin to regret releasing such lacklustre CPU updates, along with their ridiculous 16GB restriction for LPDDR3E RAM.

It’s not their fault. The Laws of Physics caught up with them. Apple will run into similar issues no matter how much money they have to throw at this.

 

[DNichter]

No, I come on here with rationale and without rosy glasses unlike yourself, which is why I am asking you what do you mean by "serious power"?

To me, serious power implies I can use terminal and do things that isn't just a simple click/run behavior. It means I can multi-task and able to see more than 1 window at a time. It means I can simply substitute the iPad for my MBP and not need to still launch the MBP to finish.

You seem to be bent on that anyone who **questions** you or your opinions implies that they hate Apple.

In another thread, a poster seemed to be convinced that the translation layer for apps run natively instead of emulated. I'm skeptical, but if Apple were able to accomplish that, then it'd be a mighty feat. How's that?

No rosy glasses at all. This is positive news for Apple. Just because you need a terminal or the like to handle your operational tasks, doesn't mean we all do. Serious power would be more power than the current best mobile processor on the market, with the potential to run both iOS and macOS apps once they are unified. I have no issue with anyone that questions my opinion, I am all for open discussion. I have never seen that from you though, it's just always negative.

 

[HobeSoundDarryl]

It’s not their fault. The Laws of Physics caught up with them. Apple will run into similar issues no matter how much money they have to throw at this.

Correct, it's not like A-Series chips can keep shrinking the architecture much further either. The limit is a HARD limit. Apple doesn't get some physics limits pass because it's Apple.

 

[0958400]

That's fine, you come on here just to hate anything positive about Apple. I wouldn't expect you to agree.

So if someone doesn't agree with you, it's hate? Seriously?

 

[lederermc]

Personally I wouldn't touch a Mac that couldn't do x86 virtualisation; but it could be very interesting for low-end, long battery life Macs.

I hope it's not a sign of Apple going the Nintendo route. ("Ignore performance, if we go our own route, the user won't care about performance").

Power users and mom/pop email/web surfers have different needs and requirements. As long as Intel Macs and ARM Macs give the same results does it really matter? I suspect "no" for the vast majority of users. Let's remember the reason for switching from PowerPC to Intel was the lack of the PowerPC's performance and lack of low energy options for laptops.

 

[DNichter]

So if someone doesn't agree with you, it's hate? Seriously?

Nope, not at all. This is specific to that person, not necessarily their underlying argument. Disagreeing and having a counter argument and discussion is one thing, acting like a know it all is another.

 

[FocusAndEarnIt]

Yeah, that's what "could have been worse" means. It means that it was better than it could have been. You might be thinking of the phrase "couldn't have been worse"?

Totally agreed, though. I was around for that PPC to Intel transition and using a Mac for my design job, and I don't remember ever being locked out of anything I needed to do. I would hope for a similar experience with this rumored transition, but let's also remember that at the time Apple was moving toward a mainstream chip architecture that everyone else was already using. I just hope this wouldn't be us all moving into some iOS-style walled garden.

Could have been worse implies a negative tone, suggesting it was rocky. Sure, anything great could have been worse. Guess it's one of those half full / half empty type of things.

 

[MacsRuleOthersDrool]

I know it's quite a hassle to switch CPU architecture, but this seems really exciting to me. Kinda feels like the old Apple.

It isn't that much for Apple's OSes and Dev. Tools. They are specifically designed to be as processor-agnostic as possible. Plus, Apple is the hands-down master at seamless CPU architecture switches.

They've already done it twice for Macs (68k -> PPC -> Intel), and then ported over to ARM for iOS/TVOS/WatchOS.

Apple gets very little credit for their expertise in this area; but it is formidable.

If anyone can do it successfully, it's Apple. And don't forget that they have more ARM experience than almost anyone on the planet, being involved in ARM Architecture Development since pretty much the beginning (e.g. the Newton).

But I think all the Haters and Hand-Wringers can stop worrying about Apple dropping the Mac now. They are just "regrouping", and getting ready to free themselves from the slow-ass and deliberately-entrapping bonds of Intel!!!

Yes, exciting times are indeed ahead; but I wonder if they are going to withhold the new Mac Pro until they get their own CPUs working? Probably not. In fact, they will likely start out with the low-end stuff first, like a non-Pro MacBook/Air, and possibly a "crossover" iPad Pro. Then the iMac and Mac mini. Then the iMac Pro and Mac Pro.

You heard it here first!

 

[bluecoast]

I really don’t understand all the people coming on here who are so wedded to Intel.

Even Microsoft are starting to make Windows work with Qualcomm ARM chips. You can buy ARM powered Windows laptops right now.

Doesn’t that tell you something?

I don’t get the ‘raw power’ argument, either. What are the people on here doing that mean that you need all that power? This is a serious question.

Can you not just get a Linux box if much of what you do is to do with the terminal?

Why are you even using a Mac? I’m genuinely curious.

Apple have always been about owning the primary technologies that they make their machines with. It’s what they do.

They create bespoke hardware and software that are tuned to work with each other so they can achieve exactly the products that they want to build.

They hate using generic parts and will only do so if it makes sense to them (for a variety of reasons).

I mean this week the company that makes power controller chips for their iOS devices has signalled that they are losing orders because Apple are making their own.

Last year they started making their own GPUs for the iPhone 8 and X.

Slowly and patiently they are bringing all of their major components in-house.

This is just them reverting to type.

Think of what they can do if they are ploughing billions into silicon for iOS already and then can now leverage that for their Macs too.

I can’t see how this can be anything other than great news.

Finally, we’ve had Intel Macs for way less time than the Mac has had Motorola & PPC chips.

Something to think about.

 

[Val-kyrie]

Keep the faith, just a couple more years and that new Mac mini will be rolling off the line.

In this matter, your faith is bigger than that of a mustard seed; mine isn't.

The magic there was from a third party- who provided Rosetta. Rosetta made the transition much less painful because it could fake the non-Apple software until maybe the developers of that software evolved a version for the Intel platform. Is there a Rosetta 2 out there to facilitate this kind of transition- something to run X86 software well on ARM hardware? I'm not aware of something like that this time. That doesn't mean it does not exist but I wish there was some Transitive-like entity out there that ALREADY has such a capability working.

If there is no Rosetta 2, there will be migration pain that could last years for those that lean on Mac software NOT created in house at Apple. Adobe? Microsoft? Etc. We should not delude ourselves into believing that such entities can simply recompile and their software will run fine on A-Series Macs. Even with the bridge of Rosetta, it took those kinds of companies a couple of YEARS to transition to Intel Macs. We should not fool ourselves into believing it's basically some kind of flip-a-few-switches proposition. There will be pain... especially if there is no Rosetta 2.

And it probably means one well-designed computer able to double as both a Mac and a PC probably forks back into needing 2 computers for that. Personally, that would be VERY unfortunate for work purposes, leading to probably having to buy a Windows machine since the bulk of the working world runs on Windows. For travel now, I can take one laptop to cover all bases (this is somewhat secretly one of- if not- the VERY BEST features of a working Mac computer). Working Mac travel in this "the future" probably means having to choose one laptop rather than lug along both. Take the Mac and likely be pretty incompatible with just about any business client's IT or take a Windows computer and likely better fit in?

But let's keep imagining that A-Series Macs will significantly up the CPU power while significantly lowering retail prices and stick it to Intel at the same time. And then lets further hope that the Apple that neglects Macs for upwards of years at a time will find new motivation to regularly upgrade them when upgrading adds the tasks of developing new CPUs too instead of just using CPUs created by an expert vendor heavily focused on exactly that. And let's imagine that all the non-Apple software that we may lean on regularly can be upgraded to "just work" in only a day or three, perhaps with simple recompiles or similar so that we don't find ourselves waiting for months or years to finally get a version- if the devs even opt to bother for such a smallish niche- that works with a proprietary platform.

I remember when the PPC transition was rumoured and there was much angst amongst the Apple faithful.

PPC Macs couldn’t even run Windows at the time (one of the big worries this time round is not being able to run Windows).

In fact, much if the angst was that Intel were seen as part of the enemy (Wintel).

However, very quickly the Apple faithful were ok with it all as the machines were fast and Rosetta worked really well.

I think that that will be the case this time too. Especially when laptops with double (maybe even triple) the battery life are unveiled.

This is going to be great.

As others have pointed out elsewhere, there is no software equivalent to Rosetta for this transition. MS is trying to move to ARM but has encountered a number of problems and limitations, including:

• x64 apps not supported: This doesn’t come as a surprise. Windows 10 on ARM doesn’t support emulation of x64 apps. However, these might be supported in the future.
• Only ARM64 drivers supported: Even though Windows 10 on ARM can run x86 apps, it can’t use x86 drivers. It wouldn’t be a problem unless you have older devices. In that case, driver support will not be available. Driver support for Windows 10 on ARM will be more limited, kind of like what Windows 10 S offers.
• Apps which modify the Windows experience will probably not work correctly: Apps like input method editors or assistive technologies will not work properly on Windows 10 on ARM. Apps which come with shell extensions such as Dropbox may not work correctly either. They will have to be natively compiled for ARM.
• Some games and apps don’t work: Apps and games which use a version of OpenGL later than 1.1 or which need hardware accelerated OpenGL will not work on Windows 10 on ARM. Games which use anticheat mechanisms will not work either.
• Apps which presume that every ARM-based device suns a mobile version of Windows may not work as intended. Some apps which were made for Windows Phone will not work correctly and might appear in the wrong orientation or have UI layout issues. However, there wouldn’t be many apps like these.
• The Windows Hypervisor Platform is not supported. You will not be able to run virtual machines using Hyper-V.

I would expect the same equivalent issues to hold true for macOS, as well.

The thing is, is that Windows is moving to ARM to - for the battery life.

In fact it has moved though I think that it only supports 32 but apps right now. Almost certainly they’ll remedy that and move to support 64 bit apps.

So by the time these new Mac ARM machines come out, this shouldn’t be an issue.

You are very optimistic.

That's fine, you come on here just to hate anything positive about Apple. I wouldn't expect you to agree.

With all due respect, you are overstating the other poster's position using the word "hate." Your posts can tend to grate on others--they do on me. Most of the posts which you write endorse the iOS ecosystem and proclaim the wonders of iOS for your workflow. That is great, for you. Please understand that there are many of us who, unlike you, are unable to accomplish our work on an iOS device and consequently require x86-64 hardware and software for our livelihood. This does not mean that we "hate" Apple when we debate a transition in CPU architecture; rather, it means that there is a bit of angst as Apple appears to be abandoning many of its users who have been with the company through its transitions and rely on a certain workflow to accomplish tasks and make a living. Simply telling others to change or leave and mis-characterizing their posts is neither an answer nor decent debate / conversation.

Yeah, that's what "could have been worse" means. It means that it was better than it could have been. You might be thinking of the phrase "couldn't have been worse"?

Totally agreed, though. I was around for that PPC to Intel transition and using a Mac for my design job, and I don't remember ever being locked out of anything I needed to do. I would hope for a similar experience with this rumored transition, but let's also remember that at the time Apple was moving toward a mainstream chip architecture that everyone else was already using. I just hope this wouldn't be us all moving into some iOS-style walled garden.

My fears exactly. And all this is even assuming an Apple computer with its A-series processor would even allow third-party apps to be used at all. Pre-iOS Apple had to keep the Mac a viable platform through that Intel transition. Now? We have all seen over the past 5 years just what kind of priority the Mac is for Apple.

Is it so hard to imagine they won't just flip this switch at some point? I personally don't look forward to the day I have to jailbreak my Mac.

We should all be bracing for this possibility. MS received push-back when they announced Windows on ARM just because of this very issue--MS intended / announced that only apps approved and purchased or downloaded through the MS Store were to be installed on ARM computers. I don't know if MS backed down on this, but I would expect in large part that the push from both MS and Apple toward ARM will lead to a walled garden for app developers and end users--all in the name of security.