ARM based MacBook Pro discussion [merged]

[Dozer_Zaibatsu]

Was looking for more info on what people are seeing/expecting for Apple Silicon. It's really quiet on the main page and even the forums. I think this is one of the more important things Apple has ever gone into.

Case being, the world has really grasped on to working (and school) remote. I work for a start-up that has recently just decided a whole lot of our workers, coders and [product] users will be permanently remote. We have an aging fleet of older Macs for the developer side, and recently have been sending out lots of Dells for workers doing standard office crunching.

My boss grumbles now about supporting Macs. I gotta say, it's a crossroads. I lobbied successfully against his plan on buying the cheap 2019 Macbook Airs for remote users because of the awful thermals. We made use of existing stock where possible, and bought 2020 MacBookPro 13s instead of Airs if we needed Macs. At least for now. If we're going to continue to drop $$$ on Macs, it depends on what Apple on Silicon looks like, simple as that.

Intel's 11th gen just dropped, and they did their usual yeoman's job in lurching along their usual improvements, with notably much better integrated graphics. AMD is just stomping everybody in benchmarks. (Although I have hesitations on AMD. They don't have drivers the way Intel does for burst-clock tasks, etc.) The T2 chips in current Macs are no slouch at improving productivity in export tasks. It is time for Apple, as basically the new third chip-maker, to lay it on the table and show us what to expect for Macs as work computers.

I feel like ten years ago when we were genuinely wondering what Apple was going to do with the Mac. We all know Apple has a great business with iOS devices, and I'd be satisfied with those products for what they are if it was all they made. But the prospects for the future of the Mac really depends on the next two months.

 

[yanksrock100]

Was looking for more info on what people are seeing/expecting for Apple Silicon. It's really quiet on the main page and even the forums. I think this is one of the more important things Apple has ever gone into.

Agreed, it does seem oddly quiet considering the Mac is about to experience the biggest fundamental change since 2006...

I do wonder how Apple is planning to position the Mac going forward for business. I work at a mid-sized company that gives Macs to developers - I think the flexibility of being able to dual boot Windows and conveniently develop containerized applications makes intel Macs a solid choice for us. My initial thought is that companies are going to think twice before upgrading the fleet to ARM Macs. I know we've seem some initial demos of Docker and linux virtualization working, but I feel like Windows virtualization (and how well it runs) is the big question at the moment...

That being said, I will gladly purchase the ARM Mac as my personal device. I haven't used Windows in years and I don't plan to anytime soon.

 

[pshufd]

Agreed, it does seem oddly quiet considering the Mac is about to experience the biggest fundamental change since 2006...

I do wonder how Apple is planning to position the Mac going forward for business. I work at a mid-sized company that gives Macs to developers - I think the flexibility of being able to dual boot Windows and conveniently develop containerized applications makes intel Macs a solid choice for us. My initial thought is that companies are going to think twice before upgrading the fleet to ARM Macs. I know we've seem some initial demos of Docker and linux virtualization working, but I feel like Windows virtualization (and how well it runs) is the big question at the moment...

That being said, I will gladly purchase the ARM Mac as my personal device. I haven't used Windows in years and I don't plan to anytime soon.

There are a lot of marketing, sales and other folks that don't need multiplatform systems and I think that they'd be fine unless they have specific software that needs porting. At the moment, I'm looking at Apple Silicon as more of a general purpose system but not for development. I think that the handwriting is on the wall - if you need to do Windows and Linux, then you need to stay on x64. Whether that's your primary machine or you get a second machine or whether you use in in-house x86 cloud system.

 

[Venderious]

What do you guys think that the design of the new MacBooks will be like?

 

[Maximara]

There are a lot of marketing, sales and other folks that don't need multiplatform systems and I think that they'd be fine unless they have specific software that needs porting. At the moment, I'm looking at Apple Silicon as more of a general purpose system but not for development. I think that the handwriting is on the wall - if you need to do Windows and Linux, then you need to stay on x64. Whether that's your primary machine or you get a second machine or whether you use in in-house x86 cloud system.

The thing is Microsoft wants ARM to take off (they have effectively been banging their head into wall for two years to get Windows on Arm accepted as a development platform)for the simple reason that PC providers could dump all the preinstalled adware which would make Windows run more effectively.

More over the mobile market is where the money is and that is ARM be it Android, Surface Pro X, or iPhone/iPad. Heck, the Raspberry Pi uses ARM.

As has been pointed out before the better development software doesn't care what the CPU is as it uses an abstract layer.. For Docker I pointed out the following links:

* Getting started with Docker for Arm on Linux
* Deploying Docker Containers on Arm Hardware Just Got Easier (Apr 25, 2019)
* Leverage multi-CPU architecture support
* Preparation toward running Docker on ARM Mac: Building multi-arch images with Docker BuildX

Other tools:
* "Swift for Windows"

Yes the "handwriting is on the wall" but it is for the x86 - it has been taken as far as is possible and has effectively lost the mobile market...the market where the future is. And that future is clearly ARM not x86.

 

[Unregistered 4U]

I think it’s because pretty much all speculation points have fairly well trodden. This close to the event, folks are willing to wait to see which one set of speculations hit closer to home.

 

[Sam Luis Obispo]

No doubt some people in Apple are sweating bullets on the final validation testing. Promise was made and it sounds like a November announcement will occur. But when will it ship?

 

[Maximara]

No doubt some people in Apple are sweating bullets on the final validation testing. Promise was made and it sounds like a November announcement will occur. But when will it ship?

The issue is not so much final validation testing but expectation vs promise. Expectations, IMHO have gotten slightly out of control.

 

[JohnnyGo]

What do you guys think that the design of the new MacBooks will be like?

Less borders = Infinity screen
Same minimalist ports
Different touchbar ?
Less weight (10% smaller battery)

 

[JohnnyGo]

No doubt some people in Apple are sweating bullets on the final validation testing. Promise was made and it sounds like a November announcement will occur. But when will it ship?

It will ship when it's ready ?

My guess is that shipment is end of November or, most likely, mid December

 

[ascender]

It will ship when it's ready ?

My guess is that shipment is end of November or, most likely, mid December

Apple have previous history of things shipping in the final weeks of December in order to make their Q4 promise on products.

I'm still hoping the ARM MBP starts with feature-parity from the iPad Pros and then adds to it.

So ProMotion screen, FaceID, industrial design, cellular option.

Improved speakers on the 13" would be nice along with thinner bezels, a 14" screen too?

Will they keep the TouchBar or will there be a touch screen given the direction of Big Sur? Having got really used to using an iPad Pro with Magic Keyboard, I think they should make an on-screen version of the Touch Bar, right at the bottom of the display, but leave the function keys intact. That way you could just reach up when typing and touch the screen to select emojis, autocorrect options etc.

 

[leman]

Was looking for more info on what people are seeing/expecting for Apple Silicon. It's really quiet on the main page and even the forums. I think this is one of the more important things Apple has ever gone into.

Well, the A14 is out and while we still waiting for in-depth analysis by Anandtech and friends, I'd say that it looks good for the Apple Silicon Macs. The big question is what core configurations we will see on them. But to put things in perspective:

- single-core Geekbench scores for 3.0 ghz iPhone A14 (5 watts) essentially match the Intel Tiger Lake at 4.8 ghz (20+ watts)
- multi-core Geekbench scores for the iPhone (2 fast and 4 slow cores, 5 watts) are only 30% below Intel Tiger Lake (quad core with hyper threading, 28 watts)
- in a gaming benchmark (3dMark Wild Life), the iPhone 12 appress to be at least 10% faster compared to the fastest 13" MBP Intel iGPU. Again, we are comparing a 5 watt phone SoC vs. a 30W laptop SoC...

Basically, a conservatively clocked 4 CPU core based on the A14 should perform similarly to the 8-core i9 in the 16" (while being significantly faster on single-threaded tasks). An Apple A14 SoC at 30 watts should handily outperform any Tiger Lake at 30 watts.

P.S. Here is a Geekbench comparison of a 5W iPhone chip vs. a pre-release AMD 5800X Zen 3.0 desktop CPU at 105W, which I will leave uncommented: https://browser.geekbench.com/v5/cpu/compare/4400063?baseline=4174823

 

[Dozer_Zaibatsu]

No doubt some people in Apple are sweating bullets on the final validation testing. Promise was made and it sounds like a November announcement will occur. But when will it ship?

An interesting theory I can get behind this days is that Apple really did intend to do this sooner. (As they've obviously been planning this transition for years.) Linus Tech Tips on YouTube always does as complete a tear-down of most consumer computers. Their review of the 2019 i3 MacBookAir was interesting. The i3 Air famously does not have the heat sink connected to the CPU. Which is outrageously poor engineering (or so it would seem) for Apple.

The theory is that the MacBookAir had been intended all along to run an Apple ARM chip. But perhaps it just wasn't ready, and they had to last-minute order a dump-truck full of i3 (i.e., Y-series chips with a new label) for the Air.

 

[Dozer_Zaibatsu]

Less borders = Infinity screen
Same minimalist ports
Different touchbar ?
Less weight (10% smaller battery)

You've just invented the Dell XPS.
/sarc

 

[pshufd]

An interesting theory I can get behind this days is that Apple really did intend to do this sooner. (As they've obviously been planning this transition for years.) Linus Tech Tips on YouTube always does as complete a tear-down of most consumer computers. Their review of the 2019 i3 MacBookAir was interesting. The i3 Air famously does not have the heat sink connected to the CPU. Which is outrageously poor engineering (or so it would seem) for Apple.

The theory is that the MacBookAir had been intended all along to run an Apple ARM chip. But perhaps it just wasn't ready, and they had to last-minute order a dump-truck full of i3 (i.e., Y-series chips with a new label) for the Air.

They can always blame COVID.

 

[leman]

The i3 Air famously does not have the heat sink connected to the CPU. Which is outrageously poor engineering (or so it would seem) for Apple.

The theory is that the MacBookAir had been intended all along to run an Apple ARM chip. But perhaps it just wasn't ready, and they had to last-minute order a dump-truck full of i3 (i.e., Y-series chips with a new label) for the Air.

There is a much simpler explanation. Its an entry-level consumer device with a 9W CPU... why bother with engineering and manufacturing a more expensive cooling solution when a simple semi-passive heatsink will cover all the intended use cases? You are reading way too much into all this.

Besides, it's not a "heat sink not connected to the CPU" but a case fan instead of CPU fan. And it was already the design of the 2018 model.

 

[Dozer_Zaibatsu]

....

Basically, a conservatively clocked 4 CPU core based on the A14 should perform similarly to the 8-core i9 in the 16" (while being significantly faster on single-threaded tasks). An Apple A14 SoC at 30 watts should handily outperform any Tiger Lake at 30 watts.

P.S. Here is a Geekbench comparison of a 5W iPhone chip vs. a pre-release AMD 5800X Zen 3.0 desktop CPU at 105W, which I will leave uncommented: https://browser.geekbench.com/v5/cpu/compare/4400063?baseline=4174823

Yes, I've heard the synthetic benchmarks optimism as well. Keep in mind that leaked performance results leaded from the developer kits show more modest results from real-world tasks. I definitely want to believe we'll see something revolutionary. It certainly would be a mistake for the debut to be a dud, considering how they set themselves up for it. I'll manage my expectations. Best case would be Apple experience with high performance and reasonable prices. I feel like we can get two of those in any mix, but it would unexpected to get all three.

 

[leman]

Yes, I've heard the synthetic benchmarks optimism as well. Keep in mind that leaked performance results leaded from the developer kits show more modest results from real-world tasks. I definitely want to believe we'll see something revolutionary. It certainly would be a mistake for the debut to be a dud, considering how they set themselves up for it. I'll manage my expectations. Best case would be Apple experience with high performance and reasonable prices. I feel like we can get two of those in any mix, but it would unexpected to get all three.

The developer kit is using a two year old iPad chip... it's performance is pretty much a known quantity.

I don't think we will see anything revolutionary per se, but the mobile A14 so far performs on equal footing with latest and greatest from Intel and AMD while using considerably less power. Scale the chip up, put it in a laptop enclosure, increase the clocks slightly, and you will have excellent performance in a form factor that other chip manufacturers won't be able to reach for a while.

Let us also not forget that the most realistic rumors point to first ARM MacBooks being equipped with the A14X — most likely the same chip to power the upcoming iPad Pro. This makes a lot of sense. A Macbook using a 10-15W 4+4+8 A14X will rival any 30W Tiger Lake laptop while being considerably lighter and having better battery life. Frankly, I don't expect big surprises here.

Surprises will come later next year when Apple is going to present their higher-end solutions. We still have no clue which path they will choose to compete with higher-end dedicated GPUs and mature CPU platforms with their plentiful I/O.

 

[MysticCow]

An interesting theory I can get behind this days is that Apple really did intend to do this sooner. (As they've obviously been planning this transition for years.)

FUN FACT: Apple kept a working copy of every single Mac OS X version (since literally Public Beta, well before 10.0) for INTEL, because they never knew if they needed to swap to Intel or stay with PowerPC. We know this because there were Sony Vaios running Mac OS X around 2001 (right when Mac OS X began and I bought my 700 MHz G3 iMac).

So that wouldn't surprise me if ASMacs could have been around sooner. It wouldn't surprise me if Apple has kept a secret Mac OS X running on iPads to test out how the chips would hold. Well, at least since 2009 when PA Semi was bought or more likely 2010.

 

[ascender]

There is a much simpler explanation. Its an entry-level consumer device with a 9W CPU... why bother with engineering and manufacturing a more expensive cooling solution when a simple semi-passive heatsink will cover all the intended use cases? You are reading way too much into all this.

Besides, it's not a "heat sink not connected to the CPU" but a case fan instead of CPU fan. And it was already the design of the 2018 model.

The simplest explanation is usually the correct one.

Laptops like the Air are very much built for the average consumer who will just buy it and use it for many years without questioning why the fans come on occasionally or what temperature the CPU is working at under load, let alone looking at thermal paste or wondering about the heat pipe.

I suspect the keyboard change and covid challenges have been the unforeseen bumps in the road for Apple. Look at the performance of the iPad Pros - I'm sure if they'd wanted to, they could have released an Air with one of those in it before now. I still think we're on-track to see a laptop of some sort this year as they said, which will ship with Big Sur on it.

 

[DanMan619]

FUN FACT: Apple kept a working copy of every single Mac OS X version (since literally Public Beta, well before 10.0) for INTEL, because they never knew if they needed to swap to Intel or stay with PowerPC. We know this because there were Sony Vaios running Mac OS X around 2001 (right when Mac OS X began and I bought my 700 MHz G3 iMac).

So that wouldn't surprise me if ASMacs could have been around sooner. It wouldn't surprise me if Apple has kept a secret Mac OS X running on iPads to test out how the chips would hold. Well, at least since 2009 when PA Semi was bought or more likely 2010.

I definitely could see that being the case. Companies like Apple have tons and tons of prototypes that never see the light of day or only elements of them make it to a consumer product. Apple's got experimental hardware and software locked away in a lab somewhere that is probably all levels of wild stuff just for the sake of testing potential options that they may or may not decide to go with in the future. I think the main reasons they held off this long on switching to Apple Silicon possibly may have just been to give Intel the chance to get their stuff together.

I think Apple would have been happy to stick with Intel long as they had delivered the consistent performance and efficiency gains Apple wanted. However, after one too many chip process delays and it affecting Apple's products thermally and reputationally, i think that's what made them decide to pull the trigger on the transition even though they could probably could have announced and started it sooner if they wanted to.

Similar to how in the recent reports that came out today that Apple's low key been working on a search engine in the event antitrust regulators go after Google (which in the U.S. is happening now and they're specifically targeting the deal where Google pays Apple to use Google Search on iPhones). I don't think Apple had any serious intention to be in the search engine business unprovoked, but they worked on it anyway knowing they may need it one day if things with Google go south.

 

[Lammers]

FUN FACT: Apple kept a working copy of every single Mac OS X version (since literally Public Beta, well before 10.0) for INTEL, because they never knew if they needed to swap to Intel or stay with PowerPC. We know this because there were Sony Vaios running Mac OS X around 2001 (right when Mac OS X began and I bought my 700 MHz G3 iMac).

So that wouldn't surprise me if ASMacs could have been around sooner. It wouldn't surprise me if Apple has kept a secret Mac OS X running on iPads to test out how the chips would hold. Well, at least since 2009 when PA Semi was bought or more likely 2010.

That’s not exactly true, at least according to what has been reported. Mac OS X for Intel apparently didn’t become a real project until 2002. Prior to that it was the work of a single engineer working on it on his own initiative:

A Bit of History Behind the Mac OS X on Intel Project "Marklar"

Apple's announcement to move their Macs from PowerPC to Intel processors in 2005 was a huge surprise at the time. Serious reports about the plan...

www.macrumors.com

But I agree it’s very likely that Apple has had this option in their back pocket for some time.

 

[Maximara]

That’s not exactly true, at least according to what has been reported. Mac OS X for Intel apparently didn’t become a real project until 2002. Prior to that it was the work of a single engineer working on it on his own initiative:

A Bit of History Behind the Mac OS X on Intel Project "Marklar"

Apple's announcement to move their Macs from PowerPC to Intel processors in 2005 was a huge surprise at the time. Serious reports about the plan...

www.macrumors.com

But I agree it’s very likely that Apple has had this option in their back pocket for some time.

The comments give some better insight and it is clear that whoever did the article didn't go deep enough with their research:

"OS X was promised for Intel; Apple shipped developer previews of OS X (as Rhapsody) for Intel. I have a copy of Rhapsody DR2 and have installed it on commodity PC hardware. It was a surprise and a let-down for developers when Apple decided to reneg on shipping the Intel version along with Yellow Box (Cocoa) for Windows. Rhapsody DR2 was released in 1998, this effort began in 2000 (before OS X 10.0, actually), so at most there is a two year gap where the status of the port is uncertain. It is likely that it was kept maintained for at least part of the gap, especially since Darwin was publicly released as an open-source project with Intel support in April 2000. What do you suppose the Intel version of Darwin was there for? No reason? It is certainly possible that in the crunch to get OS X to 10.0 for PPC the Intel port fell by the wayside and perhaps was no longer feature-complete, and this guy's job was to get it caught up, but it was still something they had sitting around the whole time." - Suboptimus

"An early Mac OS X preview release. DR2 was the last NEXTSTEP/Mac OS X to run on Intel until the Intel transition in 2006." - Mac OS X Rhapsody

Mac OS Rhapsody DR1, DR2 Intel (x86) & PPC - Year released: 1998

So the picture the article portrays is not an entirely accurate one and Suboptimus' version can be confirmed via other sources and an actual build it that version.

 

[ascender]

The key difference this time around is that Apple have years of experience with these chips, on production systems rather than prototypes, due to the iPhone and iPad.

I genuinely can’t wait to see what they have up their sleeve.

 

[nquinn]

Well, the A14 is out and while we still waiting for in-depth analysis by Anandtech and friends, I'd say that it looks good for the Apple Silicon Macs. The big question is what core configurations we will see on them. But to put things in perspective:

- single-core Geekbench scores for 3.0 ghz iPhone A14 (5 watts) essentially match the Intel Tiger Lake at 4.8 ghz (20+ watts)
- multi-core Geekbench scores for the iPhone (2 fast and 4 slow cores, 5 watts) are only 30% below Intel Tiger Lake (quad core with hyper threading, 28 watts)
- in a gaming benchmark (3dMark Wild Life), the iPhone 12 appress to be at least 10% faster compared to the fastest 13" MBP Intel iGPU. Again, we are comparing a 5 watt phone SoC vs. a 30W laptop SoC...

Basically, a conservatively clocked 4 CPU core based on the A14 should perform similarly to the 8-core i9 in the 16" (while being significantly faster on single-threaded tasks). An Apple A14 SoC at 30 watts should handily outperform any Tiger Lake at 30 watts.

P.S. Here is a Geekbench comparison of a 5W iPhone chip vs. a pre-release AMD 5800X Zen 3.0 desktop CPU at 105W, which I will leave uncommented: https://browser.geekbench.com/v5/cpu/compare/4400063?baseline=4174823

I think no matter what the case:
- Apple won't release an ARM macbook pro that is slower than the early 2020 13" intel macbook pro
- Apple likely will want the ARM chip to clock at least to similar performance to an intel tiger lake cpu to be competitive

I really wish they were further along for the 16" cpu / ~45w chip.