Why Apple will release a 27 inch Apple Silicon processor Mac as it’s first MAC With apple Silicon.

[lysingur]

It has nothing to do with clearing inventory. If you're going to stereotype Tim Cook, he's a supply chain guy, not a generic numbers guy.

Supply chain management, particularly logistics, is all about numbers. You can't improve the efficiency of your supply chain without numbers. Plus, Tim Cook comes from an engineering background. So I don't know how numerically-oriented you want him to be before you are willing to call him a "numbers guy". Dealing with numbers is practically his job.

I doubt they stockpile 6 to 9 months of any finished product.

You are making a wild assumption that they actually have any finished product (AS 27-inch iMac).

What's your clever story about why you think they specifically failed to do that for the 27" iMac, given that it's not exactly rocket science to stop making a thing according to a schedule?

It's also not rocket science to know that Apple just won't introduce a newer iMac to replace a new iMac they just introduced a little over three months ago.

Assuming we don't see a 27" iMac in a few hours, I guarantee you the real reason is that even Apple doesn't have the resources to tape out (release to manufacturing) and production ramp the entire slate of Mac Apple Silicon chips in parallel.

I guarantee you the real reason is they don't even have the new design for the AC 27-inch iMac ready because they never plan to release an AC 27-inch iMac this early. I can ask you the same thing: What's your clever story about why you think it's because "even Apple doesn't have the resources to tape out (release to manufacturing)"?

You don't have one.

 

[Jouls]

I don't really need to because that's what I actually do it for a living.

If you have a basic understanding of accounting, you wouldn't be confusing margin, revenue, and profit like this.

Maybe then you would like to educate us, how to calculate profit if revenue and profit margin are given?
And so far you have not shown that profitability has shrunk for the Mac segment. Just a off-hand statement.

 

[lysingur]

Maybe then you would like to educate us, how to calculate profit if revenue and profit margin are given?
And so far you have not shown that profitability has shrunk for the Mac segment. Just a off-hand statement.

If you really want to learn, you'd spend some time on Wikipedia rather than here parading your ignorance.

 

[CWallace]

Profitability: Desktops and laptops are not as profitable for Apple in 2020 as in 2005. How much resource has Apple reallocated and is continually willing to reallocate to make this transition as smooth and as quick as possible?

Mac profitability is clearly better in 2020 than in 2005 based on unit sales alone. Apple could spin off the Mac division and it would promptly park itself firmly in the Fortune 400.

What is true is that Mac does not contribute anywhere near the profits on a total percentage basis in 2020 as it does in 2005 thanks to the iPhone and Services / Accessories. But it still contributes a fair bit of money to Apple and that they are actually spending the resources (monetary, staff and management focus) to make this move shows the Mac is still important to Apple.

Those who use a photo editor and who really need to leverage the power of the neural engine in their workflow probably won't buy a MacBook Air or other entry-level Macs that Apple will surely introduce first in the initial months of the transition (or in less than 12 hours for that matter). The most probable timeline for an AS iMac/iMac Pro and Mac Pro is mid to late 2021. By the time the Photoshop, Affinity Photo, or Pixelmator people need to even start worrying about their customers' "confidence in [their] future support and development", we'd be standing on the cusp of 2022.

There are far more photo editors that run under iOS than macOS so the transition to ASi will open up more opportunities for users and those applications will be able to take advantage of ASi feature sets from Day One since they already support them under iOS. That should be enough to light a fire under macOS developers to "embrace and extend" ASi and it's features lest 2022 come along and they find a significant portion of their user base has moved on to other applications running on ASi Macs.

 

[TDTOMW]

Kudos to all you 13 inch MacBook Pro and MacBook Air and Mac mini people who chose the right product that would be introduced for the first Apple Silicon chips Mac. I guess I was a little premature. The iMac will come out with the I guess a M2 chip. I guess the foundry couldn’t produce both of them at the same time this early in the production stage. But if all the performance gains are true the M2 chip will Something to behold.

 

[Deinocheirus]

Kudos to all you 13 inch MacBook Pro and MacBook Air and Mac mini people who chose the right product that would be introduced for the first Apple Silicon chips Mac. I guess I was a little premature. The iMac will come out with the I guess a M2 chip. I guess the foundry couldn’t produce both of them at the same time this early in the production stage. But if all the performance gains are true the M2 chip will Something to behold.

Surely you can't miss the fact that Apple started at the bottom of the performance end of the lineup, where the first run of Apple Silicon chips would have the biggest impact.

 

[Deinocheirus]

Mac profitability is clearly better in 2020 than in 2005 based on unit sales alone. Apple could spin off the Mac division and it would promptly park itself firmly in the Fortune 400.

What is true is that Mac does not contribute anywhere near the profits on a total percentage basis in 2020 as it does in 2005 thanks to the iPhone and Services / Accessories. But it still contributes a fair bit of money to Apple and that they are actually spending the resources (monetary, staff and management focus) to make this move shows the Mac is still important to Apple.




There are far more photo editors that run under iOS than macOS so the transition to ASi will open up more opportunities for users and those applications will be able to take advantage of ASi feature sets from Day One since they already support them under iOS. That should be enough to light a fire under macOS developers to "embrace and extend" ASi and it's features lest 2022 come along and they find a significant portion of their user base has moved on to other applications running on ASi Macs.

I'm excited to see certain iOS/iPadOS apps suddenly become major Mac apps. Whereas certain Mac developers have avoided the Mac App store, others are going to rush to embrace it because it gives their already-completed apps a whole new platform to run on, overnight.

The developers who shun the app store can now adapt, or get overtaken by competitors who did nothing but click the Mac checkbox in Xcode.

 

[ADGrant]

I’m just a little confused, I thought that software companies wrote the software to take a vantage of the advancements in hardware, if the Apple silicon Chip have advancements that improve the speed and efficiency. I think they will write the software for it. If not some other company will step in and write better software and collect the money. I think that’s how it works in business.

It's not that simple. Writing good software takes time and money.

 

[ADGrant]

I'm fairly sure that a new Mac Mini will NOT be among the first AS-based Macs released on the market. While I've seen 1-2 people use the DTK as their evidence to support such a claim, I'd argue that it actually disproves it. We know that Apple is using an existing A12 series processor in the DTK, and we know from WWDC that AS-based Macs will run a "new family of processors". Given the limitations and pecularities of the DTK (different I/O layout and post selection on the back, no TB3 support, inabillity to launch into recovery mode on startup, etc.), it is safe to assume that what's inside the box is essentially "hacked" together to create a usable tool for developers. My guess is that they chose to use the Mini because it had the necessary space inside the case to accommodate the unique configuration inside the chassis. Shipping the mini (which has no screen) is also less expensive than the other models they could have used for a developer toolkit.

I was not at all surprised that they launched an AS Mini and I think one reason they did was to give developers a production replacement for the DTK.

 

[ADGrant]

I'm pretty sure that developers and colocation services are a tiny proportion of Apple's Mac customers, so why would Apple release a product to those user segments?

I agree that getting developers on-board is important to grow the adoption of Apple Silicon apps, but developers already have the DTK for this purpose, and will be happy to "upgrade" to an ASi Mac when they are available. In fact, I understand that DTK users may be given the option of swapping the DTK for the first ASi Mac (almost certainly a laptop) either free or with a substantial discount

They will want the launch to have some impact with consumers and generate significant sales, so it is far more likely to launch a low-mid level laptop, which is their biggest seller. It's a business after all!

Developers may be a small group of Mac customers but they are a very important group which is why I think a Mini was part of the first wave.

 

[throAU]

It's not that simple. Writing good software takes time and money.

Yes and no.

Apple has control of the swift and cocoa libraries and can optimise them for their own architecture. Many these libraries are used by macOS applications to do the performance sensitive heavy work.

Apple has ensured that the macOS libraries were written in a cross platform friendly way from the start. So they're fairly optimised for their own platform already and apps can take advantage of that even if the app itself is not yet optimised as most of the hard work is done by the OS libraries.

Because the quote you replied to isn't included - yes as you say writing good software takes time and money, but as far as an application is concerned - Apple has done the hard work for you in that respect. Having your app run fast on the new platform is a recompile and link to the native OS provided libraries; either by yourself or by Apple's dynamic recompiler (Rosetta2).

Windows - this is not the case. There's a huge amount of legacy crap in windows that is still embedded deep within it that was never written to be portable. They have no capability to link an x86 or x64 app to ARM native libraries like Apple does. The irony is that Java (which used to be the bane of performance0 may actually help Windows in this respect if they can make their JVM run fast enough - because that's a JIT compiler they have running that they've been working on for a couple of decades now.

Apple got a fresh start with OS X (NEXTSTEP). Nextstep was written to be portable. macOS continued that.

If Apple had just kept hacking away and tacking bits and pieces onto OS9, they'd be in a much worse position - which is basically slightly ahead of where Windows is today. Windows got a partial rewrite to be fully 32 bit with Windows NT/XP, but they didn't bother to keep it cross platform since NT4.

 

[ADGrant]

Windows - this is not the case. There's a huge amount of legacy crap in windows that is still embedded deep within it that was never written to be portable. They have no capability to link an x86 or x64 app to ARM native libraries like Apple does. The irony is that Java (which used to be the bane of performance0 may actually help Windows in this respect if they can make their JVM run fast enough - because that's a JIT compiler they have running that they've been working on for a couple of decades now.

Apple got a fresh start with OS X (NEXTSTEP). Nextstep was written to be portable. macOS continued that.

If Apple had just kept hacking away and tacking bits and pieces onto OS9, they'd be in a much worse position - which is basically slightly ahead of where Windows is today. Windows got a partial rewrite to be fully 32 bit with Windows NT/XP, but they didn't bother to keep it cross platform since NT4.

Microsoft does not use Java in Windows. They do sometimes use C# and .NET though.

Windows NT was not a partial re-write of Windows. It was a completely different operating system. The old Windows & DOS code base was abandoned with Windows ME about 20 years ago.

 

[throAU]

Microsoft does not use Java in Windows. They do sometimes use C# and .NET though.

Windows NT was not a partial re-write of Windows. It was a completely different operating system. The old Windows & DOS code base was abandoned with Windows ME about 20 years ago.

I'm not suggesting Microsoft use java in windows.

My point is that the JVM available for Windows or even the javascript execution engine could help them more than trying to emulate ARM on their platform. Because the just in time compilation is mature on windows.


Windows NT was a re-write of the KERNEL yes - but it still included a whole slew of the original 16 bit libraries and other crap to maintain compatibility with Windows 3.1 applications. I ran every version of Windows in a work environment (as a sysadmin) since 3.1. And even today still contains heaps of crap in it to maintain compatibility with 20 plus year old software.

Apple broke compatibility and started fresh with a code-base that was written with the explicit goal of being cross platform from the start. Windows was never seriously aimed at being cross platform - there were MIPS and Alpha versions of NT4 and 2000 but microsoft never properly supported them.

There's still bits of crap in Windows 10 from the Windows 3.1 days believe it or not. Just this week I got an error message about "Program Manager". https://en.wikipedia.org/wiki/Program_Manager

It's not visible on the filesystem as program manager, you can't explicitly use program manager as the shell any more, but some Windows 10 component is essentially program manager that has been hacked on and renamed in almost all places except for where it tripped up for me this week


Its not normally visible, but there's huge swathes of 20-30 year old crap code inside of Windows from things as far back as Windows 3.1 and possibly earlier. There was a 25 year old font rendering bug found in Windows last year for example. The code was the same going all the way back to Windows 3.1

To claim Windows NT was a total rewrite is very much a stretch. it may be what microsoft claim, and it may be what some people believe, but its not really true.

 

[ADGrant]

I'm not suggesting Microsoft use java in windows.

My point is that the JVM available for Windows or even the javascript execution engine could help them more than trying to emulate ARM on their platform. Because the just in time compilation is mature on windows.


Windows NT was a re-write of the KERNEL yes - but it still included a whole slew of the original 16 bit libraries and other crap to maintain compatibility with Windows 3.1 applications. I ran every version of Windows in a work environment (as a sysadmin) since 3.1. And even today still contains heaps of crap in it to maintain compatibility with 20 plus year old software.

Apple broke compatibility and started fresh with a code-base that was written with the explicit goal of being cross platform from the start. Windows was never seriously aimed at being cross platform - there were MIPS and Alpha versions of NT4 and 2000 but microsoft never properly supported them.

There's still bits of crap in Windows 10 from the Windows 3.1 days believe it or not. Just this week I got an error message about "Program Manager". https://en.wikipedia.org/wiki/Program_Manager


Its not normally visible, but there's huge swathes of 20-30 year old crap code inside of Windows from things as far back as Windows 3.1 and possibly earlier.

Microsoft already has its own JIT compiler in the .NET CLR so they don't need anything from the JVM. Apps written for .NET should be able to run on ARM based Windows without any emulation. Plenty of apps people want to run are written in C & C++ (e.g. Photoshop or Microsoft Office).

Windows NT development started 30 years ago so yes there is 20 to 30 year old code in it. It was much more than a kernel re-write though (did DOS really have a kernel to re-write?). It included a new file system and a new windowing system. Yes they did include a bunch of backward compatibility crap and yes they made it look as much like the old windows as possible with similarly awful APIs but it was still a new OS. You cannot run old DOS and Windows 3.1 apps natively on 64bit Windows and even 32bit Windows needed something called NTVDM.

It is true that "Program Manager" existed in both Windows 3.1 and Windows NT 3.1 and that they looked and worked the same. It doesn't prove the code was the same though but thanks to the C #ifdef preprocessor directive it could have been (sort of).

I have been developing software for Windows since version 1.0 so I am familiar with the ugliness of Windows and other Microsoft APIs.

 

[lysingur]

I have been developing software for Windows since version 1.0 so I am familiar with the ugliness of Windows and other Microsoft APIs.

I know little about software development but has always suspected that the underlying structure of Windows must be a mess based on my experience with Registry. Thank you for confirming this.

 

[jinnyman]

Well this thread is pretty much dead now

 

[Kostask]

it may be of interest to know that Windows NT was a ground up rewrite by one of the software architects (Dave Cutler) who designed VMS, the operating system for the old DEC Vax machines, and a lot of the concepts from VMS made their way into Windows NT. It owes nothing to the DOS based Windows, and did not need a copy of DOS loaded before it would run. it did have DOS compatibility features that were added on, just to allow legacy software to run, but those were add-ons, not part of the operating system core.

 

[Thor112358]

They JUST released the Intel version of the 27" iMac. How do you imagine the customers would feel if barely a few months later they released an all new 27" AS iMac??? That's insane from a marketing point of view. An insult and a slap in the face.

Also, it's widely anticipated, rightly so, that a new 27" iMac will sport a re-design. Given the thermal envelope performance of AS, you can do a completely different design. It takes time to do that, and they are not going to come out with a new design so soon after the Intel 27" came out. NO CHANCE.

Hi, OldCorpse. Given what you wrote, help me out on my dilemma. I am running an early 2009 27" iMac (El Capitan) that is getting more problematic each month. Part of me wants to go with a new 27" iMac with 3.8GHz 8-core 10th gen i7, 8GB memory (put more in later), 5500XT, 1TB storage and just enjoy it: all of the current programs and etc. will all work for (hopefully) another 10-11 years, and I'll be blissfully unworried about how Silicon/M1 fares for the first few iterations. The other part of me wants to hang on for another year, 2 years, 3 years(?) until the new system is glitch-free and all of the programs work flawlessly on it, and buy a new 27" then. In addition to games via Steam, Final Cut Pro and Logic Pro would be the most intense things I'd be running. What would you all recommend? Thanks.

 

[OldCorpse]

Well, I’m in a somewhat similar, though not the same boat. I have a LATE 2009 27” iMac, High Sierra, and it’s showing its age.

What I plan on doing - I think I can hang on until the AS iMac. It’s a gamble. Because what I’d like is a REDESIGNED 27” AS iMac. If the first AS 27” iMac is *not* redesigned - as has been the case so far with the first AS macs - then that puts me in a very difficult situation, because I’d really, really, really not like to wait until 2022 to buy my iMac.

Why hang on to the 27”? Well, I too need FCPX on my desktop, and without question, I think the FCPX will be optimized on AS. So, I would not like to be limited for YEARS to any Intel Mac. Second, if there is a redesign, there is a chance (a *small* chance) that if the new smaller iMac is 24”, then the bigger will move to 30”-32”, and if there’s that kind of screen size, then that’s a game changer for FCPX, watching streaming movies and so on. However, I think the chance of a 30”-32” screen is really tiny.

Now, mind you, I have a late 2009 (top of the line at the time), and you have an early 2009, so that is a fairly significant difference in my mind. If I had an early 2009, then I think I’d lean to buying the latest Intel 27”, and I’d relax about getting the AS 27” until there’s a big redesign for sure, bugs have been worked out, and the technological leap is truly giant. Even if suddenly the 2022-23-24 AS iMac proves to be a huge change, and you simply MUST have it, then in the worst case you can sell your Intel (or trade it to Apple) and buy the new AS iMac hopefully without too big a financial hit.

 

[Jimmy James]

Apple released the low end first because performance is so far ahead of OP’s expectations.