VR at Mobile World Congress 2018


#1

New Vive Trackers 2.0 compatible with the Pimax 8K


#2

Varjo prototype resolution:

From Road to VR:
Hands-on: Varjo ‘Bionic Display’ Headset is a Promising Shortcut to Retina Resolution

From Real o Virtual:



#3

From RoadToVR: “At the boundary between the focus display and context display, there’s an imperfect transition between the high resolution area and the low resolution area, which looks like a blurry rectangular halo, but it was actually somewhat less jarring than I was expecting.”

So while it sounds like an awesome idea in theory (high resolution display in the center) I don’t see this become a huge success if they can’t solve that blurry boundary between the 2 screens.


#4


#5

I like the pimax approach better. High res Display and less res rendering. Might be worth a thought in the game engine pipelines to have resolution zones with in one viewport or multiple viewports which combine on one physical…
But it’s just a somewhat uneducated guess from looking at the pics …


#6

I don’t think so. The focused resolution is overwhelming.
It’s the first prototype, a v1. Real o Virtual send Harold, in charge of the news, and he was astonished. Today goes again.

But the price is between $5000 and $10000 with 110º (at this moment 90º 60Hz).


#7

From Road To VR:
Updated Vive Tracker With SteamVR 2.0 Tracking Shown at MWC 2018


#8

I’d have to see it in person. But currently that blurring effect seems highly disturbing. In the RoadToVr article they said they’re working on minimizing it though, hopefully they can do that. Would be an awesome proof of concept if they can get that going without noticeable blurring.


#9

Point is, no matter how much they perfect the transition from the hi-res to low-res screen parts, it will always remain an immersion-killer if you gaze at the low-res part at some stage during the experience and you cannot help but notice the significantly reduced resolution. Nothing really available to fix this - unless, of course, they would manage to make the hi-res projection highly mobile within the complete FOV, i.e. allowing for replacement to match the user‘s gaze. In other words, effectively mimicking foveated rendering. In that case, if they managed to make the hi-res low-res border less noticeable, it would seem like a perfect solution. But I have not read anything about such plans yet, would require 90 Hz speed in redirecting the projection, that‘s quite an ask - the more so if it requires hardware elements to be moved at this speed.


#10

well, at least they have plans:


but, as you wrote, nothing really concrete at the moment.


#11

Here’s a look at how they’re hoping to move the high res area; they aren’t actually moving the screen itself, just its apparent position, so while it won’t be easy, it may not be as difficult as people are imagining:


#12

I don’t like the idea of a moving part… any chance of mechanical failure in a compact space is too much in my opinion.

There is a reason why Hard-drives are sealed to prevent dust and particles from interfering.


#13

True, had missed that, so they are working on it. In theory at least. On issue of course would be the distortion where the moving lens turns into a greater angle compared to the static projector (anyone who has a beamer knows the issue). On the other side - if they calibrate it for the middle of the screen, where we will be gazing at say 80% of the time, and the more you look towards the side of your FOV, the more slight distortion creeps in - perhaps it will be completely acceptable, just a bit of blur, like we know it quite heavily with the sweet spot in VR HMDs nowadays. However, all of this needs to work seamlessly, and foveated rendering is still required at the same time to reduce the computing requirements.


#14

I think production tech is further than you might think - check the DLP projector technology, they are working with legions of movable micro-mirrors…


#15

I still don’t like it, and like I said it is my opinion. Unless it is a self repairing system I will always be concerned about moving parts; especially very tiny moving parts.


#16

AdHawk (the foveated rendering system) uses mirrors too.


#17

Agreed. Besides I really wonder if this would work for VR where millisecond latencies are of essence. Moving a mirror in a few milliseconds, that seems really challenging because of the forces needed to do that and the fact you don’t want to feel things inside the HMD moving while it’s on your head. I’d be surprised if they’d get that going for VR. Seems very challenging.


#18

It doesnt say anything about that in your link? Do you have some info on this? Are you talking about mirros that move?


#19

It’s a chip with one mirror in it.
From AdHawk web site.
“AdHawk has developed microsystems that are optimized to extract eye position thousands of times per second without any computational overhead.”
At 0:40:


#20

I think DLP is worth looking into, especially if it’s LED or laser based rather than a color wheel. I’m still using an old 50" Samsung HLT-5087 LED DLP television in my living room because of its response time, reliability and lack of any possibility of burn-in. I also never have to worry about any bulb burning out or dimming.

Input lag and projectors for gaming

In Eurogamer’s 2009 tests, they found that the point where most people felt lag became detrimental was 166 milliseconds, or ten frames, of total lag a 60 FPS signal. This includes input lag, but it also includes the native response time of a game, which can be significant. When total lag exceeded 166 milliseconds, test subjects noticed the delay in their actions and felt that it was a hindrance to their performance. Therefore, your entire system needs to be fast enough to stay under this ten-frame limit for total response time.

According to Eurogamer, console games running at 30 frames per second represent the worst-case scenario for input lag. If these games have a minimum lag of 100 ms, your projector needs to add less than 66 ms to the overall response time. The good news is that 66 milliseconds is an easy target, and there are plenty of projectors out there that will hit it – some of them quite affordable.

Our testing shows that many inexpensive DLP presentation projectors have about 33 milliseconds of input lag, and many inexpensive LCD projectors stay under 50 milliseconds.

I’ve seen numbers quoted around that say DLP is generally around the 20ms to 30ms range.

Alienware once made a prototype DLP widescreen curved monitor. It never went into production, but worked fine as a proof-of-concept. Shrunk down to HMD size, it might work.

Alienware DLP computer monitor