If Apple continues with the budget Neo brand into a 12 GB iteration, I can see this becoming more realistic (rather than a novelty). That being said, Parallels may need to review its licensing with a budget tier in mind. Few will buy a cheap computer and then pay what Parallels charges for a license (regardless if one-time or subscription).
They need to introduce something below the Standard license targeting the Neo. What I'd personally consider is:
- Standard gets 16 GB vRAM (to perfectly target the base MacBook Air). But leave it at 4-6 vCPUs to not compete with the Pro (still for general computing, not power-users)
- New "Lite" tier with 8 GB vRAM max for the Neo (4 vCPUs). Increasing to 12 GB vRAM if the Neo does.
Then you target a $89 price point one-time-purchase for the "Lite" tier. Essentially three plans, targeting your three major demographics: budget, standard, and pro/power-user.
This isn't a novelty it will crush the low end of the PC market. No one cares if the next iteration will be better with 12GB of ram. The workloads that people say that 8GB can't handle will be ones that the actual users will either wait or tolerate. I've been noticing that people who review the Macbook Neo basically don't get the point [1] and just the headline of this article matters that VMs work and thats a big win. The most ridicuous thing about the laptop is that it appears to be reparable which sort of tells me this is a template similar to the M1 Air of the future laptop designs that Apple will come out with. [2]
> This isn't a novelty it will crush the low end of the PC market.
You took what I said out of context and then replied to something else. Running Parallels on a Neo is a novelty. Parallels is both what the thread is about AND what my reply was expressly about.
Nobody can reasonably read what I wrote, in context, and believe I was referring to the computer itself as a novelty.
I saw the other day people complaining about AI slop being posted on this site by new accounts - which I agree is bad.
Someone suggested that people with 10k karma and/or 10 years subscription to this site should be able to do things (such as auto-ban) to those accounts.
The account that misrepresented your comment and thus acted in bad faith is one of those 10k+ accounts.
To me, this is a data point showing the fallacy of long term subscription and/or karma accrual as evidence of their quality/good faith abilities
Windows doesn't run "just fine" on 4 GiB of RAM. I had a laptop with 6; Windows 10 became barely usable. If you want to run one, small, program at a time I think you'll be ok. Forget about web browsing; you'll get one tab and it'll be slow.
Agreed. Windows 10/11 can run just fine on 4GB of RAM. You just can't run anything inside of Windows 10/11 with 4GB of RAM.
The last version of Windows that felt like 4GB of RAM was performant for me with applications was Windows XP. Not that every computer running the 32-bit edition of Windows XP could even see/utilize a full 4GB of RAM properly, but at least it was fast.
2015 laptop, spinning rust. Nevertheless, it was at least somewhat acceptable at purchase, but crapware installed with successive system updates brought it to a standstill. An SSD might've helped, but not by much. I wiped it and put Kubuntu on it to give to my wife, for whom it ran acceptably. She gave it back when she got a shiny new MacBook Air.
Neo is powered by a fast and battery-friendly chip. It's definitely not a novelty any more than Chromebooks or Windows 11 notebooks with integrated graphics have been.
VMWare Fusion is free, even if it is a pain in the butt to download. It also has GPU paravirtualization for Linux/Windows which is the only reason I use a proprietary VMM on macOS these days.
http://tart.run works great for running macOS (and Linux) VMs on macOS if you're technical. It's free for non-commercial uses too! (Don't think there's GPU acceleration tho).
I’m excited that Apple now has a reason to keep MacOS small. Their soon to be top-selling machine has 8GB and they won’t want to make all those millions of Neos unusable by shipping a bloated OS.
I wrote about how Unified Memory, SSD directly attached to the SoC and Apple's use of real-time compression saves memory, reduces power consumption and wear on SSDs [1].
There's a difference between bloated and batteries included. From a development point of view, macOS has native system libraries for things no other platform seems to include native system libraries for. And by "native system libraries" I do not mean downloadable content, dynamic support or anything similar, even if they're first-party. Though having unremovable system apps for every one of Apple's services MAY count as bloated if you don't use them.
> Windows 11 VM requires a minimum of 4GB of RAM to function
You can give it less. It may refuse to install, but even without using any workarounds, you can change the assigned RAM after installing and it will not refuse to boot. The minimum for Windows Server 2025 is 2 GB, and it’s basically the same OS (just with less bloat).
If the A18 Pro has the same ISA as the M-series chips then this may not be so straightforward. I am still hanging on to my 2020 Intel MBP for dear life because it is the only Apple device I own that allows me to run Ubuntu and Windows 11 on a VirtualBox VM.
Would you elaborate what you mean by saying Linux on an M-series chip isn't straightforward? That's not been my experience, I (and lots of other devs) use it every day, Apple supports Linux via [0], and provides the ability to use Rosetta 2 within VMs to run legacy x86 binaries?
Clearly I'm not as knowledgable about this as I thought I was. I already have a Ubuntu x86 VM running on an Intel Mac (inside VirtualBox). Same with Windows 11. Can this tool allow me to run both VMs in an Apple Silicon device in a performant way? Last I checked VirtualBox on Apple Silicon only permits the running of ARM64 guests.
While I have a preference for VirtualBox I'd say I'm hypervisor agnostic. Really any way I can get this to work would be super intriguing to me.
> Can this tool allow me to run both VMs in an Apple Silicon device in a performant way?
I use VMWare Fusion on an M1 Air to run ARM Windows. Windows is then able to run Windows x86-64 executables I believe through it's own Rosetta 2 like implementation. The main limitation is that you cannot use x86-64 drivers.
Similarly, ARM Linux VMs can use Rosetta 2 to run x86-64 binaries with excellent performance. For that I mostly use Rancher or podman which setup the Linux VM automatically and then use it to run Linux ARM containers. I don't recall if I've tried to run x86-64 Linux binaries inside an Linux ARM container. It might be a little trickier to get Rosetta 2 to work. It's been a long time since I tried to run a Linux x86-64 container.
The instruction set is not the issue, the issue is on ARM there's no standardized way like on x86 to talk to specialized hardware, so drivers must be reimplemented with very little documentation.
As long as you're ok with arm64 guests, you can absolutely run both Ubuntu and Win11 VMs on M-series CPUs. Parallels also supports x86 guests via emulation.
How is the performance when emulating the x86 architecture via parallels?
Also is it possible to convert an existing x86 VM to arm64 or do I just have to rebuild all of my software from scratch? I always had the perception that the arm64 versions of Windows & Ubuntu have inferior support both in terms of userland software and device drivers.
Have you confirmed this? I haven't seen anyone concretely describe the boot policy of the Neo yet (it should be an easy enough check for anyone who has one in-hand).
Not surprising but good to hear. It seems that there really isn’t anything that runs on a new MackBook Air that you couldn’t run on a NEO. It might not be as fast for some things but it gets the job done.
Ish. It’s better in some ways, like single core and maybe multi, but not by a ton. At the same time I think the M1 may have more raw GPU power, though missing a few fancy features.
Hardware is mostly worse, but that’s to be expected for the price. And nothing terrible, just little cuts all over.
It will have a longer support period than an M1 based on Apple’s history of device releases. This might also mean a longer support period for the 16-series phones than typical, similar to the 4S.
I’ve been an M1 Air fan since I got mine in 2020 but recently things have become unusable. Playing 4K videos often drops frames, even at 30fps. And I can’t reliably run Notion’s transcription AI on Zoom calls, even though it’s not running locally. I’m going to do an OS reinstall soon to see if that helps, otherwise it will be time to upgrade…
Yeah, honestly not even counting. The only reason I even consider moving is that I dislike Tahoe and I know eventually I won’t be able to stall the update; hardware wise it doesn’t even cross my mind.
I have a current gen MacBook Pro for work configured with stupid amounts of ram and I feel no difference in terms of fluidity at all.
“Parallels Desktop runs on MacBook Neo in basic usability testing. The Parallels Engineering team has completed initial testing and confirmed that Parallels Desktop installs and virtual machines operate stably on MacBook Neo. Full validation and performance testing is ongoing, and additional compatibility statement will follow if required.”
Virtualization requires specific hardware support to be performant. There are ways to do complete software emulation of a virtual machine but it would be so slow that nobody would want to use it.
This is them confirming that the CPU has enough virtualization support that they can virtualize rather than emulate the guest OS
Yeah. It's the first production Mac using an A-chip and is a Mac that has had many things cut out for savings. The question is did Apple feature cut required functionality.
Maybe/maybe not (we don't know how identical the A18 chip is to what shipped in the iPhone) - but it does determine that the virtualization stuff that was added to the M1 (in the era of the A14) has now moved over to the A series, at least enough to support macOS.
Every thread about Windows on Hacker News includes claims about apps taking 30 seconds to launch, web pages taking 20 seconds to load, simple applications being unusable, and other extreme performance problems. These are puzzling for anyone (like me) who uses Windows at home without all of these extreme performance problems.
That was until I realized how many reports are coming from people talking about their work laptops loaded with endpoint management and security software. Some of those endpoint control solutions are so heavy that the laptop feels like you've traveled back in time 15 years and you're using a mechanical hard drive.
There's an unspoken rule in corporate America, colleges, etc. Laptops MUST be loaded down with terrible software, no exceptions. My last corporate laptop actually had the paid version of winzip in 2025, and it ran with a little tray icon that I couldn't disable or remove. That was in addition to all the other corporate crap I couldn't remove.
Some of this is not _just_ a corporate problem. Why would Winzip have an auto run application and tray application in the first place? Every single app seems to think they need one, and it's a classical tragedy of the commons. Perhaps on a virgin Windows install, your app with autorun and a tray icon will be more responsive. But when 20 other apps pull that same trick, no one wins.
This is actually one of the reasons I'm not excited at the idea of Linux defeating Windows. If it did, corporations would just start crapping up Linux the way they've crapped up Windows.
Our corporate linux machines have exactly the same monitoring software as Windows - even the servers.
The performance is still not even remotely comparable. Could be the hooks are more performant on linux, could be the filesystem, maybe the tools are written more sanely... But loading apps, filesystem operations... Everything is still far faster on the linux dev instance. And I have half the ram allocated to that one.
The reason every developer makes their app open at startup, is because the Windows ecosystem doesn't have a good package manager. So every app needs to be its own package manager and check for updates on a timer. So they need to run all the time so they can run that timer.
In theory the Windows Store will handle updates. In practice, I avoid the Windows Store version of applications. Also, you can't turn off app updating, only pause them for a time.
I like videogames, maybe more than I should at my age, and I prefer to play them from Steam in Linux through Proton. A couple of months ago I caved in and bought a proper Windows gaming miniPC because a game I want is not stable in Proton.
I use a corporate Windows VDI at work, so the experience is understandably subpar there, but it is still horrible on high.end hardware. Took me half a day just to herd it through update after update, while avoiding linking it to a Microsoft account despite its protests.
It's literally used to run only Steam and Firefox, and it still sucks compared to the ease of install/management of Linux. Ubuntu LTS took me about an hour to set up dual boot, apply updates, install Steam, and every other software and tool I use daily.
Why is Windows 11 still so clunky in 2026? It doesn't feel like the flagship product that many bright minds have improved for three decades. Why are hobbyists and small companies outperforming Microsoft's OS management?
I once worked on a computer for the US Government that felt slow. I counted nine (9) directly competitive and redundant endpoint protection products on it.
Not nine different/only somewhat overlapping pieces of software from companies that were competitors. Nine equivalent products. I guess defender made ten.
Corporate spyware is pretty nasty, regardless of platform. When I was at FB, they had something that forced a kernel module that was incompatible with the next big OS release; and I had accidentally disabled the FB spyware scripts. I set /etc/hosts to immutable because I was tired of them fucking with it ... didn't realize that's why things were better for the next 3 months, until I did the major update and I had to fix things from safe mode ... where everything only barely works.
Microsoft also puts a lot of crap into a default install that you may want to disable. Windows 11 with some judicious policy editor settings isn't so awful.
Outside corporate setting, it is also the fact that most windows systems you encounter are installed on cheap machines by people who just care that their word processor works a few times a month. And you were probably forced to fix it.
At the same time, as someone with a well maintained Windows gaming rig, I don't like spending time in the OS these days. Something about transparently doing stuff that puts money in their pocket while inconveniencing me gives me the ick.
No this is not just an enterprise issue. I waited 10 seconds (I counted.) for a Windows Explorer context menu to open the other day. This is on a fully decked out system with an Ultra 9 cpu and a 4090 and 32gb of memory, and basically no apps running. I think I had 2 tabs in Edge? Windows is a shitshow these days.
I just tried to open the context menu in Windows Explorer. It showed up almost as soon as I released the mouse button, and I have a much slower CPU, older video card, and way less RAM then you do. I was also running 12 windows of Firefox with collectively 1000+ tabs (though only about 36 or loaded), Steam, a Unity game, and Microsoft Teams, plus a number of background programs.
If your Explorer context menu is taking more than a split second to load, there's something wrong with your hardware.
Oh yeah no... its still terrible even without all the spyware.
First experience of Windows 11, trying to download a file through firefox caused my 18 core 10980xe to have the entire UI freeze for the full time the download was going.
Reverted back to windows 10 immediately and the problem went away.
I've said for decades that from a user perspective, malware scanners and prevention tools are fundamentally indistinguishable from actual malware. They intercept file accesses, block you from doing what you want to do, pop things up all over the place, and make your machine slow aand unreliable.
I’ve helped someone with a rather clean iMac, circa 2019, still supported by Apple. Forget 6 minutes — you can spend a full hour from boot to giving up trying to get anything done.
I think that Apple has gotten so used to having fast storage in their machines that the newer OSes basically don’t work on spinning rust.
what? on a semi modern CPU and a SATA / M2 SSD?? My Vista laptop on a spinning drive took that long to boot I am pretty sure. I am flabbergasted if this is true
corporate laptops is the key here. take 2 identical laptops one with and one without the spyware - its night and day in both performance and battery life.
My corporate spyware laden Surface ARM runs Windows faster than the Macbook Neo, but unlike the Neo can survive a fall onto a concrete floor. (Ask how I know...)
How do you know a Neo cannot survive a fall onto a concrete floor? I think it would take at least ten tests each with a new machine to get some confidence of the impossibility of that.
Geekbench 6 was around ~2600 single-core with the VM overhead for me. That's still punching above single-core power in its class for Windows machines and it makes me giggle.
Wouldn't corporate spyware equally burden the NEO? Especially more give the 8GB of RAM vs 16+ on X64 laptops? Chrome, Teams, IDEs, websites etc are equally bloated on both platforms.
My (former) corpo HP laptop with 16GB RAM had 75% RAM used at idle after a fresh boot with Outlook, Teams and all the copro shit running in the background. So the 8GB NEO CPU will spend its time swapping data from ram to disk versus the 16GB+ ones, given both being filled with corporate spyware and same heavy use cases.
Also it isn't 2-3x faster, stop with the made up nonsense please. Just checked and my 3 year old AMD laptop is on par with the NEO geekbench score I found online (slower in single core but faster in multi core), not 2-3x slower.
Man, I do wonder what the realistic lifespan of that single NAND chip will be after it gets hammered by constant swapping of running tasks way beyond the capabilities of a 8GB RAM machine.
I have a PC with a 10+ year old 256GB SATA Samsung SSD that's still in top shape, but that's different because that drive has those 256GB split over several NAND chips inside, so wear is spread out and shuffled around by the controller to extend lifespan. But when your entire wearable storage is a single soldered chip, I'm not very optimistic about long term reliability.
There was quite a bit of discussion about that when the M1 first came out, but none of it really seemed to have happened six years later. The target audience isn't in danger of wearing it out and the ones that will push the limits will grow tired of it and sell it in a year or two or move on to the Neo 2, which might have 12gb of ram due to the expected chip.
I still think it's a great machine, but I think all these worries about NAND dying really haven't come to fruition, and probably won't. I have about a hundred plus of various SSD Macs in service and not one has failed in any circumstance aside from a couple of battery issues (never charged and sat in the box for 2 years, and never off the charger).
If swapping was causing SSDs to fail on M1 Macs, we would never see the end of the hysterical articles about "NANDgate". Since we haven't seen any in all these years, it's seems pretty certain it's not happening.
Hysteria would be if all had an issue like the keyboard gate, but this isn't an issue, it's a design limitation for certain uses cases which not everyone has. Some users will wear out faster than others due to usage patterns. If their M1 dies after 6 years of heavy usage, do you think they'll investigate if it was the NAND that died and go online to tell the news, or will they chuck it and buy new one?
NAND is still the same wearable part that regular X64 laptops have, Apple doesn't use some magic industrial grade parts but same dies that Samsung, Micron and SK ship to X64 OEMS, and those are replaceable for a reason, because they eventually fail.
The reality is most 8GB M1 Macs are still working just fine 6 years later. Power users know they need more than 8GB of RAM and will buy a MacBook Air or Pro with 16GB+.
The MacBook neo is for students, grandparents, travel, etc.
Hell, even if it dies after 6 years it was still a better experience than using a $500-600 windows PC and the cost comes out to ~$8/month spread over 6 years.
>The reality is most 8GB M1 Macs are still working just fine 6 years later.
Do you think SSD drives are replaceable for no reason? Just because M1 mac aren't failing left and right doesn't mean their NAND won't fail.
Even though I like the NEO, I can't in good faith buy a machine with soldered wearable parts. That's like buying a car with soldered brake pads because "in 6 years average users don't feel like they need changing".
I still had laptops on my hands from 20 years ago that work fine simply because you can swap their drives with fresh ones. How many M1 mac will still be functional in 20 years?
>but that's different because that drive has those 256GB split over several NAND chips inside, so wear is spread out and shuffled around by the controller to extend lifespan. But when your entire wearable storage is a single soldered chip, I'm not very optimistic about long term reliability.
I thought wear leveling worked at the page/block level, not the chip level? On an SSD, if there was a failure of an entire chip, you're still screwed.
from what i seen in "low end" ssds like the "120gb sata sandisk ones" under windows in heavy near constant pagging loads is that they exceed by quite a lot their manufacturer lifetime TBW before actually actually started producing actual filesystem errors.
I can see this could be a weaker spot in the durability of this device, but certainly it still could take a few years of abuse before anything breaks.
an outdated study (2015) but inline with the "low end ssds" i mentioned.
Most flash has average wear out after 300k cycles. Let's say 64GB is used for swap. That's 19200 TB or 19.2 PETABYTES of Swap usage. Let's say you swap 12GB a day, you will burn out that 64GB of Flash Storage in 4.38 years and my guess is that amount of swap usage is extremely high that user would probably replace laptop sooner out of performance frustration.
>Most flash has average wear out after 300k cycles
No it doesn't. Most 1TB drives are rated for around 600 TBW, so enough to overwrite the drive 600 times, nowhere near 300k cycles. If you search for specs of NAND chips used in SSDs, you'll find they're rated for cycles on the order of hundreds to thousands, still nowhere near "300k".
12GB a day isn't very much. If your working set is larger than the 8GB RAM, you're swapping multiple times per second. It doesn't take very many megabytes per swap to reach 12GB if you're doing that multiple times per second.
If Apple continues with the budget Neo brand into a 12 GB iteration, I can see this becoming more realistic (rather than a novelty). That being said, Parallels may need to review its licensing with a budget tier in mind. Few will buy a cheap computer and then pay what Parallels charges for a license (regardless if one-time or subscription).
They need to introduce something below the Standard license targeting the Neo. What I'd personally consider is:
- Standard gets 16 GB vRAM (to perfectly target the base MacBook Air). But leave it at 4-6 vCPUs to not compete with the Pro (still for general computing, not power-users)
- New "Lite" tier with 8 GB vRAM max for the Neo (4 vCPUs). Increasing to 12 GB vRAM if the Neo does.
Then you target a $89 price point one-time-purchase for the "Lite" tier. Essentially three plans, targeting your three major demographics: budget, standard, and pro/power-user.
This isn't a novelty it will crush the low end of the PC market. No one cares if the next iteration will be better with 12GB of ram. The workloads that people say that 8GB can't handle will be ones that the actual users will either wait or tolerate. I've been noticing that people who review the Macbook Neo basically don't get the point [1] and just the headline of this article matters that VMs work and thats a big win. The most ridicuous thing about the laptop is that it appears to be reparable which sort of tells me this is a template similar to the M1 Air of the future laptop designs that Apple will come out with. [2]
[1] https://samhenri.gold/blog/20260312-this-is-not-the-computer...
[2] https://www.youtube.com/watch?v=PbPCGqoBB4Y
> This isn't a novelty it will crush the low end of the PC market.
You took what I said out of context and then replied to something else. Running Parallels on a Neo is a novelty. Parallels is both what the thread is about AND what my reply was expressly about.
Nobody can reasonably read what I wrote, in context, and believe I was referring to the computer itself as a novelty.
I saw the other day people complaining about AI slop being posted on this site by new accounts - which I agree is bad.
Someone suggested that people with 10k karma and/or 10 years subscription to this site should be able to do things (such as auto-ban) to those accounts.
The account that misrepresented your comment and thus acted in bad faith is one of those 10k+ accounts.
To me, this is a data point showing the fallacy of long term subscription and/or karma accrual as evidence of their quality/good faith abilities
Most people run Windows just fine on cheap laptops with 4GB of RAM.
These won't run Crysis, but they don't need to.
Windows doesn't run "just fine" on 4 GiB of RAM. I had a laptop with 6; Windows 10 became barely usable. If you want to run one, small, program at a time I think you'll be ok. Forget about web browsing; you'll get one tab and it'll be slow.
Agreed. Windows 10/11 can run just fine on 4GB of RAM. You just can't run anything inside of Windows 10/11 with 4GB of RAM.
The last version of Windows that felt like 4GB of RAM was performant for me with applications was Windows XP. Not that every computer running the 32-bit edition of Windows XP could even see/utilize a full 4GB of RAM properly, but at least it was fast.
HDD or SSD? SSD can effectively make up for SOME amount of less RAM due to faster swapping, in my experience.
2015 laptop, spinning rust. Nevertheless, it was at least somewhat acceptable at purchase, but crapware installed with successive system updates brought it to a standstill. An SSD might've helped, but not by much. I wiped it and put Kubuntu on it to give to my wife, for whom it ran acceptably. She gave it back when she got a shiny new MacBook Air.
> Most people run Windows just fine on cheap laptops with 4GB of RAM.
Windows 7. Windows 10 eats about 6GB (custom IoT with a lot of things disabled).
Neo is a parody of a computer.
Neo is powered by a fast and battery-friendly chip. It's definitely not a novelty any more than Chromebooks or Windows 11 notebooks with integrated graphics have been.
lolwut?
check your install mate
VMWare Fusion is free, even if it is a pain in the butt to download. It also has GPU paravirtualization for Linux/Windows which is the only reason I use a proprietary VMM on macOS these days.
You can also use UTM to run Windows for free and it is open source.
https://mac.getutm.app
Last I checked UTM doesn't have GPU acceleration. Parallels' proprietary GPU driver is the only reason to pay for it.
http://tart.run works great for running macOS (and Linux) VMs on macOS if you're technical. It's free for non-commercial uses too! (Don't think there's GPU acceleration tho).
Apple already sells that, it’s called MacBook Air.
I’m excited that Apple now has a reason to keep MacOS small. Their soon to be top-selling machine has 8GB and they won’t want to make all those millions of Neos unusable by shipping a bloated OS.
I wrote about how Unified Memory, SSD directly attached to the SoC and Apple's use of real-time compression saves memory, reduces power consumption and wear on SSDs [1].
[1]: https://news.ycombinator.com/item?id=47354705
MacOS has always been incredibly bloated.
9.2.2 wasn't
There's a difference between bloated and batteries included. From a development point of view, macOS has native system libraries for things no other platform seems to include native system libraries for. And by "native system libraries" I do not mean downloadable content, dynamic support or anything similar, even if they're first-party. Though having unremovable system apps for every one of Apple's services MAY count as bloated if you don't use them.
The definition of bloat is something that you don’t use, even if someone else does.
> Windows 11 VM requires a minimum of 4GB of RAM to function
You can give it less. It may refuse to install, but even without using any workarounds, you can change the assigned RAM after installing and it will not refuse to boot. The minimum for Windows Server 2025 is 2 GB, and it’s basically the same OS (just with less bloat).
So in other words... We COULD in theory run Windows on our iPhones.
In theory you can run any software on any hardware.
Oh I fully get that, what I mean is in a viable way.
Just need to download some more memory
Can it run Linux?
Yes; macOS has native container support for Linux [1].
[1]: https://github.com/apple/container
In a VM, definitely. Just like other Macs.
If the A18 Pro has the same ISA as the M-series chips then this may not be so straightforward. I am still hanging on to my 2020 Intel MBP for dear life because it is the only Apple device I own that allows me to run Ubuntu and Windows 11 on a VirtualBox VM.
Would you elaborate what you mean by saying Linux on an M-series chip isn't straightforward? That's not been my experience, I (and lots of other devs) use it every day, Apple supports Linux via [0], and provides the ability to use Rosetta 2 within VMs to run legacy x86 binaries?
0: https://github.com/apple/container
Clearly I'm not as knowledgable about this as I thought I was. I already have a Ubuntu x86 VM running on an Intel Mac (inside VirtualBox). Same with Windows 11. Can this tool allow me to run both VMs in an Apple Silicon device in a performant way? Last I checked VirtualBox on Apple Silicon only permits the running of ARM64 guests.
While I have a preference for VirtualBox I'd say I'm hypervisor agnostic. Really any way I can get this to work would be super intriguing to me.
> Can this tool allow me to run both VMs in an Apple Silicon device in a performant way?
I use VMWare Fusion on an M1 Air to run ARM Windows. Windows is then able to run Windows x86-64 executables I believe through it's own Rosetta 2 like implementation. The main limitation is that you cannot use x86-64 drivers.
Similarly, ARM Linux VMs can use Rosetta 2 to run x86-64 binaries with excellent performance. For that I mostly use Rancher or podman which setup the Linux VM automatically and then use it to run Linux ARM containers. I don't recall if I've tried to run x86-64 Linux binaries inside an Linux ARM container. It might be a little trickier to get Rosetta 2 to work. It's been a long time since I tried to run a Linux x86-64 container.
Possible catch: Rosetta 2 goes away next year in macOS 27.
I don’t know what the story for VMs is. I’d really like to know as it affects me.
Sure you can go QEMU, but there’s a real performance hit there.
> Last I checked VirtualBox on Apple Silicon only permits the running of ARM64 guests.
I used to use VirtualBox a lot back in the day. I tried it recently on my Mac; it's become pretty bloated over the years.
On the other hand, this GUI for Quem is pretty nice [1].
[1]: https://mac.getutm.app
Run ARM64 Linux and install Rosetta inside it. Even on the MacBook Neo it'll be faster than your 2020 Intel Mac.
Pay Parallels for their GPU acceleration that makes Arm windows on apple silicon usable.
The instruction set is not the issue, the issue is on ARM there's no standardized way like on x86 to talk to specialized hardware, so drivers must be reimplemented with very little documentation.
As long as you're ok with arm64 guests, you can absolutely run both Ubuntu and Win11 VMs on M-series CPUs. Parallels also supports x86 guests via emulation.
> As long as you're ok with arm64 guests
I've run amd64 guests on M-series CPUs using Quem. Apple's Rosetta 2 is still a thing [1] for now.
[1]: https://support.apple.com/en-us/102527
[2]: https://mac.getutm.app
How is the performance when emulating the x86 architecture via parallels?
Also is it possible to convert an existing x86 VM to arm64 or do I just have to rebuild all of my software from scratch? I always had the perception that the arm64 versions of Windows & Ubuntu have inferior support both in terms of userland software and device drivers.
Same Armv8 ISA. And it's the same ISA Android Linux has run on for over a decade.
In a vm, I don't see why not.
Just run WSL inside of Windows.
Likely yes, eventually
Native, no. That would cannibalise Apple services which is a huge source of revenue for them.
Nobody is moving to Linux because there’s an iCloud replacement waiting for them over there…
Have you confirmed this? I haven't seen anyone concretely describe the boot policy of the Neo yet (it should be an easy enough check for anyone who has one in-hand).
Like any other Apple Silicon Mac, you can't currently boot into Linux but Apple has native container support that Linux works on [1].
[1]: https://github.com/apple/container
I'm writing this from Linux running natively (not virtualized) on an Apple Silicon mac (M1 Pro)
I'm aware of that option, but that's not something the average user is going to do. But knock yourself out if you want to try it.
If you were aware then why did you tell me I can't???
Not surprising but good to hear. It seems that there really isn’t anything that runs on a new MackBook Air that you couldn’t run on a NEO. It might not be as fast for some things but it gets the job done.
Isn’t basically m1 air equivalent in specs?
I’ve got that one and I’m yet to feel limited.
Ish. It’s better in some ways, like single core and maybe multi, but not by a ton. At the same time I think the M1 may have more raw GPU power, though missing a few fancy features.
Hardware is mostly worse, but that’s to be expected for the price. And nothing terrible, just little cuts all over.
It will have a longer support period than an M1 based on Apple’s history of device releases. This might also mean a longer support period for the 16-series phones than typical, similar to the 4S.
Always excited to hear about fellow M1 users. I’m not limited in the slightest. 5-6 years strong now?
I’ve been an M1 Air fan since I got mine in 2020 but recently things have become unusable. Playing 4K videos often drops frames, even at 30fps. And I can’t reliably run Notion’s transcription AI on Zoom calls, even though it’s not running locally. I’m going to do an OS reinstall soon to see if that helps, otherwise it will be time to upgrade…
Yeah, honestly not even counting. The only reason I even consider moving is that I dislike Tahoe and I know eventually I won’t be able to stall the update; hardware wise it doesn’t even cross my mind.
I have a current gen MacBook Pro for work configured with stupid amounts of ram and I feel no difference in terms of fluidity at all.
“Parallels Desktop runs on MacBook Neo in basic usability testing. The Parallels Engineering team has completed initial testing and confirmed that Parallels Desktop installs and virtual machines operate stably on MacBook Neo. Full validation and performance testing is ongoing, and additional compatibility statement will follow if required.”
If Parallels can run it, UTM likely can run a fair bit too.
Nice!
The best Windows laptop you can buy is still a MacBook.
Kinda like how back in the day, the best Mac you could buy was an Amiga. :)
Was that in doubt?
It uses the iphone processor (which I think still might be one of those Mchips?) so I think it was ok to be unsure.
The odds of it not running at all were low but the performance is the real factor for whether it can _practically_ run a windows VM.
Virtualization requires specific hardware support to be performant. There are ways to do complete software emulation of a virtual machine but it would be so slow that nobody would want to use it.
This is them confirming that the CPU has enough virtualization support that they can virtualize rather than emulate the guest OS
Yeah. It's the first production Mac using an A-chip and is a Mac that has had many things cut out for savings. The question is did Apple feature cut required functionality.
The first Apple Silicon developer boxes were Mac Minis with A series chips so I wouldn’t have expected any issues.
That's why I chose to specifically mention production. The developer boxes were to get macOS native stuff going but virtualization was not a priority.
The A12Z in the developer transition kit didn't support hardware virtualization.
does that mean since this is the iPhone 16 cpu, by proxy the iPhone 16 can also run Windows in a virtual machine?
Maybe/maybe not (we don't know how identical the A18 chip is to what shipped in the iPhone) - but it does determine that the virtualization stuff that was added to the M1 (in the era of the A14) has now moved over to the A series, at least enough to support macOS.
Thats pretty cool.
I mean, in principle you already can: https://getutm.app/
Is this a trick question? Of course. However Apple imposed artificial limitations, like disabling JIT.
Funnily it probably runs Windows better than the typical corporate spyware burdened x86 laptop.
Every thread about Windows on Hacker News includes claims about apps taking 30 seconds to launch, web pages taking 20 seconds to load, simple applications being unusable, and other extreme performance problems. These are puzzling for anyone (like me) who uses Windows at home without all of these extreme performance problems.
That was until I realized how many reports are coming from people talking about their work laptops loaded with endpoint management and security software. Some of those endpoint control solutions are so heavy that the laptop feels like you've traveled back in time 15 years and you're using a mechanical hard drive.
There's an unspoken rule in corporate America, colleges, etc. Laptops MUST be loaded down with terrible software, no exceptions. My last corporate laptop actually had the paid version of winzip in 2025, and it ran with a little tray icon that I couldn't disable or remove. That was in addition to all the other corporate crap I couldn't remove.
Some of this is not _just_ a corporate problem. Why would Winzip have an auto run application and tray application in the first place? Every single app seems to think they need one, and it's a classical tragedy of the commons. Perhaps on a virgin Windows install, your app with autorun and a tray icon will be more responsive. But when 20 other apps pull that same trick, no one wins.
This is actually one of the reasons I'm not excited at the idea of Linux defeating Windows. If it did, corporations would just start crapping up Linux the way they've crapped up Windows.
Our corporate linux machines have exactly the same monitoring software as Windows - even the servers. The performance is still not even remotely comparable. Could be the hooks are more performant on linux, could be the filesystem, maybe the tools are written more sanely... But loading apps, filesystem operations... Everything is still far faster on the linux dev instance. And I have half the ram allocated to that one.
The reason every developer makes their app open at startup, is because the Windows ecosystem doesn't have a good package manager. So every app needs to be its own package manager and check for updates on a timer. So they need to run all the time so they can run that timer.
In theory the Windows Store will handle updates. In practice, I avoid the Windows Store version of applications. Also, you can't turn off app updating, only pause them for a time.
I like videogames, maybe more than I should at my age, and I prefer to play them from Steam in Linux through Proton. A couple of months ago I caved in and bought a proper Windows gaming miniPC because a game I want is not stable in Proton.
I use a corporate Windows VDI at work, so the experience is understandably subpar there, but it is still horrible on high.end hardware. Took me half a day just to herd it through update after update, while avoiding linking it to a Microsoft account despite its protests.
It's literally used to run only Steam and Firefox, and it still sucks compared to the ease of install/management of Linux. Ubuntu LTS took me about an hour to set up dual boot, apply updates, install Steam, and every other software and tool I use daily.
Why is Windows 11 still so clunky in 2026? It doesn't feel like the flagship product that many bright minds have improved for three decades. Why are hobbyists and small companies outperforming Microsoft's OS management?
I once worked on a computer for the US Government that felt slow. I counted nine (9) directly competitive and redundant endpoint protection products on it.
Not nine different/only somewhat overlapping pieces of software from companies that were competitors. Nine equivalent products. I guess defender made ten.
Can you elaborate?
Ten protection layers! This is the reverse of the seven proxies meme.
Corporate spyware is pretty nasty, regardless of platform. When I was at FB, they had something that forced a kernel module that was incompatible with the next big OS release; and I had accidentally disabled the FB spyware scripts. I set /etc/hosts to immutable because I was tired of them fucking with it ... didn't realize that's why things were better for the next 3 months, until I did the major update and I had to fix things from safe mode ... where everything only barely works.
Microsoft also puts a lot of crap into a default install that you may want to disable. Windows 11 with some judicious policy editor settings isn't so awful.
Outside corporate setting, it is also the fact that most windows systems you encounter are installed on cheap machines by people who just care that their word processor works a few times a month. And you were probably forced to fix it.
At the same time, as someone with a well maintained Windows gaming rig, I don't like spending time in the OS these days. Something about transparently doing stuff that puts money in their pocket while inconveniencing me gives me the ick.
No this is not just an enterprise issue. I waited 10 seconds (I counted.) for a Windows Explorer context menu to open the other day. This is on a fully decked out system with an Ultra 9 cpu and a 4090 and 32gb of memory, and basically no apps running. I think I had 2 tabs in Edge? Windows is a shitshow these days.
I just tried to open the context menu in Windows Explorer. It showed up almost as soon as I released the mouse button, and I have a much slower CPU, older video card, and way less RAM then you do. I was also running 12 windows of Firefox with collectively 1000+ tabs (though only about 36 or loaded), Steam, a Unity game, and Microsoft Teams, plus a number of background programs.
If your Explorer context menu is taking more than a split second to load, there's something wrong with your hardware.
Other than hardware it could also be some third-party software hooking into Explorer to do who knows what.
Oh yeah no... its still terrible even without all the spyware.
First experience of Windows 11, trying to download a file through firefox caused my 18 core 10980xe to have the entire UI freeze for the full time the download was going.
Reverted back to windows 10 immediately and the problem went away.
Windows 11 is full of spyware from the Mothership
I've said for decades that from a user perspective, malware scanners and prevention tools are fundamentally indistinguishable from actual malware. They intercept file accesses, block you from doing what you want to do, pop things up all over the place, and make your machine slow aand unreliable.
Took 6 minutes from power button to login prompt this morning. Probably even longer from login responsive desktop. So yes, probably!
I’ve helped someone with a rather clean iMac, circa 2019, still supported by Apple. Forget 6 minutes — you can spend a full hour from boot to giving up trying to get anything done.
I think that Apple has gotten so used to having fast storage in their machines that the newer OSes basically don’t work on spinning rust.
these iMacs have horrible Fusion drives (128GB SSD + 1TB HDD combo) iirc that fail often. Have you looked into that?
what? on a semi modern CPU and a SATA / M2 SSD?? My Vista laptop on a spinning drive took that long to boot I am pretty sure. I am flabbergasted if this is true
corporate laptops is the key here. take 2 identical laptops one with and one without the spyware - its night and day in both performance and battery life.
My corporate spyware laden Surface ARM runs Windows faster than the Macbook Neo, but unlike the Neo can survive a fall onto a concrete floor. (Ask how I know...)
My home laptop is even faster.
How do you know a Neo cannot survive a fall onto a concrete floor? I think it would take at least ten tests each with a new machine to get some confidence of the impossibility of that.
Geekbench 6 was around ~2600 single-core with the VM overhead for me. That's still punching above single-core power in its class for Windows machines and it makes me giggle.
https://browser.geekbench.com/v6/cpu/17011372
This was the latest UTM in the App Store, so native Hypervisor.Framework access for arm64 Windows acceleration.
Wouldn't corporate spyware equally burden the NEO? Especially more give the 8GB of RAM vs 16+ on X64 laptops? Chrome, Teams, IDEs, websites etc are equally bloated on both platforms.
The cpu in Neo is 2-3 times faster.
My (former) corpo HP laptop with 16GB RAM had 75% RAM used at idle after a fresh boot with Outlook, Teams and all the copro shit running in the background. So the 8GB NEO CPU will spend its time swapping data from ram to disk versus the 16GB+ ones, given both being filled with corporate spyware and same heavy use cases.
Also it isn't 2-3x faster, stop with the made up nonsense please. Just checked and my 3 year old AMD laptop is on par with the NEO geekbench score I found online (slower in single core but faster in multi core), not 2-3x slower.
Man, I do wonder what the realistic lifespan of that single NAND chip will be after it gets hammered by constant swapping of running tasks way beyond the capabilities of a 8GB RAM machine.
I have a PC with a 10+ year old 256GB SATA Samsung SSD that's still in top shape, but that's different because that drive has those 256GB split over several NAND chips inside, so wear is spread out and shuffled around by the controller to extend lifespan. But when your entire wearable storage is a single soldered chip, I'm not very optimistic about long term reliability.
There was quite a bit of discussion about that when the M1 first came out, but none of it really seemed to have happened six years later. The target audience isn't in danger of wearing it out and the ones that will push the limits will grow tired of it and sell it in a year or two or move on to the Neo 2, which might have 12gb of ram due to the expected chip.
I still think it's a great machine, but I think all these worries about NAND dying really haven't come to fruition, and probably won't. I have about a hundred plus of various SSD Macs in service and not one has failed in any circumstance aside from a couple of battery issues (never charged and sat in the box for 2 years, and never off the charger).
>There was quite a bit of discussion about that when the M1 first came out, but none of it really seemed to have happened six years later.
1. How do you know nothing happened? Define nothing in this case. Do Mac users check and report their SSD wear anywhere?
2. Didn't the OG 256gb M1 have 2 128MB NAND chips instead of one 256 meaning better wear resistance?
If swapping was causing SSDs to fail on M1 Macs, we would never see the end of the hysterical articles about "NANDgate". Since we haven't seen any in all these years, it's seems pretty certain it's not happening.
Hysteria would be if all had an issue like the keyboard gate, but this isn't an issue, it's a design limitation for certain uses cases which not everyone has. Some users will wear out faster than others due to usage patterns. If their M1 dies after 6 years of heavy usage, do you think they'll investigate if it was the NAND that died and go online to tell the news, or will they chuck it and buy new one?
NAND is still the same wearable part that regular X64 laptops have, Apple doesn't use some magic industrial grade parts but same dies that Samsung, Micron and SK ship to X64 OEMS, and those are replaceable for a reason, because they eventually fail.
The reality is most 8GB M1 Macs are still working just fine 6 years later. Power users know they need more than 8GB of RAM and will buy a MacBook Air or Pro with 16GB+.
The MacBook neo is for students, grandparents, travel, etc.
Hell, even if it dies after 6 years it was still a better experience than using a $500-600 windows PC and the cost comes out to ~$8/month spread over 6 years.
>The reality is most 8GB M1 Macs are still working just fine 6 years later.
Do you think SSD drives are replaceable for no reason? Just because M1 mac aren't failing left and right doesn't mean their NAND won't fail.
Even though I like the NEO, I can't in good faith buy a machine with soldered wearable parts. That's like buying a car with soldered brake pads because "in 6 years average users don't feel like they need changing".
I still had laptops on my hands from 20 years ago that work fine simply because you can swap their drives with fresh ones. How many M1 mac will still be functional in 20 years?
>but that's different because that drive has those 256GB split over several NAND chips inside, so wear is spread out and shuffled around by the controller to extend lifespan. But when your entire wearable storage is a single soldered chip, I'm not very optimistic about long term reliability.
I thought wear leveling worked at the page/block level, not the chip level? On an SSD, if there was a failure of an entire chip, you're still screwed.
from what i seen in "low end" ssds like the "120gb sata sandisk ones" under windows in heavy near constant pagging loads is that they exceed by quite a lot their manufacturer lifetime TBW before actually actually started producing actual filesystem errors.
I can see this could be a weaker spot in the durability of this device, but certainly it still could take a few years of abuse before anything breaks.
an outdated study (2015) but inline with the "low end ssds" i mentioned.
https://techreport.com/review/the-ssd-endurance-experiment-t...
The M2 MacBook Air base model has 8GB RAM and a single 256GB NAND chip. Nearly 4 years later, it doesn't seem to have caused any problems.
Most flash has average wear out after 300k cycles. Let's say 64GB is used for swap. That's 19200 TB or 19.2 PETABYTES of Swap usage. Let's say you swap 12GB a day, you will burn out that 64GB of Flash Storage in 4.38 years and my guess is that amount of swap usage is extremely high that user would probably replace laptop sooner out of performance frustration.
>Most flash has average wear out after 300k cycles
No it doesn't. Most 1TB drives are rated for around 600 TBW, so enough to overwrite the drive 600 times, nowhere near 300k cycles. If you search for specs of NAND chips used in SSDs, you'll find they're rated for cycles on the order of hundreds to thousands, still nowhere near "300k".
https://www.techpowerup.com/ssd-specs/crucial-mx500-4-tb.d95...
12GB a day isn't very much. If your working set is larger than the 8GB RAM, you're swapping multiple times per second. It doesn't take very many megabytes per swap to reach 12GB if you're doing that multiple times per second.
that doesn't maths
Now just needs to have that pre-installed by Apple, and macOS somehow hidden during boot time.