The ARM chip was so fast back then that a stunning 3D game called 'Zarch' was written in BBC BASIC!
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.
The new ARM based Mac Mini
- Thread starter dogslobber
- Start date
- Sort by reaction score
The ARM chip was so fast back then that a stunning 3D game called 'Zarch' was written in BBC BASIC!
Yup.
The archimedes was ahead of its time, and was the original machine that ARM based CPUs were developed for.
The OS it ran, RiscOS was also very ahead of its time.
https://en.m.wikipedia.org/wiki/History_of_RISC_OS
Last edited:
No.
There is no point in producing an ARM processor that will be as powerful as Intel's offerings. If you do that to an ARM chip then you're going to end up with the same things that we already have. There might be though some advantages for Apple, like increased margins and better vertical integration, but the user will ultimately end up with a machine that is not fully compatible much of the software available today. So, no, I don't need anything like that.
There is no point in producing an ARM processor that will be as powerful as Intel's offerings. If you do that to an ARM chip then you're going to end up with the same things that we already have. There might be though some advantages for Apple, like increased margins and better vertical integration, but the user will ultimately end up with a machine that is not fully compatible much of the software available today. So, no, I don't need anything like that.
HobeSoundDarryl
macrumors G5
As is, Macs are the only "off the shelf" option with a foot in both worlds: MacOS and Windows. Anyone loving Apple but needing something that only runs in Windows can kill 2 birds with one stone by choosing a Mac. If that person is trying to justify the "Apple Premium", this 2 birds concept probably helps.
Flash forward to these hypothetical new Macs that eject Intel and embrace ARM. With that goes bootcamp. And with that every "2 Birds" buyer/user finds themselves having to go back to owning 2 machines instead of one. Since Windows is often more about need (as in need to be able to run <software only for Windows>), in some purchase contest, Windows machines probably win most of the time... because need trumps want. And what the Mac business doesn't need is people who would otherwise buy a Mac needing to buy a Windows computer instead and putting their money there. Does the average Joe buy a Windows machine because they have to have one and then buys much of the same hardware in an Apple machine because they want an Apple too? Or does he/she try to make a go of it with the computer they had to buy?
Similarly, many who seem so passionate about this seem to forget how it was in the transition from PPC to Intel. Yes, Rosetta helped smooth the pain but there is no guarantee of a Rosetta-type solution for Intel-to-ARM. And Apple didn't support Rosetta for very long, which deprecated a lot of useful software that was not updated for Intel. Sure, this crowd around here blamed the developers (because it's always the non-Apple entity's fault
) but in some cases the developers were dead or the Mac market was just not valuable enough to go to the trouble.
Users who needed some of that software had to resort to workarounds, often dual boot options so they could still use some of that software AND march forward with updates to the core OS. But dual boot options won't work if Intel guts become ARM guts. In that scenario, one need to keep an "old" Intel Mac around for software that doesn't migrate to ARM and/or bootcamp.
And since bootcamp would probably no longer get much support on ARM macs, it's only a matter of time until such people need a Windows computer too. In this scenario, a working Mac environment is going to need up to 3 machines to do what one can do now.
But "we" keep thinking we want this ARM transition anyway. Apparently, we believe Apple needs to stick it to Intel and this is the obvious way to do it. Of course, that locks us Apple people even more into Apple... but we don't care. And even in this short thread, we're already spinning rationalizations that we know will NOT come to pass. For example, do we really think Apple would eject Intel, inject ARM and significantly lower the price of Macs? Or would they make the change, hold or raise pricing and then enjoy the added profits? But we'll spin lower price like it's an automatic if ARM replaced Intel when we should easily recognize that it is NOT automatic.
Obviously, I'm no fan of the idea. As a consultant, it's sooooo nice to take just one computer in the bag and be ready to link into Mac or Windows environments with the same machine, run Mac or Windows software, etc. Macs going ARM would kill that great benefit, probably pressing business people like me to carry TWO machines to be ready for about anything.
If my main driver is profit maximization (for Apple or AAPL), I might be able to get into it. But since my main driver is consumer utility, I don't see the big gains for us consumers except in fantasy- like lower prices. We grab onto some calculation scores that imply ARM > Intel, ignoring the other scores where it is not... and then let marketing spin imply that ARM > Intel in every way... that the phone is more powerful than your Mac in every way now, when you must know that is not the case. Try to do any big, Mac-intensive job on the phone and see how that "just as powerful" system does.
The only win I foresee in such a transition is more profit per unit sold for Apple and greater control of when Apple would like to roll out new Macs. The losses are likely lots of useful software, bootcamp, raw power for niche tasks, the unsteady but consistent march of CPU progress driven by Intel, the painful transition period where not-quite-compatible software is running in a Rosetta-like bandaid (if such a thing is even rolled out), etc. It would also give Apple enormous marketing spin room to claim Macs are superior to Intel-based machines with no clear & obvious head-to-head way to prove it. Remember this?
Cute commercial. Classic Apple. But not very long thereafter, Apple is on stage spinning the Intel transition and claiming most powerful Macs ever. ARM Macs get back to that spin-based superiority instead of tangible (same) guts vs. (same) guts superiority contests. In short, as is Macs have to be engineered to be competitive with (the same guts in) PCs because there is no smoke & mirrors of claiming a wholly different brain. Do we want to go back to that smoke & mirrors scenario where marketing spin is mostly casting Mac superiority?
"We" think we want this but "we" appear to forget how things were from PPC to Intel and/or "we" function almost entirely on only Apple-made software with little dependency on third party software. It's not automatically a cheaper Mac world NOR a better Macs world in such a transition... just better for Apple profitability and control. Consumers lose up to a lot in such a switch... especially those who use their Macs with software NOT made by Apple themselves.
Flash forward to these hypothetical new Macs that eject Intel and embrace ARM. With that goes bootcamp. And with that every "2 Birds" buyer/user finds themselves having to go back to owning 2 machines instead of one. Since Windows is often more about need (as in need to be able to run <software only for Windows>), in some purchase contest, Windows machines probably win most of the time... because need trumps want. And what the Mac business doesn't need is people who would otherwise buy a Mac needing to buy a Windows computer instead and putting their money there. Does the average Joe buy a Windows machine because they have to have one and then buys much of the same hardware in an Apple machine because they want an Apple too? Or does he/she try to make a go of it with the computer they had to buy?
Similarly, many who seem so passionate about this seem to forget how it was in the transition from PPC to Intel. Yes, Rosetta helped smooth the pain but there is no guarantee of a Rosetta-type solution for Intel-to-ARM. And Apple didn't support Rosetta for very long, which deprecated a lot of useful software that was not updated for Intel. Sure, this crowd around here blamed the developers (because it's always the non-Apple entity's fault
) but in some cases the developers were dead or the Mac market was just not valuable enough to go to the trouble.Users who needed some of that software had to resort to workarounds, often dual boot options so they could still use some of that software AND march forward with updates to the core OS. But dual boot options won't work if Intel guts become ARM guts. In that scenario, one need to keep an "old" Intel Mac around for software that doesn't migrate to ARM and/or bootcamp.
And since bootcamp would probably no longer get much support on ARM macs, it's only a matter of time until such people need a Windows computer too. In this scenario, a working Mac environment is going to need up to 3 machines to do what one can do now.
But "we" keep thinking we want this ARM transition anyway. Apparently, we believe Apple needs to stick it to Intel and this is the obvious way to do it. Of course, that locks us Apple people even more into Apple... but we don't care. And even in this short thread, we're already spinning rationalizations that we know will NOT come to pass. For example, do we really think Apple would eject Intel, inject ARM and significantly lower the price of Macs? Or would they make the change, hold or raise pricing and then enjoy the added profits? But we'll spin lower price like it's an automatic if ARM replaced Intel when we should easily recognize that it is NOT automatic.
Obviously, I'm no fan of the idea. As a consultant, it's sooooo nice to take just one computer in the bag and be ready to link into Mac or Windows environments with the same machine, run Mac or Windows software, etc. Macs going ARM would kill that great benefit, probably pressing business people like me to carry TWO machines to be ready for about anything.
If my main driver is profit maximization (for Apple or AAPL), I might be able to get into it. But since my main driver is consumer utility, I don't see the big gains for us consumers except in fantasy- like lower prices. We grab onto some calculation scores that imply ARM > Intel, ignoring the other scores where it is not... and then let marketing spin imply that ARM > Intel in every way... that the phone is more powerful than your Mac in every way now, when you must know that is not the case. Try to do any big, Mac-intensive job on the phone and see how that "just as powerful" system does.
The only win I foresee in such a transition is more profit per unit sold for Apple and greater control of when Apple would like to roll out new Macs. The losses are likely lots of useful software, bootcamp, raw power for niche tasks, the unsteady but consistent march of CPU progress driven by Intel, the painful transition period where not-quite-compatible software is running in a Rosetta-like bandaid (if such a thing is even rolled out), etc. It would also give Apple enormous marketing spin room to claim Macs are superior to Intel-based machines with no clear & obvious head-to-head way to prove it. Remember this?
"We" think we want this but "we" appear to forget how things were from PPC to Intel and/or "we" function almost entirely on only Apple-made software with little dependency on third party software. It's not automatically a cheaper Mac world NOR a better Macs world in such a transition... just better for Apple profitability and control. Consumers lose up to a lot in such a switch... especially those who use their Macs with software NOT made by Apple themselves.
Last edited:
Absolutely not.
But for a Kaby Lake i7 quad core with user upgradable memory and HDD at the usual quad core price point....I'd probably break a finger pulling out my wallet so fast.
Not that we'll ever see either option from Timmy's Apple.
[doublepost=1507072478][/doublepost]
Not sure why you need the rPI form factor for a home server, but there are picoITX form factor motherboards in that are only a bit larger. It is a motherboard though, so you need the rest of the computer including a power supply. For example here's a celeron-based model. https://www.digikey.com/product-detail/en/advantech-corp/MIO-3260L-S8A1E/MIO-3260L-S8A1E-ND/6011316 They also have Atom and AMD boards in that form factor.
If you're okay with a 6.7" square board, miniITX opens a world of powerful computers with socketed CPUs.
Personally though if I were shopping for me, I'd go for one of the Kaby Lake NUCs. Like this https://www.newegg.com/Product/Product.aspx?item=N82E16856102180. 4.5" square, 1.4" thick for the entire computer. And much cheaper than a base model mini. And much more powerful than a top end mini.
Last edited:
what is this 'lower price' that you speak of? This concept is unknown in AppleLand
yes, there is a point.
It puts Apple in total control of their own destiny as far as what goes into the processor, as they’re building their own design with it, not taking whatever intel put out.
It means they can gain advantages (as they have in the mobile space) through improved performance, better power consumption, immunity to x86/x64 malware, etc.
Switching to ARM would also give them economy of scale via using the same family of processors as their iOS platform.
The need to run x86 software is becoming less and less relevant as time moves on, and ARM processors are getting to the point where they’d be fast enough to emulate x86 for the few niche applications people want to run in any case.
As far as owning 2 machines goes, the main reason I have a PC and a mac is games. Boot-camp isn’t good enough for decent gaming performance because Apple has thus far been unwilling to ship a machine with a decent GPU in it at a reasonable price, and the ones they have shipped are not upgradeable.
I think ARM Macs are inevitable at this point. However, they won't be running MacOS, but rather a more feature rich version of iOS.
We're a few years off yet, but it is coming.
macos already is a more feature rich version of ios (by way of ios basically being macos with a different gui and sandboxing).
i don’t think it will change much. they’ve already got sandboxing via app store apps.
there is definitely a market for apple devices without all of the ios limitations and this is what they will be aimed at.
i don’t think it will change much. they’ve already got sandboxing via app store apps.
there is definitely a market for apple devices without all of the ios limitations and this is what they will be aimed at.
Yes, but I was referring to something like that with an ARM. (Also, the post was 2 years old)
[doublepost=1507104227][/doublepost]
Pedant point: Arthur was the original OS and wasn't really WIMP driven (and wasn't very good) - it did include a WIMP API in ROM but the only desktop front-end was a BASIC program on disk that felt more like a demo. The whole thing felt like a super-fast BBC Micro with some cool demos rather than a workstation.
RISC OS was the next version of Arthur, but was a vast improvement and really introduced WIMP as the primary UI. The re-brand was probably because of that flak directed at Arthur.
Later, there was also a RISC-IX (Unix) workstation and various free BSD and Linux OSs
I would just like to point out that Microsoft have a fully x64 windows 10 running on arm (snapdragon) now so the arguement about no bootcamp is not really valid.
Impressive. Not so sure that it signals the death of bootcamp, but it is a good proof-of-concept that you can port an x86 OS to ARM and still seamlessly run x86 software.
...but the issue is going to be running legacy Windows x86 programs: I don't think people use Boot Camp or Parallels because they just love the Windows UI and the riches of the MS App Store... and if someone is using bootcamp its usually because they want full performance.
Now, that demo showed x86 Photoshop running (apparently) fine under emulation* - which is nice - but loading up a screen-res photo and applying a swirl filter isn't exactly a torture test. The game didn't look particularly cutting-edge, either. Doesn't mean that the performance will satisfy someone who uses bootcamp because (say) they have a bit of Windows pro graphics software they want to use.
But yeah, like most of the arguments against ARM Macs, the binary compatibility issue is getting less significant with every passing year. I doubt that Windows support is top of Apple's priority list now (although it was a big deal ten years ago).
(*unless Photoshop for Windows is .net based these days, which uses a Java-like virtual machine).
Except that there are no ARM win32 binaries available. People don't run Bootcamp and boot into Windows to run Windows, they do it to run Windows software.
Hell, you could get NT4 versions for PowerPC, didn't help anyone run Windows software on PowerPC Macs.
Hell, you could get NT4 versions for PowerPC, didn't help anyone run Windows software on PowerPC Macs.
Arthur was a WIMP driven shell on top of BASIC but was superfast as it was all ROM. It was the 1980s equivalent of Windows 3 which is nothing more than a graphical shell over MS-DOS. Nonetheless, it was still a useful add on accessed via '*desktop' and had some awesome apps, was reasonably multitasking (cooperative).
RISC-OS took the Arch to a whole new level but it was always playing catch up to the likes of the Atari ST which had been running the Tramiel O.S. since 1985. A case of the brits eventually catching up with the USA.
One of us is misremembering things - pretty sure that Arthur was originally the whole OS for the Arc (a souped-up version of the BBC Micro OS), and while the WIMP API was all in ROM, the actual desktop application was very famously written in BASIC and never got much use until RISC-OS came out.
I had an Atari ST ("Jackintosh") before getting an Archimedes (I think a lot of BBC users did while waiting for Acorn to build a system around their wonder chip). The ST was a lot of bang for the buck, but nothing clever hardware- or software- wise. Tramiel OS was just the GEM GUI shell running over a CP/M 68k clone. Probably its main hardware innovation was including a MIDI interface so it became the standard system for music. Also, the Atari BASIC was rubbish (software issue, not hardware) - my dual-6502 BBC-B setup was faster at BASIC - on the ST if you wanted performance you had to use C, or get Computer Concepts' BBC BASIC clone on ROM.
Nah. The contest for most sophisticated end-of-the-80s computer has to be Archimedes vs. Amiga (and, though it sticks in my throat, the Amiga probably wins on hardware-assisted graphics and fully preemptive multi-tasking OS) but, hey, the Archimedes was the first commercial outing for the ARM, so it wins history
That said, having programmed on both ST and Arc, the WIMP system on the Archimedes was very, very obviously inspired by GEM (the way you polled the system for events, and teh events you got back were virtually identical) and quite unlike Mac or Windows (Acorn had a dual 6502/80286 system running GEM in the works at one point, so they were familiar with it). However, the way RISC-OS handled saving/loading files and inter-application data transfer (TLDNR: there was no distinction) was pretty radical, plus it had an incredible anti-aliased outline font rendering system. The first WIMP-based WP and DTP software was a port of Timeworks from GEM - and the inter-app drag & drop along with the smooth font rendering made that very impressive.
Yup. Also, another thing is that now Apple are building their own GPUs, building their own custom processors with Apple will enable them to use these GPUs in their Macs as well.
[doublepost=1507172393][/doublepost]
Windows RT had the problem that vendors paired ARM CPUs with garbage other hardware. Bad screens, SDcard storage, etc.
Also, to be blunt, running Windows on Mac hardware is fairly niche and become even less relevant, as per posts above.
Looking at the latest iDevices with their multi-core architecture, I'd not expect an either-or, but an as-well-as situation in the beginning. Arm chips will supplement Intel CPU's, becoming active for low-demand activities while saving power. Intel CPU will kick in for performance when required.
System will slowly be migrated to ARM, until - after some time - roles will reverse and ARM will do the heavy lifting, while a lower-grade Intel will be there for compatibility reasons. Once the transition phase is sufficiently progressed, lower-spec Macs will go full ARM and Intel will be only available in higher-grade Macs, perhaps even as an option only.
System will slowly be migrated to ARM, until - after some time - roles will reverse and ARM will do the heavy lifting, while a lower-grade Intel will be there for compatibility reasons. Once the transition phase is sufficiently progressed, lower-spec Macs will go full ARM and Intel will be only available in higher-grade Macs, perhaps even as an option only.