Comment by mo1ok
6 years ago
Let this be a lesson:
Working prototypes trump all theory.
I heard all silicon valley gurus stating they were "bearish on VR, bullish on "AR". This proliferated as a mantra throughout the industry. I thought they were wrong then, and believed the opposite - because I had a working VR headset that was awesome, but had only heard somewhat meh things about existing AR prototypes.
Until great AR hardware comes out, I'm still sticking with the same opinion.
I still think this write up by Michael Abrash (from 2012) is still the best argument about why AR is hard: http://blogs.valvesoftware.com/abrash/why-you-wont-see-hard-...
Though there has been some progress since then.
Interesting article. Here's Abrash's wry summary of the problems facing AR:
Leave aside the issues associated with tracking objects in the real world in order to know how to virtually modify and interact with them. Leave aside, too, the issues associated with tracking, processing, and rendering fast enough so that virtual objects stay glued in place relative to the real world. Forget about the fact that you can’t light and shadow virtual objects correctly unless you know the location and orientation of every real light source and object that affects the scene, which can’t be fully derived from head-mounted sensors. Pay no attention to the challenges of having a wide enough AR field of view so that it doesn’t seem like you’re looking through a porthole, of having a wide enough brightness range so that virtual images look right both at the beach and in a coal mine, of antialiasing virtual edges into the real world, and of doing all of the above with a hardware package that’s stylish enough to wear in public, ergonomic enough to wear all the time, and capable of running all day without a recharge. No, ignore all that, because it’s at least possible to imagine how they’d be solved, however challenging the engineering might be.
Fix all that, and the problem remains: how do you draw black?
What I find funny is that I believe he worked at Valve at the same time Jeri Ellsworth was there solving most of these issues.
Her new TiltFive system is "AR somewhere" rather than AR everywhere which allows it to provide a solid, practical, and affordable experience.
Here's a Tested review if you haven't looked into T5 before...
https://www.youtube.com/watch?v=Jse-GwkcYgI
Surprisingly enough all of the above is fairly well solved (or at least has the illusion to have been solved) in HoloLens and it turns out that not being able to draw black isn't a huge issue because AR is designed to interface with existing world and all virtual objects are ok to have grey instead of full black. If you were going to watch movie or play classic games, this would be an issue but that is not interfacing with the existing world and AR is not targetted for those use cases.
2 replies →
> Fix all that, and the problem remains: how do you draw black?
This concern is technologically narrow-sighted. We already have VR headsets with forward cameras built in. If the real world image is a projection too, you can draw whatever you want, including black.
1 reply →
Oof yeah. Crazy as it might sound, it seems like the easier path to a complete solution is some sort of neural lace interfacing with the optic nerve. We're obviously not there today, but maybe by the end of the century?
1 reply →
It depends on your definitions. You don't need a headset for pretty interesting AR. If I could just point a phone at things and get genuinely useful information as an overlay, I'd consider that a pretty decent AR application. Sure, the same thing in a pair of stylish glasses might be even better but it's not strictly necessary.
I see phone AR as a pointless gimmick. Like comparing a 1970s video game to PS4 video game. Yes both can be fun but pong is not really comparable to GTA5. Phone AR is so far off from Black Mirror AR. I can imagine every teenage girl spending all their time playing with their friends in AR, having their friends appear in their bedroom instead of just on Facetime. With AR glasses, some future version where they are no more intrusive than normal reading glasses, I can't imagine them not doing it. I can imagine all the youtube AR cooking classes will just project directly on your kitchen counter where you can either stand directly beside the chef or cover the same space, have your friends appear on the sofa next to you for facetime, etc... When it actually gets there it will be compelling in the extreme and non-geeks will flock to it like they did to iPhone. Until then it will stay in the realm of Apple Newton.
Why girls?
2 replies →
> If I could just point a phone at things and get genuinely useful information as an overlay, I'd consider that a pretty decent AR application.
Layar [0] was an attempt at that a decade ago on Android. Seems to be completely dead now though.
[0] https://www.wired.com/2009/10/layar-android-hands-on/
I think this is actually why it's so hard. I can experience AR today without buying an expensive and clunky piece of hardware.
If I couldn't get a demo, it's a cool enough concept that I might be tempted and they would get revenue and also the refinement that mass usage can help fuel.
Agreed. A reasonable model is that there's a really cool AR app that runs on a smartphone that you could imagine being cool^2 if it ran using a pair of stylish glasses instead.
But that doesn't really exist. There are some AR-ish apps like Google Translate that IMO fall into the better than nothing/sometimes useful category but there's certainly nothing in the "How did I live before this !?" camp.
Agreed. I think that Google/Apple have the right idea here. I am especially fond of Google's applications of AR in maps and translations as those use cases are both common and useful.
I've tried Hololens and it is honestly pretty amazing. The field of view really sucks but I'm sure they'll improve that, and it doesn't actually matter quite so much for AR.
But it is still a less compelling proposition than VR. The main market is games where seeing the real world is kind of pointless. Good VR is much more immersive, and being taken to another world is much cooler than seeing some floating planets or fish or whatever in an office (even though that is cool).
Gaming is like the lowest on my list of applications for AR... At the top is some kind of AR desktop environment that makes spinning up and managing arbitrary virtual monitors/windows a snap.
Hololens lets you do that. You can put windows on any walls. It's... cool. But not useful. Real screens are just much better. Nobody is going to pay much money to do something with AR that isn't particularly useful, and is better with existing technology.
They might pay money to do something that they literally can't do using any other technology, i.e. AR games.
Sounds like VR would be better for that
2 replies →
I think good AR is most important if it can be sufficiently miniaturized. I can totally see myself wearing AR glasses if they don’t look like I’m wearing Godzilla on my head.
Isn’t it the exact opposite? Only with AR can you turn everyday spaces, like a home or an office, into games... with VR you’d be bumping into walls and tripping over things
A large point of video games is escape. AR takes that away. Plus, most people’s spaces aren’t set up for a game. For example, I couldn’t play Minecraft: Earth in my apartment since there isn’t a good flat space.
No, VR lets you map a room so you can avoid walking into things. There are plenty of games that don't need a lot of space at all (the first Rift was a sit-down experience!)
Let’s be honest. Both are nothing more than niche.
What they are referring to more than likely is not the technology, but the business case. AR has more compelling / lucrative use cases than VR. While everyone is waiting for the killer Consumer VR app, Enterprise AR is blowing up.
> Enterprise AR is blowing up
citation?
The units i've used have worked reasonably well. but nobody is paying $2300 for them.
This partially comes from the SV all-or-nothing attitude: you can have AR on all the time (and thus revenue stream all the time), but you can't do that with VR.
Also the potential markets affected. VR really only affects the games & entertainment market, which is big, but best case you're revolutionizing Hollywood. AR is applicable to a wide variety of B2B markets - surgery, piloting, hazardous waste removal, firefighting, the military, mining, tourism, deep sea exploration, science, space colonization, etc. - which collectively have much more money spent on them. Your total addressable market is basically the amount of money you can capture if everything goes perfectly, which is dictated by the amount spent on substitutes. AR has many more substitutes than VR.
Same reason cryptocurrency is hot - it threatens the financial/insurance/ownership industry, which as a $13T behemoth is currently the biggest economic prize on earth.
The really big market being found for VR is in fitness, that could have a much larger impact (to general health even: than if it were just used for games and entertainment.
3 replies →
I think it's coming sooner than later but Magic Leap definitely coughed up their lead in the space and spent a ton of money in the process. Microsoft have an equivalent product and I think we'll see Apple Glasses in a year or so.
I think we will most likely see a true first generation AR device the next year or so. Apple Glasses will at the very least be 4 to 5 years out imo.