Rift, Vive, and Virtual Reality

Well how often will they iterate?

If it's every year, not a big deal, in a couple of years, they'll source something better.

If it's meant to be one every 3 years or more, they better have a good product for the prices they're asking.


would still be weird to have partnered with Samsung and built gear vr for them to in turn get galaxy s4 screens which also htc got without partnering with Samsung.


I also would expect a CV2 in 2018 . They are going to want to cost reduce the cv1 to as large of a market share as possible I would imagine.
 
And there is more vr content coming

The VR streams will put viewers in center court seats at Madison Square Garden, let them eavesdrop on the player huddles and show them dramatic "under the basket" shots when those clutch threes hit nothing but net. Also, accompanying audio and graphics will detail stats, scores and updates, so you shouldn't have much reason to unmask once the action starts.
http://www.engadget.com/2016/03/08/fox-sports-big-east-vr/

VR footage, including the opening and closing ceremonies and select sports, will be available to viewers around the world
http://www.engadget.com/2016/03/08/2016-rio-olympics-8k-vr/
 
How are they handling VR audio? I'm guessing they're not as there's no means to encode audio spatially AFAIK. Hence you'll have a spherical video feed you can look around, but only a stereo audio feed with the mics fixed on the camera.
 
How are they handling VR audio? I'm guessing they're not as there's no means to encode audio spatially AFAIK. Hence you'll have a spherical video feed you can look around, but only a stereo audio feed with the mics fixed on the camera.

I don't know what kind of camera solution they are using so at least I cannot guess on the audio part. What I do know is nokia ozo has multiple microphones(8 cameras, 8 microphones) and provides 360 degree audio as one would expect. So it is conceivable proper 360 degree audio is a possibility.
 
Regardless of which oled generation oculus used, if it's true they made it custom but still went with pentile, it's a very stupid decision. No diffuser can correctly blend a pentile pattern, that could explain what was described as a "linen" pattern on CV1. I thought it might be an artifact of fresnel scattering, but diffuser+pentile is a much better theory. It also explains the "crisp" comments about PSVR.
 
At least based on interviews from oculus employees there was a trade-off to be made. Sharper picture on oculus would have come with more pronounced screen door effect. Oculus apparently went on to minimize screen door effect.

It will be very interesting times once the proper tear downs and reviews of various headsets come out. Probably all major manufacturers decided to make slightly different trade-offs on their headsets.
 
Well how often will they iterate?

If it's every year, not a big deal, in a couple of years, they'll source something better.

If it's meant to be one every 3 years or more, they better have a good product for the prices they're asking.

It's reasonable to expect 2-3 year release cycle from oculus. Sony probably much longer cycle as I would expect them to release 1 vr headset per console cycle. Valve based products like vive could be similar to mushrooms popping up here and there where random manufacturers decide to jump in the bandwagon.

I would expect sony and oculus to work hard on cost reducing current hw. We should see cheaper prices in future. I don't expect rift to be 599 around 2017 christmas as they should have reached reasonable scale on manufacturing and optimizing the manufacturing process.
 
I read weeks ago on reddit that HTC and oculus is using the same panel but they have different approach in the lense and with HTC's "Mura correction" to reduce the screen door effect (what the heck is Mura correction).

The fov also different for those 3 headsets with Psvr being the narrowest for vertical and horizontal. While rift and Vive is win in vertical or horizonal (forgot which is which)


Btw
Crap, I keep getting bad dreams about chased by zombies, dying, and have no money, and some I can't remember.

But all of them ends with me failed to get VR headset.

And this last few days, ends with I can't live past 15th.

What the he'll wrong with me. It's like when I'm about to face final test on my bachelor years ago..
 
Btw, Gear VR using Samsung S6 "2560*1440" (you have to make quotation marks with your fingers in the air when you read this), compared to Homido VR using iPhone 6 1334*750; the iPhone had slightly pixels, but the subpixels were non-discernible, Gear VR had the weird subpixel smeared screen door effect going on; with really vague colours. My coworkers tried it as well, and nobody believed me when I explained the iPhone 6 had the lower resolution. As a matter of fact, even the iPhone 6 owner said "no way my iPhone is 1334*750, it's Full HD!!" I explained that the plus model had that type of display, I had to show the wikipedia page. This situation made me realise something:

Rift and Vive will not be able to display a 24bit color gradient without severe banding, let alone being able to display all 24bit colours, at all

Here is the explanation:

zoom_pic.gif


Galaxy phones can display all colours, but only when viewed from a 'normal'/reading distance, if you blow up the display to the point where the subpixel layout and the black space in between is visible, like with Rift and Vive, then you cannot display full colours.
 
Guess my understanding of the English language is comparable with your level of understanding display technologies ;-)

The Gear VR had some trailers preloaded in the Samsung Cinema app, and the flesh tones looked really off, in my mind I was already making up excuses: the Samsung is rendering in 16 bit to keep the image inside the frame buffer and so on. But even 16bit colours should be able to recreate believable skin tones and what not. Maybe the screen itself was set to "samsung samoled ultra vivid color rendering plus"? I don't know. Can't wait to run some tests on the consumer models!

Btw can we make a thread for all HMD's, comparing display specifics and so on? I'm guessing all this info/speculation is derailing from real discussions like "the character in game x is wearing a helmet, VR is a helmet, game x is VR"
 
You've already, rather childishly, made this one thread a neogaf-style versus discussion. We really don't need a new thread to release yet another monster.
 
It would be so immersive of you play the game 'laying over a couch' with the position resembling a real motoGP driver! :D
Though with current games you have a lot of camera shake to simulate driving an actual motor, but in VR it could be nauseating because of some disconnect.
 
Anyone here with access to a recent Vive or Oculus model who is willing to take a picture with his phone through the HMD lens and display?
Ideally displaying an image displaying a gradient, either color or black to white.

it is in my understanding that even with the diffusion layer, colours can not be reconstructed properly. But I would love to be proven wrong, seeing as I am going to build a new PC once the next generation of PC hardware comes around. A vive will probably accompany that build :)
 
it is in my understanding that even with the diffusion layer, colours can not be reconstructed properly. But I would love to be proven wrong, seeing as I am going to build a new PC once the next generation of PC hardware comes around. A vive will probably accompany that build :)
Your understanding is wrong. The color accuracy is unrelated to the pixel structure, it's a calibration issue. If the manufacturer doesn't correct for the display's primaries not being identical to the source's intended gamut, it's the manufacturer's fault of not applying a LUT.

OLEDs have different primaries than rec709, this must be corrected either in the panel controller, or a software LUT. The best solution is a color filter in front of the OLEDs, which is what sony uses for it's professional monitors, it narrows the primaries, extends the gamut, and corrects for color shifts at an angle. In all cases, there's a LUT applied for accurate colors.
 
Last edited:
When pixels are viewed so close that the entire subpixel structure becomes visible, accurate color rendering will be impossible.
In a mobile phone, when viewed while holding it in your hand 1/2 arm length away from your eyes the colours will have perfect accuracy, you are correct about that.

However.. when zoomed in, the S4 screen (I just picked that by coincidence) which has perfect 100% RGB coverage is suddenly unable to accurately render the Google Chrome logo.

rgbvspentile.png
 
I've a low quality OLED screen in my S3 Mini. The colours look the same even held a few inches from my face. I can see the 'dithering'/subpixels, but the colours don't go screwy.
 
No diffuser can correctly blend a pentile pattern, that could explain what was described as a "linen" pattern on CV1.

I wonder if the reported linen pattern (which seemed to be reported most on the early crescent bay prototypes) might in fact be the "mura" that HTC was dealing with on the Vive and seen in many different iterations of amoled panels. Whatever diffuser they're using to hide the subpixel geometry wouldn't be enough to hide any vertical and horizontal contrast banding, so that might produce a subtle fuzzy linen weave look. Weird as it may seem, but I think I'm looking forward more to the Rift tear downs than I am actually using the thing at this point :p
 
I've a low quality OLED screen in my S3 Mini. The colours look the same even held a few inches from my face. I can see the 'dithering'/subpixels, but the colours don't go screwy.

So if you look at an image of a white surface, then moving it close to your face, being able to discern the subpixels. Would you still describe the color as white?
 
Back
Top