Hacker News Reader

Comment by rubatuga | Hacker News Reader

Seeing as this is an appropriate thread to shit on VR, why does nobody seem to care about the extreme lack of Field of View on these devices? Most Oculus Rift models are around 90 degrees, while humans have more than 180, probably closer to 200. Last time I tried VR, the huge black bars stopped me from thinking it was immersive. If the optics are a hurdle to cross, why not at least light up the edges with something reminiscent of Ambilight?

View on HN

People were prototyping ambilight type stuff back in the DK1 days. Trouble is, it doesn't improve immersion because objects disappearing out of view is still jarring. At least with a black edge you can imagine you're wearing a dive mask.

If you want to know what developments are or aren't going to find their way into future VR, track down Michael Abrash's talks from Oculus Connect and other conferences. Realistically much of it is over a 5-10 year timescale, but a lot of research has been done in the last 5 years. Many things that seemed like good ideas in 2012 turned out to be dead ends, and some surprisingly simple things have been discovered (like a "virtual nose" reduces VR-sickness, which nobody anticipated).

Remember PocketPC smartphones? That's about where we are in the VR tech S-curve. They were horrible devices when compared to today's smartphones, just as the VR/AR 5 years from now will make today's tech seem. But we probably still needed the PocketPC to pave the way for the iPhone.

It's worth remembering that we have had VR for well over 20 years now, in one form or another. 5 years seems optimistic given that.

We've had affordable, consumer-grade VR for 20 years? VR that, you know, actually feels like VR? This is coming off a bit snarky, I know, but I've seen this written all over the place and I have to say, the VR we have today is nothing like the VR of twenty or even ten years ago. Not from a consumer standpoint, anyways. I'm not an expert by any means, just some guy on the internet who is excited about the future of VR and has done a bit of reading on this topic.

Just like the Pocket PC "feels like an iPhone"? It's worth keeping the context of the conversation in mind.

The tech behind VR has been available for quite some time, the fact that it's still so limited (and has shown so little growth beyond the existing headsets) doesn't encourage "full surround with high resolutions in 5 years" mentalities.

The machine that makes the machine is what hints rapid developments are coming. That machine is the global smartphone component supply chain.

It's a very different environment than when VR was build upon the desktop PC component supply chain, which brought things like the automated bank machine and interactive kiosks.

But that "machine" isn't currently putting a lot of additional effort into moving from 200 DPI displays to 400 DPI displays (most have hit their local maxima of resolution for size); the thing that's sorely needed for a better VR experience. Without that R&D being done by the phone industry, the VR industry will once again stagnate.

I don't think it actually is worth remembering. That hardware was so completely different from the hardware we have now, in terms of performance and price. Those displays were somewhere around 240x320 pixels at 30hz. The year-old tech we have today is 1080x1200 at 90hz. That's roughly 50x better, a magnitude and a half better. That's not even beginning to consider the several orders of magnitude improvement in the visuals we can push over that display.

It also comes at a fraction of the cost. Four headsets plus a cluster of SGI workstations to run them would probably set you back a million dollars. Today, you're looking at about $10,000. 100x improvement in cost. Hell, quantity has a quality all its own.

So, without even considering graphics (or ubiquitous and wireless internet access), we're looking at modern VR being at east 500x better than the tech from 20 years ago. I don't know any comparisons to make that are 500x improvements outside of tech. I don't think we can really learn anything meaningful from the VR wave of the 90s. What we have today is just a completely different situation.

50x in 20 years. I think it's fair to say that another 10 or 20x is required (displays 5x as wide with 4x more pixels per inch), that would be another 10 or so years, not a mere 5. Especially when display tech is not showing a lot of effort in getting beyond the current 200dpi.

VR display tech has piggybacked off phone tech, and phones are not actively looking for even higher pixel density. This means specialized research & design into display tech; something I don't see or even hear about being done.

You may not see it or hear it but it's actively being worked on... That's not including companies like Magic Leap and Avegant whose entire business is founded on custom light field displays for AR.

Magic Leap, the company who has been repeatedly panned on HN for providing a cool demo idea, with no actual product or research?

The "2 million mirrors" in the Avegant Glyph is roughly equivalent to 1080p; about the same resolution as the Occulus and Vive. The tech is also an adaptation of the same DLP technology that's been in use in TVs and projectors for decades.

light fields are nothing at all like the 2D image emitted by DLP projectors.

This is a printed light field - https://www.youtube.com/watch?v=6AVAzGQMxEg

This is a light field plenoptic projector - https://www.youtube.com/watch?v=blb0TUBoZwA

That's true, but is kinda like saying we've had pen computing since the early 1960s - http://www.cadazz.com/cad-software-Sketchpad.htm.

Exponentials are never easy but I think we're about to see huge strides in VR/AR over the next 5 years, which will be built upon the only recently established global smartphone component supply chain. There were no commodity retina OLED screens even 10 years ago.

I think it's fair to say the tech has improved significantly, but from my perspective, today's VR is about as interesting a proposition as it was in the 1990's. Yeah, it's an interesting gimmick that would get old fast, and there's no killer app. In that respect, I don't think VR has changed at all in 20 years.

I don't know why you think nobody cares. It's pretty obvious it will get better as display tech improves, and there are already plenty of prototypes showing that is the case. We just don't think it's important enough of an issue to wait and stay out of the field until it gets better.

It is a tradeoff. The wider the FOV, the lower the amount of detail and the more visible screen-door effect is, assuming the same display tech. Higher resolution displays require more computing power. And that power could be invested into more convincing graphics and lower latency. Comfort matters too and better displays and optics can be bulkier and heavier. The final tradeoff is, of course, price.

I know there are plans for high FOV headsets. Both cheap and expensive, so wait and see. VR tech is too early for me to invest more than a cardboard.

>And that power could be invested into more convincing graphics and lower latency.

Fortunately, that problem is pretty close to being solved on a consumer level. A foveated rendering addon module for the Vive is expected to be shipping soon in China, by Q3 in North America [0].

For anyone unfamiliar with the term, foveated rendering means only rendering high-resolution images in your field of view. Basically, you don't see full detail across your entire field of view - only at the point where your eyes are focused. Everything outside that point can be rendered in lower resolution and you'd never notice the difference. It requires high-speed eye tracking to pull off, but it can greatly reduce the required horsepower to render a convincing VR environment [1].

[0] - https://uploadvr.com/7invensun-eye-tracker-for-vive/ [1] - https://www.youtube.com/watch?v=lNX0wCdD2LA

Reply on news.ycombinator.com