Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
But yet, 99% of people buying Macs either won’t know it’s in there or won’t care. And 90% of the remaining 1% won’t be using their Macs for anything that couldn’t already be done efficiently on the last generation of Macs.
It has always been the case that 99% of computer users are not professional developers of hardware intensive software, video, games, etc.
 
A new chip that wasn't better than existing chips would have been pretty pointless.

It’s not better it’s miles better.

It’s not an evolutionary gain like previous Intel generations but a revolutionary step for personal computers.

Not only performance wise but in the extended battery life, even with heavy workloads.

And coupled with MacOS, the fluidity of day to day use is really what will set Apple Silicon apart for years to come.
 
  • Like
Reactions: NetMage and Vazor
But yet, 99% of people buying Macs either won’t know it’s in there or won’t care. And 90% of the remaining 1% won’t be using their Macs for anything that couldn’t already be done efficiently on the last generation of Macs.

Respectfully disagree.

For most performance benchmarks you would be right. Yet, the fluidity resulting from the combination of MacOS and Apple Silicon will be noticed by all.

It’s similar to the first SSD vs HD comparisons, or moving from a bogged down Windows machine to a fresh Chromebook (a good one ). Everyone can notice the difference.
 
I always wonder... what stops TSMC from leaking those designs to other companies they manufacture chips to? Not saying on purpose, I’m imagining movie levels of conspiracy, hacking, espionage, death threats and who knows what at that level... what about other manufacturers that can’t get 5nm nailed or competition, governments, etc trying to get their hands on those chip designs or the designers themselves.

Also, what does TSMC info gets “to print” the chips? Is it like a completely encrypted file with a standardized I/O machine that reads instructions of the steps to manufacture? (This would protect both Apple and TSMC). TSMC would make sure that each step in isolation works up to spec.


Totally average bit of journalism by MacRumours. For example,

What's Different About the M1​

Unlike Intel chips built on the x86 architecture, the Apple Silicon M1 uses an Arm-based architecture

Is that it? Is that really the best you can do?

EDIT:
I see a couple of people have posted negative likes. I can only assume they either haven't read my post properly or work for MacRumours. I am not being critical of the M1 chip or Apple. I am being critical of the missed opportunity to explain how ARM's RISC design will always triumph over the complex instruction set used by Intel. As a (now retired) computer journalist I followed the development of RISC during the 1980s and 1990s when Joel Birnbaum worked at IBM's TJ Watson Research Centre and then joined Hewlett-Packard to create PA-RISC. Bio: https://ethw.org/Joel_S._Birnbaum
Macrumors is maybe not the type of place to go that route, it’s mostly quick bite news, discount opportunities and general high level train of news with links to a whole lot deeper articles (like links to anadtech ones) where pertinent. A dislike is probably is still a bit unwarranted though.

That being said, with this last amazing advancement and motivation around Arm and RISC, I would be so down for a full-on documentary, article, podcast, anything that would explain the history of how it got to here... where does RISC comes from, where is it going, why, who, etc. didn’t know about this Joel guy. Seems to me that you might have loads and loads of history information on your hands. Nice!
 
Is there a way to set firmware password or any other way to protect from theft? (about Mac Mini M1)
 
  • Like
Reactions: DVD9
I always wonder... what stops TSMC from leaking those designs to other companies they manufacture chips to?
Have you considered that Apple might notice if their design was to show up in someone else's product? Also, companies that aren't able to maintain client confidentiality usually don't stay in business very long. TSMC does not want to lose Apple as a customer.
 
Last edited:
I always wonder... what stops TSMC from leaking those designs to other companies they manufacture chips to? Not saying on purpose, I’m imagining movie levels of conspiracy, hacking, espionage, death threats and who knows what at that level... what about other manufacturers that can’t get 5nm nailed or competition, governments, etc trying to get their hands on those chip designs or the designers themselves.

Also, what does TSMC info gets “to print” the chips? Is it like a completely encrypted file with a standardized I/O machine that reads instructions of the steps to manufacture? (This would protect both Apple and TSMC). TSMC would make sure that each step in isolation works up to spec.
Something called an "enforceable contract." TSMC doesn't design chips. They are a contract manufacturer (a very advanced one, but a contract manufacturer, nonetheless). Companies like Apple and AMD design chips that can be manufactured on TSMC's processes. Those companies pay TSMC to manufacturer the chips, but retain ownership of the designs.
 
It's funny, the M1 is great, but I'm starting to suspect that the UMA is more important from a performance point of view. Context switching is a big performance problem, and it gets worse in a multi-threaded multi-process environment. It's so much of a problem that Intel deliberately (IMO) punted on access control for speculative execution because of the performance hit.

If Apple figured out a good way to handle that for its use cases then that would explain the performance multiples; they're not just doing it faster, they're doing it better.

I've never seen any articles on iOS and the UMA stuff's memory management/MMU at that level. CS people used to talk about that stuff, but in the era of Intel dominance that has sort of went away. The only time I've ever seen a writer talk about that was that guy at anandtech who went to Apple a while back.
 
truly a remarkable SoC....but defects WOW....I have had this thing for a week and its jam packed with problems. In the end its worth the first adopter for the shear battery life I can get

18 hours on average on the air , doing normal work for me....crazy

Problems I have had :

when I dock it to monitor , sometimes internal display does not shut off , allowing you to move windows onto it.

bluetooth disconnecting

internet disconnecting

random reboots

freezing for a shot Period , 10 seconds than all back to normal


this is my second MacBook Air too.

My M1 MBA 16GB/1TB should be delivered next week.

If it should exhibit the issues you're seeing, I'll simply send it back and see what the situation is 3-4 months from now.
 
Something called an "enforceable contract." TSMC doesn't design chips. They are a contract manufacturer (a very advanced one, but a contract manufacturer, nonetheless). Companies like Apple and AMD design chips that can be manufactured on TSMC's processes. Those companies pay TSMC to manufacturer the chips, but retain ownership of the designs.
Plus, once word gets out that TSMC leaked info, they lose a lot of contracts, not just Apple’s.
 
  • Like
Reactions: chabig
Quote showing low power cores achieve 395/1361 in geekbench
That matches a 2011 MBA single threaded, and a 2017 MBA multithreaded. Not bad.
 
  • Like
Reactions: Henk Poley
Developing a new chip that wasn't better than existing chips would have been pretty pointless.
It really depends on what you mean by "better."

Apple's metric in the early days wasn't "faster than x86." And in fact it wasn't faster than comparable x86 chips. The metric was "low power, high performance relative to other mobile chips." And they hit those numbers pretty well, as the charts show. The A series has been the best performing and lowest power mainstream mobile CPU for pretty much it's entire existence. The same can be said for whatever that BT chip was: it's the best performing low power BT chipset in the industry.

But in a way you're right, because if there was no benefit (which is sort of implied by "better") then there's no point in doing it.
 
Being left behind already (still using iTunes 12.6.5.3), I wouldn’t mind using Apple ARM MacBook given that I am ok with renting hardware from Apple with a hefty price tag.

With that being said, Apple Silicon performance is amazing. 3DMark wise A12X still cannot directly compete with GTX1650 but I think eventually performance would be so great that even Rosetta 2 is not a problem for most people.

Of course, first gen product. But after a few years, I think hopping this ARM train might not be a terrible idea. Though, I can completely forget about Windows (and maybe Linux) support.
 
Have you considered that Apple might notice if their design was to show up in someone else's product? Also, companies that aren't able to maintain client confidentiality usually don't stay in business very long. TSMC does not want to lose Apple as a customer.
I understand, but you picked up that exact sentence out of the whole post which makes it look like that TSMC would do that on purpose and only on purpose.

Also, how long would you think it would take to realize for Apple that their designs have been leaked, stolen or an all out war espionage retrieved... they would have to get the designs of the competing products too.
 
  • Disagree
Reactions: NetMage
"It is not worth purchasing an Intel Mac mini or 13-inch MacBook Air at this point"

******** it isn't. The Intel Macs are going to be supported for a good number of years and over a few releases of Mac OS X. I just recently bought one and I'm quite happy with the purchase.
 
My only problem with the M1 and Apple Silicon and Rosetta 2 is that I remember the PowerPC experience that left a bad taste in my mouth after Apple quickly dropped Rosetta. I feel sorry for Mac Pro users, both current and all 2013+ models as they know they will be quickly discarded by Apple just like PowerPC users were. Every time Apple has changed processors, they have screwed over their customers. It's just history. Steve Jobs was proud of it, but he could get away with it. Tim Cook has no Reality Distortion Field.

You know what they say about Fool me Once, twice right? :(

PS: I'll give one exception. The Apple ][e Card. I had one of those too and I remember how cool that was. A brilliant idea from Woz to keep schools from cancelling Apple contracts during the transition to Macintosh.
 
Last edited:
My only problem with the M1 and Apple Silicon and Rosetta 2 is that I remember the PowerPC experience that left a bad taste in my mouth after Apple quickly dropped Rosetta. I feel sorry for Mac Pro users, both current and all 2013+ models as they know they will be quickly discarded by Apple just like PowerPC users were. Every time Apple has changed processors, they have screwed over their customers. It's just history. Steve Jobs was proud of it, but he could get away with it. Tim Cook has no Reality Distortion Field.

You know what they say about Fool me Once, twice right? :(
Believe me, I know what you mean because I’ve been thinking the same, however, I think Apple this time will give a longer support to Intel macs (because, hell, they are still selling those machines and will keep selling them after the MacBooks redesign, they will keep selling new Intel macs until the end of 2022). I don’t think Apple will drop support of those new machines before 5 years.

So, we have Intel macs supported until, at least, 2027. That means universal apps will be the default by then, yes, but maybe they won’t drop support for Rosetta 2 until 2025. Hopefully. Just imagine having macs capable of running x86 apps natively, and macs that cannot ruin them, supported with the same macOS releases.

Yeah, it could happen, Apple could stop including Rosetta 2 on their operating system after the transition is complete and when they start selling only Apple Silicon macs, forcing us to just stuck with native apps and ignore those which haven’t been ported, but there’s too much time invested in a flawless and super efficient Rosetta 2 to just drop it after 2 years. Hopefully Rosetta 2 will stay with us at least for 4 years (two transition years, 2 post-transition years)
 
  • Like
Reactions: BaltimoreMediaBlog
My only problem with the M1 and Apple Silicon and Rosetta 2 is that I remember the PowerPC experience that left a bad taste in my mouth after Apple quickly dropped Rosetta.
Essentially PowerPC computers were still fully supported in parallel from OS X 10.4.4 - April 29, 2005 up until OS X 10.6 - August 28, 2009. 4 years and 4 months. Which is not exactly long, but not super short either.

(Edit: PowerMac G5 was sold until August 2006, so nearer to the edge. 3 years.)

As a home user I would indeed be angry that my system would need to be thrown out at half its lifespan (average age of PCs was 4.5 years in 2006). For business users 3-5 years is sort of the expected replacement cycle. So Apple chose for the businesses.

This time Rosetta 2 is in-house software. And not something made by a third-party company that otherwise supports big enterprise companies run their legacy mainframe software. So support could be longer this time.
 
Last edited:
Essentially PowerPC computers were still fully supported in parallel from OS X 10.4.4 - April 29, 2005 up until OS X 10.6 - August 28, 2009. 4 years and 4 months. Which is not exactly long, but not super short either.

(Edit: PowerMac G5 was sold until August 2006, so nearer to the edge. 3 years.)

As a home user I would indeed be angry that my system would need to be thrown out at half its lifespan (average age of PCs was 4.5 years in 2006). For business users 3-5 years is sort of the expected replacement cycle. So Apple chose for the businesses.

This time Rosetta 2 is in-house software. And not something made by a third-party company that otherwise supports big enterprise companies run their legacy mainframe software. So support could be longer this time.

A PowerMac G5, one of which I still own, was NOT a cheap machine. It still does everything it did back then very fast. But Apple cancelled support with the next Operating system, Snow Leopard. This is history. Please don't make comparisons to PCs because nobody cares about that. People buy Apple products for different reasons than PC and ANDROID people buy things. We spend more money for a reason!
 
A PowerMac G5, one of which I still own, was NOT a cheap machine. It still does everything it did back then very fast. But Apple cancelled support with the next Operating system, Snow Leopard. This is history. Please don't make comparisons to PCs because nobody cares about that. People buy Apple products for different reasons than PC and ANDROID people buy things. We spend more money for a reason!

True!

But since WWDC 2020 Apple has announced a new transition, every one should factor that in in their decisions. No one knows the timeline for new products nor the EOL for supporting Intel Macs at this moment.

If support is kept by Apple until 2025 (6 years counting 2020 thru 2025) I would think reasonable. I understand lots of production machines do not update to latest OS for 1-2 years, so a new OS in 2025 can last for 2-3 years. That would give a machine bought in the first half of 2020 8+ years of support/work.

Given the expect advancement of Apple Silicon, any business should invest 5 years from now in upgrading their hardware to keep up with the new demands (AR/VR, 8K video, live multicam performances, ML, etc).

Not to invest is also a business decision, and for every decision there are consequences.
 
Ordered my base Mini yesterday. Couldn't bear my gradually slowing 2015 MBA any longer. Still interested in where Apple takes the redesign of the MBP, so using this little machine in the meantime. Really looking forward to the speed boost over my dual core i5 with 4 gigs of RAM ;)
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.