ARM based MacBook Pro discussion [merged]

[Omega Mac]

Note: Please merge if there is an ARM thread that is more fitting for my post but I couldn't find one after multiple searches and/or place it in the more relevant forum. Thanks.

Here are the two important reasons why ARM will make it to the various other MAC platforms IMHO.

1) Security =- Full spectrum and all inHOUSE
2) Mobile Multi processor for the PRO's

Point 1) Security.

If Apple want to up continually up the stakes in providing the best security. They are going to have to control as many of the chips in their devices across all their hardware platforms. The more Apple cross pollinate technology such as Touch ID across devices the greater the demand for full spectrum security. No longer will they have to rely on Chips that might have a backdoor left ajar.

Point 2) Multi CPU on board.

Since ARM is developing fast and very powerfully, but also cool and in tight spots.

Is it possible to imagine that to compensate on the power but in larger devices where space is ample compared to the slip of a wafer the iDevices have become, that they may introduce multiple ARM chips in one machine, for e.e.g 13" MBP, Dual A10X ARM.... 15" Quad A11X ARM CPU's and so on.

I've got to give great credit to poster who writes blogs about ARM as a chip designer I found two days ago. Whoa! Love it. I'll find the link and replace this paragraph when I get it asap. Though I don't fully understand everything you write I find it deeply interesting. Thank you and you know who you are!

 

[Toutou]

The thing about multiple CPU cores and multiple separate CPUs (each with multiple cores) is that you can't simply chop any task into, say, 32 pieces and expect it to run. There are tasks that simply can't be parallelized, no matter how hard you try. Sometimes you just need the lighting fast desktop-class i7 behemoth to crunch your numbers. ARM chips aren't quite there yet. (yet!)

 

[RoboWarriorSr]

The thing about multiple CPU cores and multiple separate CPUs (each with multiple cores) is that you can't simply chop any task into, say, 32 pieces and expect it to run. There are tasks that simply can't be parallelized, no matter how hard you try. Sometimes you just need the lighting fast desktop-class i7 behemoth to crunch your numbers. ARM chips aren't quite there yet. (yet!)

Honestly I don't think ARM will ever reach to that point, x86-64 and ARM were developed on opposite ends of the spectrum, one for performance/jack of all trades and other for endurance/low power respectively. Whether ARM can replace x86-64 for "normal" task is already happening as we see in the enormous number of devices running them successfully. But to expect ARM to crunch numbers that cannot parallelized to me doesn't seem possible due to their fundamental ideals which each were developed for. With AMD and Intel allocating part of their die to GPU (and how impressive they are getting), parallelized tasks can be easily programmed to be pass on there.

 

[maflynn]

There's been various chatter about this in the forums here at MacRumors.

Here's some of my issues with going to ARM.
1. Lack of windows compatibility. I need and use windows for my job and I'll not have the virtualization ability, never mind Bootcamp. Plus, Apple enjoyed a large spike in people buying Macs when they transitioned over to Intel. They risk alienating a large core of customers if they leave the X86 platform.

2. Performance, will the ARM perform just as well as intel running the multithreaded applications? True the iPad Pro's performance is exceptional, but its running iOS with one, maybe two apps (with split screen). How will it perform with a desktop OS and file system, with many apps running at once?

3. Developers, Apple will be asking a lot of developers to rework their apps and will we only be able to load apps from the app store at that point? Will the OS be locked down like iOS and we'll lose the ability to load Carbon Copy Cloner, or other applications that forego the MAS and apple's 30 percent slice of sales? Will large developers get on board, i.e., Adobe, Microsoft, etc?

4. Will Apple emulate the legacy code, and how will that perform, i.e., when they transitioned from PPC to Intel we still had the ability to run PPC apps, how will the transition work, can the ARM processor handle the emulation. If there's no emulation, that will make it almost impossible for apple to sell an ARM based Mac.

Personally, I find it difficult to justify spending 2k on a 15" MBP that's running a broadwell chip in 2016. I'm definitely sure, I'll not spend 2,000 dollars on an ARM based laptop. While my single sale may not mean much to apple, I'll be one of the people who chooses to walk away from the Apple computers if they move away from the intel chipset. The competition has already leaped frogged Apple in design, performance and price. Apple needs to not only catch up but surpass them, but I think embracing the ARM platform is the opposite - just my $.02

 

[whg]

...
Personally, I find it difficult to justify spending 2k on a 15" MBP that's running a broadwell chip in 2016.

And even more so with a Haswell chip! I believe only the 13" is on Broadwell and the 15" is still on Haswell.
While the CPU performance is still about the same, iGPU and power consumption improved on the newer chips. What holds me back most is the dGPU in the 15".

 

[crsh1976]

I don't think moving over to ARM is a solution, or even a good idea for that matter. So many drawbacks, so many hurdles to go through, and killing Windows compatibility is a huge step backward.

I don't think things are going to move much for the near future anyway - let's keep in mind Apple doesn't care much for Macs these days and has only put even more focus on lifestyle products and services. No point in revolutionizing Macs right now (when did it become such a hassle for Apple to use Intel chips anyway?).

 

[jerryk]

I don't think moving over to ARM is a solution, or even a good idea for that matter. So many drawbacks, so many hurdles to go through, and killing Windows compatibility is a huge step backward.

I don't think things are going to move much for the near future anyway - let's keep in mind Apple doesn't care much for Macs these days and has only put even more focus on lifestyle products and services. No point in revolutionizing Macs right now (when did it become such a hassle for Apple to use Intel chips anyway?).

Agreed. The ARM architectures does not have a good history in PC level devices. All of the emphasis on power consumption, small footprint, software cycle vs cpu cycles (ie. RISC vs CISC) is better suited to smaller packaged and potentially memory constrained systems. Interestingly Intel modern intel processors are not longer strictly CISC. They uses a RISC code for pipelining and wraps it in operators that appear as CISC instruction to the outside world.

 

[Omega Mac]

Hat tip to - Applenews.zone - http://www.applenews.zone/2015/06/feature-why-macs-will-get-armd.html

 

[maflynn]

let's keep in mind Apple doesn't care much for Mac

Macs are outselling iPads and if they don't care about macs, then that might mean more justification for a transition to ARM. I think Macs are still an important stream of revenue and profit for apple and messing up such a winning formula is just asking for trouble.

By the way, in apple's last 10k filing its reporting that Mac sales are over 25 billion dollars. I'd say any product that makes 25 billion is a product that any company would love to have.

 

[crsh1976]

Let me rephrase, Apple doesn't put much effort on Macs since sales are indeed pretty good and stable - no need to fix what's not broken, right? Switching to ARM would mean reworking products, enticing devs, enterprises, etc, and that's a creates a lot of uncertainty.

 

[poorcody]

So with reports of the A10-Fusion matching the performance of the current Mac Pro (with caveats, of course), one can't help wonder if a desktop version running at a higher clock speed might match Intel's current performance. Whether it can today or not, it seems inevitable that it eventually will, which makes me wonder, what would the implications to the Mac platform be if Apple swapped out Intel for AX-chips?

I could see these advantages:
▪︎ Cost -- since Apple is already investing in the processor for the iPhone, it seems they would save money by not buying Intel chips. (And of course Apple wouldn't pocket the profit, they would pass it to us consumers.)

▪︎ Predictability -- the ability of Apple to release a new processor every year with the new iPhone on a predictable schedule I think helps the platform. Syncing Mac's update schedule with Intel's has not worked so well recently!

▪︎ Tuning -- the ability to tune the processor to the operating system would at least in theory lead to better performance or battery-life. Intel's chips have to satisfy a wide-variety of applications which must have tradeoffs for silicon space, heat, etc. Apple could focus on only what they needed.

▪︎ Security -- having a propriety chip would probably only help increase Mac security

Downsides:
▪︎ Compatibility -- they would probably need to emulate an Intel chip for a time to make software compatible. Although this seemed to go fairly well when Apple did it before with Rosetta.

▪︎ Bootcamp/Virtualization -- would we still be able to run Windows?


Interested in what other's think...

 

[jerryk]

There already was a thread on this 2 weeks ago.

 

[Samuelsan2001]

It's looking highly unlikely to happen in the next few years so why worry about it.

 

[SuprUsrStan]

Downsides:
▪︎ Compatibility -- they would probably need to emulate an Intel chip for a time to make software compatible. Although this seemed to go fairly well when Apple did it before with Rosetta.

This pretty much. For anyone who remembers how much of a pain those IBM macs were to use sometimes knows that it is highly unlikely that apple will go back to that approach.

 

[TechZeke]

Really, Apple is only having the most problems with Intel's schedule because they are using specialized high-performance iGPU chips that 99% of other PCs don't use.

Plus, as I keep saying, if Apple wanted to release an ARM based the Mac the 12" MacBook would have been the machine to do it. They didn't, which means no ARM chips in the near future.

 

[Loge]

Downsides:
▪︎ Compatibility -- they would probably need to emulate an Intel chip for a time to make software compatible. Although this seemed to go fairly well when Apple did it before with Rosetta.

That was because the x86 chips were much higher performing than the PowerPC chips they replaced. There would be no such advantage here.

 

[chabig]

That was because the x86 chips were much higher performing than the PowerPC chips they replaced. There would be no such advantage here.

Yet.

 

[Steve121178]

So with reports of the A10-Fusion matching the performance of the current Mac Pro

Got any more jokes?

 

[poorcody]

Got any more jokes?

https://twitter.com/MattMariska/status/776495257311584261

 

[SteveJobzniak]

Nice, but there's more to the chips than just floating point and fixed point number crunching. Intel processors come with a TON (and I mean a TOOOOOON, *hundreds*) of hardware-accelerated common math functions.

There's a lot of issues: How many cores can Apple make their ARM use? How will it interface with PCI express and Thunderbolt and other Intel standards? Are they willing to ask us to throw out Windows BootCamp compatibility? Are they willing to risk their computers getting stuck in a dead end years down the road when ARM no longer goes any further and Intel surpasses them, like what happened with PowerPC? Are they willing to throw away all existing applications and force developers to recompile everything?

This is just not going to happen. At least not as long as the performance is no more than, at best, equal, for very specific tests. Let alone all the other stuff that Intel hardware-accelerates but ARM doesn't.

 

[poorcody]

That was because the x86 chips were much higher performing than the PowerPC chips they replaced. There would be no such advantage here.

But think about the cost advantages... Intel charges a premium as the number of cores increase in the chip... For Apple, the cost differential to produce a quad-core chip versus a dual-core chip would be minimal...

So Apple could put, as an example, quad-core processors in across their whole line -- something that would have been cost-prohibitive using Intel chips in most models. Heck, they could do six cores and dedicate two cores to Intel emulation.

 

[jerryk]

But think about the cost advantages... Intel charges a premium as the number of cores increase in the chip... For Apple, the cost differential to produce a quad-core chip versus a dual-core chip would be minimal...

So Apple could put, as an example, quad-core processors in across their whole line -- something that would have been cost-prohibitive using Intel chips in most models. Heck, they could do six cores and dedicate two cores to Intel emulation.

And why would any of that matter?

It is not the number of cores that matter, it what you do with them. Are the CISC, RISC, or hybrid? What is the length of the instruction pipeline, cache depth, prefetch, etc.? How about memory management function, ALU, FPU, etc. Page cache size. Then there is packaging. Are you Pad width limited? How much heat can the package dissipate, etc.

Lots of things go into selecting a chip.

 

[poorcody]

And why would any of that matter?

It is not the number of cores that matter, it what you do with them. Are the CISC, RISC, or hybrid? What is the length of the instruction pipeline, cache depth, prefetch, etc.? How about memory management function, ALU, FPU, etc. Page cache size. Then there is packaging. Are you Pad width limited? How much heat can the package dissipate, etc.

Lots of things go into selecting a chip.

Not sure of your point here... it matters because if Apple can get at least the same performance (speed/battery) and feature-parity for less money, they will likely do it. They are investing a lot of money into making the processor line for the iPhone -- if they can leverage that onto their Mac line, it will make economic sense. Buying Intel processors is a significant cost.

My question is would it be advantageous to us? My example of it possibly being yes, would be that we could, say, get a quad-core processor for less than the price of a dual-core Intel Mac.

I get the point that chips have lots of features, but many of them have been abstracted away by the compiler, ironically because the speed advantages of doing them in hardware were negated by the overall increase in performance (e.g. CISC to RISC and then the merging of the two). Intel has a lot of baggage to carry for compatibility... since Apple owns the whole stack for the Mac, they have a lot more freedom to optimize for their needs.

 

[fisha]

there is no doubt that controlling the stack makes for a very fast efficient system on the likes of iOS, but i don't see that translating across to a desktop OS yet. To me, iOS is just not there for desktop performance and the Ax is too far away from running MacOS.

No matter how they put across the iPad Pro abilities, compared to desktop/laptop tasks, its just not there for multiple windows etc etc, and I don't see it being there for a while yet. So the efficiencies of Ax chips and iOS won't transfer to desktop either for a while yet.

On the comments of intel chip cost ... is it really that much more expensive all in?

For example, how much would apple add to the cost of a desktop Ax chip as a result of development? and would that additional cost make the price similar to an intel chip. Would it be easier to then just buy an intel chip complete where intel has spent all the money on development?( i honestly don't know )

 

[Dwalls90]

there is no doubt that controlling the stack makes for a very fast efficient system on the likes of iOS, but i don't see that translating across to a desktop OS yet. To me, iOS is just not there for desktop performance and the Ax is too far away from running MacOS.

No matter how they put across the iPad Pro abilities, compared to desktop/laptop tasks, its just not there for multiple windows etc etc, and I don't see it being there for a while yet. So the efficiencies of Ax chips and iOS won't transfer to desktop either for a while yet.

On the comments of intel chip cost ... is it really that much more expensive all in?

For example, how much would apple add to the cost of a desktop Ax chip as a result of development? and would that additional cost make the price similar to an intel chip. Would it be easier to then just buy an intel chip complete where intel has spent all the money on development?( i honestly don't know )

Unsure on switching costs, but no way does the A10 come anywhere near the cost that Intel charges for their CPU's.

There is absolutely no doubt Apple is running Macs in their product labs with A chips in them. Will they ever make it to market? Perhaps, if Intel continues to remain complacent due to little to no real competition.