4 ms·

Answer is simple, explanation is complex. Wait, unless you need to run virtual Windows. x86 always sucked. It was always a dog. Everything that ever competed wi

Answer is simple, explanation is complex. Wait, unless you need to run virtual Windows. x86 always sucked. It was always a dog. Everything that ever competed with x86 was technically superior to x86, including 68k, PPC, and ARM will be no different. If you have some burning need to run WinVMs, then you need x86. Otherwise, even this first gen Apple ARM hardware will be better than the Intel Macs in performance, and this even in Rosetta2 emulation, and this has been proven by the developer kit using 2 year old processor competing with today's x86. When Apple releases their first ARM macs, they will be cutting edge, will see faster and far more efficient processing, thus expect battery life in ARM MacBooks that any x86 MacBook or AnyBook can not compete. As far as the "don't buy first gen Macs" folks are concerned, they are stuck in 1994. Apple is a different company now. Except for minor issues like avant garde butterfly keyboards that they're abandoning, the internal hardware has no competition, and you're going to pay for that, but it is not actually more expensive than any hardware comparatively configured. So save your money, get another year out of your x86 Mac, and get the ARM Mac. Otherwise, you'll have 2 gens of x86 Macs, and one will gather dust. Wait for the ARM, and you will continue to use your x86 Mac.

spronkey 6y ago
Yeah, the DTK hasn't exactly shown that Apple Silicon can outdo Intel, let alone AMD. I remain rather skeptical. I think this is a business win for Apple moreso than a technical win.
rsecora 6y ago
Osborne effect. The Osborne effect is a social phenomenon of customers canceling or deferring orders for the current soon-to-be-obsolete product as an unexpected drawback of a company's announcing a future product prematurely.
- catmistake 6y ago
  > Osborne effect. Totally. Don't buy either, Apple is obviously going under. >< Thing about Macs is, they have this incredible lifetime, annoyingly (to those that like to save with buying older hw) hold high value in resale. Try pricing a Mac Mini from 2012, or Late 2014. Compare a similar compact WinDesk from those years. Though it is interesting to see a pattern, and moreso when it is famous enough to have a name, I do not think the Osborne effect applies here. Today's current Intel Mac will still be unaffordable NIB or used when the 3rd gen ARM Macs are released. I'm not really happy about that.
  - rsecora 6y ago
    Right... And the previous transitions in early 2000s was eased by the usage of universal binaries with powerpc and intel code. Same will apply for the universal binary 2 with intel + arm code.
    - catmistake 6y ago
      1994 68k --> PPC 2006 PPC --> x86 As I recall, the first switch had the issue of the early PPC chips, 601, was kind of slow, and the boxes not designed well. The second transition in 2006 was annoying because Adobe doesn't like to rewrite their code... they carbonized, and then just kept charging for the same cruft. But frankly, I liked the Classic Environment, and I liked the first Rosetta. I'd like to run the whole shebang in MAME on whatever hw, if others were doing it to make it easier for me. If only MAME had a emulated coprocessor, I'd run A/UX on it. Apple's current ARMs and OS running on it are already quite amazing, IMHO. So I don't think the new ARM Macs could disappoint. And the same thing here, Universal Binaries, a switch in the compiler to make it easy for devs, and I don't know how it could be easier.
      - klodolph 6y ago
        > As I recall, the first switch had the issue of the early PPC chips, 601, was kind of slow, and the boxes not designed well. I had a first-generation PPC Mac. It was pretty good, I thought. The baseline 601 ran at 60MHz and had a superscalar architecture with 16K instruction / 16K data L1 cache. The high-end 68Ks were not superscalar, ran at 33MHz, and had 4K instruction / 4K data caches. If anything was slow it was probably because big chunks of code were 68K. My personal experience was that once I switched to PPC, running the same software on 68K felt slow by comparison.
        catmistake 6y ago
        That's interesting. I didn't get to touch a 601 until the 604 were common, side by side in a commercial environment, the older machines gunked with more use (old installs) and more users over time, thus my opinion is biased and worthless. Thank you for posting.
fortran77 6y ago
68k and PPC were not technically superior to x86. Why do you think Apple switched?
- mixmastamyk 6y ago
  Manufacturing fabs were a big factor in both transitions.

spronkey 6y ago

Yeah, the DTK hasn't exactly shown that Apple Silicon can outdo Intel, let alone AMD. I remain rather skeptical. I think this is a business win for Apple moreso than a technical win.

rsecora 6y ago

Osborne effect. The Osborne effect is a social phenomenon of customers canceling or deferring orders for the current soon-to-be-obsolete product as an unexpected drawback of a company's announcing a future product prematurely.

catmistake 6y ago
> Osborne effect. Totally. Don't buy either, Apple is obviously going under. >< Thing about Macs is, they have this incredible lifetime, annoyingly (to those that like to save with buying older hw) hold high value in resale. Try pricing a Mac Mini from 2012, or Late 2014. Compare a similar compact WinDesk from those years. Though it is interesting to see a pattern, and moreso when it is famous enough to have a name, I do not think the Osborne effect applies here. Today's current Intel Mac will still be unaffordable NIB or used when the 3rd gen ARM Macs are released. I'm not really happy about that.
- rsecora 6y ago
  Right... And the previous transitions in early 2000s was eased by the usage of universal binaries with powerpc and intel code. Same will apply for the universal binary 2 with intel + arm code.
  - catmistake 6y ago
    1994 68k --> PPC 2006 PPC --> x86 As I recall, the first switch had the issue of the early PPC chips, 601, was kind of slow, and the boxes not designed well. The second transition in 2006 was annoying because Adobe doesn't like to rewrite their code... they carbonized, and then just kept charging for the same cruft. But frankly, I liked the Classic Environment, and I liked the first Rosetta. I'd like to run the whole shebang in MAME on whatever hw, if others were doing it to make it easier for me. If only MAME had a emulated coprocessor, I'd run A/UX on it. Apple's current ARMs and OS running on it are already quite amazing, IMHO. So I don't think the new ARM Macs could disappoint. And the same thing here, Universal Binaries, a switch in the compiler to make it easy for devs, and I don't know how it could be easier.
    - klodolph 6y ago
      > As I recall, the first switch had the issue of the early PPC chips, 601, was kind of slow, and the boxes not designed well. I had a first-generation PPC Mac. It was pretty good, I thought. The baseline 601 ran at 60MHz and had a superscalar architecture with 16K instruction / 16K data L1 cache. The high-end 68Ks were not superscalar, ran at 33MHz, and had 4K instruction / 4K data caches. If anything was slow it was probably because big chunks of code were 68K. My personal experience was that once I switched to PPC, running the same software on 68K felt slow by comparison.
      - catmistake 6y ago
        That's interesting. I didn't get to touch a 601 until the 604 were common, side by side in a commercial environment, the older machines gunked with more use (old installs) and more users over time, thus my opinion is biased and worthless. Thank you for posting.

fortran77 6y ago

68k and PPC were not technically superior to x86. Why do you think Apple switched?

mixmastamyk 6y ago
Manufacturing fabs were a big factor in both transitions.