Are M1 Chips Going to Take Over the World?

[zeronyne]

Jump ahead a bit in the video...these guys got a tremendous amount of new subscribers after their last video running almost 1000 plug ins in Logic with the M1 Mac, so they reintroduce themselves in this video.

 

A fun, non-scientific test of third party plug ins, both in and out of Logic Pro. Lot's of issues, but these are currently unsupported plug ins. As expected, third party plug ins used in Logic Pro work marginally better than those same plug ins in Studio One. Again, no surprise, but I wanted to see it, and these guys deliver.

 

[video:youtube]

[Anderton]

The whole Mac world will churn and we will have three years of complaints about obsolescence, dead end computers, probably another round of port incompatibilities, and strange form factors that are "Almost right" for audio, but not really "just right". They will always have plenty of computers capable of making music - a phone is more than capable of multi tracking a band if you can stand the mess of adapters and small screen.

 

There was a very interesting review in Forbes by a reviewer who took great pains to point out he had not been chosen by Apple to review the MacBook Pro with the M1 chip. He really felt that we weren't being told the whole story. The bottom line was basically that as long as you're totally in the Apple world, use Apple apps (other apps run slowly under Rosetta), are happy surfing with Safari, and can get by with one port (the other is used for power), the computer is okay as long as you don't take Apple's claims for battery life seriously. There were aspects he liked, and he thought it had potential, but according to him we're not there yet, and won't be for quite some time.

[Joe Muscara]

The whole Mac world will churn and we will have three years of complaints about obsolescence, dead end computers, probably another round of port incompatibilities, and strange form factors that are "Almost right" for audio, but not really "just right". They will always have plenty of computers capable of making music - a phone is more than capable of multi tracking a band if you can stand the mess of adapters and small screen.

 

There was a very interesting review in Forbes by a reviewer who took great pains to point out he had not been chosen by Apple to review the MacBook Pro with the M1 chip. He really felt that we weren't being told the whole story. The bottom line was basically that as long as you're totally in the Apple world, use Apple apps (other apps run slowly under Rosetta), are happy surfing with Safari, and can get by with one port (the other is used for power), the computer is okay as long as you don't take Apple's claims for battery life seriously. There were aspects he liked, and he thought it had potential, but according to him we're not there yet, and won't be for quite some time.

This guy has a reputation for being completely wrong about these things.

 

https://daringfireball.net/2020/12/m1_macs_truth_and_truthiness

 

Not one of these purported 'warts' with the M1 is ever mentioned. Whatever real problems Moorhead ran into (and some were indeed real), they"re all related to software issues. Either MacOS 11 Big Sur itself, or Rosetta, or apps he"s trying to use that have compatibility issues with MacOS 11 or Rosetta or both. But: software. Those issues are real, and the fact that M1 Macs require MacOS 11 Big Sur is the single biggest reason why some Mac users can"t or at least shouldn"t buy one just yet. Even if there were no architecture transition, in normal years it"s completely reasonable for many users to delay upgrading to major new releases of MacOS.

Is Chrome '100 percent Apple software'? Because Chrome runs better on the M1 than it does on any other Mac. Yes, Safari is faster and more energy efficient than Chrome, but Safari is faster than Chrome on Intel-based Macs as well.

So on the M1 Mac rave-review side, we have every major professional reviewer, along with dozens and dozens of ordinary M1 Mac purchasers out in the real world, doing real things, who just can"t believe what the M1 is capable of.

 

And on the other side, we have Patrick Moorhead, the guy who said Apple"s 64-bit A7 chip in the iPhone 5S was no big deal; that we shouldn"t have believed the benchmarks showing the original iPad Pro pantsing Intel"s chips five years ago; and that not only was Apple unlikely to switch the Mac away from Intel, but that Intel, in fact, was this close to getting its chips into the iPad Pro.

 

One of these sides deserves more skepticism than the other.

 

 

quite some time.

It'll be a minute, but I don't think it will be years. My money is on 2021.

[Anderton]

Well, here's what I wrote for Pro Sound News in July 2018, when audio people were freaking out because Apple was going to start using its own chips:

 

The latest Apple rumor is they"ll replace Intel"s chips with their own starting in 2020. Of course, 2020 is light years away in computer time, and Apple has been known to change its mind (seen any Apple cars lately?). While this move makes sense, I"m going to go out on a limb and explain why any angst may be unjustified for pro audio.

 

There was a Mac Pro refresh in 2017: both the low-end and high-end models got two more cores and better graphics. But there was no USB-C or Thunderbolt 3, or anything fundamentally different. Now Apple has committed to a 2019 launch date for the revamped Mac Pro. But will they use Apple chips? Let"s zoom out.

 

Apple is a consumer electronics company, not a computer company. We don"t use consumer machines: they"re today"s equivalent of yesteryear"s mainframes. I may be very wrong, but I don"t see any reason for Apple to jettison Intel"s powerful chips in their most powerful computers. Creating silicon at that level for a fairly insignificant product line (single-digits revenue share) doesn"t seem worth it.

 

But given that iPhones, iPads, Apple TV, and Apple watches already use ARM processors, I could easily see non-Intel chips working their way into notebooks and iMacs. Apple would not only have better margins, but consumer device technology has to be able to turn on a dime and Apple would no longer be held to Intel"s timelines. The Mac Mini is over three and a half years old, and the MacBook Air hasn"t changed much since 2010âtime for a serious refresh. This would also facilitate merging iOS and MacOS, which despite past pronouncements about wanting to keep them separate, appears inevitable.

 

Ultimately, I see three possible scenarios:

 

⢠The worst one for pro audio Mac users is that Apple decides high-power desktops for a small professional market aren"t worth it, dumbs them down, boots the Intel chips, and cedes the market to Windows.

⢠A somewhat better option is that Apple starts moving more of iOS (which let"s face it, needs work) over to MacOS and dismantles the wall between the two. The notebooks and iMacs favored by students, companies, home users, etc. become less expensive, more friendly, and gain market share. MacOS remains enough like 'vintage' MacOS that software companies don"t have to go too crazy porting to a new processor architecture or folding the changes into their Intel code. Also, the machines will still be powerful enough for pro audio applications that aren"t excessively demanding.

⢠The best option for pro audio is that Apple follows through with their 2019 modular computers, which are so good they re-ignite the Mac Pro line. They continue to use powerful Intel CPUs even while the consumer devices become more ARM-oriented, and MacOS still does what we need (even though some will grouse about 'consumer creep,' like they did when Logic Pro acquired some GarageBand vibe). And given that pro machines are big investments, Apple decides that the computers are powerful enough for professionals, and makes them expandable so they can have a longer lifespan. They need to get the design right only once, and the computers can coast for years yet still keep people happy. Maybe the reason for launching in 2019 instead of 2018 is that Apple needed more time to get these right.

 

You gotta admit I hit pretty much all the main points:

 

⢠Non-Intel chips are working their way into notebooks and iMacs

⢠Apple indeed did this for better margins, but also, to be able to turn on a dime and not be held to Intel's timelines.

⢠iOS is being integrated with MacOS

⢠The Mac Pro didn't use Apple chips but stuck with Intel

⢠Apple made the expensive, Intel-based pro machines expandable so they'd have a decent lifespan

 

So don't read any of those other people, just pay attention to what I say

 

I'm sure the Mac Pro will have Apple silicon at some point, but Apple has stuck with Intel for now. Looking back, it's clear that Apple planned to use Intel in the new Mac Pro all along; when I said that over two years ago, the Apple fans said I was crazy, that of course Apple would be putting its chips in their Mac Pro. My prediction wasn't based on religion or politics, but what's involved in ramping up consumer silicon to insane levels of number-crunching.

 

Having gone through the Motorola to PPC transition, and the PPC to Intel transition, there's no moment when the switch flips and everything runs perfectly. It gets better over time, some things fall by wayside, some things are ready now, and it takes way too long for some things to be updated. To me, 2021 is "quite some time" in computer years. I suspect that in mid-2021, the transition will be mostly complete. But it will take additional time for the world to catch up, jettison their old machines, and have it be a 95% Mac ARM world. And I don't think we'll see a Mac Pro that matches the performance of the current model in 2021, which is why I'm planning on getting a Mac Mini at the end of 2021 to hold me over until the Mac Pros make the conversion.

[Joe Muscara]

Okay, I'll give you our differing interpretations of "quite some time." Whenever these new Macs arrive, it will be interesting. Gruber talks a bit in that article about how mind-glowingly good the M1 Macs are - we never thought we could get fast, cool, and great battery life. It's like the old engineering axiom, do you want it to be good, cheap, or fast, you can only pick two out of three.

 

One of the cool things about Apple owning Logic Pro and Final Cut is they prove professional apps can run well on these things right out of the box. There's no excuse that third-party developers can't get their stuff running eventually.

 

BTW, did you see that despite there only being two TB4/USB ports on these models, each one has its own TB controller? So instead of the two ports sharing the bandwidth, each one can have the full bandwidth. They also support Thunderbolt Hubs.

 

https://eshop.macsales.com/blog/68484-thunderbolt-on-the-m1-mac-mini/

[Anderton]

Gruber talks a bit in that article about how mind-glowingly good the M1 Macs are - we never thought we could get fast, cool, and great battery life. It's like the old engineering axiom, do you want it to be good, cheap, or fast, you can only pick two out of three.

 

Well, we already have a Mac Pro that's good and fast - but definitely not cheap However, I find it interesting that Apple is using its higher margins to keep prices in line or in some cases, lower them slightly. I'd be fine with an an Apple silicon Mac Pro that's 60% of the current price and performance. Then I could have good, fast, and not as expensive.

 

One of the cool things about Apple owning Logic Pro and Final Cut is they prove professional apps can run well on these things right out of the box. There's no excuse that third-party developers can't get their stuff running eventually.

 

What I don't understand is why other developers can't hit the ground running. Do they not get advance units from Apple to test their software? Or is it just assumed that Rosetta is good enough until they complete a port?

 

BTW, did you see that despite there only being two TB4/USB ports on these models, each one has its own TB controller? So instead of the two ports sharing the bandwidth, each one can have the full bandwidth. They also support Thunderbolt Hubs.

 

I've seen some people grouse that they can't hook up two monitors to their M1-based Mac, so they have to use Sidecar with an iPad Pro. That doesn't make sense. If it has Thunderbolt, you should be able to hook up DisplayPort monitors, or worse case, a DisplayPort adapter. Either I'm missing something, or they are.

[Joe Muscara]

Gruber talks a bit in that article about how mind-glowingly good the M1 Macs are - we never thought we could get fast, cool, and great battery life. It's like the old engineering axiom, do you want it to be good, cheap, or fast, you can only pick two out of three.

 

Well, we already have a Mac Pro that's good and fast - but definitely not cheap However, I find it interesting that Apple is using its higher margins to keep prices in line or in some cases, lower them slightly. I'd be fine with an an Apple silicon Mac Pro that's 60% of the current price and performance. Then I could have good, fast, and not as expensive.

Heh - I meant out of fast, cool, and battery life. I was comparing that to the axiom. Macs won't ever be cheap, I guess!

 

One of the cool things about Apple owning Logic Pro and Final Cut is they prove professional apps can run well on these things right out of the box. There's no excuse that third-party developers can't get their stuff running eventually.

 

What I don't understand is why other developers can't hit the ground running. Do they not get advance units from Apple to test their software? Or is it just assumed that Rosetta is good enough until they complete a port?

I don't completely get it either. When Apple announced Apple Silicon back at WWDC, they said they had Mac minis for developers running A14X chips so they could start porting their software. Some of them took them up on it and those are probably the ones that had their apps ready on day one, or close to it. What I've heard is that the issue was the moving target of Big Sur, which went through a number of betas before it shipped. The other aspect is what I believe to be pretty specific to music software companies. They seem to want to do a whole battery of tests to verify their stuff is ready, and claim that can't be done with the betas. Universal Audio did something odd, IMNSHO. One of their employees got their stuff running under an early Big Sur beta and said outside of some graphical glitches, it seemed fine. However, before or around the time Big Sur shipped, they updated their installer to not run at all under it. It just tells you it won't work and then it closes. I had an older installer lying around from a few months ago, and that one worked just fine under Big Sur in my limited testing, as did their stuff. Note that this is Big Sur on Intel, and M1 is a different ball of wax since they also are tied to hardware. I get that. But that's not what their installer complained about.

 

I used to work for a Mac-only software company (not music software), and they were the same way in that they wouldn't even bother testing their apps except maybe to have a quick look until the final release of the new OS came out. Since I wasn't one of the programmers responsible for that stuff, I'd just :SMH: and hopefully not have to fill in for tech support when the calls asking if it was compatible started coming in. I did have to update the web site with the "we are currently testing" text though.

 

BTW, did you see that despite there only being two TB4/USB ports on these models, each one has its own TB controller? So instead of the two ports sharing the bandwidth, each one can have the full bandwidth. They also support Thunderbolt Hubs.

 

I've seen some people grouse that they can't hook up two monitors to their M1-based Mac, so they have to use Sidecar with an iPad Pro. That doesn't make sense. If it has Thunderbolt, you should be able to hook up DisplayPort monitors, or worse case, a DisplayPort adapter. Either I'm missing something, or they are.

I don't know how that works but it seems like it should not be a problem though it might cost some money for an adapter.

[Anderton]

And here's more fuel for the fire...the conclusion I draw from it is that if you're not stressing out the Mac, it'll do great but for pure speed, if you don't care about battery life and such, the Intel chips handle multiple cores better, which is why they can produce better benchmarks. Then again, I'm only running "benchmarks" instead of "programs" when rendering

 

https://wccftech.com/why-apple-m1-single-core-comparisons-are-fundamentally-flawed-with-benchmarks/

 

When you zoom out, it's same as it ever was...which is why I use an Apple laptop and an Intel desktop. Looks like that's the way it's going to keep rolling for the foreseeable future, because I need a huge amount of power for the desktop stuff, and reliability/low power for the laptop.

[Groove On]

... because I need a huge amount of power for the desktop stuff, and reliability/low power for the laptop.

Could an iMac Pro with Apple Silicon replace your current desktop? Or would you still be looking for the expansion ports? I'm not saying you would upgrade, I'm just wondering if an Apple Silicon iMac Pro could handle your needs.

 

Specs for the current Intel iMac Pro.

CPU: 10 / 12 / 14-Core Intel Xeon W

RAM: 32GB / 64GB / 128GB / 256GB

STORAGE: 1TB SSD / 2TB SSD / 4TB SSD

PORTS:

- 10Gb Ethernet

- Four USB-A ports

- Four Thunderbolt 3 (USB-C) ports with support for:

- DisplayPort 1.2, Thunderbolt (up to 40Gb/s), USB 3.1 Gen 2 (up to 10Gb/s)

 

* Note: I'm talking about the iMac Pro not the Mac Pro

[Anderton]

... because I need a huge amount of power for the desktop stuff, and reliability/low power for the laptop.

Could an iMac Pro with Apple Silicon replace your current desktop? Or would you still be looking for the expansion ports? I'm not saying you would upgrade, I'm just wondering if an Apple Silicon iMac Pro could handle your needs.

 

Well, I wouldn't buy anything Apple with Intel in it at this point. The main reason I want a Mac is to be a part of my Apple personal ecosystem (MacBook Pro laptop, iPhone) and for testing, reviews, etc. For my audio/video studio, I don't want an all-in one. I'm using a large touch-screen monitor and a second monitor, four internal drives, several cards, and I need a lot of ports. There's the cost factor too; $4,999 buys a powerful Windows machine from PC Audio Labs or Studio Cat. I'm also not going to switch from Vegas any time soon, which is Windows-only and outstanding video editing software; I've been using it since V 1.0...so that's a lot of data.

 

Expandability matters, too. I've had my main Windows machine for over 10 years, it still runs rock solid and has been expanded/upgraded over the years to where it's still a very high-performance computer.

 

Aside from Vegas, Cakewalk, and Logic Pro X, all the software I use is cross-platform so if it ever gets to the point where a Mac desktop can smoke a Windows desktop, it wouldn't be too hard to switch. But I've been using the Mac since 1986 and Windows since 1995, and I don't see myself becoming a single-platform guy any time soon. I switch back and forth depending on which is better at the moment. When Apple went through its fallow period, the PC was way ahead in terms of both performance and price. The gap has narrowed, and Apple certainly seems to want to move into a leadership position. But the bottom line is I can't do what I do now on my 10-year-old PC Audio Labs machine with a Mac, at least not without spending a lot more money.

 

None of this is religion to me. I just need to get work done, and I don't care about the logo. If Linux was the answer, I'd run Linux...whatever lets me meet deadlines

 

But another point to remember in all this is that Intel and AMD aren't just sitting around...who knows what they'll come up with two years hence to compete with a hopefully mature, powerful desktop with Apple silicon. I expect I'll continue to be dual-platform for quite some time.

[Groove On]

... I'm also not going to switch from Vegas any time soon, which is Windows-only and outstanding video editing software;

That is my one regret from going all Mac, Vegas rocks I really connected with their interface and way of doing things .... nonetheless, I had to give up long ago hoping they'd ever port it over.

[Anderton]

Ultimately, it wouldn't surprise me if Apple ceded the high-end/mainframe-like space, and went after market share in the consumer space. With fewer consumers using computers since tablets and smart phones do all they need, now is the time to gain market share, because the market is smaller.

[Mike Rivers]

 

 

What I don't understand is why other developers can't hit the ground running. Do they not get advance units from Apple to test their software? Or is it just assumed that Rosetta is good enough until they complete a port?

 

I used to work for a Mac-only software company (not music software), and they were the same way in that they wouldn't even bother testing their apps except maybe to have a quick look until the final release of the new OS came out.

 

For what it's worth, I first saw Windows 7 at a software/hardware vendor's off-site demo at a NAMM show several months before it was released to the public. Obviously somebody gets it.

[Anderton]

Interesting you should mention that, Mike. Just before Windows 95 came out, I was invited (along with several other people) to a preview of the audio features Microsoft was expecting to deploy. It was not "here's what we're doing, deal with it." I made several comments; and Microsoft people were taking notes, and asking questions.

 

Then at AES, I wanted to go to an Apple developer's conference to see what was happening with the Mac. I was told that I was not allowed admittance. I assured the gatekeeper that I would not report on anything or make any public or private mention of what I saw, but I needed to be up to date on what was happening with the Mac's operating system. Never made it past the velvet rope.

 

Fast forward, and the Mac did Core Audio, which is very cool. Windows has finally come up with a protocol that rivals it, although adoption has been slow. And I should also mention that Apple has been well-represented in the MIDI Manufacturers Association, and I think Pete Brown from Microsoft will be participating in 2021. So both companies care about music ,and I have high hopes for the future, especially regarding MIDI 2.0.