73 comments

[ 3.0 ms ] story [ 143 ms ] thread
Windows mixed reality was particularly half-hearted for a Microsoft effort or maybe it’s just another indication of how Microsoft and its hardware partners are at best frenemies. (Microsoft’s partners can never seem to quite say ‘no’ but they sure can sabotage anything new Microsoft tries to do.)

There were the hololens devices which were expensive and unobtainable to many, then there were the dollar store VR headsets and nothing in between. Meanwhile, Steam and Oculus were running more successful ‘stores’ of VR applications which somebody might want to use.

> Windows mixed reality was particularly half-hearted for a Microsoft effort

That’s saying something. Nearly everything Microsoft does is half-hearted.

The "dollar store" VR headsets at least matched the resolution or beat the resolution of the other headsets at the time. For example the Lenovo Explorer I bought is 2880x1440 while the Vive at the time was 2160x1200. It was also truly inside-out tracking so it didn't need lighthouses to operate. All you have to do to use it in a new space is look left, right, down, and up.

So cheaper by a good bit, higher resolution, didn't need lighthouses, far easier to set up for a space, simpler cable setup, and lighter. And it quickly got supported by SteamVR with most games having complete support for it.

I don't have the space for a dedicated run around in VR kind of thing, so I mostly use it for racing and flight simulators where it's absolutely perfect at.

VR in general is dying a slow death. “AI” has taken its place as the next hot thing for the 2020s. VR never demonstrated enough value for the general public. Sure there are some highly specialized niches where it excels, but it’s never gonna be the next iPhone that everyone must have to function in the world.

VR was the darling of the 2010s but it crashed and burned due to both physical limitation and lackluster technology. Microsoft throw their hat in the ring, and are now walking away as they see abysmal usage.

I don’t even know if Apple could save it. Of course there is the age old conundrum of “are people not using it because the implementation sucks, or because the concept sucks”. PDAs being the poster child of this.

Sounds like the typical cycle of new technology, that even LLMs will go through at one point.

- Thing is invented and people think it's cool

- Suddenly everyone wants this cool thing to be everywhere and do everything

- People start to realize the limitations and cool thing gets used in niches where it actually makes sense.

For VR, some people thought it'd replace gaming as a whole. Instead, it just made simulator games a lot better, but the rest is kind of gimmicky (said as someone who does enjoy VR a lot).

Similar thing will happen with LLMs and also the next cool thing.

LLMs are a little different.

The consumer buy in for language based interfaces is way less. ChatGPT is literally the fastest growing product of all time. It already generates a ton of value (more than VR ever did)

LLMs are more likely to become technology that everyone uses every single day (like phones.) even after it loses its hype.

VR will never get there.

Maybe mixed reality will, but I just don’t think most people will want screens on their eyes.

> LLMs are more likely to become technology that everyone uses every single day (like phones.) even after it loses its hype.

Smartphones/tablets is another good example where the industry first were amazed by it, then industry started to shove it down everywhere even when it didn't make sense ("you can perform all your daily work tasks on smartphones/tables now!", "Now you can code on your iPad!") and eventually it went back to its optimized use cases, instead of being used for everything.

Ah, and also, every proponent for these technologies when they are at the max of the hype, have a period of "Ah, but X is different than Y" while if you hang on for some years, you'll see the exact same cycle, even for X.

I’m not seeing this cycle for phones.

For example.The most popular web browser, by far, is chrome from android. (according to caniuse) This was not the case before the iPhone. Accessing the internet on the go is a game changing feature.

People use their phones every single day for a variety of tasks.

Phones had their hype phase (the whole there’s an app for that phase), but obviously they’re here to stay.

I think LLMs will be similar.

Unlike VR, which never hit that critical mass.

> For example.The most popular web browser, by far, is chrome from android. (according to caniuse) This was not the case before the iPhone. Accessing the internet on the go is a game changing feature.

I was talking about entire categories of products, or industries/sectors/niches, rather than competitors within one industry.

> Phones had their hype phase (the whole there’s an app for that phase), but obviously they’re here to stay.

Yes, but at one point everything was trying to be an app on the smartphone, while now it's been dialed back instead.

And they're here to stay, just like VR is here to stay, just within the niches it has found. Which may or may not expand in the future depending on capabilities/availability of experiences.

There’s some miscommunication here, sorry.

I’m saying that smart phones are pervasive in the daily life of at least a billion people.

I think that LLMs will eventually reach those numbers even after the hype phase ends. I think this based on the accessibility, utility, and ease of use of them. I think the growth of LLM products is a kind of indicator for this.

I don’t think VR will ever get to that scale. I don’t think Vr will disappear, but it’ll never see the same scale of usage.

I think this because of the physical and economic barriers to using VR and the fact that people generally don’t like having things strapped in front of their eyes.

I’m not really sure what you’re trying to say.

LLMs are a fad like cellphones are a fad.
Cool that I didn't say that either are fads then, so we agree :)
Ok, my lazy use of words. Substitute the word "fad" with "niche".

VR found a couple of very specific niches where it brought value. LLMs are already starting to pervade society in both good and bad ways and will likely continue on that trajectory. The obvious uses are many, and non-obvious ones are emerging.

If that's what you're saying (it does not appear you are), then we agree.

Sorry but VR will never die, its future potential is too enormous. It just has cycles of popularity and we might see a new surge with the Q3 being a very popular holiday gift this year.
> "Future potential .. enormous"

With foreseeable technology VR is always going to need something that covers large parts of your head, covering at least your eyes and ears. This is fine in specific cases like gaming, but it seems very unlikely to be something everyone uses all the time.

No matter how you spin it it's still a gimmick and will probably stay a gimmick for a very long time. I was present at the HTC vive release in LA, none of the promises realized since then. It's the same hype cycle since the 70s or so

Everyone I know who tried VR has 30 minutes of excitement and literally never mention it ever again.

> Everyone I know who tried VR has 30 minutes of excitement and literally never mention it ever again.

Well, except of they like simulation games like flight simulators. Then they surely have already migrated to a VR setup and won't ever go back. Yes, I'm one of those.

How long do you typically spend under the goggles? Back when I played Elite Dangerous in VR, 4..5hr sessions were completely exhausting. I since switched back to traditional head tracking for flight simming (although I have to admit that at least for helicopter landings the spatial awareness and depth perception in VR is extremely useful - but it's just too damn exhausting).
Not the OP, but I spend in FS2020 up to 5 hour flying with HP Reverb G2 (2048x2048 per eye) without any issues. Frankly, it feels like being in a real airplane and one can visually recognize all their real favorite spots without any issues. Losing this would be a major blow for me to run any simulators as it's just so much better and outclasses any 3D projections on the screen.
> How long do you typically spend under the goggles?

Max around 3 hours I think, that's the longest amount of (consecutive) free time I have available.

I remember in the beginning (First version of Oculus I think?) I could play for something like 20 minutes before being tired from it. Not sure if it's because of the HP Reverb G2 I have now, or because I got used to it.

What flight simulator game would consider existing VR resolutions good enough? Reading tiny dials on a virtual cockpit? Not really. Visual IFF in a combat sim? Not a snowball's chance. That leaves us with arcadey games that put everything you need conveniently in some not-sim-at-all HUD overlay, and perhaps some paragliding sim where the only instrument is the variometer and that is mostly read through audio cues. Now racing sims, that's a different beast: two large dials and huge demand for spacial awareness. I'd be surprised if screens where still around in that scene five years down the line.
I'm using a Pico 4 with Il-2, no assists. Dials are 100% a non-issue (headset resolution has been good enough to read them easily for years). Spotting and recognition are fine.
Is the Index that far behind? I basically bought it with IL2 as my main excuse (mostly hoping for some magic hack to repurpose lighthouse for desktop head tracking, but I never got around to adding a Vive tracker) and for me it was a clear not even close to acceptable. Tolerable for some quick "blow up some AI targets" but super far using a screen with my old webcam wannabe-trackir.
I've never used the Index so can't personally compare them. vr-compare.com shows it as having about 30% less pixel density, which is substantial; the pancake lenses of the Pico 4 also offer a large sweet spot. (The Reverb G2 is perhaps the most popular headset for Il-2, and sacrifices a little FoV for a little more pixel density than the Pico 4). The only annoyance with the Pico is that it's not really intended for PCVR, so you'll need to pay $10 for Virtual Desktop. But once set up I've found it to work well.
Trading FoV for density would certainly be the correct choice for flight sim, not much value in peripheral vision while scanning horizon or instruments.

Do you spend much time zoomed in, losing the angular 1:1 mapping of the viewport? I did that a lot with a headtracker (makes you really strive for precision, I had a custom build of freetrack with generous per-pixel hystheresis), but under the headset zooming felt more wrong than expected.

FoV is still very valuable for general situational awareness, IMO.

I spend very little time zoomed in. Mostly when trying to pick up a ground target. I agree it feels very wrong in VR.

I use a HP Reverb G2, with X-plane and Microsoft Flight Simulator. I can read the dials (yes, even the tiny ones) just fine with the built-in controls. But most of the stuff I do is programmed with keyboard shortcuts/my HOTAS flight controller.

I would not be able to go back to standard 2D-view-of-3D experience again after getting used to VR. Can't even effectively look around your shoulder, and a bunch of other things.

I still remember the time when airspeed and altimeter reading on the HUD on a Rift DK1 were literally just ":::" and "::::", they just were. But that's a long time ago.
You have to admit that the promise of true pass-through capabilities have been achieved with the Quest 3 (e.g., watch YouTube videos in a screen overlaying your vision while doing laundry), and that Apple's solution promises to finally make VR something that can work for general purpose computing.

Not that I don't agree with you, as someone who sold my Valve Index. I consider it to be a technology that takes a wild amount of development money and technology to basically take solved problems and put them in an interesting perspective in front of your face.

Also, I've seen reviews of spatial videos (Apple's marketing term for stereoscopic surround-y video capture) that describe them as eerily real, like you're living inside a memory. I think that may be an interesting use case (plus, unlike Apple's silly tech demo showing a dad capturing a video with a big VR headset strapped to his face while he watched his kids play, you can capture spatial video with a phone).

By that token we had viable PDAs (Palm Pilot) since the late 90s and it took 2 decades to get smartphones. The first consumer level VR headset (the Rift CV1 ?) launched in 2016, so we're still less than a decade into it.

Comparing the windows CE devices we had in the early 2000s and the Quest 3 for instance, I think the stages are not that dissimilar: they require enough effort and compromises that the general public would not care, but there's actual real world use cases and enough companies willing to invest in the field and bringing incremental and not so incremental evolutions.

You appear to be in violent agreement with the parent commenter. What everyone in this thread is saying is: VR might be big someday, but that won't be this year, or next year, or possibly even this decade or next decade.

To continue to use the PDA analogy, if you went back to the mid-90s and told people "these will be everywhere in 20 years!", people would still shrug their shoulders and go about their daily lives same as ever, because 20 years from now ain't today. And when it comes to VR, we're all shrugging our shoulders.

I agree with your general take on the time line, but people mid-90s were definitely seeing where it was going. The move from phones in cars, to pagers, to cell phones, feature phones, and win CE phones happened in about a decade and some. That's enough evolution speed to understand that the next turn will happen within a few years.

The same way I think everyone sees smart glasses coming. We just have no idea which company will crack the nut, what will be the twist, and on what timeframe.

It'd be kind nice to replace my set of 4k screens with something physically smaller.

But that would need AR rather than VR (excessive isolation from the real world is bad), and way better resolution than headsets currently have.

The goal of VR is not VR:

- Ads that you cannot look away from

- Ads everywhere you look

- Tracking face muscle as you see ads, to build a model of your reactions and target you more aggressively

- Having more digital interactions that can be spied on

- Having more metadata about those interactions

- Seeing how you react to objects in digital environments and then serving you ads of those objects

As usual, the goals are spying and advertising.

I’ve got a Meta Quest 3 and it is great. I am playing the pack-in game Asgard’s Wrath 2 and it is the first AAA game I’ve seen in VR and succeeds in every way as a single player game.

It’s the most interesting thing this year in the game business in the sense that there is nothing special about Xbox or Playstation right now, the Nintendo Switch is ancient, they don’t plan to upgrade the Steam Deck anytime soon, etc.

Outside of that, I am interested in experiences that are more like a theme park, say 3-d models of boulders on Mars inside a natural history museum and it is not hard to make that kind of thing with A-frame.

Seems foolish to make such predictions this close to the release of the Apple Vision Pro. Maybe you are right, but also maybe we will get that paradigm shift people have been waiting for.
Apple themselves haven't said a word about Vision Pro since the demo event. If $300 VR headsets can't make a dent in the public consciousness, a $3500 bauble from Apple won't do it either.

I remember all the talk about iPad Pro's AR-capable cameras. Do you know a anybody who actually uses an iPad for that?

How is Apple the world’s largest company when they bet on the wrong horse (AR/VR)?

Satya Nadella saw the writing on the wall many months ago and dissolved their VR/AR team and focused all their attention on LLM’s instead. ChatGPT is the fastest growing product in human history. It’s changing the world. Meanwhile, Apple has spent billions of dollars and some of their brightest employees on a product that is DOA.

People praise Tom Cook but there’s a clear difference between Satya Nadella and all the other Silicon Valley CEO’s.

There isn't a single horse. VR and LLM are very different techs with different applications.

On the other hand, each company differs on it's ability to deliver in a given space. Microsoft failing on VR is just that, Microsoft failed on a category of products. Just like Google and cloud, or support. I'm confident Apple could succeed in the space, and so well that it will make it trendy again.

Yeah that's like saying Smartphones will fail because Microsoft failed at it, when Apple clearly didn't.
MS spent billions of dollars and doesn't even own the product.
In my opinion, it is very healthy for tech companies to chase different gold rushes. Would look stupid if all of them chased the same flavour of the year like blockchain, crypto, self-driving cars, AI and so on. I’d rather see them dissect the new tech and apply it to their core products.

And your argument doesn’t hold strong when you step outside tech circles. Successfully selling AirPods to 75 year old grandmas looks more impressive to me than 25 year olds using ChatGPT. I understand it sounds extremely stupid, but both of the aforementioned tech giants are making generally smart moves. Will be interesting how eventually Microsoft applies OpenAI products into hardware, because they don’t really have much (compared to Apple and Google).

Microsoft failed at the portable music player, the smartphone (multiple times), arm systems and their current flagship console is selling fewer units than the Meta Quest line.

If you’re going by things that Microsoft walked away from, you’d have a long list of products that have been extremely successful for other companies.

Implementation and longevity matter.

I am convinced AR is one of the most promising technologies since the smart phone. Maybe even more so. But only if they can solve the hardware problems. VR is a novelty where AR will be everywhere.
Apple’s headset isn’t even available for purchase yet, and you’re already doing a retrospective. Give it a few years, then ask the question.
Just like Apple bet on the wrong horse by making a smartphone with no keyboard, making a smartwatch which is just a "toy/gimmick," making a tablet when Microsoft already tried and failed with Windows XP Tablet Edition, making wireless headphones nobody asked for and removing the headphone jack...all horrible decisions for Apple!

The only thing innovative about the ChatGPT business model seems to be selling compute cycles for below cost.

The difference between Tim Cook and the other Silicon Valley CEOs are that only Tim Cook runs the world's most valuable company.

Apple didn't make a misled decision, they were just one full hype cycle late to the game.
Indeed, more than Microsoft, I don't understand how can Meta could fail miserably in their business execution. VR could perfectly fail but launching Oculus headsets with basically no apps is not following basic business rules than companies such Sony and Nintendo know for decades.

Personally I was disappointed when I discovered than the Playstation console only works with PS VR. I understand Sony classic business play here but returning to the decades when an expensive computer devices couldn't interoperate with others makes me "cry". Just a rant but I love Thunderbolt and its high-end (e.g. CalDigit) docks.

Returning to the core topic, Microsoft is well known for its business success beyond all these failures.

FTC blocking Meta acquisitions of VR apps & the refusal to have apps for kids hasn’t helped them
Sony and Nintendo call independent studios.

Where is accurate information about the apps for kids limitations? Because there are apps for kids in the device and some few popular on Internet.

I’m not sure how prevalent it is, but two of my nephews asked me to get them VR headsets for Christmas. Although my siblings didn’t allow those requests to go through because of their ages, it sounds like their peers are getting into it. Curious how well Quest 3 is doing especially during holiday seasons.

I grabbed Quest 2 in 2021, but my peers and I might be aged out from VR demographics. Maybe I should’ve spent more time to get used to the controls and UX, but it just felt extremely clunky.

Formally is for 13 years old and older (if I remember well from the instructions). It is common to experience dizziness after using it and limit its use, it is not like a game console or computer. I wonder how Apple is working on this issue.

BTW, my sideproject using lasers and Oculus in the Miami Art Week 2022: https://youtu.be/46w2XLHd9vs?si=SkfV_7Gt-MT0Y4e6&t=58

Since when is Meta failing? They have something like 90% marketshare of a >$10 billion market with ~75 million users.

The big splash with the Quest 3 is the passthrough features. 7 of the top 20 Meta Quest apps are mixed reality apps within months of the product's launch.

I think we can't call the state of the market decided when true mixed reality is just now getting off the ground.

I think we also can't determine that the market is drying up for Meta when 2023's hardware sales drop happened in a year where basically all the available hardware on the market is growing long in the tooth. Harware sales always drop for old consoles - look at the last two years of Nintendo Switch sales.

IMO Sony doesn't have a clue about what they're doing with VR (or portable consoles for that matter) while Meta, Valve, and Apple seem to understand what their customers want.

As you said it is early to talk about the success of VR but the Meta execution has concrete business non-forced errors: not motivating other game studios enough or do AAA content in-house.
Meta's Reality Labs is losing multiple billions of dollars per quarter and revenue is dropping.
I don’t think Reality Labs did as well as they thought it would, but they built a $2 billion annual revenue business and can probably cut R&D once the technology is mature. At the end of it they’ll be in a duopoly with Apple with both companies controlling all the software royalties for the major popular AR platforms.
Is HP Reverb G2 going to be useless with Windows 12? Or will Microsoft still provide bridge to SteamVR?
> Microsoft says when kills off Windows Mixed Reality, the Windows Mixed Reality for SteamVR feature will be collateral damage. This is the feature that “allows users to run SteamVR experiences on Windows Mixed Reality immersive headsets,” which suggests that when Microsoft pulls the plug, you may lose access to that functionality unless you can find (or develop) third-party apps that allow the headsets to continue functioning.
Does this mean HoloLens is dead? HoloLens 3 was allegedly axed years ago already. I wonder what happens now to the enterprise customers.
I always wondered what the enterprise customers were using it for. Is there some secret killer app or 4D excel I'm missing out on?
> I always wondered what the enterprise customers were using it for.

There are some military training research projects that I came across in a previous life. These VR apps haven't supplanted traditional PC-based training sims, though, although these are often based on commercial game engines rather than custom sims.

Hm, there's an idea.

Those annual mandatory trainings from HR. Things like "how to not harass coworkers", but now in immersive 3D.

The demo apps that microsoft shows probably are what they're using it for. If you look at the microsoft store for/on the HoloLens you'll see a bunch of training and communication apps.
At an old workplace (Traditional engineering consultancy) we used it to show prospective and current customers walkthroughs of their projects.

Take a revit model, add some textures and you've got a great demo.

HoloLens is a different category but it’s also been almost exclusively B2B based since HoloLens 2, which usually means there is a different support structure. No idea on the current status of the project/product, however.

* I used to work at Microsoft and I now work at a Microsoft subsidiary but I have no insight here other than what has been publicly stated.

Didn't Microsoft lay off the Hololens team last year?
It would be nice if companies were required to make available the source code for drivers/firmware/etc if they wish to discontinue a piece of hardware that's still in use, so that the community can take over software support and keep things going until interest/use dies. Unfortunately, we're stuck with a reality where companies don't really need to consider the negative externalities of generating massive amounts of e-waste through such a decision, or even consider the fact their customers bought and should therefore own and have the right to keep using the product they paid good money for. And so companies get to kill perfectly good hardware whenever it becomes unprofitable to keep maintaining the software themselves.
I just recently finished reading about the history of the kinescope. This device was a camera that recorded live broadcasts from a studio monitor. It was meant to distribute live content quickly for playback outside of local TV networks.

Many different kinescopes were made. All of them had slightly different film. Almost no content from kinescopes exist anymore.

So I think "consider[ing] the negative externalities of generating massive amounts of e-waste" is a pretty old problem! (When DuMont, a major TV network, folded, they dumped all of theirs into the New York Bay!)

I bought an HP Reverb G2 and it was non-functional on arrival.

It had some kind of USB 3.0 pickiness/bugginess/not-workiness. At best I had it half-working in a crazy, glitchy state. I had a perfectly mainstream, working gaming PC with an RTX 3070, the third fastest Nvidia GPU available at the time.

I wonder how many refunds they ate on the thing.

I later owned the Valve Index, which was an overall great experience. However, PC VR isn't something that is enjoyable to play for long stretches and I think those who would argue that VR is a gimmick have some sort of valid argument.

You just needed a new USB cable. They shipped one with some under-dimensioned specs, causing all kinds of problems mostly on AMD platforms.
I was on AMD! Sadly, this information seemingly wasn’t available at the time. I’m not sure if I talked to HP support or simply did an RMA with the seller, and I think that was strangely someone like CDW who was a retail launch partner. (I bought it at launch)