Apple Reality Pro: Everything You Need to Know Snazzy Labs — March 25, 2023 32 comments Tweet on Twitter Share on Facebook Google+ Pinterest Apple’s releasing a mixed-reality headset this year. Let’s talk about what it is and why I think it’s Apple’s riskiest move since Steve … source apple apple ar Apple AR headset apple glass apple glasses apple reality apple reality one Apple reality pro apple rumors Apple VR apple vr headset AR Kit arkit Best VR Games google glass Google glasses iphone ar meta quest meta quest 2 meta quest pro mobile vr oculus Oculus Quest PCVR quest pro quinn nelson snazzy labs snazzyq Steam tim cook tim cook retirement Valve Index Tweet on Twitter Share on Facebook Google+ Pinterest
Getting a lot of comments about PS VR2 not being 4K per-eye, rather 2K. Here's the thing… it's neither; however, saying 4K is "more correct." Common resolutions we refer to (HD, FHD, QHD, UHD, etc.) almost exclusively refer to consumer 16:9 aspect ratios: 1920 x 1080, 2560 x 1440, 3840 x 2160, etc. Meanwhile, "2K" and "4K" reference the official cinema resolutions that use a 1.90:1 (full container) aspect ratio: 2048 x 1080 (2K), 4096 x 2160 (4K), etc. Today, we use 4K and UHD or 2K and QHD interchangeably even though they're not, technically. Consumer, UHD 4K comes up 160 horizontal pixels short of "4K" but it's still branded accordingly. The horizontal pixel count is inseparably linked to aspect ratio, but the vertical pixel count remains constant given a resolution. DCI-4K and 4K UHD are different resolutions with rather disparate pixel counts but they're both 2160 vertical pixels.
VR headsets typically have near 1:1 (squared) displays in each eye. So to say that the PS VR2 (with a resolution of 2040 x 2000) per-eye is literally 2K pixels in each direction, you'd be right. But EVERYBODY refers to "2K" resolution when talking about the HORIZONTAL resolution—not vertical which would either be 1080p or 1440p pixels. At 2000 vertical pixels, you're a lot closer to 2160p than you are to 1440p. Even when considering total pixel count, the PS VR2 still, has nearly 500K more total pixels than a 16:9 "2K" QHD display despite having half the number of horizontal pixels. Now it doesn't seem too fair to say the PS VR2 is 2K does it? But it's also not 4K.
The reality is, each eye has 2000p display—horizontal pixel count be damned. Put them side by side and you're pretty darn close to resolution and aspect ratio of DCI-4K—closer than you would be to UHD-4K. Does that make it "4K" per eye? No. But it also doesn't make it "2K" per eye. It's 4K—split between eyes. But I don't think that's gonna get past the marketing department…
Thank you. Very good summary to get up to speed on the state of apple's headset!
When I heard about this, my thoughts were "bro what?" but what you said works too
unless devs are provided with kits (and i don’t mean just a handful of major studios) easily (both financially and practically) there’s no way this will take off. without software, any hardware is just paper weight.
To launch this without controllers will be a death sentence. Making a $2k+ headset that’s NOT for gaming?! Where are the customers?
If these adapted to your prescription they would be game changers
This headset is going to be ass lol
Huh. The PICO 4 is using the SAME pancake lens and with a similar price tag with the quest 2. Which suggests that pancake lens is NOT that expensive .😅
When/If we get a Apple VR this year, it is only a matter of time where we see videos from every channel saying "I was wrong, the price is worth it", "You have iphone, I know you have money, go buy it. You get what you paid for".
I think it is likely this will mostly be a unit for developers to use in conjunction with the dev kit that will be launched in June.
the 0:58 insert was genius
Apple may also plan to recoup some losses on the early units in other ways — accessories, software, extra warranties, etc.
I got to mess with one of the first HTC Vives in college and was definitely blown away when it came out and I have a Quest 2 that’s a lot of fun
But until the industry gets it just right being light enough and helping midigate motion sickness I’m sure apple will help make that push but it’s going to be a ways off before everyone wants one
I think if this headset is equivalent or better than the Varjo XR3 in passthrough then even at the crazy price there will be lots of interest. Can't wait to see what people come up with once tracking and passthrough becomes so good that it is indistinguishable from reality.
Its Apple. Its gonna sell
Just buy a pc and a index
Honestly VR is legitimately used in very specific business or academic types of stuff, but hasn’t become wide spread even for gamers. It’s likely a super expensive business vr unit and is mostly meant for third party developers.
Why is it so hard to make a headset with a halo band, and all the components in the back? Like with NVGs. You don’t hang everything off the front of your helmet. You place the equally heavy battery pack in the back to balance it out. You need extra IR light? Sides of the helmet. Quest is like nah, we’re putting everything in the front and using an elastic strap.
Frenel lenses, in my opinion, are the single worst issue with consumer-grade VR headsets and the thing that has held them back the most. It ruins the experience in a bunch of different ways.
If you don't know S about a topic don't talk about or do your homework first.
AirPods are not a decade old
I think you’re spot on! I too have been racking my brains to think of what experiences this headset will provide and I came to the same conclusions; an excellent virtual workspace and great virtual telepresence. With love, Ken
im rly not interested so into vr glasses so much (nauseating half baked experiences) but i am always interested into good calculated thought-through critical video essays 😉
Pico 4 uses pancake lenses and is 400 dollars…
i mean for all the "crazy" tech described …. it still sounds like a dated headset design. A thinner quest 2 is really not groundbreaking in any way.
4k per eye is pointless ( and no, psvr2 is not 4k per eye, its the same resolution more or less as quest 2, less than index), the problem is that no hardware outside gpus like 4090 can render to that resolution, though it does help to reduce barrel distortion blurriness to upscale to 4k, its really not worth the insane performance hit if you ask me.
Anyway, if the leaks in this video are true then it will be a failure, people are waiting for the next big thing in VR, not a headsets that are tiny bit thinner for 5x the cost.
I think they need to develop a display that doesn’t need separate lenses.
Something like DLP mirrors or MEMS lenses
Hmm. Everything you said seems to be 2-4 years behind the current cutting edge research, besides maybe the displays and Apple Silicon for processing all the camera and sensor feeds. I doubt they'll release hot garbage, though. It'll probably be "decent given apple's approach to software.
I remember doing real time inside out tracking in 2014, it was jank but if you could fill the room with enough "high contrast objects" for location tracking assuming those objects stayed static, it worked nearly flawlessly if you didn't move too fast and this was in 2014 (update rate was a little slow). How has it taken 10 years for that technology to be miniaturized? Guess not much demand, but the real answer is simply because the base technology for VR can't possibly be good enough given the current manufacturing capabilities that humanity possesses.
Screen technology? LCD is too slow. Or more like was too slow. It's gotten good enough in the past 2 or so years. Just horrifically expensive if you want to minaturize it at the moment. Then the contrast issue. Dual layer LCD as you can't cram a ton of dimming zones in those tiny displays. Lots of synchronization headaches. Plus the stupid high power consumption. Partially viable, but I'd look for other easier to implement solutions. MicroLED is good enough. Just needs to be able to be manufactured with reasonable yields. Not happening anytime soon. OLED? Resolution borderline too low and power draw is high. Solvable problems. Micro OLED seems to fix the resolution issue and can actually be manufactured. Power draw? Cable. Pancake lenses? Eh, so so. Fresnel lenses? I'll take pancake lenses and their drawbacks every time, thanks.
Cabled or no cable? Batteries aren't good enough/hardware is too inefficient for the current things we want it to do. Cabled it is. 2 hours of battery is pathetic for anything but a tech demo, trust me. Can bump that up to 6 hours with bigger batteries but that gets heavy and you have to wear it on you somewhere. Plus those batteries tend to explode when thrown around or punctured which means even more weight for safety… Hmm solvable problems, but cabled is just plain easier for usability all around, as you don't move terribly large distances. Cables do have the obvious downsides of lugging around a long and easily tanglable cable though. Solutions exist but involves environment infrastructure which can get expensive, pulley systems, handoff systems, etc… At least that's my opinion on this as a PCVR user who has tried out the Quest 2, pico 4, and quest pro. I tend just to use them cabled at the end of the day. Valve index is still mostly king and that thing is 2+ years old. Overall, I'd say that's a 50/50 given current technology. I could see the wireless approach working fine.
Body tracking: Pure IMU tracking sucks (it's gotten better in the past year thanks to better IMUs, better estimation, and noise rejection algorithms. Who would have thought, huh? But the base limitations still apply). Camera tracking works well but stops working if it can't see you (IE dark environments/poor contrast/occlusion). Maybe some sort of infrared hybrid camera tracking like they use with motion capture suits? Sounds horrifically expensive. Lighthouse based tracking? (Basically cameras, but just the infrared part with way higher "resolution") Rather limited area, and instantly stops working the moment the lighthouse stops being able to see the tracker. Hybrid imu + lighthouses seems viable and probably what I would use if it existed and worked to solve the lighthouse tracking drawbacks. How does this not yet exist? Possibly just software? Of these, I would choose the hybrid camera based inside out tracking with a few points of tracking on your body, as it's the most robust (theoretically) and can track way more stuff simultaneously. Seems like that's what apple chose as well, but using just 1 point, the headset. I can bet you they'll eventually sell external trackers (and controllers) that wirelessly sync up to provide extra reference points for tracking (they probably use them internally for testing). But again sounds horrifically expensive in terms of both hardware and software costs, especially getting it to be inside out. The horsepower apple seems to be giving this thing backs up that assumption and my own experience. Requires basically ASIC processing to handle that in real time. An m2 chip might be enough so maybe this is a solved problem right now. Sort of doubt it though.
The only thing I can say is that this apple headset better be leaps and bounds better than the valve index with tundra trackers and lighthouses. Something my gut tells me just isn't true. It'll be better sure, but not like "woah, game changing!" Maybe the screens "look better" but looks better is just 1 part of the experience, isn't it? I wouldn't even say the quest pro is "better" all around than the valve index even in screen quality.
Regardless, I'll probably still shell out the multiple thousands of dollars on it. Lol. I like VR and have seen the potential, but it's a little demoralizing to see such slow progress for things that actually come to market. Lots of innovation occuring, just slowly.
Conclusion: Not good enough. Unlike GPT-4 in the AI space, even apple's attempt at VR isn't good enough. Not even apple can pull it off without strings attached, given current technology with the amount of turmoil that has occurred with the development of this headset. If engineers are saying it's not ready, then welp, it's not ready. Not because they can't make it, but because what they end up with "isn't good enough" to meet the lofty vision they likely share with me. My mid 2020s prediction I made when the Oculus quest originally launched seems to have been very wrong. Looks like early to mid 2030s at this point.
pancake lenses used also by pico 4
I absolutely do not agree with what you say at minute 21:35. The iPhone was already expected… and dreamed about, by those who followed Apple closely. When it arrived, it exceeded even the best expectations. At the time, only the dinosaurs on Wall Street didn't understand the product. Everyone else was rushing to buy the phone Steve had dreamed of.
The iPhone reinvented something that people could no longer live with: the mobile phone. In contrast, today I don't see anyone trying to use anything that might resemble the prehistory of a VR set in their daily lives. This dream of turning VR into something democratized (which started so many years ago) has proven to have many fundamental problems. Serious problems of lack of ergonomics that have kept people away from using it. With the exception perhaps of gamers.
To consider that it can be a success just because it has an Apple logo, is nothing but wishful thinking.
In my opinion, it will be another product for the "Valley of the Fallen" of failed Apple products.
pico 4 had pancake lenses before quest pro
I have ZERO (yes ZERO) excitement about this ever elusive, eventual, Apple product. They can leak it to the usual suspects all they want…my excitement level is still nil. My excitement level about Apple has been nil for the past half dozen years. Thank you Tim Cook. 👏👏 The "dent in the universe" is (certainly) closer to becoming a reality. So… now seriously, no hope for anything really revolutionary here.
By the way, in what way is this better than the aberration Meta is trying to develop?
Comments are closed.