Apple Transitioning Macs to Custom Chips With 'Incredible' Performance Starting Later This Year

[smulji]

But, does that mean FULLY transition, or just that they'll have a full range of Apple Silicon models?

Good question. I took it to mean the former.

 

[djjeff]

Because in those days, Apple was the underdog, with a brighter future in the potential of compatibility, and being dependent on others to make chips that would fuel their laptops, etc.

Things are MUCH different today.

Today, Apple *could* just move forward, leaving all the legacy behind. We'd (a small group of us) hate that, but they could do it.

None of this answers the question. Apple was the underdog back then and moved to PPC. Why wouldn't they have moved to the vastly superior ARM processor instead? They're making the transition anyway so why go with the inferior option? Then, as with now, Apple has left the legacy behind. Apple has always left the legacy behind. Today is no different.

 

[Gerdi]

That's not what he said. He said "Software compatibility. " That's it, no qualifications. As for this being a "superior" architecture I repeat my question: Why now? Why not when they made the switch to PPC?

I did ask when you would do the switch if not now?

 

[smulji]

IMO, the look is positive (in the little I've seen)

I actually like the consistency in the UI across all of Apple's devices.

 

[SteveW928]

Exactly! This is PPC versus x86 all over again. The difference this time, at least at the time I write this is, we haven't seen any supporting data. People keep touting ARM as the best thing to hit the processor world. I have no doubt these processors will be competitive and, in some instances, faster than x64. Just like PPC was back in the day. But to listen to people tell it Intel is doomed. They may as well liquidate their processor business and focus on something else.

I wouldn't go quite that far (re: Intel), but in many ways the situation is kind of reversed. Back then, Apple was the underdog and Intel a leader. Today, Apple is the leader, and Intel more the underdog.

The real key, though, as pointed out before, is software development. How much of the world is tied to Windows legacy. In the business market, a big percentage still is. In the consumer market, it is mostly irrelevant. And, even in the business market, most of that dependence is just perceived.

 

[djjeff]

I did ask when you would do the switch if not now?

1995

 

[SteveW928]

None of this answers the question. Apple was the underdog back then and moved to PPC. Why wouldn't they have moved to the vastly superior ARM processor instead? They're making the transition anyway so why go with the inferior option? Then, as with now, Apple has left the legacy behind. Apple has always left the legacy behind. Today is no different.

No, because back then, Apple needed that legacy compatibility to survive. They no longer really do (as much at that would piss many of us off).

(And, but legacy compatibility, I'm not talking backwards support of previous Mac stuff so much as the 'legacy computing world' which was/is Windows.)

 

[Wando64]

Windows RT was windows 8.x built for arm32.

That it was a commercial flop is irrelevant. My point is that Microsoft started “releasing” “the real windows” (as opposed to the previous windows ce, windows mobile etc which were completely different systems) on arm, a decade ago, and still haven’t managed to ship native office for it.

My point is that they didn’t.
Windows RT is not “the real Windows“. It never ran “real windows” applications.
The first full Windows release on ARM was Windows 10 in 2017.

 

[djjeff]

I wouldn't go quite that far (re: Intel), but in many ways the situation is kind of reversed. Back then, Apple was the underdog and Intel a leader. Today, Apple is the leader, and Intel more the underdog.

Intel is no underdog. As for Apple leading that has yet to be demonstrated. Mac advocates are hyping the hell out of Apples yet to be released / benchmarked desktop processor. They're doing so based on comparisons between an iPad and desktop systems. IMO a completely worthless comparison given all the variables.

I am not against Apple making this transition. I assume they're doing so for good reason. However I find it very hard to believe Apple's desktop ARM processor is going to be oh so much better than x64. Maybe better for Apple but better overall? That has yet to be demonstrated. Since I've heard all of this before when they transition to PPC I think people need to be tempered with their expectations.
[automerge]1592946871[/automerge]

No, because back then, Apple needed that legacy compatibility to survive. They no longer really do (as much at that would piss many of us off).

(And, but legacy compatibility, I'm not talking backwards support of previous Mac stuff so much as the 'legacy computing world' which was/is Windows.)

I fail to see a distinction. PPC was no more backwards compatible with "which was/is Windows" than ARM is today.

 

[Gerdi]

None of this answers the question. Apple was the underdog back then and moved to PPC. Why wouldn't they have moved to the vastly superior ARM processor instead?

There was not a single high performance ARM implementation on the market at that time much less any Apple in-house design. Back in the days it was also not clear if ARM would become the leading architecture - this is importan because you invest in an ecosystem. The PPC ecosystem was shrinking as Apple was the single biggest user of PowerPC.
Today the situation is quite a bit different, as ARM is literally ubiquitous.

ps. Or are you talking about the switch from 68000 to PPC?

 

[seclusion]

Seemed like last year we just got everything on 64 bit... Geez.
However let's hope that some of the issues (like in Logic Pro maxing out one core) while the others are at 20% gets solved.
My 2017 iMac is chugging along just nice but maybe by 2022 there might be some sense of stability.
I would suspect that Apple must have a CPU that is superior to what they have access to with Intel. The 2 year roadmap I assume meant that all their Desktop products would then be sold with Apple Silicone CPU's exclusive.
Intel processors only for products that AApSil can't compete in. e.g. maybe the MacPro, iMac Pro etc. to cover until then.
We need a shortform Apple Silicone... APlSil?
App makers likely could do much testing on an iPad Pro now.
I'm hopeful the transition won't be insane, also I suspect the crew that support current OS's on unsupported Mac's will be digging in well after Apple stops supporting

 

[cmaier]

I assume you are referring to RISC here, am i right?

Yep, sorry!

 

[djjeff]

There was not a single high performance ARM implementation on the market at that time much less any Apple in-house design. Back in the days it was also not clear if ARM would become the leading architecture - this is importan because you invest in an ecosystem. The PPC ecosystem was shrinking as Apple was the single biggest user of PowerPC.Today the situation is quite a bit different, as ARM is literally ubiquitous.

ARM existed back then. Today we hear how superior ARM is. ARM is ARM is it not? So again I ask: Why did they go with PPC instead of the superior ARM?

 

[MyopicPaideia]

There's been an Office for ARM version since the Surface RT launched in 2012, and Office on the Surface Pro X is a native ARM build. Put down the cool-aid. It's not hard to get office to run on ARM MacOS when it already runs on WinRT.

As for Adobe, the Lightroom app they showed isn't the one professionals actually use with all the tools needed - that's Lightroom Classic. This is the port of Lightroom mobile, so it's basically the iOS version ahyways, and Photoshop already runs on iOS too. None of the apps which will actually present challenges, and NEED big performance (Premiere, After Effects) were shown or mentioned, and the ommission of real Lightroom I think is a clear indicator it's not as smooth as portrayed.

Throw in that most of us in this industry rely heavily on plugins and codec add-ons from smaller devs, and it doesn't matter if Adobe and Apple are both onboard (with the former being questionable) if we can't get our plugins. It's a chicken and egg problem - these devs can't afford to support both platforms, and won't support ARM until we switch... but we won't switch until they support it.

The Surface Pro X runs an emulated x86 32 bit version of MS Office...why?

 

[cmaier]

There was not a single high performance ARM implementation on the market at that time much less any Apple in-house design. Back in the days it was also not clear if ARM would become the leading architecture - this is importan because you invest in an ecosystem. The PPC ecosystem was shrinking as Apple was the single biggest user of PowerPC.
Today the situation is quite a bit different, as ARM is literally ubiquitous.

It could just as easily have been mips or SPARC or PowerPC or pa-risc or power or Alpha that we ended up with today instead of Arm as the leading architecture. It all came down to business methods and licensing policies. So much in technology happens for reasons other than technical merit. It’s really all quite fascinating.

 

[Gerdi]

It could just as easily have been mips or SPARC or PowerPC or pa-risc or power or Alpha that we ended up with today instead of Arm as the leading architecture. It all came down to business methods and licensing policies. So much in technology happens for reasons other than technical merit. It’s really all quite fascinating.

Absolutely agree. Arguably Alpha was by far the most advanced architecture of its time. And you see many features from Alpha were introduced later in ARM - like load-link-store-conditional or the weak memory model.

 

[djjeff]

It could just as easily have been mips or SPARC or PowerPC or pa-risc or power or Alpha that we ended up with today instead of Arm as the leading architecture. It all came down to business methods and licensing policies. So much in technology happens for reasons other than technical merit. It’s really all quite fascinating.

SPARC, like ARM, is a specification and can be licensed. SPARC is RISC. MIPS was RISC. PPC was RISC. Alpha was RISC. If RISC is so superior why is it that most of them are gone in favor of x64? SPARC is the only one on that list left and I don't even think Oracle is developing it further, only Fujitsu. One might argue PPC is still alive in IBMs POWER processors but those are not the same. What is it about ARM which makes it so superior to all of these other RISC processors which are, for the most part, gone?

 

[sirozha]

Indeed, it is not yet confirmed so i can give no guarantees apparently. I can tell you 2 things:
1) XBox app works on Windows ARM including game streaming
2) Windows ARM runs on a Linux VM (KVM) for instance on the Raspberry PI4.

So it is mostly up to Apple at this point. And there could be more issues potentially as for instance the question if there will be GPU and other drivers for Apple HW for Windows?

Windows for ARM runs on the raspberry Pi, but the steaming from the
Xbox One doesn’t work on the Raspberry Pi. I have no other ARM platform to test it with.

 

[Gerdi]

Windows for ARM runs on the raspberry Pi, but the steaming from the
Xbox One doesn’t work on the Raspberry Pi. I have no other ARM platform to test it with.

I just did test it on my Surface Pro X - working. Did you try to run it on PI 4? What happens?

ps: I am using the Xbox console companion app.

 

[sirozha]

I just did test it on my Surface Pro X - working. Did you try to run it on PI 4? What happens?

ps: I am using the Xbox console companion app.

Refresh rate is one frame per second.

 

[Gerdi]

Refresh rate is one frame per second.

Ah yeah, thats what can happen if there is no GPU/video driver for your device. Maybe it is also a memory problem on top, since 1fps is still slow even in the absence of a video driver. Do you have the 4Gbyte version of the PI4?

Thinking about this, i consider it not very likely that Apple will write video drivers for Windows. Of course if they are ambitious and want to have the best Windows Laptop on top of having the best MacBook they might - at least financially it would be a peanut investment for Apple.

 

[Alan Wynn]

Cinebench for me personally is an important example, since my company uses the engine extensively. Of course we’ll want to use the native version, I just hope it’ll exist.

Given that they are currently on macOS, and consider the Mac market to be important for them, I expect they will port. Again, unlike Windows, Apple has made it clear that all their machines will be ARM, so if they want to stay, they will need to be on ARM. Again, for companies that are on Apple’s tool chains and use Metal, porting will be very easy.

 

[SteveW928]

Intel is no underdog. As for Apple leading that has yet to be demonstrated.
...
I fail to see a distinction. PPC was no more backwards compatible with "which was/is Windows" than ARM is today.

Well, in the big picture they (Intel) are starting to become an underdog considering Apple's success with iOS devices and competition from AMD.

Apple is leading, just not in the segment of the world Intel still holds. I suppose you're right to consider Apple's success in that realm until it is proven. We'll see.

re: backward compatible - I agree. The point is that backward compatible was highly needed back then, where as it isn't now (at least in the big picture).

... The 2 year roadmap I assume meant that all their Desktop products would then be sold with Apple Silicone CPU's exclusive. ...

As previously mentioned, I'd question that exclusive aspect. I see no reason Apple couldn't have a full-line of Apple Silicon machines (even Mac Pro), while still maintaining a segment of Intel-based machines (probably pro models) for people who still need x86 performance.

They'd still have fully transitioned in 2 years, just not exclusively (which wasn't claimed). For the people who don't need x86, quite possibly, they'd get benefits from picking the Apple Silicon route.

ARM existed back then. Today we hear how superior ARM is. ARM is ARM is it not? So again I ask: Why did they go with PPC instead of the superior ARM?

Because they needed x86 compatibility to survive in the market reality of that time. They also didn't have the clout and expertise to force things to happen with IBM/Motorola.

... If RISC is so superior why is it that most of them are gone in favor of x64? ... What is it about ARM which makes it so superior to all of these other RISC processors which are, for the most part, gone?

IMO, software/compatibility. x86 just isn't as important any longer (in the big picture, I personally need it).

 

[Alan Wynn]

It’s very clear from the wording used at things like the SOTU that Apple is making a big push into gaming.

My best guess, we’re going to see an AppleTV using Apple silicon that is every bit on par with the next gen consoles in 3ish years. They will tout that you’ll have the ability to run any apple ecosystem game on any device.

I think you might be right here. They have a potentially large advantage that they do not have to pay anyone’s profit on their silicon, so they can build a competitive console at a lower cost, especially if they make 30% on games delivered to it.

 

[ssgbryan]

As others have said... if they take the tech and are able to apply it to desktop designs. This is out of my wheelhouse, for sure, but some seem to think they can do it. It isn't about a jumped up iPad chip.

Why do you not think they could make 20-core/40 thread system, for example? Maybe there is some reason, but I don't think AMD or Intel have some secret Apple can't have.

Isn't part of the problem that they HAVE either not had the resources to do their own thing (early days) or have been held back by the advancement of others. When they broke those ties on the phone, they've blown others away. I guess the *assumption* is they might be able do that on the desktop, too. Maybe, maybe not, but I haven't really heard good reasons why they won't be able to.

AMD and Intel have built and shipped multi-threaded CPUs - Apple hasn't. AMD and Nvidia have shipped discreet GPUs - Apple hasn't. Intel will beat them to the punch also.

I am not saying that they couldn't build a 20-core/40 thread system - (although I would point out that they don't have any multi-threaded SoC at this point. AMD has developed 4 way SMT, that will end up in either Zen 4 or Zen 5. Most folks here wouldn't have any use for a 16 core/64 thread computer, but I certainly do).

They could - I just have no confidence in ability - Apple hasn't been able to walk and chew gum for a while now.