Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Mac Pro's are no where even close to 1M. 0.1M maybe
Right, that’s actually why I said perhaps FAR less than a million. 0.1M is, indeed FAR less than a million.

Apple sales are only in the 20's M (measuring the pandemic mania buying levels as some 'norm' is dubious.)
I didn’t say it was the norm. There exists, in reality, a year where analysts estimated Apple sold 28 million Macs. As it’s been done once, it could recur.. So, as 28 is so close to 30, indicating that Apple may sell somewhere between 20-30 million Macs in a given year is accurate… unless the analysts come back and estimate that they actually sold 31 or 19 million. :)
 
This is why more corporations still predominantly order computers that have Intel cpu’s over AMD.
Yet another reason why Intel can claim to be on the cusp of producing the coldest processor this side of sliced ice, if they go to the vendors saying, “Hey, um… yeah it’s not as cool as we thought and consumes more power than we expected… it’s a dumpster fire, really. So, how many can we put you down for?” they’re all still going to place those orders. :)
 
  • Haha
Reactions: Detnator
You are aware that ARM is almost as old as x86, right?
Ah, but it’s what you do in the intervening time that matters. Do you:
a) continue to support all manner of edge cases and poorly compiled code in order to ensure you don’t lose any of your legacy customers such that your decoder HAS to take up about 20% of the processor? OR
b) dump legacy cruft at every opportunity in order to ensure that your processor doesn’t have to do anything but execute the code provided?

Intel is free to dump legacy cruft, they even tried it once with Itanium. However, their overriding goal will ALWAYS be backwards compatibility, so, while Intel MIGHT have the capability to ship a cool and fast 64-bit only Intel processor that ACTUALLY rivals Apple Silicon, they know they don’t have to, because that’s not what their market wants. More than anything else, they want to be able to “run all of yesteryear’s code” not “last all day on battery” or “run cool”. They’ll always spin a good yarn, they’ll always miss their goals and it won’t matter.

In 3 years, x86 ecosystem will still have orders of magnitude more business applications than ARM. Support for legacy code is paramount for business.
Absolutely no question, no doubt and businesses have GOT to LOVE that fact. And, the x86 ecosystem, as a result, will still be less efficient than the stuff Apple’s producing. Garbage in, garbage out.
 
  • Like
Reactions: 2Stepfan
As of now: All speculation. No one knows ...or WILL KNOW..until Apple releases its product. End.
We do have a strong indicator from, of all sources, Apple, though. Anyone that looks at, say, the base configurations of the Macs shipping with slower drives and a HomePod with an older version of WiFi coming out of Apple (those products that are likely expected to sell in the millions) can’t think that the SAME Apple is going to create a one-off, esoteric, incompatible with the code Apple’s been telling developers to write for the last few years configuration for a system that will likely sell less than 1 million units across it’s entire lifetime.
 
Last edited:
  • Like
Reactions: 2Stepfan
In 3 years, x86 ecosystem will still have orders of magnitude more business applications than ARM. Support for legacy code is paramount for business

Business/government do take a long time to change technologies, demonstrated by the Cobol crisis a few years ago.

But in terms of unit sales it is a different story:

“By 2026 — not 2050, but 2026, four years from now — half of the cloud processors will be ARM-based, and 30% of PCs will be ARM-based."

Maybe optimistic, but could happen particularly if the Microsofts' license problems get resolved and ARM is let out of its cage.

 
  • Like
Reactions: aytan
Apples current strategy works wonders for laptops and all-in-one integrated solutions and they will keep it like that forever if they can. I think they learned a lot from Studio experiment and they will improve upon it because Studio still doesn't play well with many things you throw at it. It's not as versatile $10K machine as it should be, just yet.

Downside of Apple's scaled up mobile chips is that they will never be a true alternative to x86 workstations. I'm sorry but that just how it is. Both Intel and AMD already have leverage in workstation CPUs and within two years both will come up with a leap generation in this segment. I am not sure how Apple can answer to that unless they come up with some super scalable stacking.

One interesting fact that probably few people know is that Apple designs in-house their x86 motherboards and it's one the only PC vendor in the world that does so. All other vendors rely on a few outsourced x86 motherboard design companies.
 
Business/government do take a long time to change technologies, demonstrated by the Cobol crisis a few years ago.

But in terms of unit sales it is a different story:

“By 2026 — not 2050, but 2026, four years from now — half of the cloud processors will be ARM-based, and 30% of PCs will be ARM-based."

Maybe optimistic, but could happen particularly if the Microsofts' license problems get resolved and ARM is let out of its cage.


That all depends what you count as a PC after all. You could say iPad is a PC as well and so is a Chromebook. Sure. That generalization standard has not been set yet. I do not count Chromebooks as PCs at all. I've never counted Windows netbooks as PCs either since they had a hard time running Word let alone Excel or PowerPoint.

There is a market for ARM based notebooks and tablets sure. Will they deduct from PC sales? Sure, just like mobile phones and tablets did. But are they PCs? No. Maybe better analogy is electric scooter vs a bicycle. Did scooters take a chunk out of bicycle sales? They sure did. Are they the same thing as a bicycle? Nope.

Maybe sometime in the future there will be a strong and powerful ARM chip that will run Windows machines at the same speeds as current x86 CPUs and GPUs but that is not coming out anytime soon. AMD and Intel will soon be on 2nm nodes which should mean performance will be carried over while efficiency will improve drastically.
 
Maybe sometime in the future there will be a strong and powerful ARM chip that will run Windows machines at the same speeds as current x86 CPUs and GPUs

The future may not be that far away ....

"For the first time, an ARM-based SoC has performance closer to an Intel Core i5 chip."

 
Maybe sometime in the future there will be a strong and powerful ARM chip that will run Windows machines at the same speeds as current x86 CPUs and GPUs but that is not coming out anytime soon. AMD and Intel will soon be on 2nm nodes which should mean performance will be carried over while efficiency will improve drastically.
Pro workstation is one battle Apple already lost. They went all in on Apple Silicon without considering all the ramifications. Or maybe they did and decided pro workstations are one segment of the Mac market they can afford to lose. As iPhone sales slows, a lot of the ill-advised decisions Cook made now started to come back to bite him.

Most of the major projects he started haven't yielded anything tangible and likely won't for the foreseeable future, e.g., Apple Glasses, Apple Car, 5G chip, and the delayed ASi transition. This is not to mention getting Apple deeply entangled in China.

ASi has great potential in gaming, but Cook decided it's not worth the effort. Apple's pitiful gaming platform, Apple Arcade, is a case in point. Another is its half-hearted push for the Metal engine.

Siri feels like it's never come out of beta and HomeKit is in tatters (just check all the complaints on Reddit). But Cook never deemed it important enough to fix. Should I go on?

If you analyze Apple's revenue stream, you will soon come to the realization that without another blockbuster product soon, it will become a phone/service company. Perhaps it already is.

17862.jpeg


Apple's overdependence on iPhone as a revenue stream and iPhone's overdependence on China as a manufacturing hub. This is a dangerous combination that can spell Apple's downfall.
 
Last edited:
Downside of Apple's scaled up mobile chips is that they will never be a true alternative to x86 workstations.
Yes, but that’s because they are, in reality, not x86 processors. For anyone that needs an x86 workstation, there’s literally nothing Apple can provide them because they’re not even an x86 licensee.
 
If you analyze Apple's revenue stream, you will soon come to the realization that without another blockbuster product soon, it will become a phone/service company. Perhaps it already is.

17862.jpeg


Apple's overdependence on iPhone as a revenue stream and iPhone's overdependence on China as a manufacturing hub. This is a dangerous combination that can spell Apple's downfall.

FINALLY someone who can see things the way I do!
You said everything.
The issue with Apple is prioritizing their phone market at the expense of everything else (desktops, tablets, games...).

Is Apple in trouble NOW?
Of course not. Apple is doing quite well NOW.
But if ANYTHING goes wrong with the phone market, boom! They will be in serious trouble.

Or a more likely scenario could be happening: Android phones get better and better. And if their competitors do everything Apple does, then the smartphones become a commodity. It starts not to make sense to buy an iPhone if Android phone cameras are just as good, Android is just as responsive, and it is just as stable (or close to the stability of OS X).

So, Apple is doing well NOW, but what about in five years? Will they still be able to remain relevant, considering their decisions now?
 
Or maybe they did and decided pro workstations are one segment of the Mac market they can afford to lose. As iPhone sales slows, a lot of the ill-advised decisions Cook made now started to come back to bite him.
Pro workstations were lost prior to Apple Silicon. Apple’s actions up until now were a preparation for Apple Silicon… the reason why they cut off 32 bit support while they were still on Intel processors (which were 32-bit capable) is an example of this. iPhone sales sliding was expected, which is why they’ve been providing services that have been bringing in high profit money (900 million subscribers across all their different offerings, Apple One, iCloud, etc.).

This is not to mention getting Apple deeply entangled in China.
Yes, Steve jobs got Apple deeply entangled in China. Thanks Steve?

If you analyze Apple's revenue stream, you will soon come to the realization that without another blockbuster product soon, it will become a phone/service company. Perhaps it already is.
Yes, it is. Though I’d modify it slightly to say phone/tablet/service company, because they sell almost twice as many iPads as Macs in any year.
 
Pro workstations were lost prior to Apple Silicon. Apple’s actions up until now were a preparation for Apple Silicon… the reason why they cut off 32 bit support while they were still on Intel processors (which were 32-bit capable) is an example of this.

Actually, they cut on 32-bit support to get rid of legacy code.

It IS possible to support 32-bit code even if the processor itself is 64-bit only. Proof of that is Box64, which allows ARM v8 to run 32-bit x86 code.

If Apple really wanted, they could modify Rosetta 2 to run 32-bit MacOS code and make the support permanent. There WOULD be a performance penalty, but that's only around 15%-20% (or even less, if you add hardware acceleration for 32-bit code emulation).
 
  • Like
Reactions: DavidSchaub
The question some should be asking is not, “Is Apple going to be relevant?” It should be more like “In five years, will I be relevant to Apple?” It depends on whether or not that individual looks at the landscape of the things Apple has created and can say to themselves, “I like these products an their price is worth it for the value I would receive from them.” If so, then the will be.

At any one time, the products and services Apple produces are not made for the entire computing market, they focus on a slice of it that can be profitable to serve. For the folks in that slice, it feels to them like these products are made for EVERYONE if only everyone would see it (or if Apple made their products cheaper… which is against the point ;)) There are folks that, for a number of years, were IN the slice as Apple defined it at the time. And, when Apple redefined it (which they do regularly), some folks found themselves no longer in the slice Apple’s putting their attention on.

As long as Apple can continue to find those profitable millions every year (they’re pretty good at it by now) and continues to offer a mix of goods and services an ever changing slice of folks find valuable, they’ll be ok.
 
Pro workstations were lost prior to Apple Silicon. Apple’s actions up until now were a preparation for Apple Silicon… the reason why they cut off 32 bit support while they were still on Intel processors (which were 32-bit capable) is an example of this. iPhone sales sliding was expected, which is why they’ve been providing services that have been bringing in high profit money (900 million subscribers across all their different offerings, Apple One, iCloud, etc.).


Yes, Steve jobs got Apple deeply entangled in China. Thanks Steve?


Yes, it is. Though I’d modify it slightly to say phone/tablet/service company, because they sell almost twice as many iPads as Macs in any year.
Agreed with most points. But it was really Cook, not Jobs. https://bit.ly/3DnUFwa
 
The question some should be asking is not, “Is Apple going to be relevant?” It should be more like “In five years, will I be relevant to Apple?” It depends on whether or not that individual looks at the landscape of the things Apple has created and can say to themselves, “I like these products an their price is worth it for the value I would receive from them.” If so, then the will be.

At any one time, the products and services Apple produces are not made for the entire computing market, they focus on a slice of it that can be profitable to serve. For the folks in that slice, it feels to them like these products are made for EVERYONE if only everyone would see it (or if Apple made their products cheaper… which is against the point ;)) There are folks that, for a number of years, were IN the slice as Apple defined it at the time. And, when Apple redefined it (which they do regularly), some folks found themselves no longer in the slice Apple’s putting their attention on.

As long as Apple can continue to find those profitable millions every year (they’re pretty good at it by now) and continues to offer a mix of goods and services an ever changing slice of folks find valuable, they’ll be ok.
It's not really that easy to find the next big thing and be at the right place at the right time. Apple really hit a home run with the original iPhone. Everything else after was built on the moment the advent of that particular device created.

It's not like there wasn't any precedent either. Look at Nokia and Sony. One of them pretty much disappeared and the other morphed into an entertainment-focused company that makes movies and games, which will probably be Apple's fate if no blockbuster product is introduced before they run out of iPhone cash.
 
  • Like
Reactions: gusmula
Actually, they cut on 32-bit support to get rid of legacy code.
Yes, guess why? Because that legacy code would have caused problems during the transition to Apple Silicon that Apple had on their internal roadmaps for years. Case in point, Rosetta. That the current one works as well as it does seems amazing until you think that Apple set it up to make its job as easy as possible. Rosetta today doesn’t have to translate ALLLL of the cruft that’s been created in x86 over the years. It’s JUST focusing on the 64 bit stuff and, of that, ONLY the stuff that’s been created by Apple’s compiler (Apple’s deprecated a lot of functions in their API’s over the years, again, that wouldn’t align with the Apple Silicon transition).

So, having excised all the problematic legacy cruft, Rosetta’s being given this a subset of a subset of all that history (mostly just the recent bits) and being tasked to be performant. Which it turns out is an effective way to do things. It performs well compared to similarly performant PC’s ONLY because those PC’s are having to spend cycles upon cycles weeding through those instructions that have been moved out of Rosetta’s way. All in preparation for the future that is today.
 
Agreed with most points. But it was really Cook, not Jobs. https://bit.ly/3DnUFwa
Mr. Jobs angrily held up his iPhone, angling it so everyone could see the dozens of tiny scratches marring its plastic screen, according to someone who attended the meeting. He then pulled his keys from his jeans.

People will carry this phone in their pocket, he said. People also carry their keys in their pocket. “I won’t sell a product that gets scratched,” he said tensely. The only solution was using unscratchable glass instead. “I want a glass screen, and I want it perfect in six weeks.”

After one executive left that meeting, he booked a flight to Shenzhen, China. If Mr. Jobs wanted perfect, there was nowhere else to go.

Now, it’s POSSIBLE that Tim Cook, using his masterful arts of mind manipulation, was actually controlling Steve Jobs from the shadows. I can’t not say with 100% certainty that Tim Cook didn’t force Steve Jobs to deal with China against his will.

It’s possible that everyone at Apple kept Steve Jobs in the dark as to where those glass panels were coming from. And where iPhones were being built. For the entirety of his run as CEO.
 
Last edited:
Mr. Jobs angrily held up his iPhone, angling it so everyone could see the dozens of tiny scratches marring its plastic screen, according to someone who attended the meeting. He then pulled his keys from his jeans.

People will carry this phone in their pocket, he said. People also carry their keys in their pocket. “I won’t sell a product that gets scratched,” he said tensely. The only solution was using unscratchable glass instead. “I want a glass screen, and I want it perfect in six weeks.”

After one executive left that meeting, he booked a flight to Shenzhen, China. If Mr. Jobs wanted perfect, there was nowhere else to go.

Now, it’s POSSIBLE that Tim Cook, using his masterful arts of mind manipulation, was actually controlling Steve Jobs from the shadows. I can’t not say with 100% certainty that Tim Cook didn’t force Steve Jobs to go to Shenzhen, China against his will.
You should really take the time to read that article.

There is a fine line between going to China to procure unscratchable glass and signing an agreement with the CCP. And I really hope you think so too because I like the stuff you write so far.
 
You should really take the time to read that article.

There is a fine line between going to China to procure unscratchable glass and signing an agreement with the CCP. And I really hope you think so too because I like the stuff you write so far.
Apple has been doing business with China for years. This is, effectively, more business with China. Which… one can’t have “more” business with China unless they’re doing business with China to start with. If Steve Jobs hadn’t set that direction for years PRIOR to 2016, then Tim Cook wouldn’t be signing a deal in 2016. (OR Gil or Scully, I should look into who moved manufacturing to China first)
 
It's not like there wasn't any precedent either. Look at Nokia and Sony. One of them pretty much disappeared and the other morphed into an entertainment-focused company that makes movies and games, which will probably be Apple's fate if no blockbuster product is introduced before they run out of iPhone cash.
I don’t doubt that Apple’s going to go through changes. Someone used to the “Apple II” Apple wouldn’t get how Apple makes more money on services than they make on the Apple II, which isn’t even sold anymore. Actually that’s a good parallel. Apple, if run like most of the rest of the industry, would have JUST removed the final bits of AppleIIGS legacy code from the latest OS in 2021. You could still use a legacy IIGS installer, but it would route you to the appropriate 32-bit installer. :) And the iPod would be entering the terabytes of storage by now.

Steve Jobs died in 2011, but in 2018, we still have Apple making decisions focused on the future (to the detriment of current users) by ending 32-bit support. If this is part of the culture of Apple, it’s the thing that will keep them future facing and relevant primarily TO that future. I would not be surprised if there’s a roadmap to the iPhone’s demise just as there was a roadmap to the iPod’s demise when it was at the top of it’s game. At that time, iPhone people are going to be pissed and wondering if Apple will be relevant in another 5 years…
 
Apple has been doing business with China for years. This is, effectively, more business with China. Which… one can’t have “more” business with China unless they’re doing business with China to start with. If Steve Jobs hadn’t set that direction for years PRIOR to 2016, then Tim Cook wouldn’t be signing a deal in 2016. (OR Gil or Scully, I should look into who moved manufacturing to China first)
A lot of American companies were doing business with China at that time. A lot of Japanese and Korean companies too. But to get to the level that Apple is? That's of Cook's own volition. https://on.ft.com/3Y7TKaO

"It is the result of corporate diplomacy led by chief executive Tim Cook, whose regular visits to Beijing in pre-pandemic times, including meetings with Xi and Chinese tech executives, have helped avoid the fate of other western tech companies. The likes of Alphabet, Meta and Netflix have been locked out of the country."

He could've diversified but he didn't. In fact, he doubled down. Even now, he's one of the most reluctant CEOs of a Fortune 500 company to move its supply chain away from China.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.