Choose Your Own VR: Onix VR Is Building A Modular HMD With A Curved 120Hz 4K Display

Status
Not open for further replies.
Hey thats pretty cool, dig the curved screen and it makes sense.

Not so sure about the "upgradeability" part however.

Tier'd versions of this product would probably make more sense.
 
Nice! I swear, I was suggesting this idea to someone the other day. Make VR HMDs like PCs...want inside-out tracking? Buy that module. Need beefier CPU/GPU? Ditch the mobile compute module and connect directly to the PC. Demand the highest resolution possible? Pop on the 4K display instead of the 1080p display. Etc.
 


I don't even think you'd be able to run it at all on anything available today.
That question was asked but I had to cut it because someone walked into the shot.
 
Even Nvidia will tell you that graphics technology (Including their latest 1080) are not powerful enough to crank out VR in 4K. It will be a while before that happens especially using high quality graphics in the VR app. You throw in that low res crap like half of the apps coming out these days then maybe.
 


I would have to see this as headroom and future proofing.
I am looking at a 4k hmd that is 60 right now just to watch video on because that would be a full 1080p per eye, good enough for video.
 
Actually, given that VR games are usually very "light" on graphics, this is not such an unthinkable idea. If we can run AAA games at 4k 60fps high/ultra with 1080sli, for example, I don't think it would be hard to do the same with many VR games.
 


Technically, 4k could work if it were combined with a fast eye-tracking system and foveated rendering. If the headset knows where you are looking, the simulation can render what appears at the center of your field of view using the headset's full resolution, while everything in your periphery could be rendered at a much lower resolution, then those two images could be softly blended together. That way, everything you look at can appear sharp, while the areas covering your peripheral vision, where your eyes can't see much detail anyway, can be left blurry.

As a simplified example, with a setup like that, offering around 2048x2048 pixels per eye, you might render a 1024x1024 area at full resolution centered on what the viewer is looking at, filling around one quarter of the display area. For the other three quarters of the display covering the peripheral vision, you might only need to render at a fraction of the display's resolution, perhaps equivalent to 512x512 pixels in total. Combined, your system would only need to render about the same number of pixels as the current Oculus Rift or HTC Vive, only the viewer would see roughly 3 times the pixel density. The eye-tracking tech already exists, and it seems fairly likely that we'll see it incorporated into the next-generation of headsets from companies like HTC and Oculus.

Of course, these guys didn't mention anything about eye tracking or foveated rendering. Without that kind of tech, I kind of doubt 4k would be a particularly usable feature by the end of next year. It would probably be a good idea for them to look into incorporating some kind of eye-tracking, since their headset could be a difficult sell if another company comes out with a competing 4k headset that can run fine on mid-range hardware.
 
For whoever interest, there is already 4K resolution HMD released some days before.
http://en.pimaxvr.com/
And it is very cheap because its made in China.
However, today's graphic cards still can't run any game nicely in 4k.
It also doesn't have special input devices like HTC / Occulus provides.
Some review say user really don't quite see the "pixels" in this high density screen.
I believe People mainly buy this to watch 3D po..em... 😀
 
For whoever interest, there is already 4K resolution HMD released some times before.
http://en.pimaxvr.com/
And it is very cheap because its made in China.
However, today's graphic cards still can't run any game nicely in 4k.
It also doesn't have special input devices like HTC / Occulus provides.
Some review say user really don't quite see the "pixels" in this high density screen.
I believe People mainly buy this to watch 3D po..em... 😀
 


porn is one reason, youtube 360 videos are another, and not everyone has a big monitor, or a big tv. You have to imagine vr as being able to sit you anywhere lets say in a movie theater to watch whatever you want that day. This for video would be amazeing because like I said, 1080p per eye and the screen is an amoled so better then 99.99% of peoples monitors and tvs.

Would love to get my hands one one to try it out, but would cost me at least 330$ for this and a dp 1.2 to hdmi 2.0 for something im not 100% sure on... not the kind of money i can spend on a whim.
 
Hopefully a curved display should make them feel lighter since the weight/mass will be closer to your face. Reducing inertial forces for head movement.

The Vive with one (1920x1200)*2 per eye is a total of 4,608,000 pixels compared to a single 4K is 8,294,400 pixels. This does not tell us how many of those pixels actually get used though as pixels that fall outside the field of vision are usually not rendered.
 
Curved display pointless?
I know what he said about "chromatic aberration" etc, but the Rift and others use a FLAT display then use LENSES.

His product still uses LENSES as well, but wants to start with a curved screen? Okay, but now he's going to need to have SOFTWARE to properly compensate for these changes. His company isn't very big and he's already introducing issues that need to be solved by OTHER PEOPLE who have no vested interest in doing so.

NVidia for example already has warping and other technology to properly process the image to look optimal (and perform optimal) through the Rift.

Other:
They want a consumer product a YEAR from now and haven't even got the details worked out for where to get the screen?

They want to mount a PC as well that can drive this 4K/120Hz HUD?

I can't see any way for them to succeed.
 


Actually the Rift and Vive both use a pair of screens that each have 1080x1200 resolution (not 1920x1200), so in total they have less than one-third the resolution of a single 4k display (2,592,000 pixels compared to around 8,294,400).

As for making the headset feel lighter, it's possible a curved display could help a bit, but you could just as easily position two individual screens at an angle for a similar effect. A single curved display might also make it difficult to adjust for IPD (the distance between pupils) which varies from person to person, since someone with a wider IPD would be closer to the edges of the screen, likely putting them out of focus, and the same would go for those with a narrow IPD. With two individual screens, or one flat screen, that's not a problem.

Also, as far as their mockup designs shown here go, the headset would probably be rather uncomfortable for extended use, since unlike all of the current major consumer headsets, there's no support on top of the head. The Rift and Vive each have a strap over the top, while the PSVR has a design that suspends the display from over the forehead, shifting much of the headset's weight off the face and onto the head. Despite the PSVR being the heaviest design, it's been widely said to be the most comfortable because of that. The renders for this headset don't show anything to transfer weight over the top, which might look nice in concept art, but probably wouldn't be very practical or ergonomic (the same goes for the white cloth, which would get discolored pretty easily). Going forward, I think we're going to see more designs closer to what Sony uses. The Windows 10 VR headsets that Microsoft revealed the other day use a very similar head mounting system to the PSVR, and I suspect Oculus and HTC will follow in the next versions of their devices.



I kind of agree. At least when Oculus announced their device, they already had working hardware to demonstrate, rather than just some CG renders. They were also announcing something that was relatively unique, since the only other consumer VR headsets at the time had narrow fields of view, imprecise tracking and practically no official support from game developers, and the VR field had been largely stagnating for many years. This guy, on the other hand, is coming into a market where a number of other, much larger companies are already developing similar hardware, and yet he's hoping to have a premium device intended to be better than the others available in a year. The timescale seems a bit too optimistic, and by the time such a product would likely launch, the specs would probably be nothing special compared to what other companies will be offering.
 
Status
Not open for further replies.