The end of Apple?
Even if apple stopped making macs altogether it would hardly be a blip.
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.
Intel Expecting Apple to Transition to Custom ARM-Based Chips Starting in 2020
- Thread starter MacRumors
- Start date
- Sort by reaction score
The end of Apple?
Even if apple stopped making macs altogether it would hardly be a blip.
Well, I think that Cook has addressed that point quite nicely, with a resounding "Never".
[doublepost=1550881418][/doublepost]
I thought there was a Downloadable Win 10 for ARM 64 from MS, but I couldn't find it.
As for .NET, MS seems to have a really spasmodic "policy".
However, it seems things are more rosy for the JDK:
JDK 8 for ARM via Oracle:
https://www.oracle.com/technetwork/java/javase/downloads/jdk8-arm-downloads-2187472.html
Last edited by a moderator:
You can count on Apple using this opportunity to require all MacOS apps go through their app store.
There haven't been for quite some time now. Ajit Pai is a liar.
[doublepost=1550882005][/doublepost]
You're wrong about that.
Due to WoW64 layer in Windows 10 ARM 64, no "ARM Port" of x86 is needed. It is generated "on the fly":
https://docs.microsoft.com/en-us/windows/uwp/porting/apps-on-arm-x86-emulation
They CALL it "Emulation"; but it is more correctly called "Just In Time Cross-Compiling". The code that actually RUNS is ARM 64 NATIVE, NOT EMULATED.
[doublepost=1550882278][/doublepost]
Based on what, other than rank paranoia?
[doublepost=1550882421][/doublepost]
I think it will be "low-end to high-end" Transition, not the other way around.
[doublepost=1550882527][/doublepost]
And why, oh, why would Apple ever mess with that flexibility?
An iPhone is not a Mac, and never will be, and iOS is not macOS, and never will be.
This was me. my old iMac was a pain in the arse to use. then I put a ssd in... new machine. Night and day. Then a month later Apple announces that my iMac can't run the new os. Booooo
Wrong.
You are thinking of Windows RT or 10s. Windows 10 on ARM is NOT THAT.
Read THEN Post:
https://www.pcmag.com/news/353637/windows-10-on-arm-runs-all-win32-apps-unmodified
You (as a Software Dev.) certainly CAN recompile/rewrite Windows Applications for ARM, and can develop Native ARM Windows Applications; but, thanks to their JIT-Compiler-Based "Emulation", you don't HAVE to!
See:
https://docs.microsoft.com/en-us/windows/arm/
[doublepost=1550883885][/doublepost]
Why will it be a mess for a few years? Because the past TWO Architecture changes were?I like this but I know it will be a mess for few years. So probably not a bad idea to buy something and wait for them to bridge it. Probably 2019 products will be the best for next few years
Oh, wait...
[doublepost=1550884084][/doublepost]
And of course REAL COMPUTERS ONLY RUN INTEL, right?
Please.
[doublepost=1550884628][/doublepost]
Wrong.
Hackintoshes remain a rounding-error as far as percentage goes. Apple could care less; so long as they remain a rounding-error.
The market has changed so much since the PPC to Mac transition.
Serious question: what things have changed that have made system architecture less relevant?
I didn't pay much attention to this and then sort of "woke up" a couple years ago and realized that absolutely everything I use daily is either built-in to every OS or is part of a subscription I pay for that allows me to download the software on whatever computer I use. I am accustomed to using Mac OS and like it. It would slow me down with my work flow for a few weeks if I switched to Windows. But otherwise it would not matter.
Same goes for Android vs. iOS, except for all the digital media purchases... and even those are becoming less important now as the music subscription model fixed half of that for me.
To a large extend, the only thing that matters is a diverse app ecosystem. Apple will instantly enjoy that with this transition where the software packages that people really need will be ported over almost instantly as far as 99% of the population is concerned.
I could be wrong. But I think the 1-2% of specificity users who are more likely to come to sites like this one are not very indicative of the overall market. There will be a 12-18 month period of minor pain. After that, it'll be fine and no one will talk about it.
Serious question: what things have changed that have made system architecture less relevant?
I didn't pay much attention to this and then sort of "woke up" a couple years ago and realized that absolutely everything I use daily is either built-in to every OS or is part of a subscription I pay for that allows me to download the software on whatever computer I use. I am accustomed to using Mac OS and like it. It would slow me down with my work flow for a few weeks if I switched to Windows. But otherwise it would not matter.
Same goes for Android vs. iOS, except for all the digital media purchases... and even those are becoming less important now as the music subscription model fixed half of that for me.
To a large extend, the only thing that matters is a diverse app ecosystem. Apple will instantly enjoy that with this transition where the software packages that people really need will be ported over almost instantly as far as 99% of the population is concerned.
I could be wrong. But I think the 1-2% of specificity users who are more likely to come to sites like this one are not very indicative of the overall market. There will be a 12-18 month period of minor pain. After that, it'll be fine and no one will talk about it.
No. Because I think iOS does not equal macOS.
Read again the sentence from the original article to which I was referring. It came immediately after this paragraph:
This is the transition to which I was referring, not the rumored transition from Intel to Apple CPUs.
KazKam already explained the point I was making perfectly clearly:
I assume what he’s trying to say is that Apple is not allowed to implement chips using the x86-64 instruction set. That would be true unless they used a fab that had a license. Used to be IBM had one, for example, so maybe Global Foundries inherited it from them or AMD - no idea. Or they would have to get Intel to license its contributions and AMD to license the AMD64 extensions. There used to be some other licensees floating around, and other clones (Cyrix, rise, national, etc.) Some had licenses, some used licensed fabs.
I think they’ll start with the little MacBook and work their way up.
[doublepost=1550887612][/doublepost]
I agree—I do not want to be locked into an App Store in a Mac, the way Windows is doing with its Arm Windows thingy.
Here we go again, back to the endless incompatibilities of the 90's where everything always worked perfectly on the Windows machines and nothing but incompatible this and that on the Macs.
Lost count of how often I couldn't open, or play or see or hear, email attachments that Windows users opened up just fine.
Lost count of how often I couldn't open, or play or see or hear, email attachments that Windows users opened up just fine.
Apple had better be really careful with this one. I transitioned to Mac from Windows in part because I could virtualize Windows and run all of my old programs while I found Mac-based replacements. If corporate software gave me issues, I could always virtualize Windows and run things there. Once I had settled on the Mac, I went all-in with Apple to get the full benefit of the "Apple ecosystem" - iPhone, iPad, Airport router, Watch, Apple TV...
But at the end of the day, my computer is still #1 among all of those devices in its importance. If Apple goes full ARM and it's a messy transition that doesn't offer any major benefits, I'll be forced back to Windows. And once that's done, the hold that Apple has on me with all of their other products will be at risk, because my most important computing device will no longer be in their ecosystem.
I hope it doesn't happen, or that we're somehow pleasantly surprised.
But at the end of the day, my computer is still #1 among all of those devices in its importance. If Apple goes full ARM and it's a messy transition that doesn't offer any major benefits, I'll be forced back to Windows. And once that's done, the hold that Apple has on me with all of their other products will be at risk, because my most important computing device will no longer be in their ecosystem.
I hope it doesn't happen, or that we're somehow pleasantly surprised.
I only used the info I could find readily on Google. I made a point of noting that the Intel pricing was retail b/c I have literally no idea what Apple is paying them & wanted to make that clear, as not to obfuscate.
As to your other few queries: I don’t think the A12 in the iPad Pro has an inordinate sized heatsink, based on the slim/sealed nature of the device, yet the latest I’ve heard is that it runs faster than 90% of all notebook computers sold in the last two years. As such, I scarcely think 8 of them would be a necessity. It seems that 2 would come in (with 5nm die shrinkage) the same size or smaller than an i7 & handily best it in raw power (are there any Xeons that bench twice as fast as an A12??). Further- I think, even not knowing Intel’s pricing given to Apple, we could assume it would be greater than $60, which would be roughly Apple’s cost for a dual A13 chip setup on a 2020 MacBook.
[doublepost=1550893631][/doublepost]
Who’s talking about an iPad???
The article is talking about 2020 Macs running Apple’s A-Series chips...
Yes, we all know that they’d be thermally throttled in an iPad, but given the vastly larger amount of space in a MacBook, why would they also be thermally throttled there?
That is a Geekbench at an equivalent standard in iOS vs macOS; that is not real world, or user experience and is only a measures at peak, not sustained performance.
The laws of physics will always thermally throttle any processor-it's more important that the machine doesn't become thermally saturated so the performance is sustained.
Last edited:
Nah. The first releases are likely to be developer Macs. Which means that any Mac developer can write and load any app they can build from code onto their ARM computer. (Same as with iOS devices currently.)
Plus Unix/server terminal command-line application development.
When I was studying for my Ph.D with a concentration in solid state physics they didn’t teach me that one. Care to elaborate?
You want me to describe what a TDP of a chipset is? Quit trolling. Mac Pro 6,1, need I say more?
Last edited:
You said all chips, by the laws of physics, must throttle. That is certainly not true. It is only true when the ability of the cooling solution to remove heat is less than the maximum heat flux generated by the chip when it is running at its maximum speed. In other words, in SOME systems the maximum clock rate is determined by cooling. But certainly not in all systems. In many chips there is a critical delay path that limits clock frequency to something less than the cooling ability of the system. In other words, the chip inherently has a maximum frequency F, and power dissipation is F times the square of the voltage times the effective capacitance switched each cycle times some constant. If the power dissipated by the chip can be removed by the cooling system, which is often the case particularly in RISC workstation class systems and the like, and was always the case in the old days before throttling was a thing, then the CPU does not need to throttle ever. It’s actually not that hard to cool things if you have enough volume - 10W/cm^2 of heatsink was easily doable with a light breeze even back in the 1990’s.
You make these grand pronouncements about what physics demands and that all chips behave a certain way, but your physics is simply wrong.
Yup. Many real world usage situations do not saturate processor functional units nearly as much as benchmarks, which allows the machine to seem even faster under that real work usage to the user than indicated by the benchmarks.
And yet here we are at an agreement heat must be cooled at an effective rate or even the most powerful processors available will falter-you just seem to take the attitude of puking your degrees all over the thread like it matter which processors we are talking about. It doesn't. I hope they cool these new ARM's well so it's not a repeat of the 6,1. Better?
When you’re wrong you’re wrong. And the fundamental point is an A-series processor may very well NEVER have to throttle when placed in an enclosure as big as a MacBook or iMac or the like. If you don’t want to be corrected, don’t say one thing, get it wrong, and then pretend you said something completely different.
Not true by example.
The 6502 processor in the Apple II did not throttle.
Even the 150 kiloWatt Cray 1 did not throttle (although it was capable of an emergency shutdown if a leak in the liquid freon cooling system was detected). All you need is enough cooling to meet specified performance.
Off hand I can’t recall any of the 8-bit or 16-bit systems that throttled. Early 32-bit systems didn’t either. Many industrial 32-bit systems still don’t.