Slashdot Mirror


Google and LG Unveil World's Highest-Resolution OLED On-Glass VR Display (androidauthority.com)

A couple months ago, Road to VR reported that Google and LG were planning to reveal the "world's highest-resolution OLED on-glass display" for virtual-reality headsets on May 22nd. Well, that day has arrived and the two companies unveiled that very display. Android Authority reports: As expected, the 4.3-inch OLED 18MP display has a resolution of 4,800 x 3,840. The display has a pixel density of 1,443PPI and a 120Hz refresh rate. Google and LG referred to it as the "world's highest-resolution OLED on-glass display." For comparison's sake, the HTC Vive has two 3.6-inch displays with resolutions of 1,200 x 1,080. The higher-end HTC Vive Pro has two 3.5-inch displays with resolutions of 1,600 x 1,440. The Vive Pro maxes out at 615PPI, making this new LG panel about 57% better than HTC's best offering. However, there's already one display that's better than anything on offer, and that's your own vision. A person with great vision sees in an estimated resolution of 9,600 x 9,000 with a PPI density of 2,183. In other words, this new display from Google and LG is about half as good as our own eyes. Unfortunately, there are no plans to use them in any consumer products yet. Google rep Carlin Verri told 9to5Google that the companies started this project to push the industry forward.

55 comments

  1. So I can get a 4.3" OLED display on a phone by Anonymous Coward · · Score: 1

    And at the other end of the spectrum I can get a 55" OLED TV.

    But if I want something somewhere in the middle, I'm SOL.

    1. Re:So I can get a 4.3" OLED display on a phone by viperidaenz · · Score: 1

      Samsung do a 9.7" tablet with an OLED display

    2. Re:So I can get a 4.3" OLED display on a phone by Anonymous Coward · · Score: 0

      I avoid OLED because they wear out much faster and are prone to burn in.

  2. Re:So I can get a 4.3" OLED display on a VR headse by Anonymous Coward · · Score: 0

    Er, not a phone. On a VR headset. My bad.

  3. Impressive by 110010001000 · · Score: 1

    Imagine all the advertisements that would fit in that!

  4. Please square all numbers by Anonymous Coward · · Score: 0

    about 57% better [...] about half as good

    A screen is two-dimensional, not 1-dimensional. Why make comparisons on 1 dimension only? The new screen has 8x more pixels than the 'best offering' making that number 800%. And even in 1 dimension, measured as PPI, this 57% is wrong, 1443:615 is not `57% better`, but way more.

    Next sentence it reads 'half as good as your own eyes' where it should read 'a quarter as good', since good eyes apparently see 4x more pixels as this new screen provides.

    I don't like inflating numbers at all, but credits are where credits are due.

    1. Re:Please square all numbers by Anonymous Coward · · Score: 0

      A 4.3-inch, 4,800 x 3,840 resolution display will need a microscope or magnifying glass to see anything (T&A) worth looking at.

    2. Re: Please square all numbers by Anonymous Coward · · Score: 0

      The idea is for vr headsets. It'd be 1 inch from your eye

  5. Re:So I can get a 4.3" OLED display on a VR headse by ShanghaiBill · · Score: 2

    Er, not a phone. On a VR headset. My bad.

    There are several OLED phones available. Also, tablets, laptops, etc.

    They have brighter colors, and darker blacks. They are also thinner and lighter. The main drawback is shorter lifetime.

  6. Eye don't think that's correct by Anonymous Coward · · Score: 0

    The human eye might have a PPI density of 2,183 at the center of vision, but it goes down by well over a factor of 20 off center. Also, our rod *(Black/White contrast) spatial resolution is higher by a factor of 120:1 to our color spatial resolution, but that doesn't fit what displays are optimized for.

    That said, I'd love a VR headset that makes virtual pages of text readable at a virtual arms length. Coupled with high quality prescription lenses, this opens up VR for real productivity and truly immersive applications, not just first person gaming and nausea inducing rollercoaster demos.

    1. Re:Eye don't think that's correct by Anonymous Coward · · Score: 0

      Coupled with high quality prescription lenses, this opens up VR for real productivity and truly immersive applications, not just first person gaming and nausea inducing rollercoaster demos.

      You should try out Farpoint on the PSVR. It's a nausea-inducing first person shooter and so far it's the only PSVR title that gives me nausea. It probably ignores the user-specific IPD configuration on the console and assumes everybody has 63mm IPD.

  7. Rubbish metric - IOW it is probably crap by Anonymous Coward · · Score: 0

    Stating the PPI of the display itself is pointless. So what's the PPI that the user actually perceives?

    I.e. how big the screen and at what distance does it appears when I put it on? If the 4800 horizontal pixels is going to look like 10ft wide and appear 6ft away from me, then the resolution would be crap compared to a real 4ft wide 4K TV at the same distance.

  8. Fix the links please Beau by Anonymous Coward · · Score: 0

    The first three links all go to the same slashdot story - even the headline, despite being labelled as androidauthority.

    1. Re:Fix the links please Beau by bill_mcgonigle · · Score: 2

      Should be this one: https://www.androidauthority.c...

      That links to the more interesting: https://onlinelibrary.wiley.co...

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
  9. Re:So I can get a 4.3" OLED display on a VR headse by Anonymous Coward · · Score: 0

    I guess I should have been explicit. I want a 32" OLED TV.

  10. Resolution is half the problem by mykepredko · · Score: 4, Interesting

    That's an amazing pixel count and density but what is going to be the method of generating those 3-D images and passing them from the computer into the headset?

    4,800 x 3840 x 120 x 32bits/pixel = 64 Gbps.

    I guess for practical operation, this would be a fibre optic connection but, as I understand it, current single mode fibre optic tops out at 10 Gbps which is a fraction of the speed required.

    So propeller heads and prognosticators, how will these VR headsets be connected to their controllers?

    1. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      This is where the other critical vr tech comes in, eye tracking.

      Our eyes are garbage in terms of resolving fine detail outside of 10 ish degrees from centre, so once reasonable tracking is available, you can render the centre of vision at high/full resolution, then drop that off drastically as you move outward.

      What they need to do with panels like this is team them with an fpga that can take *any* resolution input, with multiple variable resolution layers, and resample to the native/ideal pixels.

    2. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      SMF can carry 100 Gbps+

    3. Re:Resolution is half the problem by rtb61 · · Score: 2

      Size of display ie distance to the eye, governs that. Closer the fewer number of pixels but the smaller they need to be and the more complex the lens needs to be to fit it to your eye. A compact curved screen where that curve, together with lenses is match to your eye, correct fitting is going to be quite fussy. Why does this story feel like an ad.

      --
      Chaos - everything, everywhere, everywhen
    4. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      That's an amazing pixel count and density but what is going to be the method of generating those 3-D images and passing them from the computer into the headset?

      4,800 x 3840 x 120 x 32bits/pixel = 64 Gbps.

      Human optic nerve = 10 Mbps.
      How many orders of magnitude of waste is sufficient?

      current single mode fibre optic tops out at 10 Gbps which is a fraction of the speed required.

      HDMI currently offers 2x this bandwidth. Display port 3x.
      Next gen DP has capability of handling this.

      So propeller heads and prognosticators, how will these VR headsets be connected to their controllers?

      The answer is VR specific foveated display driver. With ideal configuration limit of vision with any detail is about 1000x1000 not the bullshit in TFA.

      Current generation of HDMI interface is way overkill for what is actually necessary.

    5. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      24bits/pixel and compression

    6. Re:Resolution is half the problem by bill_mcgonigle · · Score: 3, Informative

      So propeller heads and prognosticators, how will these VR headsets be connected to their controllers?

      The paper covers this ground nicely.

      https://onlinelibrary.wiley.co...

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    7. Re:Resolution is half the problem by ClarkMills · · Score: 1

      They don't need to generate the whole image full resolution, most of the image can be low-res blurred. It's just the fovea that needs to be painted hi-res. They will need eye tracking though...

    8. Re:Resolution is half the problem by DamnOregonian · · Score: 2

      current single mode fibre optic tops out at 10 Gbps

      That's a big negative. Common single-lane PHY rates of 10,25,50 and uncommonly- 100 (I think just Juniper, and a draft spec).
      Multi-lane (WDM) solutions are more common though, they're cheaper. Multi-lane 100Gbps is commodity now.

    9. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      Single mode fiber is currently limited at roughly 10000Gbps with DWDM multiplexing. Individual interfaces exist for 100Gbps over single mode fiber (using four colors). All of this is a bit pricey at the moment, but that will of course change rapidly. Also, these technologies are optimized for long-haul transmissions. Short range transmissions for VR are much simpler. Display Port can carry 26Gbps at the moment.

    10. Re:Resolution is half the problem by thegarbz · · Score: 1

      They will need eye tracking though...

      Yes and no. There's only limited amount of eye movement in VR in general thanks mostly to the thin lenses causing large amounts of CA outside of the sweet spot. Render the edges blurry and people will naturally move their heads rather than their eyes.

    11. Re:Resolution is half the problem by Namarrgon · · Score: 4, Informative

      As the original paper describes, they don't send the full 18 Mpix to the display - they use a foveated transport system, where the displayed image is a much lower pixel density (e.g. 1280 x 1600 pixels, upscaled to fit the display resolution) except for a small window (640 × 640) of high-density pixels located where the eye is actually looking (as determined by an eye-tracking system).

      They pack the high-density image data into a few extra scanlines of the low-density image, with a little metadata to describe where it should go, then send the resulting 1280 x 1922 image to the display, where an onboard microcontroller does the bilinear upscale of the low-density image and composites the high-density window in place.

      --
      Why would anyone engrave "Elbereth"?
    12. Re: Resolution is half the problem by Anonymous Coward · · Score: 0

      That is an incredibly stupid "solution".

      PROTIP: You are not a FPS avatar. You can move your eyes, head, torso and feet independently!
      Forcing people not to is bound to lead to problems. Apart from being so incredibly stupid.

    13. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      Yeahh...highest single channel (single wavelength, as in excluding multiplexing) that's readily available is over 25Gbps. Considering delivering pixels is a highly-parallel process multiplexing comes into play. We're talking 100Gbps easy on current hardware. Expensive, but available.

    14. Re:Resolution is half the problem by TomGreenhaw · · Score: 1

      You wouldn't transfer raw uncompressed data to the display. MPEG compression includes the capability of only sending a compressed stream of pixels that have changed from frame to frame dramatically reducing the bandwidth requirements. Additionally they may utilize short distance low power point to point RF chips an order of magnitude higher than 5Ghz where bandwidth will not be an issue at all.

      --
      Greed is the root of all evil.
    15. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      FTA:
      "Given the high-bandwidth of the resolution and data involved, Google and LG fabricated a custom driver integrated circuit (IC). Another trick to power this display is foveated rendering. This technique renders only what the user is currently looking at in high-resolution while the periphery fades out around that view."

    16. Re:Resolution is half the problem by iampiti · · Score: 1

      Several recent video standards (I don't recall if it's HDMI, DisplayPort or both) already use compression to be able to support big resolutions so that might help a little. I don't think that would be enough in this case though

    17. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      Is this a serious question?

      Six channels of LVDS at 12Gbps can do it. Or, in English, one and a half HDMI 2.1 48G cables.

    18. Re:Resolution is half the problem by Anonymous Coward · · Score: 0

      10Gbps max from fiber? Has this guy been on the moon the last ten years?

      Considering a top-end HDMI 2.1 cable can carry 48Gb/s it's not unreasonable to expect 64Gb/s and above from HDMI 3.0, although it might need a wider connector now rather than just faster lanes.

      It's certainly achievable in consumer electronics without going anywhere near photonic transmission if the vendor uses a custom cable assembly, which is definitely an option for PSVR 2.

    19. Re:Resolution is half the problem by DamnOregonian · · Score: 1

      Definitely. There is no need for fiber optic, and in fact, being you can do multi-lane (QAM) easier (cheaper) than you can over fiber, copper is a superior solution in this regime anyway. As long as you don't need to go long distance- copper is the way.

  11. Original paper by Namarrgon · · Score: 1

    Link.

    Karma whoring like I just don't care.

    --
    Why would anyone engrave "Elbereth"?
  12. Re:So I can get a 4.3" OLED display on a VR headse by ShanghaiBill · · Score: 1

    I guess I should have been explicit. I want a 32" OLED TV.

    You could buy a 65" and cover half the screen with black duct tape.

  13. Eye don't think that's correct-analog. by Anonymous Coward · · Score: 0

    Not to mention our entire visual system is ANALOG. So trying to fit digital terms to it will not work.

  14. Re:So I can get a 4.3" OLED display on a VR headse by mnemotronic · · Score: 1

    Hillary Hahn is not interested in politics.

    --
    The Russians have won. They have made the world a cesspool of distrust, greed, fear and hate.
  15. Re:So I can get a 4.3" OLED display on a VR headse by mnemotronic · · Score: 2

    You could buy a 65" and cover half the screen with black duct tape.

    Or use a magic marker to obscure every other pixel.
    Or wear an eye patch.
    Or tune it to "Fox" and leave it there. Half the IQ; half the truth; twice the number of ambulance chasing lawyer commercials.

    Ok. Enough wise-assities. I have to agree and say I want a 40-ish inch OLED. They say facebook nudes look really nice on OLED.

    --
    The Russians have won. They have made the world a cesspool of distrust, greed, fear and hate.
  16. this is tech reporting? by RhettLivingston · · Score: 3, Insightful

    Tech reporting has really, really gone downhill. And Slashdot editors chose to quote it?

    57% better than [615ppi]

    Let's see. That would be 615 * (1 + 0.57) = 966 ppi, not 1443 ppi.

    1443 ppi is (1443 / 615) - 1 = 135% better than 615 ppi. Apparently this author thinks that because 615 is about 43% of 1443 that the 1443 ppi display is 57% better. Wow. Percentages are taught in 5th grade math.

    there's already one display that's better than anything on offer, and that's your own vision

    WTF? The human eye is a display? Does the author maybe think that some people really can read other people's eyes? Like you look in there and there is a tiny little screen?

    1. Re:this is tech reporting? by nasch · · Score: 1

      The human eye is a display?

      That was so bad. Does that writer even think about what they've written?

    2. Re:this is tech reporting? by Anonymous Coward · · Score: 0

      What else to expect from millenals than a half assed, fucked up, job.

  17. Better than 'half as good as our own eyes' by ET3D · · Score: 1

    As Google says, a headset using this is expected to have a smaller field of view than the eyes have, so acuity will be 40 ppd vs. 60 ppd for the eyes (just 50% for the eyes). Also, the eyes don't have the same acuity over the entire range of angles.

    In short, this is close enough to what's needed.

  18. Bullshit. We can see at 36k*24k(?) at least! by Anonymous Coward · · Score: 0

    Actually, due to the Nyquisz frequency, probably twice that is needed!

    I've verified it myself. You can too, with your recular screen, by backing up, away from it, until you can't distinguist two parallel lines white of one pixel on black ground with at native resolution, for example. (Or whatever ability you want to measure.)
    Then you can take your distance, the screen's true DPI, and the angles of your field of view, and calculate your eye's best resolution from it.

    We can't see that well over the entire eye, but we can look everywhere in ouf FOV. So unless you attach it directly to the eye, that's what's needed.

    For me, that turned out to be the above mentioned 36k*24k (Or was it 28? I forgot the angle.)

    Ergo I'm waiting out for the 72k screen with 3,456 gigapixels. ;)

  19. Wow by robi5 · · Score: 1

    > However, there's already one display that's better than anything on offer, and that's your own vision.

    Since when is our vision a display?

    > A person with great vision sees in an estimated resolution of 9,600 x 9,000 with a PPI density of 2,183

    It's not measured this way. First of all I assume we talk about display resolution - which then gets focused onto the retina - rather than density of rods and cones on the retina which is way denser in the fovea. Second, the spatial, temporal and color resolution of human sight is hugely dependent on the part of the retina, so a matching display must either match (probably by a multiple of 2 at least) the foveal resolution, or foveated rendering must be implemented. Then let's not get started on color spectrum, temporal resolution and how all these depend on the actual input, eg. the eye can detect I believe a few photons in otherwise complete darkness, but isn't as sensitive when flooded with light.

  20. Re: So I can get a 4.3" OLED display on a VR heads by Anonymous Coward · · Score: 0

    I want a 32â OLED phone!!!

  21. Now what by Impy+the+Impiuos+Imp · · Score: 1

    What good does this do for porn, in the sense of "all the best skin is wrinkled", if all pictures are computer bot airbrushed to the texture of a balloon?

    Porn can no longer lead the bleeding edge!

    And now, to hit the anonymous check and submit...

    --
    (-1: Post disagrees with my already-settled worldview) is not a valid mod option.
    1. Re:Now what by Anonymous Coward · · Score: 0

      4K did it, really. Bringing fantasy closer to reallity but not in the direction you'd wish for.

  22. Re:So I can get a 4.3" OLED display on a VR headse by iampiti · · Score: 1

    Facebook nudes? What's that? I thought Facebook didn't allow nudes. Twitter on the other hand...

  23. Re:So I can get a 4.3" OLED display on a VR headse by kelemvor4 · · Score: 1

    We will be required on pain of imprisonment to watch Hillary (or her successor) win the "20/20" election on any display, glasses, screen, projector or display!

    Updated with the more likely scenario should that extremist regime come back to power.

  24. Re:So I can get a 4.3" OLED display on a VR headse by jtgd · · Score: 1

    I think you meant to say cover 3/4 of the screen.

    --
    J