John Carmack on Meta Horizon OS
twitter.com
twitter.com
On top of this, hardware partners have very different goals from the core OS developer (i.e, Google or Meta) who want to grow the ecosystem. So the hardware vendors add various features to their hardware and the OS devs need to add hacks to work around it. In theory, the OS developer writes a "compatibility" doc and requires hardware makers to follow it. In practice, the hardware maker fails to do so and the OS maker has to put in software hacks to work around hardware bugs. And then engineers on the OS team have to waste time chasing these bugs which affect a tiny portion of users but are high priority due to business contracts.
Some of those business contracts made no sense from a VR ecosystem perspective, but Google went through with them for other reasons such as preventing a key phone maker from jumping ship to another OS or company. That's probably what's happening here. Meta doesn't want third-party vendors to build VR solutions on top of AVP or Google's XR OS, so they're offering crumbs to distract other companies.
Since Meta can't win over google, i guess the next best thing to do is make sure meta store can be on other hardware since G isn't allowing that.
But none of the VR vendors was focused on user demands. Valve probably had the most user orientation, but their devices had still a huge cost barrier.
I had a Rift S (or still have) and I looked at how I could interface it for dabbling. Wasn't all bad, but I still lost interest immediately because of artificial vendor lock-in.
Other than a few games (Alyx, Beatsaber, maybe a few others), VR gaming is awful. "Productivity" software is even worse. I regret spending $1500 on my Quest Pro given that I've only used it for like a dozen hours.
VR Operating Systems and UI/UX concepts need to be re-thought from the ground up.
For a along time I had the idea to create some FPS rogue type game (because I was more interested in the mechanics to follow than the game itself) for VR, where your skill level in a specific aspect affected your perception of the world. High sword skill? While you're swinging a sword the world slows down. High block skill? Same when blocking with a shield. High evasion? When you move while there's a projectile/weapon swinging within a certain distance, same thing. You can apply the concept to many things.
I know I'm never going to have the time and attention to do it, but it would be cool to see the idea explored at least, and it's something it's hard to do unless you're actually doing a good approximation of the fidelity of actual movements of a person so the time time difference actually gives them a chance to take advantage of it.
I think people focus too much on the 3D aspects of VR and not enough on the interactions, and ways in which those interactions open up completely new gameplay mechanics, not just aping what people can do in real life now that you can see arms and hands.
It's no surprise to me that the most popular VR game is a rhythm game, because the latest hype cycle of VR is following the same trajectory as "novelty input device" arcade games whose most lasting hits were rhythm games too.
It seems that the group of people who a) are interested in physically moving their body around in their spare time, and b) prefer indoor environments to outdoor ones, and c) prefer virtualized spaces to physical spaces, and d) prefer the privacy of their own home to third places, and e) also have a home big enough to dedicate space to this hobby... is a relatively small group with a rather specific taste in games.
A much larger group of gamers already has an exhausting physical job, or a preexisting workout schedule, or prefers to go hiking in the woods, and so when they get back to their home to game they just want to plop down and zone out.
I'm not looking to see something change the industry, I'm looking for interesting and novel.
> It seems that the group of people who...
I don't disagree, but people are writing games for the platform. I just want to see people lean into the strengths of what they're targeting. Some of the most fun and interesting games for the wii did exactly that. Boom Blox for the Wii was extremely simple in concept and execution, and was also more fun than most Wii games I played.
> A much larger group of gamers already has an exhausting physical job, or a preexisting workout schedule, or prefers to go hiking in the woods, and so when they get back to their home to game they just want to plop down and zone out.
All the more reason to lean into new and novel ideas, because as I know, when your time for playing games is limited you're often going to stick with the known quantity that can provide enjoyment (which is likely some AAA title that is good enough) unless there's a very good hook.
Try Toy Trains VR!
It’s reeeeeaaally simple simulation-wise, like it says on the box - it’s mainly a toy, but it’s very cute
> games that play with size/perspective differences
Fisherman’s Tale!
They aren't interesting enough to buy hardware to use/play
They are novelties but not true game changers
The whole thing is just a bit "meh". If you dig deep enough, you can find some interesting stuff, but it feels so far less interesting than it could be.
Just look back at all the stuff that happened back when Oculus was still in charge, we had Oculus Medium and Quill for content creation, we had Oculus Story Studios, we had numerous good games released in short order and so on. It felt like VR was expanding, the last few years in contrast it felt like the thing was just shrinking down. And a lot of that was not by accident, but Meta's conscious effort to turn VR into "Metaverse" and move away from games. Which in turn was a flop, so Quest is back doing games, but it all feels very half-hearted.
VR has been hyped up since the 90s and when you look at what Quest delivers today, 35 later, it just doesn't feel like it's actually delivering on the potential of VR.
Yes, for the most part VR gaming is awful. Sometimes in a kinda fun way. I wonder if we'll look back with nostalgia on some of these weak games like we do with weird old turbografx platformers.
* The resolution in commodity headsets is just low enough that text is barely legible. Productivity apps become suddenly practical when you can render good-looking text.
* It's been years since the announcement of holographic lenses, which will remove the headache-inducing fixed-focal-length and pupil misalignment; significantly reduce headset size/weight; and increase the brightness and color gamut, and make the headset cooler, because the display backlights are replaced with lasers.
Most of the tech stack for SteamVR just needs to be semantically moved from "game engine library" to "HID & UI/UX framework".
To me, it's just a gaming console with mediocre games in the form of an awkward hat, and I don't like hats. The UI is definitely a place I could be won over. Maybe someone will do something undeniably revolutionary without drilling diodes into my head. Who knows? Until then, my wallet will certainly override any peaked interest, so it better be cheap enough to not have buyer's remorse like described.
I think people at Meta are aware of that and the Horizon-compatible devices will be less diverse than the ones from the Android ecosystem.
Still, I’m surprised to see Carmack taking a position against opening a platform. Although it sounds from his tweet that he thinks it’s more of a proprietary partnership than true open sourcing. If that’s the case, I see the argument for it being a distraction. But if the roadmap includes open sourcing the OS, then surely the “distraction” is worth it to capture the majority of the market.
2008: https://en.wikipedia.org/wiki/Android_(operating_system) 2010: https://en.wikipedia.org/wiki/Google_Nexus 2016: https://en.wikipedia.org/wiki/Pixel_(1st_generation)
The timing of this is probably just a manifestation of the f-you back to google from Zuck and Boz.
Android chose an “open” model and ended up failing on its primary goal as a mobile OS experience, even if some components of android went on to be popular for other applications. It led to chaotic device fragmentation, nearly every major Android vendor going bankrupt , and Google running effectively a charity unit (Pixel) to keep it alive. It only carries on with search engine subsidies.
Apple chose “closed” to the chagrin of hacker news “experts”. But the truth that “closed” iOS is a very lucrative, high-margin business.
“Open source” sounds nice, but as we’ve seen, many dedicated open source developers end up burned out and under-compensated.
“Open” can be a good thing, but is not a guarantee of producing positive outcomes.
I do think that stand alone VR is where it's at because it frees you while still being completely capable for PCVR so i'm hopeful some PCVR "first" headsets can join a program like this and deliver on stand alone while still keeping their bread and butter.
This is of course scary and why the VR market is now pretty much a monopoly. Perhaps the next versions of the vision pro will be lowered cost and have more games, but Zuck is just throwing money at each headset so how can groups like HTC compete? HTC and Valve never really had a chance when a headset that costs $300 to make is sold for $300.
Zuck and Carmack running victory laps now trying to Android-ize VR is probably not super surprising, but all of this show what happens when there's no real regulations to stop this kind of monopolization.
imho, Apple certainly saw this coming and fears a new Android-like competitor in a space they arguably could do well in. So the Vision Pro was pushed out before this got traction. Now its a matter of titan vs titan because smaller players are probably not going to enter this space anymore outside of hardware partners for Meta.
HTC never really iterated like Quest did beyond big bulky headsets that required a full room set up did they?
I can't say Meta spending 10s of millions to push the technology forward is monopolistic unless you want to say Nintendo, Sony and Microsoft consoles or any device with a walled garden (ipad/iphone) ecosystem is monopolistic.
[1] https://www.nbcnews.com/tech/tech-news/meta-stock-plunges-15...
Mobile gaming (iOS and Android) dwarves other gaming.
This is the issue that “gamers” have a hard time grappling with because they often disregard mobile games as an inferior product.
But then it leads to the fact that the mobile platforms don’t actually have to cater to their needs other than as halo products. They optimize for the majority of their customers and the majority are mobile gamers who are happy with the range of mobile games available.
Have you seen the type of mobile game which accrues the vast, vast majority of the revenue? They are literally designed as addiction engines first, and games a distant second. They're "games" in the same way that a baited hook is fish food.
The relative handful of actual good mobile games have often struggled, partly because Apple is very happy to promote and take a cut of the enormous revenue generated by this predatory business.
They have a different definition of gaming than we do. The games they're interested in are the ones with in-app purchases.
Except the design department, "good enough" is their motto for everything. "iPhone is the most popular gaming device" and "iPhone is the most popular camera" are two technically correct statements that don't sit right with me, but that's just me.
Apple literally pays developers of mobile games an up front premium to strip out all the Skinner Box nonsense and create an Apple Arcade version of their app.
Also regarding the Vision Pro, as long as Apple doesn't give up on it, it absolutely should come down in price over time. The original Macintosh retailed for $2,495, which is approximately the equivalent of $7,250 in today's dollars adjusted for inflation.
[1]: I'm mostly thinking about constrained space, weight and power delivery here. Obviously for proper VR there's a bit more that goes into it, but they're definitely not clueless.
Invest incthat, not VR headsets
But I was wrong: Zuck also said "if they (Google) are up for it." If Zuck has to resort to spitballing that idea out loud, I'd say it is far from happening any time soon.
That means HorizonOS is def not an Android equivalent to VisionOS. It's just AOSP plus the Horizon SDK.
Google should do a deal with Meta. But they guard Android with a very risk-averse approach.
VR isn't being "held back" at all. There's simply absolutely no demand for it. It solves no problem for which there is a mass market. It is the ultimate solution looking for a problem.
> ... Meta as a company, as well as the individual engineers, want the shine of making industry leading high-end gear.
Do they? I suspect this claim is colored by his own experiences at Oculus/Meta but it's not necessarily true. I suspect Meta would be giddy if they could sell 100x as many units of a cheaper unit because it means they would've found a product-market fit of some kind.
Maybe twitter was right about 140 chars.
Just give people a damn VR headset with the same compute flexibility as a Windows, macOS, or Linux desktop and do away with these stupid walled gardens.
The units would fly off the shelf because, whoa! You could actually do something with the damn devices.
That's why, a few generations ago at least, game consoles are so much more efficient. Consoles have fixed and specialized hardware, and developers can tune their games to take advantage of every single bit. No need to accommodate for different specs, no costly abstraction layers, no random tasks running in the background,...
http://web.archive.org/web/20240425211438/https://nitter.pri...
I'll repeat the usual sentiments of Glass being a failure, the Vive is nearly ten years old, the Quest is a VR Chat/Beat Saber machine, and the Apple Vision is a Black Mirror style immersive nightmare machine.
I want my ideas to be challenged on this, but I really believe that Horizon OS will be a "Did you know that Meta released a VR operating system?" fun fact in 10 years, probably when Apple releases a $5000 Vision Pro 4 Ultimate.
Who on Earth is using these things? I realize where I am posting, but who outside the tech world is getting excited about and actually buying/using VR Headsets?
Obviously I've been wrong before about tech trends but this one seems to be so blatantly companies sniffing their own farts in regards to "we are the future" sentiments.
I think this is the biggest hurdle: getting your VR legs nice and strong. It took me about two weeks of reasonable use before smooth locomotion was possible. I know other people that tried it once, felt sick, and were done with it. I think this is one of the reasons why there's such a big youngster population on Quest games; they don't seem to be affected as much.
The enterprise and general industry mindset is very different. These are already used for product design, medical procedures, training, vehicle development, and more.
Parents who want to buy a toy for their children, they are not exited about it they just find it cheap enough.
If you want proof just play any game/app that has voice chat on the quest. VR chat, among us, any shooter, ....
I really wish there was a way to filter them out.
What do I use them for? 360/stereo movies are incredibly cool. It is just another way of experiencing your personal history. Also there is Oculus Labs where they have some indi games and software which does not show up in the official Oculus Store. There are some gems, like some really cool games and some scientific applications, like a protein modeller.
I have also written VR programs by myself for scientific purposes (mainly biophysics) but also data mining and 3d CFD simulations. The 3rd dimension makes so much difference when you look at objects and you have a real feeling for the objects.
What I miss: Easily exporting 3d Models to VR (e.g. Blender), a good VR web browser. No Chrome is just the 2D version on a virtual screen. Not very impressive. Firefox VR is aready dead. And a good standard fiel format is still missing. VRML was quite nice in the 90s but hey that was 30 years ago.
just my 2 ct
He's at the stage of his life/career where he doesn't have to actually worry about a successful product. He's probably content to just hack on cool tech - regardless of outcome.
So, take his predictions for all things VR-related as wishful thinking. Maybe they'll become true, but probably not. He'll have a ton of fun either way.
Ignoring that, like covid test kits, if you make something (useful) available for actual free, there is unlimited demand and all stock disappears immediately.
This is locally bounded. My community can provide free dog-poo bags. I doubt one could do that in e.g. rural Alabama.
As for ads, so far the Quest is still ad free.
Other VR companies won't need to make their own OS, directly reducing the cost of producing the software for their headset. That's going to lead to cheaper headsets from those companies. Maybe not as cheap as Meta's, but still cheaper than current prices.
"VR is held back more by software than hardware. This initiative will be a drag on software development at Meta. Unquestionably. [It] will steal the focus of key developers that would be better spent improving the system."
You just said it's held back by software. Software is part of the system. Having a reusable VR OS frees up resources to focus on other parts of the software that build on top of it.
Meta isn't doing it out of the goodness of their heart... they want 3rd parties to use this OS, so that when apps are created for their headset, it'll work with Meta's. And since Meta gives away the hardware at cost, they'll win the race.
Carmack is so focused on the view of winning through depriving competitors of resources, he thinks there's tension when Meta executes a strategy that grows VR and puts them at the center, ensuring their long term success.
It is a bit ambiguous, but it sounds like they sell the Quest at hardware cost and take the loss in all software costs and R&D.
So it's irrelevant that other VR companies won't have the costs of building the OS, the headsets cannot be cheaper unless they optimize the production, which they are unlikely to do better than Meta (except perhaps for Chinese companies).