For what it's worth, on the off-chance that anyone at MS who worked on this views the thread, I thought it was cool. I liked what you did. I'm sure that seeing people review the Vision Pro is frustrating, seeing how many things you already did being declared revolutionary. I'm sure it's like that for Oculus developers, too.
In retrospect, people like John Gruber and Tim Urban will mock the industry prior to the Vision Pro for not Getting It Right, or for fundamentally-flawed interaction models, ignoring where things were right. Eye-tracked selection was already a thing in the Sony headset. Tracking windows fixed throughout your environment has been a thing in the Oculus inside-out devices since release, good enough to keep track even on different floors. Varjo does reprojected AR better than Apple, for a cheaper price point (though still a steep one).
Microsoft, for their part, captured a sort of mundane magic:
I still dispute that eye tracked selection will prove to be the correct solution. I think it might be part of the correct solution but eye, hand and controller based interactions all have distinct advantages and disadvantages and I'm very sceptical of anyone claiming that only one of them should be the only option across a wide range of applications.
Personally speaking I have used all three and if someone put a gun to my head and insisted that I had to pick just one, it would be 6DOF controllers. (and that's me also trying to take into account future improvements in implementations for each)
I agree. I do think eye tracking selection is part of the solution, but combined with at least one physical button on a physical controller. The worst part of my Vision Pro demo was the finger gesture tracking; it's simultaneously impressively good and not nearly good enough. The reliability and tactile feedback of buttons can't be beat.
But there are a lot of ways to do minimal controllers that don't look like gamepads. They could be tiny to fit on your keychain, or built into a ring the size of a normal wedding band. You really only need one solitary button when combined with eye tracking. Since you're already carrying your phone everywhere, perhaps phones could add a physical button on the back for this use case. (The existing buttons wouldn't be good for this because they are intentionally hard to press.)
I love controllers, personally. I think the actual solution is a lot closer to making controllers into hands than the other way around. For now, though? People are going to incessantly bask in the Sony-Apple approach.
I want a little track nub like thinkpads have that’s glued to my thumb so I can click as if doing a bomb detonation movement or do basic 2D movement inputs
What about that track pad you use with your tongue? [1]
Or maybe some wires implanted into your tongue nerves that activate cursor movement when you think about moving your tongue and shock your throat muscles when you become apneic at night (a la inspire device).
This happens with every Apple product, and people need to get over it.
iPod: "No wireless. Less space than a Nomad. Lame."
iPhone: "it doesn't appeal to business customers because it doesn't have a keyboard"
iPod did not invent MP3 players, iPhone did not invent multi-touch or smartphones.
Yet both became the defacto standards for every future mp3 player and smart phone.
The same is going to happen with the Vision Pro - for better or for worse. Everyone trying to do only AR or only VR will give up, and try to do what apple is doing.
Is there some law of nature that Apple always has the right ideas? I very much doubt it.
They've had plenty of misses, not just hits. They got the mouse wrong (three buttons are now universal, one was never even close to good enough). The touch bar was a failure. The Apple Watch has not redefined how smartwatch UIs work. MacOS has plenty of UI ideas which have not been adopted by any other OS, most notably the top bar and the left-hand side window controls.
they only succeeded when the market expanded such that you don't have to be technically right, but just have the larger marketing budget.
ipod was advertised around the world like absolutely nothing before it. there was no international launch of consumer device before it.
iphone destroyed the media because att did not have the best network but saw they could sell more data than they could sell minutes. so it also had the most marketing of any telecom device in history.
all that in a time consumer eletronics left the enterprise+niche market.
hardly either are the case with the headset. it will flop. and if not, it's because they undertood fomo influencer marketing better than everyone else.
But at least the touch bar was honestly pretty great. It happened to arrive with a generation of MBPs that killed all their ports and ruined the keyboard, which is what people more hated.
The Apple Watch seems to be the only smart watch I see around - though granted smart watches are not as ubiquitous as smart phones.
MacOS is tricky because it just feels like if it wasn't for Office at The Office, it would have just fully won...
> The Apple Watch seems to be the only smart watch I see around - though granted smart watches are not as ubiquitous as smart phones.
From some quick Googling, while they are the biggest player by far, they have around 23% of the smartwatch market. And as far as I have seen, other smartwatches are not so similar to the Apple Watch as Andoid is to iOS.
> MacOS is tricky because it just feels like if it wasn't for Office at The Office, it would have just fully won.
MacOS is tied to extremely expensive hardware, so it was never going to win. And Apple never actively pushed for training on Macs like MS did, so there are many orders of magnitude more pieces of software exclusive to Windows than to MacOS. Office may be MS's crown jewel, but there is giant long tail of Windows software that dwarfs even Unix, not to mention MacOS in particular.
You see more apple watches than other because it's one of the smart watches that can't be mistaken for a normal watch.
I see quite a lot of people using smart watches due to actively going to gym (where a lot of people use them for fitness tracking etc).
A lot of them are hard to recognise as smart watch unless you know the specific model already, or catch someone interacting with it. The ones I and my partner uses are similarly "invisible" (Garmin Fenix series).
In comparison, Apple Watch is the kind where it's immediately obvious you see an apple watch or something that apes its design.
I won't miss it. WMR was by far and away the worst experience available on the market.
Putting aside the numerous bugs in the VR environment, the software itself was typical Microsoft garbage. It absolutely infests your system and clobbers any and all other VR runtimes you have. Difficult to remove, and once you do uninstall, your other runtimes like OpenXR or SteamVR are still left broken and you have to reinstall those too.
I had to develop an app for WMR at the same time we were working on OpenXR/SteamVR apps. I simply could not do my job until I ripped out WMR. We relegated it and the headset that only works with WMR to an isolated machine used exclusively for testing WMR.
The industry is much better off without Microsoft trying to elbow their way in to disrupt standardization around OpenXR.
No, because Windows would recognize when a WMR headset was plugged in and throw up a big screen inviting you to install WMR at every boot and at random times between.
> Windows Mixed Reality is deprecated and will be removed in Windows 11, version 24H2. This deprecation includes the Mixed Reality Portal app, Windows Mixed Reality for SteamVR, and Steam VR Beta. Existing Windows Mixed Reality devices will continue to work with Steam through November 2026, if users remain on their current released version of Windows 11, version 23H2. After November 2026, Windows Mixed Reality will no longer receive security updates, nonsecurity updates, bug fixes, technical support, or online technical content updates.
> This deprecation doesn't affect HoloLens. We remain committed to HoloLens and our enterprise customers
--
Some more HN discussion about it when it was announced back in December:
That's a real shame, but it's clear they stopped investing in it years ago. This is just an acknowledgement of that fact.
I have a WMR collecting dust as we speak. I loved the concept for a usable VR workspace! And it had a LOT going for it. But too many issues, bugs, and QoL issues were never addressed in the workspace:
* Allowing a user to exclusively use a physical keyboard (the floating keyboard would automatically pop up if you clicked inside a text box or app)
* Only about 4 windows/apps floating in your view could actively refresh (a damn shame as having a bunch of floating SSH sessions felt quite cyberpunk, and it put tabs for browsers to shame)
* Only one desktop view at a time, even if you had multiple monitors
* Bad performance on most mobile devices
* No support for non-WMR headsets (i.e. I could never use one of my Quests with it)
* They made a tease of a Halo demo that never panned out (OK, I admit this was sour grapes, but it was just a tease at the end, just when it got interesting)
So much potential. But they basically created version 1 and either never listened to the audience, or never invested in it again.
Wow, is there any indication of an alternative? Or are they just going to brick everyone's WMR headsets (except HoloLens of course, because that's their one)?
Edit: As far as I can tell they're just yanking support from Windows and tough bikkies. I asked Bing about it and it basically said to buy a different headset or hope a third party releases alternative software that can run the headsets. Oh and also no you can't prevent Windows from auto-updating and disabling it.
Extremely disappointing behaviour on MS' part. I can understand them sunsetting development, but actively rendering fully functional devices unusable is criminal.
After what MS did with Windows Phone and the first Lumia headsets*, I think you need to be crazy to release hardware that depends on a not-yet-established MS product. They're very good at keeping their successful products backwards compatible and at least somewhat supported for years, but they can dump unproven products like led disregarding any kind of marketing push.
* Nokia's first Lumia smartphones released with an older version of Windows Phone, while MS was publically working on the Windows 8 push for a common OS across all form factors. They promised everyone in huge marketing campaigns that the Lumias would be upgradeable to the new Windows Phone 8, which all the marketing was pushing for. Fast forward two months from the release, they announce that no, they would not upgrade and you'll have to buy a whole new phone if you made the mistake of buying a first generation Lumia.
Announcing this long ago doesn't make it any less shitty that a fully working still relevant and expensive VR device is now a paperweight. Keeping a minimal framework alive so that the devices can be used with e.g. steam would have been the right thing to do.
Also, no one using WMR devices on Windows wants to bother with Linux. Most of us were using the device with things not available on Linux, and even if they were we don't want to run Linux on a desktop. It's not a solution.
> Also, no one using WMR devices on Windows wants to bother with Linux. Most of us were using the device with things not available on Linux, and even if they were we don't want to run Linux on a desktop.
Cheaper headsets for Linux users, then! It's a solution in one way.
Most users of WMR headsets had a Reverb G2 because it was by far the cheapest headset with its featureset, and were willing to bother with the issues that WMR brings and tinker around.
> WMR headsets work on Linux pretty well nowadays.
Works pretty well as in 'apt-get install wmr-for-linux' or as in 'spend three weeks setting up a build environment for some hairy proof-of-concept code which isn't compatible with SteamVR games'? WMR is one of the last things really tying me to Windows, if it's practical to use on Linux I could see myself switching back.
Works pretty well as in, "Install Monado using your distribution's package manager, set it as your default OpenXR runtime, hit three buttons to get Valve to begrudgingly respect system defaults instead of the closed-source 'Open'VR runtime."
Optically-tracked WMR controllers work, depending on how new of a build your distribution has and the type, but WMR controllers are generally considered to be pretty bad to begin with. You can use a different style of controller with a WMR headset if desired.
That said, building Monado isn't hard if your distribution doesn't currently have a version that supports WMR controllers. The repo lists all of the dependencies you need, and what they're called for popular distros. It also tells you the exact things you need to run to get it to build; you can just copy and paste the build instructions after installing its dependencies. You can probably get it done within ten minutes.
Optically tracked WMR controllers do not work yet in Monado. Some people have code to do the optical part of the tracking, but there is a bit of reverse engineering left to synchronize the IMUs and control the LED brightness to enable sensor fusion.
I wouldn't call it pretty well, not quite yet. The Monado driver has come a long way, but doing anything useful with it still has too much legwork involved for the majority of WMR users, not including the switch to Linux. Although since the announcement, work on WMR-specific stuff seems to have sped up, so hopefully the situation improves.
I also recall seeing a working demo of hand tracking for WMR (maybe just the G2?), so that's neat.
> Is WMR discontinued? "As of November 1, 2026 for consumers and November 1, 2027 for commercial customers, Windows Mixed Reality will no longer be available for download via the Mixed Reality Portal app, Windows Mixed Reality for SteamVR, and Steam VR beta, and we will discontinue support." Dec 27, 2023
The official deprecation was made in Dec 2023. Removing it in the next feature update is too soon. Lots of people still having working WMR units, so this feels super cheap.
Why did Microsoft give up on AR/VR just when Apple and Meta have shipped the fruits of their massive R&D investments and are spending to create the market?
Sure, Windows Mixed Reality wasn’t great. Maybe in the current marketplace it was the Windows 2.0 to Apple’s Mac. But the old Microsoft would have persisted when underestimated and come back with a sneakily good 3.0 product.
Because Microsoft doesn't get a 30% cut of products sold for it's platforms. Also Windows on ARM hasn't caught on and x86 has only recently been closing the performance per watt gap with ARM.
Meta's headset is Android based so it has 18+ years of low power, ARM performance and power tweaking. And Apple is using their in-house M series chips.
Microsoft's bread and butter is B2B; Azure, Office, etc. The Xbox and consumer divisions are just too small to take on high cost, novel devices. They've historically not been great at branding or marketing direct to consumer either. (Zune, those Microsoft store commercials, etc.).
Fundamentally those are excuses based on Microsoft’s current market position.
When they invested in Windows 1.0, their existing users were not asking for it and didn’t have a use for it. MS-DOS wasn’t a good foundation for a GUI, the IBM PC hardware was misaligned (e.g. terrible graphics), and obviously Apple was miles ahead.
The product didn’t fit any of Microsoft’s existing strengths and felt like a toy rather than a credible business tool. By the same logic that’s applied to AR/VR today, Microsoft absolutely should have exited the GUI market in 1988. (They were even a leading Mac software vendor! Why bother making their own worse GUI?)
It's ok to make short sighted investment decisions based on current market position, especially when one of your biggest bets from the last decade is striking gold in a big way, and you need all hands on deck to reap maximum strategic value during the temporary opportunity window.
In that phase, it makes absolute sense to reduce or cut anything that is not CURRENTLY performing well, in favor of the area that is taking off and needs all the push you can give it. When the gen AI market stabilizes, ideally with Microsoft in a dominant position, it will again be appropriate to diversify bets. That includes pushes for markets where MS can grab 2nd or even 3rd place.
The old Microsoft wasn't so myopic which resulted in not one but two of the biggest software dynasties with windows and office and in more recent times Azure.
You can't get a winner if you leave the game early.
> IBM PC hardware was misaligned (e.g. terrible graphics)
In 1985 it was terrible for games due to no sprites, but business applications run on the Hercules graphics card (monochrome but 720x348) or EGA, which had decent resolution and was way better than the competition for business applications. It had worse color resolution than say the Amiga, but by 1988 the VGA had fixed that as well.
How was it better for business? Text mode?There were plenty of terminals available at that time with better monochrome and color text modes and fonts, even line graphics support. It was about various attractive business software available for PC, not about EGA being superior somehow.
I recall in the late 2000s Microsoft had some of the best mice and keyboards on the market, but you'd never know they existed unless you happened upon them.
Gamers definitely did know. For a good part of the decade, the basic IntelliMouse Optical (2001 revision) was one of the very few devices that didn't have pointer assistance gimmicks that stand in your way, like angle snapping. It was sought after well into 2010s, despite it being limited to the measly 125Hz polling rate, having high SRAV and slippery sides.
Apple thinks it has a new and different approach which will be popular, Meta is hoping their current approach will lead to widespread popularity if they keep going at it, and Microsoft had an approach they tried but it didn't really work out and they don't want to throw their hat in the ring again quite yet.
Which of these decisions end up working out or not requires waiting 10+ years and looking back. Companies were hyped about 3D TVs, some held out longer than others, it turned out to be a bad idea. Companies were hyped about e-readers in the 90s but it didn't pan out at first only to be successful with future innovations and iterations and it turned out to be a good idea.
They spent a lot but they have about 20 million units sold which is pretty close to the latest Xbox generation. Content wise they have a lot of defacto exclusives. I wouldn't call that nothing.
I would call it less than nothing given it's not profitable[0][1] until there's evidence of pay off in future earnings from content and the operating losses are overcome.
Meta has sold far more VR headsets than all other the other brands combined. Unlike Apple's headset, they also have a clear use case, a "killer feature": VR games. Something you can't have without VR. Apple seems to just offer somewhat improved(?) experiences for watching TV and working with multiple monitors.
Yeah I 100% agree. No matter how impressive AVP is, it's still clearly worse than just a normal laptop/TV. You can do all the same things but just more painfully and at lower quality.
Quest games make much more sense because it's a completely different thing to non-VR games. You can't just say "well I'll play beatsabre on my laptop".
That said, they seem to be losing money hand over fist anyway.
Maybe read a few comments here and try to understand the reasoning instead of just riding on the hype train and automatically assume Apple will succeed in every endeavor.
Microsoft's strategy was a mess. They were pushing a lot of vaporware so they could imagine a world where Azure rendered XR frames and streamed it to headsets. Definitely a lot of thinking about how to charge for something and working backwards not meeting the actually cool but fledgling tech being built.
Teams integration was pushed and is a mess with or without XR. They finally launched something that should be able to connect web cam and XR users in a teams session but why... AFAICT it's in a special teams app. Apple's face time windows that are not a full overlay are a better fit for a productive experience. I haven't checked out the new Teams yet, though. Maybe its just not in the marketing material.
They had the first inkling of the exciting spatially anchored app stuff that the Vision Pro is pushing but it just was not fleshed out or really pursued beyond proof of concept.
MR for Windows was just not a good value add in general. It didn't supplant SteamVR or Oculus and Microsoft didn't (and doesn't) have a way to capitalize on their commoditization so what even is the point?
Right now, making XR apps is their only coherent strategy.
I never used WMR, but I never understood why it had to be part of the operating system. Why load that onto millions of computers when so few people use it?
IIRC, it was one of the things that Microsoft made difficult to remove. You could uninstall it through some PowerShell incantation, but chances are it would be back in a couple of Tuesdays.
I own two HP Reverb G2s and am not at all happy about this. I shouldn't be forced to choose between being able to use my hardware and getting security updates. This behavior on the part of Microsoft should be illegal.
They have a great relationship with Apple and Meta and have already brought their full suite of applications and services to their respective XR platforms. With Google/Samsung aiming for third place there just isn't room for another OS.
They developed MSFS 2020, which is the killer app for VR headsets, and announced a successor already. At least there seems to be some strategic misalignment.
They're display devices, they don't need their own operating system.
Microsoft's problem was actually that they did what you're talking about - Mixed Reality had all of these nonsense things microsoft built into it that didn't improve the experience and just put barriers in front of using the device to play games or whatever.
When you plugged in the headset it would automatically pop up mixed reality whether you wanted it to start or not. I couldn't keep mine plugged in and not running reliably. It would restart mixed reality at random times.
When you logged in instead of a well designed menu system, they made it stupidly spatial and you had this virtual house where functions were in different rooms you had to navigate to instead of just letting me start the application I wanted. It had a whole new set of widgets and interactions. A lot of designers got paid to create little toys that didn't improve the experience at all and just complicated things.
And microsoft's own device wasn't aimed at consumers and cost more than the vision pro, so you were buying HP or some other 3rd party's implementation microsoft's of mixed reality on windows.
It was fine to play games but not an improvement over steamvr.
On the brighter side, the Monado project is not too far away from a fully featured experience for WMR devices on Linux, and maybe eventually on Windows, too. Currently, the headsets even work with positional tracking and hand tracking.
Microsoft really has graveyard of dead projects similar to Google. Just to highlight
What I'm missing from these sites though is a short, snappy lesson and a link to a similar product that succeeded. Basically turn it into a counter argument lookup page.
Like this gravestones :
"
hololens, given up 2025.
Succesor: Apple vision pro.
Reason: better execution, endurance, clarity of vision.
Loss: whole ar/vr market"
Make the failure into a monolith throwing shade at incapable management, that left market on the table by half heartedly abandoning side battles. And if your MBA tech career has a gravestone in it and your decision was pro burrial, your career should be over.
> We are trying out a new experience for Copilot in Windows that helps showcase the ways that Copilot can accelerate and enhance your work. This experience will show when you copy text – since Copilot supports helpful actions that you can take with text content. In this scenario, the Copilot icon will change appearance and animate to indicate that Copilot can help (there are several different treatments so you may notice a different visual effect)
Full disclosure: I like AI. I also like my clipboard being mine…
Surely this will never go wrong with a password manager copying stuff to a clipboard.
Also they probably never tested that with pornographic search queries / clipboard contents, which will be very common in the real world. And even if they did test, they probably just did so with vanilla terms that are easily identified as pornographic by AI. Cue Copilot regurgitating porn queries because it mistakenly believes they are relevant to some current work task.
Personally I'm ready with popcorn for Microsoft unleashing chaos.
Somehow I doubt it would work correctly for my setup. My middle monitor has a KVM switch to switch from my Linux work/main setup (laptop in a docking station) to my gaming Windows tower computer (+a bunch of audio hardware to merge output and split mic input). I'm using barrier[1] to share mouse and keyboard between my Linux and Windows PC. My password manager runs on Linux because the Windows setup is bare-bones for obvious reasons, with barrier syncing the clipboard over the network. If I had to guess, that stuff is going to synced to the cloud and ran through AI by default, because Microsoft won't have the foresight to make cloud-sync opt-in as opposed to opt-out by the software.
Obviously I will make damn sure that Microsoft thing isn't running on my setup, but if they considered barrier/synergy users, I'll be surprised.
For Linux, Android based systems clipboard content can be marked as sensitive. For systems using X and Wayland the informal standard is to add an extra format to the clipboard contents with the MIME type "x-kde-passwordManagerHint" and have the contents of it be "secret" to signal that it is sensitive.
If these aren't being properly translated between systems when using barrier than that is a security issue with barrier itself.
Of course you're right, but bugs in third-party software using your APIs should not cause security issues in stuff you introduce later. Using whitelisting/opt-in vs blacklisting would solve this.
I already don't like how many operating systems retroactively dealt with passwords on the clipboard. "Sensitive" should've been the new default. Passwords aren't the only thing either. If I'm using some app or software that is either messaging, financial, or something medical, even if it just tracks my period, anything I copy originating from there should be treated as sensitive.
Soon you won't be able to write family members "hey, the results are in and I've got testicular cancer" without some data-kraken like Microsoft gobbling that up and feeding it to their AI.
Next thing you know some AI sneaks that into a company-wide mail "Quarterly Results" at your mother's place of work, because your mother has trouble using computers and is happy AI can write mails for her now. If that sounds unlikely to you, then to illustrate here's ChatGPT making that exact mistake:
I get that no every use is supported on Linux but at this point most everyday uses are. I don't understand how people don't at least seriously consider using Multiboot and Windows only were it is absolutely necessary.
Modern Windows is little more than an appliance with all it's modern day adware. It's closer to one of those Amazon buttons designed solely to get you to buy more stuff than a real computer where the user is in control.
Must be some kind of mass learned helplessness or something.
Multi-booting means stopping what I am doing and rebooting.
Unless I use two computers of course.
But since I can only use one computer at a time (for many practical definitions of "use"), I will probably use the computer that has all my tools on it.
And of course, Excel, etc. Which is why I went back to Windows after almost a decade of Linux as a daily driver. I am sure there are hills worth dying on, but Linux turned out not to be one of them for me.
Well there’s WSL to run Linux in Windows. It’s just a feature to turn on.
And before I started driving Linux daily, I ran Linux in VM’s for a couple of years to sandbox web browsers. This was before common browsers supported multiple identies.
But your question highlights why I went back to Windows. I finding solving many classes of ordinary problems easier than reading Archwiki.
Office Online doesn’t solve Photoshop, AutoCad, Ableton, utilities for configuring miscellaneous hardware, etc. etc.
Like I said, I drove Linux daily for close to a decade. I found and used work-arounds. Going to Windows removed a lot of “arounds” and lets me spend more mental cycles on the work.
Some contrast to this comment: this is steamVR data which is not only specific to steam (which as pointed out in the article, is not representative of the entire market) but also mostly biased towards gaming.
Now I will concede that VR has mostly been “successful” with that demographic (gaming), in fact, oculus (now quest) is mostly making money solely on that sort of content… some people at Meta while I was there would even argue that this product should limit its ambition to be the “Nintendo switch of VR”.
Apple on the contrary has been setting expectations much more clearly that gaming is not in the forefront (but not excluded).
Whether it will be successful is another story, but we should still welcome new challengers in an industry / product category.
What happened to the excitement in tech everyone :-)
oculus always wanted the walled garden approach, but steamvr had so much more content that users were unhappy that their vr display device wouldn't play steamvr games even though it technically was possible (in fact people hacked it in before it was officially enabled).
Now that oculus has made the device cheaper so more non pc gamers are playing I'm sure they're selling more software themselves. Valve also hasn't launched a headset in a while and wasn't targeting low budget headsets at all.
I'm not sure 6 million MAU is really "successful" when game consoles are approaching 400 million MAU and they're still mostly nerd boxes. Let's reserve "successful" for products that break 10 million MAU or perhaps 100 million MAU. A few hundred thousand devices, as Apple will sell this year, or a few million devices, as Meta has deployed successfully over the last decade, that's just not mainstream and not even close to any other Apple release in terms of sales or fashion factor (AVP is actually anti-fashion, a real departure for Apple.)
These statistics are complicated. They're looking at the evolution in percentage relative to a Steam's total numbers.
For instance if Steam had a sudden influx of new comers because of some standard popular franchise's game, the total VR usage could slowly grow yet still be declining in percentage relative to Steam as a whole.
Then users not relying on Steam (someone playing recrooms through Oculus Link for instance) also won't be counted. I have no idea how many people that would be, but VR is not that mainstream, so people doing that kind of stuff feels par for the course
The author speculates that PCVR users are switching to the PSVR2. It seems much more likely that they are switching to the standalone Quest headsets, which have far higher sales numbers and exclusive AAA games like Asgard's Wrath 2. The graphics aren't as good as on PC or PS5, but not everything is about polygons.
PSVR2 is an excellent headset, I’m not buying PCVR until those headsets are on the level of PSVR2. OLED or micro LED is a must, and foveated rendering is good for performance on pixel dense displays. Plus, inside out tracking instead of extra boxes to mount. The main problem is you can’t use it except on PS5.
It's probably subsidized by Sony, otherwise it is hard to explain how they ship it at $550.
Quest 3 (500 USD), from Meta, is definitely subsidized, given that their hardware bill alone is almost as high as the end price of the product [1] without any inclusion of development cost, and considering that the latest losses of Meta's Reality Labs exceed 4 billion USD [2].
So I don't think there will be anything soon in that price range that can compete with the headsets above. The Quest 3 seems the best deal currently, given that it has a lot of games and doesn't require a PS5 or even gaming PC.
Also with a quest you get the best from both worlds. If you're coming from pcvr you already have a suitable pc and can even play wirelessly. With the native games to boot. It's a really nice combo.
OP was arguing that it was decreasing. Steam is a bad source of info for a lot of reasons. Meta is the largest platform and they don't release numbers. My personal feeling (Apple Vision Pro hype not withstanding) as a VR developer is that the industry is in the slow growth phase and will continue to grow as real usecases are found and hardware continues to improve year over year.
Sure, the source isn't perfect but May 2021 still being higher than this peak is no more an argument it's actually growing in absence of other actual data nearly 10 years into the current cycle. Particularly since this is for the much more popular overall VR market, not just AR. That could always change in the future, but there is nothing beyond hopes and feelings suggesting it's already changing at the moment (not that those can't ever turn out to be right).
Typically anything in ${latest-steam-survey} is more unreliable than ${not-latest-steam-survey}. People often get hyped up on a big month change in e.g. Linux that ends up being a detection issue or the like.
thnx, there is "aces of thunder" dropping for PSVR2 in a quarter or two which should get that moving again, either way great to see the trend only accelerating.
For what it’s worth, my VR usage is greatly down because I used it for iRacing but grew frustrated with windows and switched to Linux where I can’t use it anymore for the time being. Blame anticheat.
The only thing I want PCVR for is simracing; I don’t particularly enjoy other games. I have a PSVR2 from when I planned to just use a console but now I use an IR head tracker in iRacing, which is like a poor man’s VR.
Too bad they won’t get anti cheat working in Linux, I really hate Windows too.
I use my WMR for sim racing as well as Elite: Dangerous, with a wheel and pedals for one and HOTAS for the other. It definitely was "good enough" for that purpose, and my IPD is within the spec to use it, it I really don't use it often enough to warrant upgrading anything.
They've been in the process of pulling out since before the Vision Pro was announced; HP was the canary in the coal mine, firing or moving around everyone they had working on WMR devices a year or two ago. When HP pulls out of a Microsoft initiative, it's over.
Yeah that. They enter every single market, deliver a half-assed broken product with dubious gains and expect to pick up business and when it doesn't they shelve it, even if the market is heading in that direction. At the same time they compromise their core business with the fads they pick up because attention is diverted. Can you imagine what it's going to be like getting all the unused AI crap out of windows and office when the inevitable incoming AI winter hits.
Dear MS, can you please just concentrate on your core business. We know it's dreary but we just want Windows, Office and AD that doesn't suck balls.
They meant "Mixed Reality" in general. The Apple Vision Pro has just launched. You can debate whether it will be succesful or not (please - do that in another discussion thread as it a topic I grow weary of) but the timing is worthy of note.
So the entire opinion of "Microsoft leaving as a field starts growing" rests on the Vision Pro coming out and therefore VR growing, but we should not question the fact that Vision Pro is unquestionable proof VR is growing in this thread...
No. VR is not growing. Apple was supposed to put out Vision Pro in 2021 when everyone was aboard the VR hypetrain, and AI was not in the public's mindset. Vision Pro is not the VR industry growing, it's an echo from yet another dead VR hypecycle.
Well the metadiscussion upends the whole discussion.
Microsoft should've wrapped up their HoloLens research and offered a nice low-volume product for researches, engineers, and designers. We need such products, and it's a shame they abandoned it.
But they keep trying to find the "next big thing". And VR is absolutely not the "next big thing". So if they only see value in that, they were right to cut losses and move on to focus their resources on other things like AI.
I rather agree than chasing mass-market VR (with "mass" defined as "the new smartphone") is a huge part of the problem.
However I do think VR potentially has a large niche to fill amoung consumers, hobbyists and various semi-professional roles. Something bigger than you're suggesting but smaller than say - consoles or smartphones.
Whether this is a niche big enough for Microsoft, I genuinely don't know. I think it probably is - but it would require aggregating several quite diverse market segments.
They were outcompeted in mobile and tablets. They had the most popular web browser for a good decade in the late 90s and early 00s, until they were outcompeted there too.
But how are they abandoning AI? They're a major part of the leading AI company.
They're not abandoning AI yet. They are however compromising it pretty heavily with overblown claims about how it's going to make people's lives better. They also have little to no profitability model around it and the development is operating at a heavy loss with the hope that they can break even with silicon advancements and user count down the line.
The thing is 99% of the people I know either played with their proposition and found it irritating or don't give a shit. They literally just want to be left alone and for people not to keep changing things. The machine is a means to an end and that end is well defined for them already. Only the tech and investment industries are gushing all over this and the latter are only riding the hype.
Typical MS this. Give it 5 years and read this comment again...
It's hard not to see it from the perspective of "for the person with a hammer, everything is a nail", Azure and already having the capability to build data centers is their hammer and the nail is a compute-heavy task. There's an element of speculation if whether it will be a lasting wave of the future people will pay for, but they have the levers to pull in case it is. It wouldn't be the first time they or any other company has made a bet so they don't get left out, time will tell if AI becomes a part of the furniture commodity or doesn't prove broadly valuable and fades away after a change of direction.
The interesting thing for me is the push for AI hardware as a baseline in consumer chips, right now I'd assume training and inference is done on their servers, but as it moves from a development/exploration phase to regular operations and scales up they want to offload inference to save costs, and probably provide a better experience running a model locally.
Well they failed miserably at that first thing. Azure's only leverage is really AD, InTune, O365 etc for corporates and then shovel everything else in there around the edges. Even in the MS house I work in, other than AD, all the product stuff is in AWS.
And the Azure hardware situation is so crap they had to buy in capacity from Oracle!
> They are however compromising it pretty heavily with overblown claims about how it's going to make people's lives better. They also have little to no profitability model around it and the development is operating at a heavy loss with the hope that they can break even with silicon advancements and user count down the line.
You've just described every AI company. AI is not even their core business, but MS is certainly in the best position out of any of the Big Tech companies to deliver a successful consumer product. Whether someone else will step in and steal their lunch, or they just fumble the product, only time will tell, but none of this is an indication that they're "abandoning AI" as GP mentioned. Quite the contrary―they're heavily invested in it.
The rest of your post is pure speculation, but I'll set a reminder for 5 years from now in case your crystal ball is legit.
Not every AI company. There are two approaches to solving problems. One is finding problems you didn't have to solve with the latest hammer. The other is having existing problems and solving them with the latest hammer. The subtle difference tends to bisect product outcomes accurately and thus can drive investment decisions. My crystal ball is mostly fairly well proven heuristics and financial modelling (used to be the day job). I could be wrong but the chance of a favourable outcome is less than one I'd put money on.
> ... overblown claims about how it's going to make people's lives better.
That sounds exactly like how they did Cortana too -> "It'll be GREAT!!!". Fast forward a few years and they're ripping it out of everything as it's actually not useful.
In retrospect, people like John Gruber and Tim Urban will mock the industry prior to the Vision Pro for not Getting It Right, or for fundamentally-flawed interaction models, ignoring where things were right. Eye-tracked selection was already a thing in the Sony headset. Tracking windows fixed throughout your environment has been a thing in the Oculus inside-out devices since release, good enough to keep track even on different floors. Varjo does reprojected AR better than Apple, for a cheaper price point (though still a steep one).
Microsoft, for their part, captured a sort of mundane magic:
https://www.youtube.com/watch?v=97mqPUn-4x0