I see a myriad of comments that boil down to “duh”, but I think these comments are missing the point.
Consider an alternate possibility: “sales of Apple’s new ARM macs outpaced by older models”.
Had Apple not knocked M1 out of the park, this was a real possibility. Is it surprising that a well executed architecture change is now selling well? Not really.
But the more interesting point is that the rest of the market agrees with us early adopters and tech nerds.
This would be a very different headline if devs were struggling to port their software, or if Rosetta 2 didn’t live up to its promise.
I wonder how many of those who got an M1 mac recently are aware of what exactly that means, besides that it's new and fast. I could imagine a large part of those sales to fall into the "I need a new computer, I'll get another Macbook Air" category. Those buyers wouldn't have much reason to care anyway; I guess you'd have to do pretty specialized work to really notice a big difference besides "cool" and "fast".
I'm using an M1 MacBook Air right now. It's more nuanced than speed. That thing's responses are insane. It's an insanely fast computer which wakes up like an iPhone and has an endurance like a vintage, entry level Nokia phone.
It's ready to work before I finish opening up the screen and, they forgot to tell the battery that it needs to deplete after some time.
It doesn't matter for ordinary people to know all the cool tech and architectural stuff running below that keyboard, because at the end of the day, your ordinary applications are running, using way less power and it responds so fast that you can't even think.
This matters way more for end users than for us, developers. Because we the devs & admins know the nature of the beast, and the ordinary users have unmoored expectations (and it's not a bad thing TBH).
I have a personal intel Mac and office gave me an M1 one. I reach for M1 without even knowing.
I've had top of the line Intel/Windows laptops since the dawn of time, and the latency and responsiveness of my Air destroys them all. Simply navigating through the OS is crisp as is dealing with apps.
My XPS always had tiny bits of lag here and there, enough so that I hated using it. I hated doing any intensive work with it because the fans were annoying. My Air completely stomps it in speed and is silent while doing it.
Surfing with Safari is noticeably faster than on any other computer I have, even my much more powerful Ryzen 3700. Even my Ipad Air 4 is much faster than any browser on my Ryzen machine.
Rosetta 2 just works.
Compare the M1 Air to the new ARM Surface (which is 50% more expensive), and you see how far behind MS is. Their ARM cpu is 5 years out of date, and their x86>ARM translation software is flat out broken. Keep in mind MS has been fiddling with ARM devices, so they have zero excuse for being so far behind.
I'm not a platform evangelist, but Apple is firing on all cylinders and nailing execution across the board. MS seems disinterested in software, and Intel is lost at sea.
> Because we the devs & admins know the nature of the beast,
I would not lump all devs under 1 group. While the technological aspects of M1 are impressive, the compatibility issues and Apple "being Apple" for lack of a better word are significant enough for some of us to steer clear away. My work issued MBP just doesn't recognize the external display 80% of the time without me having to plug and replug it or usbc hub back in, and the fact that its an issue is simply unnacceptable when my 5 year old Dell, running Linux, has no issues.
Especially for the devs that use desktops on a regular basis, as the speed that you talk of is something that we have had for the past 3-4 years.
Honestly, I've seen only one application crash on me due to using an M1 Mac. TBH, didn't try to port my scientific applications on M1 though (most of the time, they run on Linux servers, so no rush).
> Especially for the devs that use desktops on a regular basis...
In fact, I'm a desktop first user. Using M1 daily, since we're WFH. Maybe you can be this responsive on a desktop, but I'm talking on the mobile space. If we take out form factor considerations, I can get the same speed and responsiveness from 6-7 year old servers we use at the office.
I watch youtube before I sleep (I know, bad habit :P) and sometimes I forget to put it on charge before I start watching, and when I fall asleep, youtube autoplay continues to play random crap over and over (usually ends up with https://www.youtube.com/watch?v=YKAblynZYhI somewhere in the chain)
OK the point is, more often than not I've woken up in the morning with youtube still playing. Absolutely insane.
In January, I switched completely to Mac from PC, and I did it with an M1 mini. It was time to upgrade my ancient Dell XPS (bought it in 2012 or something), and decided what the heck. I'll give it a go. For only $700, I got probably the best computing experience of my life. Rosetta has been amazing, and a lot of the Mac software I used to use on my top-of-the-line 2019 Intel MBA just fly.
When I first started using it there were some glaring issues that have all disappeared as software has been updated/optimized for the M1 architecture.
Notably back in February I attempted to write a .net core console app, and it would never enter Main() {}. A week later there was an update to .net, and that went away.
I still use my Intel MBA for Parallels. A client of mine is Windows only, and I have to VPN in, then RDP into the development machine, but I'm hoping that Parallels has released the new ARM version officially, it might just work (there were issues regarding connectivity while I was beta testing it).
You won't be the M1 early adopters in 2020 that had their software break in front of them this time when getting an M1X or M2 device this year.
Not only it will be a lot faster thanks to it being possibly going down to 4nm, the software ecosystem will be much improved and supported by it from day 0.
Do we know if the silicon shortages will affect the M2 or what comes after? I wonder if they booked enough volume with TSMC to get to 4nm or whatever the next process shrink is or if they will be trying to shove M1s into everything.
AFAIK the silicon shortages (of high end nodes) are partly because Apple has all the capacity (which they have because they're swimming in cash and have high margins and thus have been able to outbid everyone else).
Apple fused a high performance core to a platform which used to run a latency critical platform, the iPhone. Actually it happened over the years.
- They learned to design a phone platform.
- Then, shoehorned a modified version of macOS on top of it.
- Then, they started to iterate on performance. Added iPad. They increased the performance while keeping latency criticality of iOS platform. Trying to have performing CPU/GPU and trying to keep power consumption low paved the way, year over year.
This iteration over and over resulted in M1. A descendant of iPhone and iPad, but with unbelievable performance. They kept the best parts of every platform since they can design from the processor to the latest bit of the software.
At the end of the day, M1 is greater than sum of its parts. An M1 Air behaves a lot of like an iOS device in some parts. Connect to power, screen comes on, a big battery appears with "XY% charged" message and you hear iOS charging chime. It's funny, yet hints on the heritage.
Apple fused everything together. A unified ARM platform. Variety of OSes with the same roots. One optimization for all. With a big blow, they theoretically reduced the software work they have to do dramatically, so they can build magical software again with last effort and can transfer technologies from one to another virtually with no effort.
See: iPad Pro with M1. iMac with M1, MacBook Air and MacBook Pro with M1. They are the same device. Same performance. A slightly different OS on iPad. That's just it.
Apple is now able to codesign the software and hardware top to bottom. I have no inside knowledge but using one of these things it's basically obvious what's going on. They only have to write drivers for their own hardware, and for their own GUI library. So I think they've been really aggressive about making the interactions low friction.
It's not just that the hardware is fast. It's that the sort of hitches, pauses, etc that we're used to are just gone. They've made the way the software and hardware interact a nearly perfectly smooth racetrack. They've done this to such a degree that it must have been a very overt high level goal.
> They only have to write drivers for their own hardware, and for their own GUI library.
They’ve always had limited processor and GPU options though (relative to MS where Windows had to support an insane amount of options), so wouldn’t they have optimized for what they knew would be in the case?
They'd still be constrained by Intel/nVidia's expected API design and almost certainly some key code from them as well. With apple controlling the hardware too, they can make the interface work however they want. That introduces the possibility of making some radical simplifications or optimizations.
ARM Windows laptops have been around for a while, and have not taken off.
Some of that could be that these computers don't represent a shift in the wind - x86 remains the primary architecture for Windows. But I think that another big part of it is that every review of a Windows ARM laptop is full of caveats. Some programs will run slower, some programs won't run at all. Knowing which is which requires technical knowledge that a lot of people don't have, such as the ability to distinguish native executables from Java or .NET apps, or the ability to distinguish 32-bit programs from 64-bit ones.
Apple, on the other hand, really outdid themselves here. The story for most consumers is just, "It will run all your old apps, no problem." The worst-case-scenario story for a certain niche is, "Docker may be flaky until later this year."
IOW, the fact that "those buyers wouldn't have much reason to care anyway" represents an enormous accomplishment on Apple's part.
Microsoft don't control the whole ecosystem unfortunately. They have a much harder time making that architectural shift than Apple does (although they could have done it better). There's all sorts of backwards compatibility guarantees with Windows that Apple doesn't have to care about.
Rosetta is the key piece here to 'fill in the gaps'. IMO, Rosetta only really works because of Apple's silicon and so how insanely fast the machine is compared to the Intel equivalent. The M1 has such a performance advantage over Intel that you can use it up in translation and your options become "x86 version of program running at roughly the speed it always used to" or "ARM version that is lightning quick".
Whereas the Windows ARM machines weren't a performance leap over Intel so even if you had translation it would be "x86 version that is very sluggish" or "ARM version that is just about OK".
Unfortunately, Windows ARM laptops were and are poorly executed.
Apple is carrot and stick, where the carrot is performance and the stick is obsolescence.
With Windows, there's neither carrot or stick. ARM isn't that much more performant, the battery life isn't that much better, and they're certainly not abandoning Intel.
OTOH, Microsoft could invest a huge amount of their money on making ARM attractive, but the Wintel ties clearly run deep.
My gf and I each earn FANG salaries and could upgrade a,couple of times a year if we were so inclined but haven’t bothered until the M1 arrived. IPhone 6s & 7, watch 0 & 3, 2016 Macbook + 2017 MacBook Air (TBF she has a big 2019 Intel MBP from work though)...upgrading that old gear didn’t seem worth it. Patches and upgrades continued to arrive (except for the watch 0).
But we finally upgraded everything and now I’ll upgrade my 2014 ipad that I’m typing on now.
I doubt most people upgrade simply because something new has arrived. Apple doesn’t really drive you to it.
I think the "stick" the GP was referring to is applied to developers, not consumers. Developers who don't get on board with Windows ARM get... to save themselves some effort. Developers who don't get on board with M1 will have their apps running slower than their competitors' apps across the entire ecosystem within another year or two, period.
Whilst you can run Intel code via Rosetta indefinitely(), Apple's strategic position is no more Intel hardware.
I'm typing this on a 2013 MBP. I'm personally waiting for an M2 chip, which I expect will follow Moore's law, albeit for ARM. Which backs up your point.
() indefinitely as defined by Apple is until they change their mind.
Depends on who you’re asking. My 2012 Thinkpad is on the latest Windows 10. It works very well with its SSD and it is really quiet (new thermal paste). Meanwhile, my 2011 MBA cannot even get GarageBand from the App Store because its OS won’t upgrade. Otherwise it’s fine...
The decade of hand wringing about Tim Cook’s deficiencies as CEO compared to Steve Jobs have got to be put to rest by even the biggest skeptics. The timing, execution, and scale of this transition has been so well done.
Jobs was focused on vision and product design. Cook is on operations and execution -- it also shows in his presentations. He follows the relatively safe "Apple-formula" that was put in place in the 2000s. Even the transition to Apple Silicon mirrors that of PowerPC to Intel.
Looking at the products it is hard to see where Cook has made his impression. It feels like it is about refinement and improvements year after year. That isn't a bad thing, because over time those things add up, and Apple is probably a much more well-oil and stable machine behind the facade today than 10 years ago. This is what Cook does (even before he became CEO).
I suspect Cook also knows that this is his strength, and that's where he should focus, but I would like to see something more risky and exciting come out from Apple.
I'd argue that the fundamental worry of the skeptics is correct. Apple hasn't been innovative in the same way. Apple's not brought out any products that created or disrupted an entire market niche since Cook took over, and it's arguably taken fewer risks. Instead, Apple's big wins over the past decade have been things like this: Really impressive operational successes.
Personally, I'm OK with that. But I've never really stayed abreast of the latest trends, anyway. I can also see where people who enjoyed the excitement of those famous "one more thing" moments might feel like some of the magic has been lost.
An old app in Apple's ecosystem isn't the same as an old app in Windows.
You can still run executables compiled under Windows2000 today. Meanwhile, programs as old simply won't work in MacOS (at the time, Apple was shipping OS9).
I have no doubt that a Windows release with legacy components removed can run fast on ARM (it was already demonstrated with Windows 10 mobile).
But that's just the tip of the iceberg: Nobody is shipping a chip comparable to the M1. All the ARM vendors are focused on phone chips, with completely different power optimization goals.
He "likes" it, as opposed to "loves" it; because he doesn't actually care that much (he's not a tech nerd). He likes his iPad a lot more.
He would probably feel the same way about a new Dell or Lenovo. It's just not that important to him. Computers are a "necessary evil." He's a musician, and he asked my advice, so he got a Mac.
For myself, I am waiting on the M1X/M2, later this year. I have a couple of big external monitors, and I know the M1 has issues with those. I've already weaned myself off of my BlackMagic eGPU Pro, in anticipation.
> He would probably feel the same way about a new Dell or Lenovo.
Not sure if this was ever true for someone who used a Mac before. Going to a Dell or Lenovo feels like a huge downgrade. Everything about these laptops feels cheap in comparison to a Mac. They are bigger, thicker, heavier, more plastic, hotter, and less elegant. Also the track pad and other screen on the Mac is so much better. Even if some Windows laptops come with a crazy high resolution, the OS looks shit on it and the user has to go into the Windows menu and play around with all the UX controls until the proportions of the start bar, windows and other UI elements somehow make sense on the high resolution again. Of course there is a million Windows apps still out there which don't scale at all. I cannot imagine that a user who was on Mac for more than a year could go back to Dell or Lenovo and feel excited about it. Now with the M1 the contrast is even worse. I have an M1 and Intel MBP and even going back to the Intel MBP feels really crappy because I need to charge my device when working for a full day. I got so used to the M1 not needed a single charge during the day that it's really odd to think about how other laptops still do. I treat my M1 like my iPhone. I plug it in over night, that's all it needs.
> Everything about these laptops feels cheap in comparison to a Mac. They are bigger, thicker, heavier, more plastic, hotter, and less elegant.
This is completely wrong. For example, a Lenovo X1 Carbon is 2.5 pounds, which is lighter than a MacBook Air at 2.8 pounds. It also offers 32GB of ram, does not have a glued-in battery, does not use glass and the keyboard is much less likely to fail. I am tempted by the M1, but from my experience Apple machines are heavier, fragile, less reliable and less user-serviceable. I also don't understand the aversion to plastics. Aluminium makes MacBooks heavy and can attenuate radio signals.
> For example, a Lenovo X1 Carbon is 2.5 pounds, which is lighter than a MacBook Air at 2.8 pounds. It also offers 32GB of ram, does not have a glued-in battery
.. and yet if feels heavier, looks shittier, is less durable, full of Lenovo spyware and overall still much worse quality. I have had so many Lenovo laptops in my life that I can say for sure that there is nothing good that comes out of this company.
It's like buying a Wilkinson razor with 5 blades, plastic covering and which weights only a few grams in total. Yes, on paper it's lighter, has more blades and has super easily replaceable parts and is also much cheaper, but if you buy a thiers issard carbon steal cut-throat straight razor you will only have one blade, it will be much heavier and more expensive but you will still have the best shave in your life with zero irritation and not look like a cheap red faced pig afterwards. That's the difference between Lenovo and Apple in computer hardware.
If the same screen sized laptop has bigger bezels, the grip is bulkier, less sleek overall form factor makes it feel heaver in your hand even if it's a few grams lighter. That's why it's called "feel" and not "is".
> I've heard MacBook keyboards tend to self-destruct when encountering small specks of dust though?
Citation needed. My MBPs haven't self destructed yet despite being daily in use for years.
I agree with you, but one thing that I have learned, is that non-nerds tend to hate "computers."
By "computers," they mean things with screens, mice and keyboards that don't have a simple, fixed user interface. I know that some of the iPad Pros are starting to look a lot like "computers."
Me? I love "computers." But I'm a nerd. We're supposed to love them.
I am really looking forward to getting my grubby little mitts on a new laptop, when they come out.
I have a Pro Display XDR hooked up to a MacBook Air and the only glitches I have had would be unplugging and plugging in the display while in sleep mode which just requires me to unplug the XDR and plug it back in.
I have a 49-inch LG Ultrawide (Split into 3440 X 1440, 1680 X 1440), and a 4K monitor that I use for presentations and classes (so it is not on all the time).
I have become addicted to the vast real estate of this big honker. I did have it set up, once, using the whole screen width (5120 X 1440), but that was too wide for me (also, Apple does not yet natively support that resolution. I had to use a third-party screen res tool).
I've never squeezed more than 4 or 5 hours out of my two year old Macbook Pro. A 20 hour battery life in addition to the speed gains make upgrading a tempting prospect.
My wife has an ~year-old Intel MBA. I have an M1 MBA.
Her battery's dead after 2 hours on Zoom or Discord or any of those battery-hungry web tech multimedia apps.
I've not tried taking it that far, but judging from where the battery on my M1 MBA gets to after a couple hours, I could probably do 6ish hours under that load (it's very dumb that some of these "productivity" programs are high-load and eat battery like crazy, but that's where we are).
If I avoid the webshit (Slack's unavoidable, for me, but I can ditch most of the rest, at least temporarily) I can work a whole day plus another half a workday on battery, no problem.
[EDIT] oh, and it feels faster than my hex-core AMD desktop with the badass graphics card and 64GB of memory does, under Win10 or Linux. Jank, jitter, and pauses galore, when doing basically nothing. Not so on the M1, unless I really abuse it. Granted that's largely the software's doing, but it doesn't really matter why it's better, in the end.
My parents got an M1 mba a few days ago. When they asked my advice about getting a new laptop they got an earful, and then the favour was returned when I listened to my flabbergasted dad explain to me how his bank website now renders instantly compared to 15 seconds on his old mba.
It’s hard to quantify what portion of the so called tech-illiterate have a tech literate person in their lives who they go to for purchasing advice, but anecdotally I must provide decisive “wait until September” or “nows a good time to buy and here’s why!” advice to at least a dozen or relatives and friends.
It’s also why the butterfly keyboards or the mdns debacle can end up being a drag on sales. I’m very bullish on Mac market share in the next couple of years due to pent up demand from family members being advised to hold on to older machines.
Right, even if it's distilled down to "I've heard great things about these new Macs" that's still reflective of an incredible technical and marketing accomplishment.
You can't really buy an equivalent since Apple focused on price, performance, absolute silence and battery life with virtually no deck flex due to the aluminum unibody.
You can get laptops that can almost match it on price and performance but the responsiveness and battery life is going to be extremely hard (including the MacBook Pro and reviews have stated that you can get it to above 18 hours with light use. Also, the MacBook Air is totally silent with great performance and I can't find many silent laptops at all. The biggest downside to the new M1 Macs are probably that they rely on swapping from RAM to disk but we have to see when those parts will wear out.
I bought a Macbook Air M1 a few months ago. It does not take specialized work to notice the difference between this laptop and every other intel based laptop.
The Tinkpad X1 Carbon which I bought multiple time over the years would run hot, make noise and the battery would deplete fast when watching youtube videos. Before M1 this was just normal laptop behavior and the X1 is a great laptop. With the M1 though, it's now been 2 hours of watching youtube and you're at 95% in a perfect silence and the laptop is cool to the touch.
Also I don't find a Macbook Air to be "cool". To me it feels mainstream just because it's an obvious choice.
In some ways, it reminds me of the Intel transition in that regular consumers might not have been aware that this move was coming until they saw it in front of their eyes.
I often see that transition dated to 2006 at MacWorld, but the reality is that it was previewed at WWDC the previous year and there was only a minor spec bump to - I think - one model before the iMacs appeared to kick things off.
On a related note, I was not surprised - but definitely disappointed - to hear them describe the iMac changes as "making the computer disappear." Such 'appliances' have always seemed cheap and purposefully limited (not in a good way), yet this is what the market seems to want.
Regular people continue to not understand how computers work and they probably never will.
Regular people also don't really understand much about how cars work, and they love that modern ones start with the push of a button and that they don't have to fiddle with the carburetor or the choke to get it going.
The problem space is much smaller and with each big shift in hardware and software, we've missed the opportunity to do this well.
A rough timeline: Desktops, the Internet, Laptops, and - for probably the last decade - Mobile. Is "Ambient" next?, I'm not sure, that seems like it carries other baggage along with it.
It may look like an appliance, but it certainly doesn't work like one, and if users had just a bit more knowledge about it, I wouldn't feel like we're going back to square one with each step forward in technology.
I understand how both work and I LOVE both push button starting modern cars and computing appliances. I'd rather have an iPad and all the extra time it saves me to learn new things than fuddle with a desktop PC. Just like owning a Tesla will save me time vs owning/repairing an older car.
Apple has a long, long history of pulling off well executed architecture changes. If you are a Apple computer shopper then you would know by now that by the time they release a new architecture, internally, that platform has long been retired for their internal users.
It would have been an even bigger headline if Apple released the M1 and it didn't live up to expectations.
I'm actually more surprised about the iPad Pro going from the A-series to the M-series. And now, the new iMac looks very iPad-like.
The original Intel Mac's ran the 32-bit Core Duo, although a mid-season refresh took them to the (64-bit) Core 2 Duo. That generation of purely-32-bit hardware being out in the wild had some real impact on moving the OS forward, later down the road.
We had purchased one, when 10.5 released. And it died shortly after to be replaced by the Core 2 Duo Mac Mini. The x86 era was so short I barely remembered it.
Also probably worth dividing PPC into the 32-bit architectures (601, 603, G3, G4) and 64-bit architectures (G5), although the PowerPC architecture did that transition in such a civilized manner that it wasn't really traumatic. In some ways the jump from the 601 (Power-ish) to 603 (true PowerPC) was bigger than the jump from the G4 to the G5.
They also never fully went 64-bit on PPC, so there was never really a transition there, just some 64-bit machines that could do some slightly different things and I don't think there was ever a full 64-bit version of OS X for PPC. So there weren't any huge painful breaks.
Also the jump from 68000 and 68010 to 68020-030-040 had a similar transition (if anyone remembers having to worry about wether your software/system was 32 bit clean)
I thought about mentioning it but that doesn't really count as a transition like the others. You could not even run the card and MacOS simultaneously:
> When emulating the Apple IIe, only a full-screen mode is available and all native Macintosh functions are suspended while running (a proprietary graphical control panel, running outside MacOS, is available for configuring the virtual Apple II slots and peripherals; however, both native and emulated computer function are suspended during this activity). Macintosh functions and control resume only once emulation is completely shut down and exited.
Arguably, the transition from MacOS to Mac OS X does belong in the list. Apple did a ton of work to provide the Carbon compatibility layer to make it easier for developers to port to Mac OS X. And there was also the ability to run MacOS 9 under Mac OS X using Classic.
Agreed. It is sort of helping users to make the move though.
It's almost like a reverse transition: instead of getting a new cpu and making it work with the old ecosystem and software you take an old cpu and software and and get it to work with the new hardware (I know the analogy isn't perfect!)
I hated my Surface Pro when I had, granted this was 3 years ago. It had terrible battery life, windows was horrible with the touchscreen, very few apps responded to touch events at all let alone correctly. All together one of the worst laptops/tablets I've ever used. I have an M1 macbook air and it's the best laptop I've ever used, same with my iPad Pro. Apple has really nailed the Arm device/os combo IMO.
People held on to their 2015 15" MBPs for as long as they could because they didn't like the upgrade paths. The 16" was well received and helped fix that. The M1 laptops were incredibly well received and people are excitedly jumping on them instead of feeling forced.
My 2015 MBP is still going strong and Apple support were even pretty great at helping when I did have one issue, even though they classify it as Vintage. But you're totally right, I'm really impressed by the M1 and am just holding out for the M1X to upgrade.
Another 2015MBP stalwart here. I wasn't planning on upgrading to an ARM Mac.
I use BootCamp (daily lately, since we're in lockdown and it's snowing) for gaming on Windows with an eGPU. Neither of which are supported on the M1.
I'm considering selling my eGPU and just building a gaming PC (the M3 will likely be out by the time I pull trig though :/), but it was nice to have 1 machine do it all (work + play), especially since I didn't need the eGPU to play some couch co-op games (Cuphead, Castle Crashers, very cottage-friendly!)
I looked into the eGPU for the same reason, to have a single machine to do it all. In the end I got a separate machine as it was cheaper to do so (!!) and saw people having lots of issues with eGPU support. I couldn't believe the price on those eGPU enclosures.
How was your experience? I bet it was nice just plugging in and getting all that extra power.
Good point, but really it's more about usability than anything else.
Imagine that the M1 wasn't actually faster than just about everything out there. But also imagine that the new design decisions, the improved memory allocation methods, the significantly better power usage, and the materially lower price meant that it was faster, took less memory and less power for lots of things, or even for most things.
Would it be compelling for people who already have Intel to move? No. But would it be preferable to choosing Intel for a new machine? Certainly, at least for low-end models and applications.
And look at what Apple has done - the M1 is in the mini, the Air, the smaller (usually lower end) MacBook Pro, and now the 24" iMac (which is the lower end iMac). So, even though the M1 kicks Intel's ass, Apple is still playing it safe.
There'd be plenty of reason to choose ARM over Intel even if performance wasn't clearly better :)
It's also worth observing that a lot of IT departments aren't necessarily rushing into the M1, especially to the degree that there isn't a full range of laptops with the new architecture yet.
IT departments do often offer a small/light and a non-travel-optimized model but I imagine many companies are waiting until they can have a common architecture on new systems and any software teething pains are worked out (which seems to be happening but probably not quite there yet).
Do IT department ever rush? I think their ambition is to be slow, but steady. If you don’t want to change course often, you can’t react immediately to every thing that happens.
When they have to. Like when everyone went remote a year ago. But, to put it mildly, providing people with access to the newest Apple laptop does not constitute an emergency.
Even w/o the perf of the M1 I think sales had been overwhelming, part because yes it's the "entry level" models but mainly because nobody wants to be left with a machine that won't get supported by the OS (even if I think low-mem M1 machines will be phased out of OS updates at some point)
I think you vastly overestimate how much the average consumer thinks about or cares about OS updates.
What you say is something that could affect the decision making process for highly technical people, but even then... we have no firm commitments from Apple on which platforms will get the longest support, so it's a gut-feeling decision, which isn't very technical.
As you hint, the M1 could be treated as a pathfinder platform and deprecated much sooner than subsequent platforms, kind of like the original iPad.
The M1 is not ARMv9 and lacks SVE2... ARMv9 is coming soon, and that may become the long-term supported standard for Apple's Mac-on-ARM platform.
The original iPad was, in terms of compute specifications, a minimum viable product. The M1 Macs are not—they are the combination of a mature platform with a mature silicon chip. Given the sheer volume of M1 sales there’s no chance they’re going to be so prematurely deprecated as the original iPad was.
At the very least I’d put money on the M1 not being deprecated before all Intel-powered Macs (present and future) are deprecated.
I think the M1 will easily last for a few years - but we don't know what they have left in stock. Was the original iPad regarded as MVP when it came out?
>But the more interesting point is that the rest of the market agrees with us early adopters and tech nerds.
No one outside of early adopters and tech nerds cares one wink about the components of their laptop. The days of normal people caring about specs are long gone.
“Duh” comments come from all the wins they’ve had in the mobile space. They have been and likely will continue to be 2 or more generations ahead in performance per watt on mobile. So when they were talking about doing a desktop and laptop chip it wouldn’t be a stretch to think they’d “knock it out of the park”.
BS, there is no reality where the M1 failed with a company like Apple. They have too much loyalty for people to even consider other options. So much in fact that no one can be convinced of its flaws it’s all “Apple knocked it out of the park, most things work!”
We all know “most things” is subjective to how you use your PC.
Tell me a world where that wouldn’t ever happen and I’ll consider your Intel leading model scenario.
Seems to me that the statistic is designed to hide other possible stories. Consider that those unhappy with the M1 move wouldn't buy older Macbooks, they would leave the Apple ecosystem all together.
I can't speak to the quality of the M1 since I haven't used one yet, but I don't think the sales numbers mean that much. Basically just new shiny model outsells old standard model. Just look at the website. What would you rather buy, a standard model, or the one "supercharged by the Apple M1 chip?"
I assume Apple sells vastly more Airs and 13” Pros than they do any other machine they make. The lower price is a huge contributor. If the M1 were only on the top end machines, this thread wouldn’t exist.
OTOH, people who might have otherwise upgraded to one of those older and more expensive models are holding off on purchasing until new higher-end models are available.
M1 is amazing and I bought my partner one and she loves it.
However, when it came to buying a new engineer on our team a laptop, we had to go with the last 2020 Intel model.
We still have some uncertainty over whether or not our development tools/environment will work on M1. Many python packages will take longer to install I imagine (if there are no wheels for the M1 mac, not sure if that's an issue) and I suspect some won't build at all. I also have some concerns over our Docker environment building.
Is there anyone doing python development and using Docker who can share their experience so far?
Edit: please also share non-python development experience, it's really interesting to hear
As an engineer on an M1, I have to say you made the right choice. Half the stuff that I try to work on is just straight up broken, especially Python/AI related tools. I have to switch to my Linux laptop regularly to get anything done. I've wasted days just trying to get certain dev tools working.
Apple should really hire a couple smart devs to just focus 100% on getting working M1 wheels on PIP for all the most popular Python libraries (same for conda forge). Seems like a no-brainer given their dominance in laptops for software developers. I'm sure the community would just do it on its own, but if they can speed up the process by 6 months they will be able to sell enough extra units to pay for the devs.
Maybe they have, but whoever does the work, there is a lot of ground to cover in the space of developer tool chains. It's going to be a minute to get to a point where devs don't feel like they have to make compromises.
I have hit the same problems with Python deep learning tools. Apple’s version of TensorFlow was an OK start. If you like conda, they now have pretty good support for M1. I have the usual Docker problems, and I don’t think that they will get sorted until many more pre-built containers are also built for ARM.
That said, I love my M1 MacBook Pro and for now I don’t mind too much keeping my old MacBook nearby.
I've found the experience better, most stuff I've used works out of the box, but there are often weird lib problems that require various hacks. You then can't check them into source control as it'll break intel users, and it becomes a bit of a headache remembering all this.
I think the situation should continue to improve, but there is an extremely long tail of stuff to get working.
The other thing I would mention is that Rosetta is very very slow. A lot of the initial impressions made out it was basically a bit slower, but for a lot of stuff it's slow as hell. You do not want to for example run an IDE in Rosetta.
I just default to an x86-emulated terminal and everything I use on my M1 works exactly as it did on my Mid 2015 MBP, except faster. The only exception is virtualization, which only works if there’s an ARM image available (which there isn’t, mostly).
Totally agreed and tensorflow is pretty broken (I guess the alpha release tag is fair). But the newest brew released has fixed maybe 90% of the issues. With miniforge + pytorch, you can run most things.
I do a lot of Python on my M1 and now that NumPy compiles correctly, I have no complaints. Docker also works fine presuming your base image has a ARM version (I haven’t tried their x86 emulation, but presumptively it’s slow since it can’t use Apple’s Rosetta speedup).
It was bumpy until March or so, but now I really really like it for Python.
I don't have any Python experience but Docker has been a mixed bag but mostly fine.
ARM-native images are 100% fine, and x86 images are surprisingly usable via emulation (albeit considerably slower). During the technical preview stage some x86 images simply didn't work at all (MySQL being a big one that springs to mind). However, I've just checked it in the stable release and it's working now. Not sure if there's still images that don't work under emulation though.
That's good news. I was having trouble with some MySQL images using the Docker Preview. I have installed the final version but frankly, I never use that machine unless I have to test something in Safari. I just can't stand having to always remember what arch I need to use for a particular tool. (I do have an alias that opens another terminal in "emulate x64 mode" but, that's annoying too"). I think in a few years, when there are more toolchains built out, the ecosystem may be more attractive to me.
Current lineup has too little RAM for many dev-tasks. Like running multiple docker containers, or developing in a ram-hungry language (java+intellij), or running emulators (mobile dev).
If the next MBP16 lineup with M1 (M2?) supports 32+ GBs of RAM other vendors will have to step up their game, though.
I have a M1 Mini with 16GB memory. The machine's SSD is incredibly fast, so it swaps very well.
Whereas my 2019 MBP will suffer incredibly when I start running out of memory (it literally becomes unusable and requires a hard reboot), the M1 will transparently start swapping, and I generally don't even notice.
Just the other day I noticed a python process was running slower than expected, and it was a bug in my Pandas code so I had allocated 85GB of memory. The rest of the system was operating normally - Chrome, VSCode, Music were all fine.
I regularly kick off a data pipeline process with 8x concurrency, even though each process uses between 4-6GB memory. It's probably inefficient in some way, but overall is still much faster than running at 2x concurrency (where everything would fit in memory and it wouldn't need to swap).
At least my coworker's former 16GB MBP (until he got a new one now) basically halted when the emulators were running, and he had to close lots of programs. Not strictly appdev, though, so he was then running and developing our backends locally, running some other needed services in docker, and then working on the android and ios app at the same time to match and test the changes in the backend. So basically all of my examples above, but at the same time.
So not strictly necessary, but when you're used to it going back to having to work around the limitations is burdensome. People not having lots of ram seldom complain, but they don't know how much time they're wasting on an inefficient workflow.
I'm not a mobile developer, but between docker, my IDE, Chrome, and whatever other crap I've got running on my 16 GB MBP, I'm using 14.8 GB of RAM and 20 GB of swap.
Not python, but I have had hiccups with both Go and Java when targetting x86 linux docker images. The Go stuff would make qemu choke when building multiple go binaries in parallel, and the java testcontainers I use had some sort of an issue.
In general though, I love my M1 air, but I am sticking to my workstation (5950x, 64GB, debian) for dev work for now.
Being able to control the base platform is sometimes important - in this case the target is aws lambda, and even building on Ubuntu 20.10 we ended up with links to incompatible GlibC library versions compared to the lambda runtime environment.
Quick fix for all of this - whatever platform you run the build on, do it in docker on the exact same base image that lambda uses at runtime.
I know, you’re going to tell me there’s an easy way around this with cross-compilation. But when you already have docker available , quite often everything looks like a nail.
I've recently gone through the process of getting our stack, which includes python, go, and docker, running on an M1. Docker for Mac only recently released M1 support but it runs well enough. We upgraded to Python 3.9 to incorporate some M1 compatibility updates as well.
Overall, developing on the M1 Macbook Pro has been great so far.
> However, when it came to buying a new engineer on our team a laptop, we had to go with the last 2020 Intel model.
There is no rush into purchasing an M1 for developers since, the software ecosystem at the time (and still to this day) is somewhat immature and not ready. There is plenty of time to switch to Apple Silicon until the ecosystem is mature.
By then, you would not even consider an M1. It might be an M1X, M2, etc. On top of the already existing issues that plague the first generation, the best thing to do for now is to stay on Intel for now. I am long term Apple Silicon, not long term 'M1'.
This is what I have been saying for weeks [0][1] and weeks [2][3] and now just yesterday. [4] I even said it a day after 'launch day'. [5]
I know it's not what you asked, but it might help someone else. Rust and JS/TS work well, no complaints so far. Performance is great. The only issue I had an issue where some package depended on Chromium for something and I couldn't get it to work. But it was a relatively minor thing so I let it go.
> Many python packages will take longer to install I imagine (if there are no wheels for the M1 mac, not sure if that's an issue) and I suspect some won't build at all. I also have some concerns over our Docker environment building.
Just use an x86-emulated terminal and you don’t have to worry about this. The only limitation is x86 docker images, which I can’t get to run on my M1. Everything else works as it did on my 2015 MBP (when executed in an x86 terminal).
Have you tried the newest docker? I was on RC2 and managed to run a couple of x86_64 containers minus the issue of networking (for some reason uploads times out)
Will currently rock my 2017 MBP until it shits the bed. While there have been huge strides in getting tools compatible with apple's ARM, I am not a fan of spending huge amounts of efforts to get a tool(s) to work on something that was already working sufficiently enough on x86.
Plus, I imagine in most corporate environments M1 will break many internal tools making it nearly impossible to access protected resources.
Roughly the same scenario. Was buying new laptops (3) for me and a new team. We looked into M1, but need to do some things like run Parallels/Windows, and underlying software, that we didn't have enough confidence yet would work as needed on the M1. I figure we'll wait a year or so for the M1 ecosystem to get a little better understood and then re-evaluate.
So I'm not using docker or Python so my opinion isn't so worthwhile but I would say you made the right choice for right now.
I have an M1 Air and ended up listing my iMac I use for work as the Air was just wholely faster for my tasks. For the average dev, I think there's a weird hap right now of "M1 now vs M1X/M2". It's getting later in the cycle and while M1 support has greatly gotten better, it's now closer to new models with drastic changes.
Now in my use (TS/Node), the M1 in native Node is unbelievable, it's twice as fast as Intel machines. But that's my use case.
I'm working a lot on M1 and few months ago it was unusable for Deep Learning stuff but now its mediocre not good but also not terrible. (even coreml stuff wasn't working properly on M1 2 months ago)
I got Solidworks (CAD program) running on a Windows ARM VM with 64bit X86 emulation on my M1 Macbook Pro and it's still as fast as my dedicated Windows laptop.
Some of that speaks to Solidworks performance in general, but I don't know how to make sense of this other than the virtualization on M1 Macs being so good it feels like magic. I don't know how I could recommend any other laptop in good conscious (Price to Performance). Happy to be kindly corrected in comments :)
The interesting thing is that they've only moved their consumer models over—presumably we can expect an even more impressive chip later in the year. (I know the low-end MacBook Pro has it, but that model has never really been a "pro" machine in the traditional sense, it's always had a low power CPU and RAM constraints.)
And yet I know tons of developers, myself included, that have made the jump. I think that's indicative of both how well they've done with the M1, and how lacking the Intel models were/are in many regards. This M1 is a good bit quicker than the 16" it replaced, but I notice the improvements to thermals, noise, and battery life way more.
Yeah, i'm very close to just buying an M1 because it's just so good, but think i'm gonna wait a few months because i like having one single primary computer and bigger screen.
But damn it's already so tempting. Really hope we'll see updates in September.
Also "cheap models" sell more than expensive models, which was always the case. Apple has stated that the MacBook Air is their most popular computer on multiple occasions.
Perhaps he's trying to communicate that most of the initial issues with various pieces of software are now fixed. Without having to explicitly call it out.
What’s baffling to me is why the M1 doesn’t natively support multiple monitors.
I’m holding off for this reason, and I’ve not been able to find why this is the case, particularly when they seem so much more powerful compared to previous MacBooks.
Good job Apple, I really do love the idea of deprecating the old x64 architecture entirely.
It's as if we have decades of legacy code wasting billions of dollars of power per year. I strongly believe within the next decade the vast majority of computers, new computers consumers buy anyway, will run on arm( or whatever replaces it). This will also be amazing for servers, imagine a data center that can run off 30% of the power of an Intel based data center.
I really do love my m1 Mac Mini, I only use it for music but so far I've been nothing but amazed by it.
The $999 Macbook Air has long been Apple's best selling computer by a pretty wide margin. The Air being M1 now (and retina, and still $999!) has probably accomplished most of this.
That was actually not the case for a while... OS X was originally a LOT slower than OS 9, was missing basic features (at the time) like a DVD player, etc. 10.1 fixed some issues, but it wasn't until 10.3 or 10.4 until OS X really got momentum and more apps were available outside the 'Classic' environment.
so I have an old MacBook Air at my office and the new M1 Air at home. Last night I was trying to run something from our internal "ruby cli" and that meant the dreaded "bundle install" command. I'll be honest, I was nervous. But I thought back to that line in Godfather "i want you to use all your powers, and all your skills" and I got into a headspace of, I was going to get this to run or die trying. There was issues with many gems h3, openssl, pg, etc. I finally got past them but then running the cli still had openssl issue. I tried to switch to using rvm and then rbenv but both failed with zlib issues. So then I tried VirtualBox but oh yeah, on M1 you need https://mac.getutm.app/ and I got a few linux distros running. But all had different issues with ruby. Long story short, i failed, but omg any complex ruby project with a big gemfile and a new processor == issues. +1 vote for this cli being writen in go or rust and a binary distributed vs. making each consumer of this cli have to "compile" the entire project.
Now that Parallels, Crossover and UTM have solved the boot camp issue there is little reason to let x86 compatibility get in the way. Most of the major Mac apps have released Universal Binaries as well. It's not even been six months yet and most Macs have already transitioned. Now all we need is the Mac Pro and Macbook Pro 16" to switch.
I don't know if they've completely solved the boot camp issue. I was excited to hear the news that Parallels can run ARM Windows, though not all software is supported on ARM Windows (for example IIS isn't available on the ARM version of Windows). That's a Windows issue more than a Mac one, but a software gap with Windows on M1 does still exist unfortunately.
The current top end M1 Mac Mini tops out at 16GB of unified memory (meaning, it's shared with the GPU). That's OK but I'd prefer 32GB. More annoyingly, the store only stocks (up to) 8GB/512GB.
Part of the appeal of a Macbook Air (which IMHO is the best M1 form factor thus far) is it's not a lot of money (compared to a Macbook Pro 16" anyway) and you can just walk into a store and get another one if you need it.
This applies to accessories too. I know it's easy to get a Macbook charger. But a Dell or Lenovo charger? I'd probably have to wait days for that.
Also the upgrade prices are kinda unjustifiable. Upgrading to 16GB/1TB is $1299. That's a lot for the specs ($200 for each of 16GB->32GB and 512GB->1TB). I'm really hoping the next version gets better.
I do think the M1 has been received incredibly well and Apple has done a great job here it seems.
Exactly this - if they had released the documentation and made it available through usual vendors, we could see excellent linux machines coming, probably best in history.
The new M1 machines seem cool, and I almost considered grabbing a MacBook or Mac Mini last year, but I'm worried about the complete lack of upgradability.
From my understanding, the older Intel models of some Apple devices at least had some level of upgradability (memory, etc.), but the the memory being on the SoC, I can't buy a cheaper model and hat upgrade it later as I see fit. So if I need something new later, I have to purchase a brand new, expensive machine.
This has been the case for years now, and other manufacturers are following the same trend. I'm starting to find that I rarely upgrade things anyway because by the time I need to upgrade, things like the CPU might be too slow anyway, or ports have drifted away to USB-X and Thunderbolt 12 and the whole device falls behind.
I say this as the owner of a maxed out Thinkpad X1 Yoga 1st gen. I tend to buy used top of the line and forget about upgrades. The ram is soldered in this one as well, but I can change SSDs and WiFi cards, which is very nice.
It's been running MacOS Catalina for more than a year now and rock solid stable, so basically a $300 i7, 16GBRAM /512GBSSD 1440p 14in Macbook that is lighter than the 13in Pro.
The cons are the damn speakers, if this thing had good speakers I wouldn't even consider a Macbook.
Duh. This is a classic case of Osbourne Effect. Who's going to buy an Intel Mac when M1 is just as good, and appears to be the direction Apple is headed? The point at while an Intel Mac goes obsolete is likely to be sooner, because at some point (probably soon) you'll encounter some app that won't run on Intel (oops we only tested it on M1), or at best it runs in some kind of emulation mode.
I will miss the Intel Macbooks. I use a 2013 Macbook with Bootcamp for freelance work. It's close to perfect. I don't want to maintain 2 laptops for Mac/Windows work. If I could work on an ipad I could imagine having an ipad and a Windows machine but currently my next setup will be worse due to "progress". Am I using it wrong?
I hate Apple anti-competitive behaviour and that they claim they are environment friendly while trying to fight the right to repair, but I need a new laptop for work and everything that is currently on the market feels outdated. The biggest offender are of course Intel based laptops - eve n the very latest ones are just simply a reheated and vintage at this point, whereas AMD ones seem to be either not in stock or paired with sub-par peripherals.
So it looks like I will hold my nose and get a M1 or its successor when it comes this year.
In Q1, Apple took 8% of the PC sales worldwide. This means that Intel lost 4% of the market share in one quarter since Apple only uses Intel CPUs before.
The branding is arbitrary. If you want to read the tea leaves, it seems to indicate that Apple is going to incorporate more Mac-like elements into the iPad Pro, which from its inception has had some users struggling to make full use of it's power.
The ad they showed it with involved Tim Cook "stealing" the M1 from the Macbook and putting it in an iPad. So if you want to predict, perhaps there will be some big software news at WWDC.
But that's just speculation. There's nothing really different from what the M1 is and what people expected the "A14X" to be.
I’ve had a pet theory that the company has spread its wings after Jonny Ive left.
More practical designs
Thicker, more performing machines
Higher quality and durability at the cost of thickness
Colors again?!
It’s possible, perhaps even likely that this was planned before Ives departure. But we are seeing a massive sea shift in the way Apple designs their gear. The old mantra was “thin and light before all else”
Now it seeks back to “make stuff people love”
Thinness at the expense of all else is what I meant.
I don’t see where they compromised to get that thin.
However compared to MacBooks with the butterfly keyboard, Apple introduced more than one laptop that had useability issues due to the drive for shedding weight and centimeters.
I suspect that new iMac design has been in the can for years, waiting for the technology to catch up. And it’s a consumer device anyway; the real test will be whether the “M1X” serious iMac is inflicted with arbitrary thinness or if it will be designed with serious users in mind.
I was very skeptical about the M1 macs but then I got an M1 mba as a complement to my 16 inch mbp and well it's just fast and the temperature is much lower for typical tasks.
Now I'm just looking forward to the 16 inches arm macs.
We're able to run iOS apps on the M1 computers.
The iPad is essentially the same hardware and has a proper keyboard / trackpad. I see no reason why it won't be like that.
The new iPads have up to 16 GB of RAM, and... well, it's a lot beefier a machine than iPadOS can do anything with. So it's not a given, but people are hoping.
Proprietary was the point. There is no open source support (in the Linux kernel). There are attempts to make it run, but before it becomes even barely usable Intel will catch up and AMD will be far ahead in performance and catch up with power consumption.
Then there will be new models from Apple, but you the cycle will repeat itself, again no linux support etc..
Not sure if directly, but to distributors who then sell to shops or consumers.
You can also get datasheets for Intel CPUs and create your own motherboard if you wish so. You cannot do that with M1.
>known for dropping macOS support on perfectly capable older hardware
?? My 5+ year old systems are still getting updates. Yes, they're probably getting near the end of their support window but that doesn't seem unreasonable.
2013 was only 8 years ago. Maybe you upgrade hardware frequently, but for many people it's unnecessary unless artificially required. I have a family member who uses an iMac from 2011. With 16 GB upgraded ram, etc it still works great for her needs... but it cannot run anything more recent than High Sierra, from 2017. Which means that certain third-party software (coughMSOfficecough) also refuses to run the most current version. In the midst of environmental crisis, why should we normalize a desktop OS only supporting < 8 y/o hardware, especially when the same company is moving toward proprietary hardware that is (so far) unsupported by alternative operating systems?
Great marketing. I wonder the sales difference if they were also selling Macs with current generation AMD Ryzen chips manufactured on a modern TSMC node.
That seems to miss the point that they’ll have one chip family across all their devices - the cost savings to APPL will be tremendous compared to tailoring to intel/amd/iPhone_chip/ipad_chip.
So we can bemoan a missing cpu option but profitwise (ie price*volume-costs) it makes more sense.
He's basically saying to investors "We're executing this bet-the-company's-lap/desktop-market change to our CPUs flawlessly."
It's got nothing to do with the merits of the new architecture vs Intel, it's about Apple not only transitioning, but also consolidating its market edge with yet another moat of smart technology adoption.
What is great marketing? That they successfully sold the merits of their new architecture? Or are you saying that talking about the difference in chip sales is a marketing tactic?
It seems highly likely given past performance; if they had put any half-reasonable Intel or Ryzen chip in their $999 Air then that chip would be their best selling CPU.
Saying the M1 "isn't just an upgrade, but a breakthrough," and supporting that with sales figures that don't really show anything is "great marketing".
I think they are so good at building CPUs that fit their designs why do they need to rely on AMD keeping pace with them? I bet you Apple increase performance 20%+ per year for a decade while at some point AMD will slow down their improvements. Never mind the whole issue with heat and battery life where AMD is miles behind.
> Never mind the whole issue with heat and battery life where AMD is miles behind.
That's just false. AMDs newer U models have TDP and performance per watt at the same level as the M1.
Apple has other advantages with the M1 (like running iOS programs, more optimised for their software and so on), so they still benefit from having their own chips. M1 is great, but it's not magic.
The AMD CPU wattage != Apple’s SOC including graphics card, large parts of the motherboard and memory. If you’re right where can I buy an AMD laptop with 20h battery life?
Yes I've noticed something about Javascript and video seems to consume huge amounts of battery life. What do we think the battery life of an AMD 5800U laptop would be at Google Meet if you could buy one of course...
You are mixing battery life (= battery capacity / power consumption at low digits cpu usage) and performance per watt (= max performance / power consumption). M1 is better at the former, high end laptop Ryzen models are better at the latter.
If the numbers are accurate, then it is close. However, each company measures TDP in different ways. Plus AMD can boost to 30W for short periods of time. Obviously the boost is good when you need it. The big question, how does this affect battery consumption?
That said, AMD's biggest issue is supply. I don't know if it is supply issues from TSMC, or component issues with the OEMs. But it is nearly impossible to buy a 5800u.
Consider an alternate possibility: “sales of Apple’s new ARM macs outpaced by older models”.
Had Apple not knocked M1 out of the park, this was a real possibility. Is it surprising that a well executed architecture change is now selling well? Not really.
But the more interesting point is that the rest of the market agrees with us early adopters and tech nerds.
This would be a very different headline if devs were struggling to port their software, or if Rosetta 2 didn’t live up to its promise.