I owned one and this article is perfectly accurate. This guy hit the nail on the head on everything.
Passthrough is massively oversold, even if it's technically impressive. They set expectations way too high.
The comfort is the #1 reason I don't own a Vision Pro anymore. I feel exactly as the author put it, "relieved," when I take it off.
I feel significantly more disconnected from my family when I have it on because I have no easy way to share my content with them. A massively improved guest mode, better casting, or something else would go a long way here.
The eye tracking + tap is incredible, but Apple tried to shoehorn this into everything. It should have been the primary mode of interaction with a detailed/precision interaction when needed. Eye tracking + tap is simply not good enough for power user use cases. It was such a relief to go back to my Quest after Vision Pro because the controllers were so precise and easy to use.
And finally, I'll mention that the OS + standardization of UX is HUGE. The Quest feels like a crappy Chinese clone in comparison. Every single window has a completely different way of moving, adjusting, etc. Sometimes you click in the center, sometimes you click under, sometimes you can't move it at all. On the Vision Pro, everything is standardized. I'd love to see Meta fix this.
> I feel significantly more disconnected from my family when I have it on because I have no easy way to share my content with them. A massively improved guest mode, better casting, or something else would go a long way here.
This whole paragraph sounds downright dystopian.
“A family that shares content together, stays together”
If I had a family member who insisted on wearing a headset all the time I’d probably be calling them a glasshole (Vision Pro is so much more obnoxious than Google glass ever was)
You're right that we need to be aware of this because we can and WILL be exploited via these avenues, but for now at least, it's kinda the other way around - a family that stays together naturally wants to share content with each other.
"Hey, let's play a game!" What sort of game? "One we can play together" Welp looks like this multi-thousand-$$ toy is out then, lets go back to traditional games.
I also feel like the lack of precise/physical controls is going to make gaming a non-starter. Which is really the main reason I’m excited about VR in the first place.
Owning a Rift and a Quest 2 has pretty much convinced me I'll never be into VR. VR makes for incredible immersion but, for me, terrible gaming. Instead of putting the cool stuff from games into reality, it puts the worst parts of reality into my games. Instead of walking and crouching with a touch of a finger, now I have to walk and crouch. To look behind me I have to actually turn around. Instead of everything being in focus all the time, I have to look at things one by one. Not for me.
I would assume someone could easily integrate a controller for their app using the Game Controller framework [1]. So its not impossible. The only complaint I could see is there not being a standardize visionOS physical controller. But I assume that you could assume the PlayStation DualSense or SteelSeries Nimbus+[2] is probably what you should target controls for.
> I owned one and this article is perfectly accurate. This guy hit the nail on the head on everything.
Strangely enough, I don't own one, but just assumed this guy had hit the nail on the head with everything. Everything else I've read felt like it was pushing an agenda, this guy's writing just seemed like it was accurately weighing the pros and cons as he saw them.
I've been complaining that flutter is the next flash, meaning that by rendering their own UX instead of using the platform's they're basically making dead end apps. They'll claim devs just need to reship their apps with the latest version but that's means all old flutter apps are abandonware. It also ignores that the next platform they'll run into the same problem again.
agree very much here. apple - we're missing many intuitive gestures. lets ditch the cursor paradigm eye tracking is set to work on. in the real world we have proprioception that lets us accurately interact with the environment without looking at what we're doing.
>Passthrough is massively oversold, even if it's technically impressive. They set expectations way too high.
As a Vision Pro owner, kinda, but kinda not?
I have a really low-power glasses prescription (-1/-1.5) and I've come to the conclusion that everyone saying this has the priviledge of being born with perfect (irl) vision. It's intentionally a little bit out of focus to obscure the screen door effect, which I understand might be disorienting to people who've never had to wear glasses.
But taking that into account, it's incredibly clear. If you've used any other pass through, it's night and day. I will frequently put on my headset having forgotten that the cover is on, and when I take it off my brain genuinely reads that as uncovering a transparent lens. The slightly reduced HDR, slight desaturation, and almost imperceptible visual snow all added up still don't detract enough from that to say it is not close to "lifelike".
It's not perfect, but it's incredibly close - and probably better than a good chunk of the population's uncorrected vision.
I mean, the comparison images in the article almost make the point for me. The left image, when adjusted for brightness/contrast (which the human vision system does automatically absent other stimulus, e.g. when all you can see is the headset) has a greater resemblance to what a cloudy day looks like IRL compared to the completely oversaturated, amateurly HDR'd image on the right.
I’m -3.25 (or right around there) and I found the AVP pass through lackluster. Better than anything else I’d tried but I completely agree with the “oversold” point. The quality just isn’t all the way there yet and the blur when moving isn’t great.
The passthrough camera has fairly low resolution compared to the inner displays, so if it looks "lifelike" to you, it's probably time to consider getting your glass prescription updated. Comparing the VP to the Quest 3, it's a significant upgrade (and should be, for the 7x price), but neither of them is anywhere close to "lifelike."
I bought Vision Pro and used it as is for three weeks, then put it away for a couple weeks because of degree of difficulty for me, definitely NOT a techie.
On a whim I went to the optometrist, got examined and a prescription for glasses (I use over-the-counter 1.25x readers from CVS), then uploaded the Rx to Zeiss using their iPhone app, and ponied up $149 for their optical inserts.
Since the inserts arrived and I put them in, my experience overall with VP has been better: still way too difficult and confusing, but movies and TV shows, which were the only redeeming factors in my initial three weeks of use, seem really, really crisp and vivid, more so than initially.
Take these observations with a grain of salt, after all, it's a series of one 75-year-old retired neurosurgical anesthesiologist's experiences.
But if you've already dropped 4k, well, that's a lot of sunk cost to write off without trying everything possible to make it work.
> The only way to check progress over slow hotel wifi was to look at the progress circle, which is also the cancel button. I was aware that every time I checked progress, I was one finger-spasm away from cancelling the whole download.
oooof
> Every app on Quest has to reinvent how buttons work, how a scroll view works, how far away from the user the content should be etc.. and every app works differently.
kinda shocked there's not a Material Components equivalent for Quest. I guess it's designed like a game system (where custom menus are standard), but that's a reinforcing loop. As long as there's no standard component system, it'll continue to be game-centric.
Now I'm not sure what their army of designers were doing?
This is 101 of product design - put in place a standard UI Guidelines. This is particularly important for a 'platform' player like what Meta is aiming for (Ex. from other platform/OS - Windows UI Guidelines, Apple's Human Interface Guidelines, Google's Material Design, etc..)
There’s a game on my Quest - I don’t remember which one - where when you hold the controllers up close to your face it pops up the controls, with arrows pointing to each button.
That should be all but standardized, other than exceptions in games where that could be a problem.
It’s not so bad when you’re used to the system, but giving it to someone and trying to explain what button does what isn’t exactly fun. You usually end up needing to move their fingers for them, and playing a weird version of that scene from Ghost with my father-in-law isn’t my version of a good time.
> I was downloading Star Wars for my flight back to London. The only way to check progress over slow hotel wifi was to look at the progress circle, which is also the cancel button. I was aware that every time I checked progress, I was one finger-spasm away from cancelling the whole download
Oh god this.
The last 10 years of UX trends where everything has to be two or three different actions behind a single tap or tap and hold does NOT translate well onto AVP. Most iPad apps are nigh unusable, esp those with high information density. Vision is not quite as precise as a mouse cursor but it definitely not as imprecise as a tap, but only when the UI is predictable. Because where you are looking is also how you read, every informational field also needs to include either both the information and the action, or they need to be separated out.
So far there is no bigger culprit than the native Music app. The bottom of the player has like three nested buttons that all do different shit. The same place you look to see what song is playing is also a hidden progress/playback scrubber and also a shortcut to switch to the miniplayer if you happen to look at the eye catching album art icon directly next to the title of the song. It’s maddening.
Try hanging up a call on your smartphone and your counterpart hanged up a a fraction of a second before you. Now you call someone (pretty deep down) in your recent call list.
Or click "Connect to Bluetooth headset" in Windows. Congratulations, your headset just connected itself, so you now you disconnected it.
My favorite is watching TV and pausing a video in Netflix or whereever, then resuming and wanting the controls to disappear. If I 'cancel/return', they are removed. If I wait a while, they disappear by themselves. But if I 'cancel/return' the splitsecond the app removed them by itself, boom I just stopped the player. Just ignore input for a moment after the auto-remove ffs.
> Try hanging up a call on your smartphone and your counterpart hanged up a a fraction of a second before you. Now you call someone (pretty deep down) in your recent call list.
I do this daily and it infuriates me.
I am also a fast typer on both touch and physical. As a result I often type into an auto complete box (e.g. kagi, gmail, etc) and see the result I want come up midway through my type. As I'm going to click or tap it, I am often in the 200ms window where that result is replaced by another or the list is reordered in some way.
Dynamically loading webpages are also a plague of the last decade for the same reason. They load an initial layout, often interactable. Then as you're going to click or tap, they dynamically load other elements or content, reordering and moving things around the page!
If you don't have anything n̵i̵c̵e̵ ̵t̵o̵ ̵s̵a̵y̵,̵ ̵d̵o̵n̵'̵t̵ ̵s̵a̵y̵ ̵a̵n̵y̵t̵h̵i̵n̵g̵ ̵a̵t̵ ̵a̵l̵l̵ ready to load, don't load anything at all.
YES! I counted 5 discrete actions that I had to perform in sequence in order to turn off Vision Pro. Why not an On/Off switch? Or are they saving this for Gen 2?
The Apple Music team is off doing their own BS. The Music app is also trash in other platforms with terrible UX.
The Home tab in the music app shows exactly 1 tile of information on my 15 Pro Max. That's 3.6 million pixels and they couldn't be bothered to fit more than 1 item without scrolling.
One of the most important and key points I hope Apple iterates on in an upcoming version is some type of universal undo action.
It's immensely frustrating to be watching a video in the photos app and swipe a tiny bit wrong only to lose all current progress in the video even after you swipe back. Even scrubbing controls for long videos (1+ hours) can be super finnicky and it's tough to select accurately. The traditional scrubbing method of holding while moving vertically to scrub at slower speeds doesn't seem to work either.
As a result, I'm often catching myself in frustrating situations where I'd like to just __UNDO__ whatever action I previously took was. Either jumping around a video to the previous timestamp, or undoing the close safari tab button I didn't mean to select, or any number of other things I accidentally press due to the options being just a little too close to each other.
It's just a mild annoyance, but it's something that would massively improve my experience with using the device, as someone who has spent a LOT of time in it nearly every day since release.
But I've seen plenty of people being confused by the popup about undoing when they just as shake their phone a little bit for other reasons (moving around for example).
> Every app on Quest has to reinvent how buttons work, how a scroll view works, how far away from the user the content should be etc.. and every app works differently. On visionOS, all of this is handled by Apple, and every app looks and feels the same.
I remember the days when EVERY application in Windows had "File" in the upper left hand corner. You could make a good bet that "Edit" was next and "Preferences" was somewhere in there as well.
Being "good with computers" back then had a lot to do with knowing the standard layout of applications to help guide you through where things ought to be or, at least, could plausibly be found.
I applaud Apple for keeping this going in this modern world we live in of SAAS websites that each do their own thing.
> The biggest innovation with Vision Pro is visionOS. visionOS provides native app frameworks, so developers can build apps for it. That sounds ridiculously obvious, and yet its something Meta have failed to offer for years. Every app on Quest has to reinvent how buttons work, how a scroll view works, how far away from the user the content should be etc.. and every app works differently. On visionOS, all of this is handled by Apple, and every app looks and feels the same.
Meta does have standardized utilities for translating movement to touch/drag/etc. interactions on arbitrary virtual surfaces:
But it doesn't seem (AFAIK) to answer the other side of this, which is the UI design system so apps have a consistent look and feel. Which is perhaps more common coming from a game development perspective, but ever since the Mac OS shareware days, Apple's understood that it's empowering to a certain kind of developer if you make it easy/the default path for them to build experiences that match a standardized look and feel. I'm honestly surprised that Meta didn't at least make an optional SDK for this.
I think this is inevitable when you look at the original team at oculus, there were a lot of gaming folks (John Carmack was their CTO!) and I have no doubt that lives on in the culture today, for better or worse.
I doubt a UI toolkit or standard OS primitives were even on their radar.
From what I've heard, they've just recently elevated some really talented leaders who understand the interplay between the gaming world and line-of-business-app needs, so I'm hopeful they're moving in the right direction!
I agree that Meta has tried but their solution is really completely different than a more traditional OS UI. Meta provides low level convenience functions SDK for the various native app development frameworks (mostly Unity and Unreal) to access the hardware and facilitate common movements, but the actual UI implementation is up to the app author resulting in a helter skelter hodge podge of varying UI from app to app. Even in the first link you provided above, after introducing many alternate concepts then it shows screenshots of wildly varying interfaces across multiple apps.
With a total of 10 buttons on the controller it's a (usually) minor annoyance that one app chooses to use one button to choose an action and another chooses a different button. Sometimes it can be really frustrating to try a bunch of buttons and see what works. I've put quite a few people through the AR demo that shipped with the Quest 3 and noticed that VR-naive people vary a lot in how quickly they figure out the idea that you can use a controller to shoot a gun.
Because apps on Quest are universally games, and those pretty much never use standard controls. They don't even all use the same engine, so Meta would have to provide a UE SDK, a Unity SDK, etc.
I never felt this was remotely an issue with the Quest, and more than it is with desktop games.
It's annoying as hell for applications like Immersed and for media players. If every app had a different way to position windows and the UI was good in 50% of them it would be one thing but every app I've seen with floating windows is terribly awkward, although anybody can think up a number of alternatives that ought to work better.
That is completely psychotic blacking out single frame screenshots for copyright reasons. Copyright is truly out of control and just comically ridiculous at this point.
It's not blacking out on purpose, rather it's a side-effect of the DRM chain. The DRM content is rendered outside of the UI chain, so it cannot be captured as part of a normal screenshot or video. The same happens on macOS (and even windows) if you try to screen record while watching DRM L1 content such as Netflix, Apple TV+, etc.
Readit News
Passthrough is massively oversold, even if it's technically impressive. They set expectations way too high.
The comfort is the #1 reason I don't own a Vision Pro anymore. I feel exactly as the author put it, "relieved," when I take it off.
I feel significantly more disconnected from my family when I have it on because I have no easy way to share my content with them. A massively improved guest mode, better casting, or something else would go a long way here.
The eye tracking + tap is incredible, but Apple tried to shoehorn this into everything. It should have been the primary mode of interaction with a detailed/precision interaction when needed. Eye tracking + tap is simply not good enough for power user use cases. It was such a relief to go back to my Quest after Vision Pro because the controllers were so precise and easy to use.
And finally, I'll mention that the OS + standardization of UX is HUGE. The Quest feels like a crappy Chinese clone in comparison. Every single window has a completely different way of moving, adjusting, etc. Sometimes you click in the center, sometimes you click under, sometimes you can't move it at all. On the Vision Pro, everything is standardized. I'd love to see Meta fix this.
This whole paragraph sounds downright dystopian.
“A family that shares content together, stays together”
"Hey, let's play a game!" What sort of game? "One we can play together" Welp looks like this multi-thousand-$$ toy is out then, lets go back to traditional games.
[1] https://developer.apple.com/documentation/gamecontroller/
[2] https://www.apple.com/shop/smart-home/accessories/gaming
Strangely enough, I don't own one, but just assumed this guy had hit the nail on the head with everything. Everything else I've read felt like it was pushing an agenda, this guy's writing just seemed like it was accurately weighing the pros and cons as he saw them.
How about flutter apps like https://flutter.github.io/samples/web/material_3_demo/
I've been complaining that flutter is the next flash, meaning that by rendering their own UX instead of using the platform's they're basically making dead end apps. They'll claim devs just need to reship their apps with the latest version but that's means all old flutter apps are abandonware. It also ignores that the next platform they'll run into the same problem again.
As a Vision Pro owner, kinda, but kinda not?
I have a really low-power glasses prescription (-1/-1.5) and I've come to the conclusion that everyone saying this has the priviledge of being born with perfect (irl) vision. It's intentionally a little bit out of focus to obscure the screen door effect, which I understand might be disorienting to people who've never had to wear glasses.
But taking that into account, it's incredibly clear. If you've used any other pass through, it's night and day. I will frequently put on my headset having forgotten that the cover is on, and when I take it off my brain genuinely reads that as uncovering a transparent lens. The slightly reduced HDR, slight desaturation, and almost imperceptible visual snow all added up still don't detract enough from that to say it is not close to "lifelike".
It's not perfect, but it's incredibly close - and probably better than a good chunk of the population's uncorrected vision.
I mean, the comparison images in the article almost make the point for me. The left image, when adjusted for brightness/contrast (which the human vision system does automatically absent other stimulus, e.g. when all you can see is the headset) has a greater resemblance to what a cloudy day looks like IRL compared to the completely oversaturated, amateurly HDR'd image on the right.
On a whim I went to the optometrist, got examined and a prescription for glasses (I use over-the-counter 1.25x readers from CVS), then uploaded the Rx to Zeiss using their iPhone app, and ponied up $149 for their optical inserts.
Since the inserts arrived and I put them in, my experience overall with VP has been better: still way too difficult and confusing, but movies and TV shows, which were the only redeeming factors in my initial three weeks of use, seem really, really crisp and vivid, more so than initially.
Take these observations with a grain of salt, after all, it's a series of one 75-year-old retired neurosurgical anesthesiologist's experiences.
But if you've already dropped 4k, well, that's a lot of sunk cost to write off without trying everything possible to make it work.
At least that was my thinking.
Dead Comment
oooof
> Every app on Quest has to reinvent how buttons work, how a scroll view works, how far away from the user the content should be etc.. and every app works differently.
kinda shocked there's not a Material Components equivalent for Quest. I guess it's designed like a game system (where custom menus are standard), but that's a reinforcing loop. As long as there's no standard component system, it'll continue to be game-centric.
This is 101 of product design - put in place a standard UI Guidelines. This is particularly important for a 'platform' player like what Meta is aiming for (Ex. from other platform/OS - Windows UI Guidelines, Apple's Human Interface Guidelines, Google's Material Design, etc..)
I think there's an emphasis on shipping a feature over shipping a cohesive product. Shipping simple vs functional one.
and churn seems to be getting in the way of mature products.
That should be all but standardized, other than exceptions in games where that could be a problem.
It’s not so bad when you’re used to the system, but giving it to someone and trying to explain what button does what isn’t exactly fun. You usually end up needing to move their fingers for them, and playing a weird version of that scene from Ghost with my father-in-law isn’t my version of a good time.
Oh god this.
The last 10 years of UX trends where everything has to be two or three different actions behind a single tap or tap and hold does NOT translate well onto AVP. Most iPad apps are nigh unusable, esp those with high information density. Vision is not quite as precise as a mouse cursor but it definitely not as imprecise as a tap, but only when the UI is predictable. Because where you are looking is also how you read, every informational field also needs to include either both the information and the action, or they need to be separated out.
So far there is no bigger culprit than the native Music app. The bottom of the player has like three nested buttons that all do different shit. The same place you look to see what song is playing is also a hidden progress/playback scrubber and also a shortcut to switch to the miniplayer if you happen to look at the eye catching album art icon directly next to the title of the song. It’s maddening.
Or click "Connect to Bluetooth headset" in Windows. Congratulations, your headset just connected itself, so you now you disconnected it.
Just brilliant, overall.
I couldn't blame her. You touch ANYTHING, even the spam caller in recent calls, and it immediately calls.
why oh why can't there be a setting - on by default - confirm before dialing?
I do this daily and it infuriates me.
I am also a fast typer on both touch and physical. As a result I often type into an auto complete box (e.g. kagi, gmail, etc) and see the result I want come up midway through my type. As I'm going to click or tap it, I am often in the 200ms window where that result is replaced by another or the list is reordered in some way.
Dynamically loading webpages are also a plague of the last decade for the same reason. They load an initial layout, often interactable. Then as you're going to click or tap, they dynamically load other elements or content, reordering and moving things around the page!
If you don't have anything n̵i̵c̵e̵ ̵t̵o̵ ̵s̵a̵y̵,̵ ̵d̵o̵n̵'̵t̵ ̵s̵a̵y̵ ̵a̵n̵y̵t̵h̵i̵n̵g̵ ̵a̵t̵ ̵a̵l̵l̵ ready to load, don't load anything at all.
The Home tab in the music app shows exactly 1 tile of information on my 15 Pro Max. That's 3.6 million pixels and they couldn't be bothered to fit more than 1 item without scrolling.
It's immensely frustrating to be watching a video in the photos app and swipe a tiny bit wrong only to lose all current progress in the video even after you swipe back. Even scrubbing controls for long videos (1+ hours) can be super finnicky and it's tough to select accurately. The traditional scrubbing method of holding while moving vertically to scrub at slower speeds doesn't seem to work either.
As a result, I'm often catching myself in frustrating situations where I'd like to just __UNDO__ whatever action I previously took was. Either jumping around a video to the previous timestamp, or undoing the close safari tab button I didn't mean to select, or any number of other things I accidentally press due to the options being just a little too close to each other.
It's just a mild annoyance, but it's something that would massively improve my experience with using the device, as someone who has spent a LOT of time in it nearly every day since release.
I always disable this feature.
I remember the days when EVERY application in Windows had "File" in the upper left hand corner. You could make a good bet that "Edit" was next and "Preferences" was somewhere in there as well.
Being "good with computers" back then had a lot to do with knowing the standard layout of applications to help guide you through where things ought to be or, at least, could plausibly be found.
I applaud Apple for keeping this going in this modern world we live in of SAAS websites that each do their own thing.
Meta does have standardized utilities for translating movement to touch/drag/etc. interactions on arbitrary virtual surfaces:
https://developers.facebook.com/blog/post/2022/11/22/buildin...
https://developer.oculus.com/documentation/unity/unity-isdk-...
But it doesn't seem (AFAIK) to answer the other side of this, which is the UI design system so apps have a consistent look and feel. Which is perhaps more common coming from a game development perspective, but ever since the Mac OS shareware days, Apple's understood that it's empowering to a certain kind of developer if you make it easy/the default path for them to build experiences that match a standardized look and feel. I'm honestly surprised that Meta didn't at least make an optional SDK for this.
I doubt a UI toolkit or standard OS primitives were even on their radar.
But, historically, it was a misstep not to build these kinds of developer tools and design systems the moment https://forwork.meta.com/quest/business-subscription/ became a glimmer in their eye.
I never felt this was remotely an issue with the Quest, and more than it is with desktop games.
The page you linked to is full of links to Unity APIs / SDKs. So meta doesn't have them, Unity does.
1 - https://www.youtube.com/watch?v=1Jwo5qc78QU