It's truly heartening that you think that MacOSX is worth fifteen hundred dollars.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Plans to Ditch Intel and Use Custom Mac Chips Starting in 2020
- Thread starter MacRumors
- Start date
- Sort by reaction score
It's truly heartening that you think that MacOSX is worth fifteen hundred dollars.
2) While things like CPU/GPUs may not be upgraded, RAM and HDD upgrades are quite normal.
3) Do you have any actual data to back that up?
I can tell you first hand I wouldn't be typing on a mac while waiting for another mac to be delivered if it wasn't for being able to run Windows because that runs software macs can't.
If Apple didn't have dual booting, I would have just bought Windows machines. Simple solution.
While I respect that these are your thoughts on the matter, my decision and actions are mine regardless. Have a good day.
2) Are you joking ? Why da hel * apple support egpu ? Just put one usb c for charging hoping bluetooth wifi solve everything ?
3) People dual boot to get perfomance . A BIT rude saying just get 300 dollar windows machine.
I'm late in my fifth year of ownership of a fully specced Retina MacBook Pro which shows no obvious signs of getting long in the tooth yet; mainly thanks to superb I/O performance.
Meanwhile, my colleagues with upgradable Lenovo T-series laptops have replaced their computers at least once during the same period due to real or perceived performance issues.
Out of almost 50 people, exactly 1 tried doing something about their computer's performance or lifetime by installing a larger and faster SSD; that's how popular upgradability is where I work...
Granted, this is a very limited dataset, but except for the usual lust for the latest-and-greatest, a machine that's properly configured for its intended tasks from the start and which doesn't have an accident shouldn't be a "throwaway computer" for most people no matter its brand. Serviceability, on the other hand, is definitely a problem with the current generations of thin-and-powerful laptops, but that is true for all brands. The main difference is that you've got no choice from Apple; they simply don't sell any user serviceable models.
As for the necessity of x86 compatibility:
Perhaps I'm just lucky, but since several years I would do just fine on anything with a decent package manager, as long as a handful of proprietary software houses beside Apple recompiled their products for whatever new architecture was chosen.
Edited to add: As you stated, your use case is your own; I just wanted to give a real-life example of where other use cases may create very different needs.
Maybe I'm misunderstanding your post, but are you saying Apple's A series CPU running an X86 emulation will be faster then intel's chip, i.e., emulation is faster then hardware?
In his defense, Remember that anything that’s cocoa doesn’t need to be emulated, and the arm hardware may in fact have a higher CPI than intel’s X86 by then. Of course, Intel would sue the crap out of Apple unless Apple takes an x86 license - they’d also need a license from AMD, which is easier to get.
Apple has proven many times that it's got lots of "courage", thus they couldn't care less about upgradability and supporting legacy applications. so i won't be surprised if they ditch intel chips.
Lets see if they can actually come up a design that is competitive first. Until they do, it is all a pipe dream, vapor ware, and a gnashing of teeth.
I doubt I would buy another Mac if it didn't have X86 compatibility. It is too useful for me.
Like the removal of floppies, legacy ports like ADB, and optical drives - this transition from x86 to ARM, if it ever happens, will begin slowly with consumer machines and over a few years titrate up to the "pro" machines that most folks on MacRumors buy.
I bet Safari, iTunes, Photos, and most of the other apps most consumers use on their Macs run great on ARM.
ARM, the Mac App Store, and sealed (not just not upgradable, but actually sealed) units (like iPads) are the future for the Mac I think.
I bet Safari, iTunes, Photos, and most of the other apps most consumers use on their Macs run great on ARM.
ARM, the Mac App Store, and sealed (not just not upgradable, but actually sealed) units (like iPads) are the future for the Mac I think.
It never will be with apple current resources. Microsoft fail miserably on surface rt.Arm didn't have power horse like intel ,amd,ibm . Unless amd , intel and ibm want to do major investment then it will be success.
Further and further down the rabbit hole they go. I'm now stuck on a Mid-2015 15" MacBook Pro that's about to go out of warranty, and have already set my foot down as it being the last MacBook Pro I buy until they return to a full keyboard (I'm a Unix guy, I need my escape key in the PROPER spot and to be a physical key). Switching away from Intel would be another reason for me not to buy another Apple. It's like they truly refuse to understand their user base, and instead are focused on the people who simply want a desktop sized iPad with keyboard and mouse.
The Touch bar is a solution to a problem that doesn't exist. The TouchID button is nice but yeah Touch bar is useless to me.
ARM has improved a lot since Surface RT, and Macs in general have far fewer legacy apps to support than Windows so Apple could be more successful, whether or not they actually do it remains to be seen.
For developer like me, yeah great i don't want fake simulator .
But for some video editor, photographer here deal with 40 GB RAM or want 128 GB RAM it will be super bad. NVME ssd tablet already in the market so doesn't much issue now.
Yes, the is linux flavour with arm proc, apple just take some code from open source community. I also see some server sell using arm proc also.
The main problem is , if apple move to arm, are premier Software Developer like Adobe,Da Vinci ,Microsoft want to recode to arm proc ? It take years to migrate and testing all those again unless APPLE have mention in 3 years ahead they want it.
In an ideal world, making an ARM version of these apps is just a recompile. For Adobe for example, I do know that at least some parts of their code is quite independent from the machine architecture. Now it's not an ideal world, there are some differences between Intel and ARM that actually do bleed through to the application, and some rewriting may need to be done for performance. There's a fair chance though, that it won't actually be all that difficult from a technical perspective.
I think the difficulties are more administrative. It's yet another platform and architecture to keep track of. Instead of one download button for macOS you need two. It gets confusing for users. Maybe there will be licensing issues from some developers. Other devs will have discontinued their products or have folded, and won't be doing even a simple recompile. And same for Apple. They will need to keep parallel versions of the OS, and this is extra admin and confusing to non-technical users.
On the topic of upgradeability, I think it's more of a security blanket for most users than anything else. Actually, most users probably won't even know that upgrades is a thing. For those who do, I really find it hard to believe that it's a major issue. Computer specs aren't changing as fast anymore, and it's really not too hard to have a little think at purchase time what specs you think you'll need for the lifetime of the device. Even on desktops, which are presumably built for upgrading, I find it rare to be a good decision, because by the time you need to upgrade, you need to upgrade multiple parts anyway. Disk space maybe excluded, but that's easy to add externally to laptops as well.
It's a bit the same with dual boot to Windows. It's a nice feature, and there are indeed some who do benefit a lot from it. But for the majority, I suspect just getting a separate cheap Windows box is a much better solution. I've been sceptical to this ARM move myself, but when I think about it I don't actually dual boot my Mac now, and I don't see that I would really need to in the future either. It would be nice, but Apple have priced themselves out of it anyway.
I was just thinking about how if I was speccing out a system today, bootcamp would not get me anywhere.
The three tech aspects that intrigue me at this time are
4k media like 4k bluray and paid streaming-- Apple doesn't use the appropriate chipsets to make the vendors happy
nVidia RTX--don't imagine that it will be in any mac products soon.
hexcores-- Apple's throttling fix, iirc, is macos only
At present I use some niche, closed source windows products that are not replicated on the mac platform. Some of those products take advantage of my imac's large screen. If, for some godawful reason, apple renders my bootcamp installation useless, a 300 dollar walmart special isn't going to restore that lost functionality.
The three tech aspects that intrigue me at this time are
4k media like 4k bluray and paid streaming-- Apple doesn't use the appropriate chipsets to make the vendors happy
nVidia RTX--don't imagine that it will be in any mac products soon.
hexcores-- Apple's throttling fix, iirc, is macos only
At present I use some niche, closed source windows products that are not replicated on the mac platform. Some of those products take advantage of my imac's large screen. If, for some godawful reason, apple renders my bootcamp installation useless, a 300 dollar walmart special isn't going to restore that lost functionality.
The Apple ecosystem is only as strong as its developers. It's how they got here today along with the x86 adoption. Imagine if they had stuck to PPC. Would they even have a prolific iOS environment?
The part I am most skeptical about is the application support. All my C/C++ modules will need to be recompiled again and are dependencies for other applications/libraries. If the maintainers of those modules don't feel like making the modules compatible with Apple's new architecture, then as a dev I am kind of forced to jump back to Linux/Windows if I wanted to use those modules again.
Re-compile is the hardest part cum the cost of time of it. Who want to re-code and test for small market like APPLE ? With i see track record IOS , introduce new future and abandon old thing and force user to upgrade . I would said, apple will be dead.
[doublepost=1535662606][/doublepost]
if it ain't broke, don't fix it .
They're already re-coding and testing for the small market like Apple. Switching to ARM doesn't change that, aside from library dependencies and things.
[doublepost=1535675282][/doublepost]I have zero need and zero interest in running Windows on Mac hardware. Not being able to run Windows is 100% okay in my house.
Apple is planning to transition from Intel chips to its own custom made Mac chips as early as 2020, reports Bloomberg.
Apple's initiative, reportedly code named "Kalamata," is part of an effort to make Macs, iPhones, and iPads work "more similarly and seamlessly together" according to unspecified sources that spoke to Bloomberg. Apple already designs its own A-series chips found in iPhones and iPads.
The Mac chip plans are said to be in the early stages of development and the transition from Intel chips to Apple chips could involve multiple steps, starting with the "Marzipan" initiative coming in iOS 12 and macOS 10.14 to allow developers to create a single app able to run on both iOS and macOS.
With its own chips, Apple would not be forced to wait on new Intel chips before being able to release updated Macs, and the company could integrate new features on a faster schedule.Apple has already begun using custom designed T1 and T2 chips in its MacBook Pro and iMac Pro machines, and the company is said to be planning to integrate additional custom co-processors in Macs coming later this year. The custom chips will also be used in the upcoming Mac Pro, which is in development.
The T1 chip, included in the MacBook Pro, powers the Touch Bar and authenticates Touch ID. The T2 chip, in the iMac Pro integrates several components including the system management controller, image signal processor, SSD controller, and a Secure Enclave with a hardware-based encryption engine.
Previous rumors have suggested Apple is interested in creating its own ARM-based core processor chips for its Mac lineup in order to reduce its dependence on Intel. Apple is also rumored to be pursuing development of its own modem chips to also reduce reliance on both Intel and Qualcomm.
A move away from Intel would have a major impact on Intel, with Apple providing approximately five percent of Intel's annual revenue. Intel stock has already dropped following the news.
Article Link: Apple Plans to Ditch Intel and Use Custom Mac Chips Starting in 2020
I do think Apple needs to supply x86 compatibility to keep developers (like me). It is too essential... one reason we like Macs is we can do everything on them (Windows, x86 Linux, and Mac, etc.)
However, emulation doesn't necessarily turn me off like it used to. I think the marginal cost of emulation decreases so much as performance increases, that it may just fade to the background most of the time. I can't notice any difference (outside of gaming) between Boot Camp and Parallels anymore, for example. (I know that's not the same as processor emulation, but there used to be more of a performance penalty for the emulation it does do.)
I think of it like hard drive encryption: it will always be slower than not having it, and for a long time, the performance price was too high. But it is so small now, it is not significant. Emulation will be slower, but maybe no more so than, say, having an extra core in an Intel processor anyway.
Also, with Apple making the chips, they might be able to do interesting things... like putting a separate co-processor in "Pro machines" to handle x86 emulation (like how they added a T1/T2 chip for some functions). If you offload core OS functions to the main chip, you might get close to native speeds with well-designed parallel processing. Or maybe you can double or triple the cores in the main chip, since you don't have to pay the "Intel tax" for such a move anymore. Emulation on an 8-core may be no slower than native on a 4-core, for example.
Another good option would be for "Pro machines" to keep Intel co-processors as an option.
However, emulation doesn't necessarily turn me off like it used to. I think the marginal cost of emulation decreases so much as performance increases, that it may just fade to the background most of the time. I can't notice any difference (outside of gaming) between Boot Camp and Parallels anymore, for example. (I know that's not the same as processor emulation, but there used to be more of a performance penalty for the emulation it does do.)
I think of it like hard drive encryption: it will always be slower than not having it, and for a long time, the performance price was too high. But it is so small now, it is not significant. Emulation will be slower, but maybe no more so than, say, having an extra core in an Intel processor anyway.
Also, with Apple making the chips, they might be able to do interesting things... like putting a separate co-processor in "Pro machines" to handle x86 emulation (like how they added a T1/T2 chip for some functions). If you offload core OS functions to the main chip, you might get close to native speeds with well-designed parallel processing. Or maybe you can double or triple the cores in the main chip, since you don't have to pay the "Intel tax" for such a move anymore. Emulation on an 8-core may be no slower than native on a 4-core, for example.
Another good option would be for "Pro machines" to keep Intel co-processors as an option.
People do not really understand ARM vs Intel. ARM is a RISC chip, Reduced Instruction Set Computer. The ARM CPU is less complex and has to break up instructions/code into small pieces.
Because it is less complex it is easier to shrink to 10/7/5nm. It benefits from from using less power, which produces less heat and requires less cooling. This makes it the best choice for mobile devices.
Intel chips are CISC chip, Complex Instruction Set Computer. The Intel CPU's are way more complex and thus harder to shrink. At the same time they way more powerful in many ways. Being more powerful requires more power, which produces more heat which requires more cooling. This makes it better for computers in terms of space/power/cooling (minius the throttlebook of course). The CISC CPU's can run way more complex software and do it faster. The IPC of a Intel CPU over a ARM CPU is many times greater. This is why we do not see full blown Photoshop, CAD apps, games like Witcher 3 etc running on ARM. We do see light versions of all of this.
If Apple or Qualcom (Windows 10 on ARM) keep making their ARM CPU's more powerful to take over for Intel x86 they will eventually run into the same issues with power, heat, and cooling. There is no way around it. Unless of course they never want to run anything but light/less complex applications which might be fine for the majority of people. Powerful, complex applications will require more than ARM can deliver today.
yup, they will be limit proc when they try to shrink it unless they found new material can sustain more core . Maybe we will see again 2 actual core cpu again.
This is very misleading. I designed many x86 chips. They have risc cores. They have complicated circuitry on board which breaks up instructions into smaller, more regular instructions. They include microcode roms and circuitry for converting variable-length instructions with lots of operands into a sequence of reduced, fixed length instructions.
You say risc chips need to break up the instructions into smaller instructions but You have it backwards. In CISC chips like x86 the processor does this. In RISC chips the compiler does it. The compiler can often do a better job because it can analyze the full program to determine how best to break instructions up.
The IPC of arm vs Intel isn’t a thing. You can make ARM chips have higher IPC than Intel if you want ... it just wouldn’t do much good on a phone.
Remember, if you strip out the complex instruction decoder and microcode from an Intel chip, what you are left with is a risc cpu. Many risc cpus, by the way, have very high IPCs. Look at Sparc, PowerPC for workstations, etc.
I’ve designed PowerPC, sparc and x86 chips, and your explanation is pretty much nonsense.