Most persons on this thread seem to be approaching the topic as though it is an all-or-nothing proposition - that Apple either stay entirely with Intel CPUs forever, or they have to switch all Macs over to an AX unit at one go. A few posters have said, as I have before, that Apple has the option of creating a new "Mac category" that would use ARM running OSX. This category could target education/enterprise/mobility/etc, offering benefits of an AX design:
- lower price points (while maintaining same or higher margins for Apple). A BOM cost reduction of $100 can make a significant impact.
- lighter/thinner/silent
- more stability (better integration of Apple HW and Apple SW - IMO iOS devices are significantly more stable than OSX)
- AX functions not on Intel like TouchID & secure element, additional DSP packages (HW acceleration of functions), all within the SOC package
- better battery life
The "con" of course is the incompatibility with legacy 3rd party software. That is why this new Mac category would be initially targeted at sectors where that could be mitigated more easily. A small number of partnerships (e.g. IBM, top Mac App Store developers, education partners) could provide initial set of ARM compiled applications, along with Apple's native apps. Many others would join in over time. Students and education have a limited set of app requirements (as do the majority of consumers really). Over time the application compatibility would reduce.
The traditional Mac line (Pro, iMac, Macbook Pros) would remain with the x86 compatibility for the majority of apps, and Windows capability. It would be a differentiator of the "pro" devices over the AX line.
The long-term benefits of Apple controlling the silicon development are too much to ignore. As I think Wizard69 put it, silicon now is like the logic board was. More functions continue to get integrated into SOCs, and GPUs do more of the heavy lifting. The future of innovation in HW is there.
Of course Apple may not take this approach, but I fully believe they could pull it off (providing a compelling differentiation between the two lines). It would provide a meaningful increase in overall Mac business (imagine impact of a $699 highly portable Mac).
That is rather old and unrelated to the discussions at hand. The fact of the matter is that these modern Intel cores do operate well above 3GHz approaching 4GHz all the while running on a far more performanc core. The only problem Intel has is that they speed step to those high frequencies and can only do so as long as the thermals remain in spec.
In any event you missed the idea that new technologies are on the horizon that could displace silicon and lead to higher speeds.
That is largely due to Intel focusing on other weak spots. It will be most interesting to see if SkyLake lives up to the rumors.
Intel Pentium 4 3.73GHz released in Q4'04 (90nm 115 W) is the same performance as a Intel Core i3-4020Y @ 1.50GHz released Q3'13 (22 nm 11.5 W), the market is determined by it's weakest parts that OEM's are willing to buy. Now consider that the top-of-the-line part with the highest single-thread performance is Intel Core i7-4790K 4Ghz(22nm 88 W) is 3 times as powerful but consumes 8 times the amount of power.
So single core performance has improved dramatically. If a chip running at 1.5 GHz beats a topped out old Pentium 4 then obviously single core performance has increased dramatically.
So if you apply the scaling logic to the A8, it needs to be clocked 2.5-3 times higher(4Ghz to 4.5Ghz) to come close. But we don't know if it actually scales that way.
While it is unknown how far the A8 can be clocked, I'm pretty sure it won't go 3X higher. Mainly because it is running on a low power process. Since a high clock rate A8 (A9 Super) needs to run on a different process there is nothing to stop Apple from tweaking the chip to increase performance. Varying the cache size is a common ploy that Intel uses. You don't have to rely upon scaling only to move performance forward. Keeping the CPUs feed is part of the performance solution. I don't see any problem at all when it comes to designing a much higher performance A series chip.
Most persons on this thread seem to be approaching the topic as though it is an all-or-nothing proposition - that Apple either stay entirely with Intel CPUs forever, or they have to switch all Macs over to an AX unit at one go. A few posters have said, as I have before, that Apple has the option of creating a new "Mac category" that would use ARM running OSX.
Maybe they call it something besides "Mac".
This category could target education/enterprise/mobility/etc, offering benefits of an AX design:
- lower price points (while maintaining same or higher margins for Apple). A BOM cost reduction of $100 can make a significant impact.
I think people underestimate just how important this can be for Apple.
- lighter/thinner/silent
It may take a few more years but I do see a future where you can get MAC Pro like performance out of a fanless laptop. It is just a matter of some new concepts making it out of the labs.
- more stability (better integration of Apple HW and Apple SW - IMO iOS devices are significantly more stable than OSX)
Maybe. Generally Macs are pretty good, however there was an unacceptable number of missteps with the last Mac OS release. If there is a real difference it probably has more to do with IOS getting most of the companies focus.
- AX functions not on Intel like TouchID & secure element, additional DSP packages (HW acceleration of functions), all within the SOC package
Speaking of Touch ID, this really needs to make it into Apples new laptops.
- better battery life
The "con" of course is the incompatibility with legacy 3rd party software.
This is always a problem, especially if you use legacy software. However Mac OS updates have broken apps on i86 so it is a problem even without an architecture change.
That is why this new Mac category would be initially targeted at sectors where that could be mitigated more easily. A small number of partnerships (e.g. IBM, top Mac App Store developers, education partners) could provide initial set of ARM compiled applications, along with Apple's native apps. Many others would join in over time. Students and education have a limited set of app requirements (as do the majority of consumers really). Over time the application compatibility would reduce.
IPad has largely proved that Architecture isn't the big problem it was in the past.
The traditional Mac line (Pro, iMac, Macbook Pros) would remain with the x86 compatibility for the majority of apps, and Windows capability. It would be a differentiator of the "pro" devices over the AX line.
I'm not sure why people are bothered by this idea.
The long-term benefits of Apple controlling the silicon development are too much to ignore. As I think Wizard69 put it, silicon now is like the logic board was. More functions continue to get integrated into SOCs, and GPUs do more of the heavy lifting. The future of innovation in HW is there.
This is a big issue. I'd have to say that how long Apple stays with Intel is directly proportional to how long they can apply influence over Intels processor designs. It is fairly well accepted that Apple is responsible for Intel taking focus off the CPUs to beef up its GPUs. I would have to imagine that Apple and Intel have long and serious talks about product directions.
Of course Apple may not take this approach, but I fully believe they could pull it off (providing a compelling differentiation between the two lines). It would provide a meaningful increase in overall Mac business (imagine impact of a $699 highly portable Mac).
This is the thing, a $700 Mac laptop that doesn't suck.
Is there something in Swift that binds it to ARM? Perhaps it's ARM exclusive, and a first hint of what's to come. Speculating. Another idea is to kill the Mac and introduce a more mature iOS platform for a new generation of ARM trucks iPhone, iPad, iTruck. In any case I bet the A processors are being developed simultaneously with a more powerful version for trucks with 4x the performance, or something.
Odd way to end the article, undermining everything written before. The whole point is that "good enough" isn't Apple's M.O. They smoke the competition, which tries to make ends meet with cheaply made, off the shelf products. If Apple powers its computers to be "good enough" for most consumers in order to save on costs, while undercutting their premium edge, they'll suffer the same fate as the others. I think the rumored 12" Air will say a lot about Apple's direction. It's hard to imagine it coming close to the performance of current, i7 Airs, assuming it doesn't have a fan.
I got the same vibe from the last sentence.
Apple's design philosophy is not about making a "good enough" product, but they do like to streamline a product and get it down to its essence and maybe that's what the writer was aiming for... How much computer and graphic power do I need? How much may my needs change over the life of owning a specific product? All those questions asked while assuming I'm not a candidate for a professional-level product.
I don't want to go on a rugged wilderness trip with a "good-enough" tent or boots, but I don't want to drag along any more weight then possible, so I fit my gear to my needs.
In the past, where each year's passing meant over-buying to accommodate ever bloating software that required more and more resources. That's the old way of planning. When you buy a mobile computer or device, part of that purchase decision relates to being mobile. With iOS, being free of the power connector is a big part of Apple's design. So multi-tasking is only allowed where it benefits the user experience. The iOS is designed to be available to only one user, and highly secured to that user.
That the computer isn't fully multi-tasking and that the device can't serve multiple users (like a Mac or PC) aren't "good enough" decisions, they are decisions that weight heavier on portability and security.
Is there something in Swift that binds it to ARM? Perhaps it's ARM exclusive, and a first hint of what's to come. Speculating.
Another idea is to kill the Mac and introduce a more mature iOS platform for a new generation of ARM trucks iPhone, iPad, iTruck. In any case I bet the A processors are being developed simultaneously with a more powerful version for trucks with 4x the performance, or something.
I think you are seeing the possible future clearer than most people on this forum. And, if I might add, think of the generations of Macs to come more as airships and less like trucks. They both can do the heavy lifting but not both in a brute-force way.
That's because this is what Apple does. And for good reason. One of them is that Apple has to get the message out to the developers for them to jump on the wagon. And that message should unambiguously say "All of our Macs will run ARM in 2 years".
no, the message to get apps is really, really easy. Developers, new and revised apps submitted to the app store after x date must be universal binaries.
They don't need a rosetta. They don't need every legacy app. They don't need mac pro performance. All they need is developers to build their apps to support both platforms. It's really no different that requiring 64 bit support for iOS apps in the app store (and really no harder for developers). The Mac can be smart enough not install the binary that isn't needed on it's hardware. The compliers do the work.
They don't need to switch all mac's to ARM. But the ones that do switch will be huge sellers because the clear first thing to switch is the MacBook Air. Apple is really close to having ARM chips with MacBook Air level performance and can build them lighter, and cheaper, and with longer lasting batteries using their own ARM chips vs using intel chips.
For the most part it should be irrelevant to the user what CPU is inside the machine. not having intel chips in the iPhone and iPad hasn't been an issue. Other than some small number of people who need to run some windows apps, it really shouldn't matter either.
edit: for most people, if they see a $799 MacBook air sitting beside a $999 MacBook air, both with run apps at the same speed, only the first one is lighter and has better battery life, both have the full iLife and iWorks suites, with the only difference is one can run a windows emulator and the other won't, 9 out of 10 people will buy the first one.
iOS is a subset of OS X. They share the same Darwin-based code base and share a lot of the fundamental framework, including the same file system. The differences are the front-end levels of the OS, where they have different framework such as Cocoa Touch for iOS and Cocoa for OS X among others.
iOS does have a file system, it is just not user-accessible. You can jailbreak and gain access to its file system via ssh, you can change root password, etc etc. You would be able to create folders, files, delete, etc etc like you can via Terminal on OS X. In fact, if you don't want to jailbreak your device right now, look at iFunBox or iExplorer to get some basic file system access to your device.
Also, considering you probably are not programmer, you can develop OS X and iOS apps with the same code base while "forking" the different interfaces for each platform. I know, I work for a company that does this.
You also missed the part about iCloud drive, where apps store their data files and can be accessed by OS X versions of the same app. iOS 8 now has Document Picker, that replicates a file system via iCloud Drive for apps. So, yes totally possible.
Yes, there's a big difference, they're not the same architecture. That's what I just said. You can build x86 version of iOS apps actually, it's done all the time via Xcode's iOS simulator which runs the x86 native code of iOS apps.
It does have everything to do with architecture. You can code the same way, a string is a string in the code. However, when you compile the code, it translates to different code for different architecture such as x86 and ARMv8, because they do not not process a string of "text" the same way. A code compiled for ARM can run on both iOS and OS X and Xcode when compiling the code will know the difference in interfaces that the developers ask for. So yes, you can run the same app on both Mac and iOS but you don't see the same app, it will behave differently to fit the screen they're for. Touchscreen on iOS and keyboard/mouse for OS X.
It is exactly how Windows 10 is going to do it. Microsoft demo'ed the entire process already at the last event they had. Same codebase, runs on all Windows devices. Note, this requires using the specific runtime, you can't use the old win32 runtime.
Must say I really enjoy your posts, not sure if you are new here or a reincarnation like Soli but glad you are here either way.
Are you even a developer? I'm not asking to mock you or anything. It's just that you got things really mixed up.
yes, I've been a Mac developer since 1984 (back in the day where you had to have a modified LISA to develop with). I lived through both previous transitions and have watched Apple eliminate nearly all of the major issues from the prior transitions (the obvious exception is people who need windows emulation). If developers are doing their jobs correctly, there shouldn't be any CPU dependencies in their code. And the core of many apps already works on ARM because so many apps today run on both MacOS and iOS with the key differences being the UI classes (i.e. Cocoa vs Cocoa Touch). And there is nothing processor dependent in the Cocoa APIs (or any of Apple's APIs).
Some developers will have serious issues (Adobe always does), but people who depend on photoshop aren't the target audience for $799 MacBook Airs. For pretty much any app in the App store, the complexity is going to be roughly the same as moving to 64 bit.
I currently wouldn't buy an ARM based mac for my own personal use. But I would have no problem buying them for sales and support staff in my company.
Don't even get me started on Adobe lol. Understood where you're coming from. But now I see a possibility to wrap my iOS apps in Cocoa and release them for ARM OS X. Didn't though about that one. Still I think it not a good idea to split the Mac line-up for several reasons. One being upkeep. Adding more complication to a system always leads to more problem solving.
I suspect for Apple it will come down to Intel being willing to play ball and give Apple reasonable access to the silicon. Eventually Apple will need to place its IP on die. Let's face it Apple has already had significant influence upon Intel driving the significant increase in GPU performance. As such I believe Intel will do whatever is required to keep ARM out of Apples Macs. it could get very interesting in the next couple of years.
One thing everyone here is missing is moore's law is breaking down at 4nm. My limited tech guess is that 7nm will be the limit for silicon. So be it Intel or arm the current semiconductor limited will be reached by about 2018. Now of course more cores performing parallel task has what has happened. I believe it won't be for another 5 years before the carbon solution will arrive which means the chips will stagnate until about 2021. The battle will be in the cloud, services and ecosystems and of course content
One thing everyone here is missing is moore's law is breaking down at 4nm. My limited tech guess is that 7nm will be the limit for silicon. So be it Intel or arm the current semiconductor limited will be reached by about 2018.
Performance has gone up over 5x between 2008 and 2014 for both CPU and GPU. If the same thing happens between 2014 and 2020, a Macbook Pro will be 2x the CPU speed of the current 12-core Mac Pro with a GPU close to the current GTX 980. They can go further than that into 2022. They can possibly stack chips into layers.
It will slow down upgrades but it won't be an industry-stopping problem. Regular users just won't use that much power. RAM and SSD will be much cheaper too.
I have a hard time imagining that Apple will translate it's desktop and laptop devices away from Intel and into ARM. Of course there are strategic reasons like how Apple could produce products on its own schedule and have more flexibility.
The biggest reason that I don't see it happening is because I fail to see a serious strategic value over the long-term by switching away from Intel. In the words of Steve Jobs about how "desktops are going to be like trucks" I can't help but feel it was directed at laptops as well. I don't see the laptop market drastically shrinking are going away anytime soon but, on the flip side I see Apple doing everything it can to make their iPads as capable as possible.
In a lot of cases I feel like we're starting to see that shift as people in the office move from laptops two iPads with accessory keyboards or devices like the surface.
When I hear about Apple switching to the arm architecture I definitely have these weird cloudy feelings of back in the day when Macs still ran the PowerPC architecture which was such a pain in the ass in the realm of compatibility.
If over the next generations iPads become so capable and competitive that they continue to eat into the laptop market then I would imagine Apple would just stick with Intel. They'll need to keep pouring millions and millions of dollars into the arm architecture to ensure the future iPads are up to the task.
By the way I'm well aware that iPads and tablets in general can't replace traditional laptops. Some of us have to drive trucks. Including me :-p
In any case the future is long and will just have to sit back and see what happens.
Comments
Most persons on this thread seem to be approaching the topic as though it is an all-or-nothing proposition - that Apple either stay entirely with Intel CPUs forever, or they have to switch all Macs over to an AX unit at one go. A few posters have said, as I have before, that Apple has the option of creating a new "Mac category" that would use ARM running OSX. This category could target education/enterprise/mobility/etc, offering benefits of an AX design:
- lower price points (while maintaining same or higher margins for Apple). A BOM cost reduction of $100 can make a significant impact.
- lighter/thinner/silent
- more stability (better integration of Apple HW and Apple SW - IMO iOS devices are significantly more stable than OSX)
- AX functions not on Intel like TouchID & secure element, additional DSP packages (HW acceleration of functions), all within the SOC package
- better battery life
The "con" of course is the incompatibility with legacy 3rd party software. That is why this new Mac category would be initially targeted at sectors where that could be mitigated more easily. A small number of partnerships (e.g. IBM, top Mac App Store developers, education partners) could provide initial set of ARM compiled applications, along with Apple's native apps. Many others would join in over time. Students and education have a limited set of app requirements (as do the majority of consumers really). Over time the application compatibility would reduce.
The traditional Mac line (Pro, iMac, Macbook Pros) would remain with the x86 compatibility for the majority of apps, and Windows capability. It would be a differentiator of the "pro" devices over the AX line.
The long-term benefits of Apple controlling the silicon development are too much to ignore. As I think Wizard69 put it, silicon now is like the logic board was. More functions continue to get integrated into SOCs, and GPUs do more of the heavy lifting. The future of innovation in HW is there.
Of course Apple may not take this approach, but I fully believe they could pull it off (providing a compelling differentiation between the two lines). It would provide a meaningful increase in overall Mac business (imagine impact of a $699 highly portable Mac).
In any event you missed the idea that new technologies are on the horizon that could displace silicon and lead to higher speeds. That is largely due to Intel focusing on other weak spots. It will be most interesting to see if SkyLake lives up to the rumors. So single core performance has improved dramatically. If a chip running at 1.5 GHz beats a topped out old Pentium 4 then obviously single core performance has increased dramatically.
While it is unknown how far the A8 can be clocked, I'm pretty sure it won't go 3X higher. Mainly because it is running on a low power process. Since a high clock rate A8 (A9 Super) needs to run on a different process there is nothing to stop Apple from tweaking the chip to increase performance. Varying the cache size is a common ploy that Intel uses. You don't have to rely upon scaling only to move performance forward. Keeping the CPUs feed is part of the performance solution. I don't see any problem at all when it comes to designing a much higher performance A series chip.
Another idea is to kill the Mac and introduce a more mature iOS platform for a new generation of ARM trucks
I got the same vibe from the last sentence.
Apple's design philosophy is not about making a "good enough" product, but they do like to streamline a product and get it down to its essence and maybe that's what the writer was aiming for... How much computer and graphic power do I need? How much may my needs change over the life of owning a specific product? All those questions asked while assuming I'm not a candidate for a professional-level product.
I don't want to go on a rugged wilderness trip with a "good-enough" tent or boots, but I don't want to drag along any more weight then possible, so I fit my gear to my needs.
In the past, where each year's passing meant over-buying to accommodate ever bloating software that required more and more resources. That's the old way of planning. When you buy a mobile computer or device, part of that purchase decision relates to being mobile. With iOS, being free of the power connector is a big part of Apple's design. So multi-tasking is only allowed where it benefits the user experience. The iOS is designed to be available to only one user, and highly secured to that user.
That the computer isn't fully multi-tasking and that the device can't serve multiple users (like a Mac or PC) aren't "good enough" decisions, they are decisions that weight heavier on portability and security.
I think you are seeing the possible future clearer than most people on this forum. And, if I might add, think of the generations of Macs to come more as airships and less like trucks. They both can do the heavy lifting but not both in a brute-force way.
That's because this is what Apple does. And for good reason. One of them is that Apple has to get the message out to the developers for them to jump on the wagon. And that message should unambiguously say "All of our Macs will run ARM in 2 years".
no, the message to get apps is really, really easy. Developers, new and revised apps submitted to the app store after x date must be universal binaries.
They don't need a rosetta. They don't need every legacy app. They don't need mac pro performance. All they need is developers to build their apps to support both platforms. It's really no different that requiring 64 bit support for iOS apps in the app store (and really no harder for developers). The Mac can be smart enough not install the binary that isn't needed on it's hardware. The compliers do the work.
They don't need to switch all mac's to ARM. But the ones that do switch will be huge sellers because the clear first thing to switch is the MacBook Air. Apple is really close to having ARM chips with MacBook Air level performance and can build them lighter, and cheaper, and with longer lasting batteries using their own ARM chips vs using intel chips.
For the most part it should be irrelevant to the user what CPU is inside the machine. not having intel chips in the iPhone and iPad hasn't been an issue. Other than some small number of people who need to run some windows apps, it really shouldn't matter either.
edit: for most people, if they see a $799 MacBook air sitting beside a $999 MacBook air, both with run apps at the same speed, only the first one is lighter and has better battery life, both have the full iLife and iWorks suites, with the only difference is one can run a windows emulator and the other won't, 9 out of 10 people will buy the first one.
Must say I really enjoy your posts, not sure if you are new here or a reincarnation like Soli but glad you are here either way.
Are you even a developer? I'm not asking to mock you or anything. It's just that you got things really mixed up.
yes, I've been a Mac developer since 1984 (back in the day where you had to have a modified LISA to develop with). I lived through both previous transitions and have watched Apple eliminate nearly all of the major issues from the prior transitions (the obvious exception is people who need windows emulation). If developers are doing their jobs correctly, there shouldn't be any CPU dependencies in their code. And the core of many apps already works on ARM because so many apps today run on both MacOS and iOS with the key differences being the UI classes (i.e. Cocoa vs Cocoa Touch). And there is nothing processor dependent in the Cocoa APIs (or any of Apple's APIs).
Some developers will have serious issues (Adobe always does), but people who depend on photoshop aren't the target audience for $799 MacBook Airs. For pretty much any app in the App store, the complexity is going to be roughly the same as moving to 64 bit.
I currently wouldn't buy an ARM based mac for my own personal use. But I would have no problem buying them for sales and support staff in my company.
I suspect for Apple it will come down to Intel being willing to play ball and give Apple reasonable access to the silicon. Eventually Apple will need to place its IP on die. Let's face it Apple has already had significant influence upon Intel driving the significant increase in GPU performance. As such I believe Intel will do whatever is required to keep ARM out of Apples Macs. it could get very interesting in the next couple of years.
He's right.
The question is: how will A chips be to Intel in a few years?
Intel is going to hit a physical wall, it seems, as you can only get so small.
A series, too, but it is a different kind of chip.
Can Apple upset the apple cart and have the A series blow away Intel? I don't know.
Me neither.
Intel has a roadmap going to 4nm in 2022:
http://uk.hardware.info/news/16468/intel-shows-roadmap-of-up-to-4nm-in-2022
Performance has gone up over 5x between 2008 and 2014 for both CPU and GPU. If the same thing happens between 2014 and 2020, a Macbook Pro will be 2x the CPU speed of the current 12-core Mac Pro with a GPU close to the current GTX 980. They can go further than that into 2022. They can possibly stack chips into layers.
It will slow down upgrades but it won't be an industry-stopping problem. Regular users just won't use that much power. RAM and SSD will be much cheaper too.
The biggest reason that I don't see it happening is because I fail to see a serious strategic value over the long-term by switching away from Intel. In the words of Steve Jobs about how "desktops are going to be like trucks" I can't help but feel it was directed at laptops as well. I don't see the laptop market drastically shrinking are going away anytime soon but, on the flip side I see Apple doing everything it can to make their iPads as capable as possible.
In a lot of cases I feel like we're starting to see that shift as people in the office move from laptops two iPads with accessory keyboards or devices like the surface.
When I hear about Apple switching to the arm architecture I definitely have these weird cloudy feelings of back in the day when Macs still ran the PowerPC architecture which was such a pain in the ass in the realm of compatibility.
If over the next generations iPads become so capable and competitive that they continue to eat into the laptop market then I would imagine Apple would just stick with Intel. They'll need to keep pouring millions and millions of dollars into the arm architecture to ensure the future iPads are up to the task.
By the way I'm well aware that iPads and tablets in general can't replace traditional laptops. Some of us have to drive trucks. Including me :-p
In any case the future is long and will just have to sit back and see what happens.