Honestly, this is what I'm most excited about. The new MacBooks are relatively straightforward, but it'll be exciting to see what happens when they start including things that higher end machines...
Honestly, this is what I'm most excited about. The new MacBooks are relatively straightforward, but it'll be exciting to see what happens when they start including things that higher end machines need - like external GPUs and insane amounts of cores.
So I this is something I posted on reddit, but didn't really get much discussion. I really wonder what the longevity of these early high end systems will be. I feel like there's a good chance that...
So I this is something I posted on reddit, but didn't really get much discussion. I really wonder what the longevity of these early high end systems will be. I feel like there's a good chance that either:
A) Apple arbitrarily revises the architecture, claiming "new and improved design makes it incompatible with our previous versions". Forcing early adopters to upgrade at a huge loss if they want continued support.
B) The platform fails to be popular enough to receive widespread compatibility beyond a few "killer apps" that make the platform viable in the first place. Ultimately Apple kills off the platform, either entirely, or at least in its current form to make a cheaper equivalent.
C) Apple gets cold feet, and cancels the platform once it becomes clear that it's not an instant success; goes back to x86. Fortunately Apple isn't Google, otherwise I'd be sure this would be the case. Still, it's not out of the question.
And that's just all the things Apple might do based on past behavior (NeXT, PowerPC, planned obsolescence philosophy). Not to mention, this is exactly the kind of behavior that 80's brands exhibited back when proprietary non-x86 platforms were common. And as a brand new, completely closed platform, owners would have no recourse either. Sure, there might be legacy support for a while, but there wouldn't be any standard version of Windows or Linux you could install. I'd be surprised if there wasn't a forward thinking hardware lock on installing some ARM compatible distro anyways, effectively limiting you to a compatible Mac OS.
The only silver lining here is that I feel like the only people buying these systems will be huge studios with money to burn, so perhaps they wouldn't be nearly as miffed by such actions as an ordinary consumer. Regardless, seems like a risky decision to be an early adopter. What do you think Tilderinos? Would Apple actually do something like the above, and would you care?
Edited in "either" for clarity so it didn't seem like I was assuming all these things would happen simultaneously, rather one or the other.
I think those concerns are bit too pessimistic. For "arbitrarily revises architecture" and "chips fail", that would be a concern if these were the first CPUs Apple makes, but they're not. They're...
I think those concerns are bit too pessimistic. For "arbitrarily revises architecture" and "chips fail", that would be a concern if these were the first CPUs Apple makes, but they're not. They're actually Apple's 10th year at designing their own chips, it was simply that this was the first year where their consistent linear growth in performance hit Intel's logarithmic growth such that they thought it was reasonable to make the switch.
And from that, you can also see how Apple approaches their ARM CPUs. Do they "arbitrarily revise the architecture"? No. Will Apple abandon it if it isn't an instant success? Also no. This isn't a moonshot, it's the continuation of 10 years of hard work. Additionally, Apple will never cease to make better CPU architectures so long as the iPhone uses them (and considering how Qualcomm performs, I doubt this will change anytime soon).
The platform fails to be popular enough to receive widespread compatibility beyond a few "killer apps" that make the platform viable in the first place. Ultimately Apple kills off the platform, either entirely, or at least in its current form to make a cheaper equivalent.
This I also doubt will happen in practice. Apple successfully whipped everyone through the transition twice already, and now they are an even bigger force than before, being the world's most valuable company. Already Adobe has released a native lightroom, Google a native Chrome, Microsoft a native office suite.
I don't think Apple will have any trouble bulldozing 3rd parties into getting on their platform. Apple's no Google; it's their way or the highway.
Sure, there might be legacy support for a while, but there wouldn't be any standard version of Windows or Linux you could install.
This is like half true half not. There actually are ARM windows and linux distros (that's what all the Pi's run). Microsoft currently is only licensing the ARM version to specific vendors (Apple said that they're happy to have windows run if Microsoft allows it), and as for Linux, the kernel should run, although you need driver support to be useful.
However, you can actually disable the bootloader lock on the new Apple Silicon macs, there's even an Apple help page for it, so theoretically if people get the driver situation down, Linux shouldn't be a problem.
As I understand, this is also thanks to Apple's decision not to fully lock the boot process: https://twitter.com/marcan42/status/1333126018068955136 Marcan's twitter is a really good source of...
as for Linux, the kernel should run, although you need driver support to be useful.
The main problem with the M1 for me is the GPU and other devices around it, because that's likely what would hold me off using it because it wouldn't have any Linux support unless Apple opens up
A) Guaranteed. 68000 to PowerPC. PowerPC G3 to G4 to G5. Of course the classic, PowerPC to Intel. Within the Intel era there were a few snags along the way, like with 32-bit only Core Solo Mac...
A) Guaranteed. 68000 to PowerPC. PowerPC G3 to G4 to G5. Of course the classic, PowerPC to Intel. Within the Intel era there were a few snags along the way, like with 32-bit only Core Solo Mac minis. Apple never should have shipped an Intel machine that didn't support x86_64 because it was a massive footgun when they finally phased out 32-bit support over a decade later.
B) It already is popular. Apple has been shipping their own high performance silicon since the A9.
C) Steve Jobs famously said during the PPC-to-Intel transition keynote "Why? [...] because we want to be making the best computers for our customers going forward." They see where the ball is going with the Arm ISA. If the proverbial ball starts heading back towards x86, maybe they will switch back. I can't see it happening any time soon though. I'd sooner bet on a fully custom Apple ISA than a switch back to x86.
I agree with you on lock in. That's a big software risk by choosing Apple hardware.
Of the three, I think that A is the most likely scenario, but even so I don't think that it's terribly likely to happen. The fact that Apple has chips that are competitive with Intel and AMD was...
Of the three, I think that A is the most likely scenario, but even so I don't think that it's terribly likely to happen. The fact that Apple has chips that are competitive with Intel and AMD was no overnight success; it took more than a decade of development. This hypothetical high-performance chip is just scaling those existing designs up, so I don't see them coming up with something dramatically different any time soon.
B and C are just incredibly unlikely given how good their compatibility solution is reported to be. There is far too much momentum to believe that they would want to go back to purchasing from Intel. Being able to have their own processors is potentially a huge savings for them. The only outside company they have to pay is for the licensing of their IP, and AFIAK that's just limited to ARM.
And who knows? Perhaps apple will transition their silicon to remove/replace ARM's features overall in the future just so they have one less licensor to pay. But either way, they'll still be making their own chips - at least until the point that they are dramatically behind their competition.
Not only that, if we look at their other product lines (and even the Mac to a large extent), they have continued supporting their devices for far longer than other manufacturers. iOS 14, for...
Of the three, I think that A is the most likely scenario, but even so I don't think that it's terribly likely to happen. The fact that Apple has chips that are competitive with Intel and AMD was no overnight success; it took more than a decade of development.
Not only that, if we look at their other product lines (and even the Mac to a large extent), they have continued supporting their devices for far longer than other manufacturers. iOS 14, for example, still supports the iPhone 6S which was released 5+ years ago.
Honestly, this is what I'm most excited about. The new MacBooks are relatively straightforward, but it'll be exciting to see what happens when they start including things that higher end machines need - like external GPUs and insane amounts of cores.
So I this is something I posted on reddit, but didn't really get much discussion. I really wonder what the longevity of these early high end systems will be. I feel like there's a good chance that either:
A) Apple arbitrarily revises the architecture, claiming "new and improved design makes it incompatible with our previous versions". Forcing early adopters to upgrade at a huge loss if they want continued support.
B) The platform fails to be popular enough to receive widespread compatibility beyond a few "killer apps" that make the platform viable in the first place. Ultimately Apple kills off the platform, either entirely, or at least in its current form to make a cheaper equivalent.
C) Apple gets cold feet, and cancels the platform once it becomes clear that it's not an instant success; goes back to x86. Fortunately Apple isn't Google, otherwise I'd be sure this would be the case. Still, it's not out of the question.
And that's just all the things Apple might do based on past behavior (NeXT, PowerPC, planned obsolescence philosophy). Not to mention, this is exactly the kind of behavior that 80's brands exhibited back when proprietary non-x86 platforms were common. And as a brand new, completely closed platform, owners would have no recourse either. Sure, there might be legacy support for a while, but there wouldn't be any standard version of Windows or Linux you could install. I'd be surprised if there wasn't a forward thinking hardware lock on installing some ARM compatible distro anyways, effectively limiting you to a compatible Mac OS.
The only silver lining here is that I feel like the only people buying these systems will be huge studios with money to burn, so perhaps they wouldn't be nearly as miffed by such actions as an ordinary consumer. Regardless, seems like a risky decision to be an early adopter. What do you think Tilderinos? Would Apple actually do something like the above, and would you care?
Edited in "either" for clarity so it didn't seem like I was assuming all these things would happen simultaneously, rather one or the other.
I think those concerns are bit too pessimistic. For "arbitrarily revises architecture" and "chips fail", that would be a concern if these were the first CPUs Apple makes, but they're not. They're actually Apple's 10th year at designing their own chips, it was simply that this was the first year where their consistent linear growth in performance hit Intel's logarithmic growth such that they thought it was reasonable to make the switch.
And from that, you can also see how Apple approaches their ARM CPUs. Do they "arbitrarily revise the architecture"? No. Will Apple abandon it if it isn't an instant success? Also no. This isn't a moonshot, it's the continuation of 10 years of hard work. Additionally, Apple will never cease to make better CPU architectures so long as the iPhone uses them (and considering how Qualcomm performs, I doubt this will change anytime soon).
This graph from Anandtech really says it all
This I also doubt will happen in practice. Apple successfully whipped everyone through the transition twice already, and now they are an even bigger force than before, being the world's most valuable company. Already Adobe has released a native lightroom, Google a native Chrome, Microsoft a native office suite.
I don't think Apple will have any trouble bulldozing 3rd parties into getting on their platform. Apple's no Google; it's their way or the highway.
This is like half true half not. There actually are ARM windows and linux distros (that's what all the Pi's run). Microsoft currently is only licensing the ARM version to specific vendors (Apple said that they're happy to have windows run if Microsoft allows it), and as for Linux, the kernel should run, although you need driver support to be useful.
However, you can actually disable the bootloader lock on the new Apple Silicon macs, there's even an Apple help page for it, so theoretically if people get the driver situation down, Linux shouldn't be a problem.
As I understand, this is also thanks to Apple's decision not to fully lock the boot process: https://twitter.com/marcan42/status/1333126018068955136
Marcan's twitter is a really good source of information about Apple Silicon without the hype, I think. He's also going to work on a proper Linux port to them, and has raised a good chunk of funding already: https://tildes.net/~tech/tob/notable_developer_hector_martin_marcan_starts_patreon_to_fund_apple_silicon_linux_port
Linus Torvalds has given a few less optimistic thoughts: https://www.zdnet.com/article/linus-torvalds-would-like-to-use-an-m1-mac-for-linux-but/
A) Guaranteed. 68000 to PowerPC. PowerPC G3 to G4 to G5. Of course the classic, PowerPC to Intel. Within the Intel era there were a few snags along the way, like with 32-bit only Core Solo Mac minis. Apple never should have shipped an Intel machine that didn't support x86_64 because it was a massive footgun when they finally phased out 32-bit support over a decade later.
B) It already is popular. Apple has been shipping their own high performance silicon since the A9.
C) Steve Jobs famously said during the PPC-to-Intel transition keynote "Why? [...] because we want to be making the best computers for our customers going forward." They see where the ball is going with the Arm ISA. If the proverbial ball starts heading back towards x86, maybe they will switch back. I can't see it happening any time soon though. I'd sooner bet on a fully custom Apple ISA than a switch back to x86.
I agree with you on lock in. That's a big software risk by choosing Apple hardware.
Of the three, I think that A is the most likely scenario, but even so I don't think that it's terribly likely to happen. The fact that Apple has chips that are competitive with Intel and AMD was no overnight success; it took more than a decade of development. This hypothetical high-performance chip is just scaling those existing designs up, so I don't see them coming up with something dramatically different any time soon.
B and C are just incredibly unlikely given how good their compatibility solution is reported to be. There is far too much momentum to believe that they would want to go back to purchasing from Intel. Being able to have their own processors is potentially a huge savings for them. The only outside company they have to pay is for the licensing of their IP, and AFIAK that's just limited to ARM.
And who knows? Perhaps apple will transition their silicon to remove/replace ARM's features overall in the future just so they have one less licensor to pay. But either way, they'll still be making their own chips - at least until the point that they are dramatically behind their competition.
Not only that, if we look at their other product lines (and even the Mac to a large extent), they have continued supporting their devices for far longer than other manufacturers. iOS 14, for example, still supports the iPhone 6S which was released 5+ years ago.