Assuming you're an engineer, have you thought about what handing over a "window" from one computer to another actually entails? CRIU can do checkpoint/restore at container/process level - but you actually want it to run on both, no? So you need to split off just the I/O, but at the OS-level per window.
Apple has been doing a lot of work in this direction and they have stuff that actually work (like video calls and to some extent windows. These are processes running on different OS-es with a matrix of hundreds of devices.
It's not something you vibe code over the weekend.
There may be simplifications to this, but I suspect it eventually lands in a distributed system problem involving state and strong consistency semantics. I’ve done this for most of my engineering career and all I can say is that they are fun problems to work on because they are some of the most difficult ones :)
the revenue-driven decision was choosing to make the OS more like iOS, locked down with an app store, rather than macos, which allows third-party applications, browsing the filesystem, dropping into a terminal, etc. with built-in first-party support.
instead of making a computer in an AR form-factor, they made an iPad in an AR form-factor.
And it's easier and more power-efficient (because of hardware video encoding) to use screen sharing instead of sending drawing commands etc.
I'm very eagerly looking forward to Valve's headset coming out.
It is priced to be a pack-in with seats of Dassault 3DEXPERIENCE, but Meta's experience has been that the VR consumer is price sensitive which is why they followed up the MQ3 with the cost-reduced MQ3S.
I think Apple is looking at this the way they look at AirPods (something you stick in your ears to modify and augment your hearing) which is good and monitors, which is bad. Apple's always sold a tiny number of monitors with astronomical margins which looked like a good business because they didn't have to invest in product innovation to make them the way they do for the AVP -- and they didn't need software developers to invest in product innovation for their monitors, but the AVP absolutely requires it. And if they aren't shipping enough units, who is going to make software for it?
The biggest issues left in VR sim racing are still arguably comfort and weight in longer races, but even that is getting solved now with really compact new headsets like the Big Screen Beyond 2 etc.
I'm honestly not sure if Apple Vision would fare much better if they had a device that costs $100. Like, sure, they'll sell more units that way. But how many of those units would end up collecting dust?
My take is the MQ3 is pretty good for UX. I regularly use my image sorter and RSS reader in a web browser with a huge number of windows open, bringing my click targets up to the WCAG AAA standards made them very usable in this environment. The Apple Magic Keyboard pairs perfectly, as does a mouse. The controller work great for games and other immersive applications.
I'd say though that my game backlog on the MQ3 is long as is my game backlog in Steam, and there is just a huge amount of competition from flat content so I don't play as many VR games as I could.
The metrics are brutal, for Quest 3 and for every other headset. Meta worked hard to bring the costs down and improve the UX, and it still isn't enough. They wanted "the next smartphone" and fell so short it's not even funny.
It's less comfortable, inputs are much less precise, inputs take longer, output is less legible, output has less information... The list goes on.
https://www.kandaovr.com/qoocam-ego
It eats batteries but you can get a charger and more batteries the way you would have a few mirrorless generations ago.
My trouble is not really having a system to show these to people, red/green anglyohs sometimes work fret but aren’t consistent, there doesn’t seem to be an ‘instagram’ for sharing stereo photos, it ought to be easy to make a WebXR application to show stereograms but I haven’t seen a fully realized one and found texture memory limits are a bitch in the MQ3 so my first attempt to make one got stuck.
I wrote this to take the .MPO files (a common file format on the earlier commercial digital stereo cameras) and convert them to print-ready stereograms: https://github.com/EngineersNeedArt/Stereographer
You'd also have more of a critical mass of people with accessories. A used market for lenses and accessories, people with spare batteries about, etc.
Not to mention an app ecosystem. For every person like me lamenting what could be, there are a dozen people with more mainstream desires, like Instagram 3D or Temple Run 3D or whatnot.
I hope they swap out Haskell though, probably can make your own compositor
But Valve is reliably rumored to be releasing an untethered successor within the next year for a third of the cost of the AVP. The appeal is having a dozen monitors anywhere.
Vision Pro M5 PR says something about 120mhz. So the M5's "40% faster GPUs" is being used to double the frame rate?
Being noob, I'm guessing each desktop requires 3 additional cores and 16Gb VRAM.
Of course, we can think of many strategies (using current hardware) to support additional desktops, spaces, windows, etc. If Apple really wanted to.
60mhz and 120mhz are your typos, because that's an insane refresh rate. It's 60Hz and 120Hz, not mhz...
> So the M5's "40% faster GPUs" is being used to double the frame rate?
You always have to read full marketing BS when you read it, it almost always says "X times faster than Y in Z workfload".
> "X times faster than Y in Z workfload"
Mosdef.
Also, reducing energy use (heat generated), mostly thru process improvements, allows each new generation to run that much faster.
It was incredible how much Vision Pro marketing was based on stuff Microsoft had demoed for HoloLens a couple of years before.
I still remember the good old days of Gary's mod movie theater servers.
https://github.com/saagarjha/Ensemble
This is what is looks like in action:
https://twitter.com/TheOriginaliTE/status/175125156764134634...
Disclaimer: I don’t know how well it works in practice because I don’t have a Vision Pro.
I do wonder if that person had ever tried any of the myriad of VR device available a decade before beforehand and I do wonder how popular Apple's product would be if other companies were given the opportunity to integrate with the OS on the same level that Apple can (I feel like Apple is breaking anti-competitive laws constantly but nobody really cares about making them open up).
I also don't think think this breaks anti-trust but percentages doesn't stop market manipulation.
My point in saying the usefulness of it being tied to the ecosystem was more of a negative one than a positive one if it wasn't clear. It is personally useful for me because I have a lot of Apple products at home (though I also have PC and linux stuff too), but I wish my primary use-cases for it were more platform agnostic.
I'm also very in support of aggressive anti-trust legislation and it's probably my biggest point of contention with Apple.
Despite all that, I still like the Vision Pro and think it's an incredible piece of tech that blows every other headset I've tried out of the water for the things I like to do.
Microsoft had and still has this opportunity, they even have game console and much of game developers to bootstrap VR/AR ecosystem.
> (I feel like Apple is breaking anti-competitive laws constantly but nobody really cares about making them open up).
lol okay
Even before it came out, I naturally assumed it was going to be able to do this. Major flub IMHO. Well that and the completely superfluous frontfacing screen for your "virtual avatar". Because the Vision wasn't already expensive enough...
That's just an hyper expensive huge screen.
I would love to be able to have a dedicated spatial location for each part of my Django app: look over here for the CSS, step here for the views, scroll through the logs over here...
1) It's been doable for years and yet AR/VR doesn't take off
2) It will only ever be a hyper niche use case of an already hyper niche market. People do not want a giant screen glued to their face.
Have you actually tried "working" in VR? It is awful and pointless and zero percent of my work is limited by the size of my screen FFS. Are you one of those people that keep 100 terminal windows open with random TOPS and log watchers but never actually uses them? Those people always seem to have lots of monitors. They don't actually seem more productive though.
I even downgraded from two external monitors to one because all the extra monitor was doing was making me hurt my neck from looking back and forth.
A 30 inch monitor with 1440p resolution can do everything you want from django and is like $200
And no I haven't tried working in VR. I do actually just work with one monitor and only have a few terminal tabs running at any one time, but honestly do think there are so many improvements I could make to my workflow before even a perfect VR solution would be worth considering.
Any monitor recs? My current one has started flickering and it's slowly driving me insane
Pimax and Bigscreen I think work hard for that market? Bigscreen has those neat goggles which I'd like to try. Rumors are that Valve is building another headset, with better specs. That might put pressure on that market.
For monitors, I must confess I am somewhat wrong. My 32inch (or something) LG monitor might have been $400, not $200. But I used a giant TV for years thinking I didn't need "nice" monitors and I was so wrong. 1440p and over 120hz refresh rate are so worth it, even for work tasks. When I bought it a couple years ago, the caveats were that it's a VA panel, so it has really bad ghosting. It's noticeable in games, but somehow has not ruined my experience. The other types of panels have their own drawbacks I think. OLED is the best non-compromise for image quality but then you will have to keep burn in at bay.
I like my LG monitor. My girlfriend's ASUS monitor is nice. I hear nice things about BenQ, but you're better off with review sites than my anecdotes.
My window management "trick" is that no window is full screen, and every window is overlapped such that there's always a sliver of the "under" window I can click even if that window is otherwise not in view. I also make significant use of alt-tab. I think people who use the multiple desktop feature of modern windows and mac have a much better system. One desktop per "Task" I think works really well.
To a first approximation, Apple is a manufacturer of locked-down handheld entertainment appliances whose primary function is to psychologically condition children into siphoning off money from their inattentive parents. There's no reason to suspect their vision for the AVP to diverge significantly from this user story.
Any other hardware is a non-starter.
When you say something like "whose primary function is to psychologically condition children into siphoning off money from their inattentive parents", it's rhetorically effective. It paints a vivid picture. It encourages your readers to have a certain attitude towards the company you're talking about.
In other words, that bit of what you wrote was load-bearing. It served a purpose for you. That means that it isn't exempt from criticism. We should reject conversational norms according to which it's OK to throw in these little barbs but not OK to object when someone points out that what you're saying is flatly false.
"My goat-fucking opponent wants to raise your taxes and use the revenue to subsidize tobacco companies. You should vote against him." "Excuse me, I am absolutely not a goat-fucker. How dare you?" "Look how he avoids the central argument about his policies!"
As to the "actual argument": no, actually, that clearly isn't your actual argument, or at least if it is then your argument is unsound.
You can't get from "Macs are <10% of Apple's revenue"[1] to any prediction about what Apple will do with the Vision Pro. For that, you need to (1) classify it as "like an iPhone or iPad" rather than "like a Mac" -- which I agree is a reasonable classification, though you haven't bothered to argue for it at all and it is at least a bit debatable -- and then (2) look at what sort of thing Apple does with its iPhones and iPads. This bit you have done, kinda ... and this bit is exactly the bit where you said something obviously false. "whose primary function is to psychologically condition children into siphoning off money from their inattentive parents", remember? That is, or claims to be, a description of what sort of thing Apple want their devices to do. It's exactly the sort of thing that's directly relevant to supporting what you say about what features we should expect them to give the Vision Pro and its software. And, once again, it's plainly false.
[1] Perfectly true, though "an insignificant historical afterthought" is obviously false -- once again, you're festooning what you say is your "actual argument" with little untruths that make the "actual argument" feel stronger, and I wish you wouldn't -- and, also "historical afterthought" is kinda nonsensical, no? Being a historical relic and being an afterthought are opposite and incompatible varieties of insignificance.
There absolutely is an argument to be made along the lines that the VP is kinda like an iPad, and despite their impressive hardware capabilities iPads are designed for entertainment much more than for getting useful work done, and so we should also expect the VP's software to pass up opportunities to make the device useful for serious purposes in favour of making it an entertainment-consumption device. You could totally have done that. It would have been pretty similar to what you wrote. It would have been rather a persuasive argument. But it amused you to go way the hell over the top and say that Apple's non-Mac devices are mostly intended to manipulate children into wasting their parents' money and, once again, that's obviously not true and you put it right where the core of your "actual argument" should have been.
I'm aware that I'm making rather a big deal of a small lie. But this sort of thing is everywhere in online discourse at the moment, and I am getting extremely fed up of it. It's never enough to make a reasonable argument; it's always necessary to throw in all these playing-to-the-gallery jabs, which no doubt get you a bunch of likes and retweets and other forms of Meaningless Internet Points. It's yet another form of the optimizing-for-engagement that is eating our societies alive, and the HN crowd is supposed to be smarter than that, and I wish we would all collectively Do Better, damn it.
Personally I find a Mac to be a better development environment than Windows even for Microsoft tech like the dotnet stack.
This is both true and completely irrelevant, because the point of the above comment is that Apple is not in the business of catering to this use case.
Look at the annual revenues: $225 billion from iPhone/iPad, $100 billion from "services" (which Apple mostly characterizes as "app store stuff"), $40 billion from accessories (watch, airpods, etc), $30 billion from desktops. The Mac segment comes out to 7.9% of their overall revenue. And this number is shrinking in both the absolute and relative sense, as "services" continues to grow and as Mac units shipped peaked in 2022.
Even without that, calling a $30B business an "insignificant historical afterthought" is a bit of an exaggeration, no?
This is both entirely true and still manages to miss the point. Yes, Apple keeps the Mac around exclusively to accommodate the creation of iOS apps. No, they are not financially incentivized to create new categories of hardware that cater toward productivity when instead they could lock them down and milk that cow for all it's worth via the app store tax.
That word right there means that we can both be correct, because I personally find MacOS to be frustrating for Dev compared to every other alternative. Great for creativity, but not so much for productivity.
But the caveat to my statement is that _everything_ added to their ecosystem to business reasons is useless and counterproductive. For example I can plug my Android phone into a Windows machine (two different companies inb4 someone uses flawed logic) and it just works. If I plug my Android phone into my Macbook it doesn't work at all...but an iPhone does! ;)
They only very recently got decent-ish Window management, basic snapping that Linux/Windows has had for at least a decade or longer. And even then their implementation is "pretty" but slow to respond. It's like just expand and snap the fucking window for fuck's sake.
In terms of the "development environment" they enjoy having had the OS built on top of FreeBSD (which yes, they have contributed to - bet Apple management hated that).
To me it's a machine that gets stuff done; they could literally strip the thing down to the bare minimum, removing all of the "magical wonderful Apple stuff with cutesy fancy sounding names" and I couldn't give a shit.
"Retina" screen? Fuck offff Apple.
That doesn't mean MacOS is good, though.
As a film maker and editor I'm enormously more productive on my dual screen (used) M1 Max Studio machine than I was on a variety of PC setups with high end GPUs. Even just the missing overhead of not having to keep graphic drives and constant Windows updates is great. Reliable renders were never a thing on Windows. The time lost doing things again because window had some strange colour issue, render crash, font issue and on and on was ludicrous.
The idea of having to use Windows for daily productivity sends a shiver down my spine.
Software is definitely more debatable.
I've not seen a professional desktop operating system show you adverts and click bait during worktime before and I am not impressed.
I'm taking it as an alarm bell that I should get out as soon as I can. For me and what I do, macOS or Linux are my platforms of choice.
Even billionaires get burnout.
> how does windows and Linux waste your time?
Windows is deep down the enshittification curve.
My Linux-using coworkers seem to spend a lot of time wiping things and switching distros. I'm very happy with it for servers, but not for desktop.
- 15 minute boot-up times
- Complete lack of power management ( close your laptop screen )
- Random forced updates
By what measure? As I understand it, more Macs sell now than in any other point in history.
https://www.apple.com/newsroom/videos/2025/autoplay/10/apple...
My workaround for this is using applications that work comfortably in a safari window (like vscode) and you can have as many safari windows free floating as you want
(I returned it)
Do you feel you're more in focus when using Vision Pro like this, as opposed to using an external monitor?
There was this program called Immersed for Quest and some other VR headsets (apparently Apple's one too?) that ages ago was quite brilliant: I could connect my laptop or PC and have as many virtual desktops in addition to the real ones displayed around me. Even on my Quest 2, until the strain set in, it was cool to just kinda tune out the room around me and be in a black void surrounded by just the screens of whatever I'm working on.
Sadly, in one update they randomly removed it as a "legacy feature nobody uses", which ruined the program for me, and Virtual Desktop for the Quest also has limited screens you can display, so I can't even do all 4 of my physical ones, nor virtual ones - despite it having worked previously in the other program.
For a little bit, it was a really cool mode of working, but sadly I only got glimpses of it and would need more capable hardware like AVP for that in the first place (Quest 2 very much had its limitations).
Also, this was a pretty cool idea, even if a novelty: https://www.youtube.com/watch?v=Js7Y1H5D8cY
This gave me the impression it wasn't ready for general use, and they would have to control your impression of the thing while selling you on it.
I know that VR is hard, and AR is much harder, but this made me think apple still hasn't cracked it.
As to single app - I figure the thing has to be hard real-time to do things without giving you a headache or nausea/vomiting. that probably doesn't leave room for a lot of sharing of resources.
In the end, I think this is a iphone 1.0 type device, and only continued development with deep pockets will make it viable, eventually.
There isn’t any prep I saw that they would have to do before they could except a walk-in.
If they'd focused on maximizing the device's usefulness instead of its revenue stream, maybe things would have worked out better.