Samsung HMD Odyssey+ bursts through the screen door

Samsung's HMD Odyssey Windows Mixed Reality headset was already among the highest-end of its type, and the company is refining that head-mounted device with the Odyssey+. Most notably, this pair of goggles includes a technology that Samsung calls an Anti-Screen Door Effect Display.

While the Odyssey+ keeps its predecessor's OLED display panels with a 1440×1600 resolution per eye, the Anti-SDE display apparently places a diffusion (read: softening) filter of some sort over the pixel grid of those panels that's claimed to reduce the perceived space between pixels to a negligible level. Though the actual pixels-per-inch figure of this headset remains the same as the non-plussed Odyssey at 616 PPI, Samsung claims the new HMD has an "effective PPI" of 1233. No telling whether there are other visual artifacts from this diffusion grid, however.

Beyond that change, the Odyssey+ doesn't mess much with what worked in the company's first Windows Mixed Reality headset. A refined controller can act as a flashlight in the event you need to duck back into the real world and find something in your VR cave, but past that, the headset sticks with the AKG audio hardware, dedicated interpupillary distance adjustment, and inside-out tracking that Samsung introduced with its first Odyssey. No word on pricing, but the company says the Odyssey+ should be available starting today at the Microsoft Store and its own online storefront. I wouldn't imagine a new filter layer in the display stack would add much to the older Odyssey's $500 price tag.

Comments closed
    • End User
    • 1 year ago

    Anything that reduces/eliminates the screen-door effect is a step in the right direction.

    • Chrispy_
    • 1 year ago

    As a Vive and Oculus DK2 user, I feel that the problem with VR isn’t the Screen Door Effect, but that people confuse SDE with low resolution.

    My DK2 has terrible SDE, but it’s possible to ignore it. You’ll never not see it when you’re specifically looking for it, but just like the way your brain tunes out the fact that you can always see your nose, my brain just tunes out the fact that the Vive and Rift have visible SDE.

    The thing your brain can’t tune out is the resolution that’s too low to read text properly. Distant detail, HUD elements, a lot of text and inventory elements. Is that a target to shoot in the distance or is it just some aliasing artifact caused by the low resolution?

    When you think that the effective resolution of most HMDs is somewhere in the region of two megapixels it sounds okay, but that’s two megapixels spread over an effective field of view nearly six times larger than a typical 16:9 monitor with the usual 59° vertical FOV.

    Suddenly, trying to stretch the seemingly reasonable two megapixels so that they cover your entire field of view right out to peripheral vision results in a very low pixel density and the effective 16:9 window that you’re used to focusing on is actually only represented by a tiny little 720×480 block of pixels in the middle.

    Yeah, try running your favourite games at 720×480 and telling me that gameplay/UI/text/chat isn’t affected 😉

      • Voldenuit
      • 1 year ago

      I still play DOS games in 40 and 80 column mode, or 320×200 graphics, on DOSBox*.

      It’s the screen door effect that bugs me on my 2x1440x1600 Odyssey.

      * I’ve been watching a lot of IBM PC 5150 modding videos on youtube lately, and getting a bit of a retrocomputing itch.

      • DPete27
      • 1 year ago

      I think we can only physically go so far on the resolution front though. How many pixels can they cram into a 2″x2″ screen?

      Also, if games are STILL rendering both eyes/screens individually, that’s extremely prohibitive on hardware requirements. What happened to Nvidia’s “Simultaneous Multi-Projection” mojo they introduced on Pascal 2 years ago? Why render two screens worth of pixels when both screens are showing mostly the same image?
      With resolution comes cost. And cost is a BIG factor in home adoption. That 8k HMD sounds awesome until you realize you have to buy two $700 GPUs just to run the thing, not to mention the cost of the rest of the tower. What niche slice of the market has the $$ to pony up $3,500 or more for a gaming system?

      That leads me to my next challenge. Aside from a VERY small number of VR games, most of them are still demo quality. I can play a game for 20 minutes and I’ve exhausted the content. Time to move on to another $30-$60 game? Not only have I now bought an insanely expensive PC, cleared a whole room in my house, and bought this chorded HMD, but I’ll have paid for and played through every half-decent game there is on the platform in a few months. Then what?

        • Chrispy_
        • 1 year ago

        I don’t know how easy it is to do, but perhaps it’s possible to shape lenses for HMDs such that resolution isn’t linear across your whole field of view – so that rather than only 20% of the pixels being in the centre of your view, 60% of the pixels are in the center of your view, and the remaining pixels are in your peripheral vision.

        Hell, your peripheral vision is so poor that it’s monochrome and largely made of stuff your brain remembers was there the last time you looked that way. Obviously a non-linear lens would mean that the HMD would need a very distorted image fed to it to look correct through the lens, but I don’t think that’s compute-intensive since the existing HMDs all project distorted images anyway.

          • Voldenuit
          • 1 year ago

          To an extent, fresnel lenses already do that, and so does warping the square edges of a LCD/OLED screen into the circular FOV of a HMD.

          It’s definitely not as efficient as designing a custom metamaterial lens, but it’s a hella lot cheaper.

          • Waco
          • 1 year ago

          If we could somehow lock people’s eyes in place, this could work. Unfortunately you’d notice it RIGHT AWAY if you move your eyes around (which is one of the cool things about VR anyway, you can look around without moving your head).

            • Chrispy_
            • 1 year ago

            Yeah, there would be drawbacks to this approach – namely that the resolution gains in the centre of the FOV would be at the cost of resolution losses at the edge.

            I just feel that it’s an acceptable tradeoff.

            Yes, I do love that you can look around without moving your head, but realistically I’m looking within a reasonably narrow, central field of view about 80-90% of the time. I’d gladly trade in some of the peripheral quality for extra central quality.

            At the moment, with ‘linear’ lenses, if something is too low-resolution to see clearly, you’re SoL.
            With ‘foveated’ lenses, or whatever the term would be for them, at least you could turn your head to focus on something that was unclear in your peripheral vision.

            • Waco
            • 1 year ago

            Oh I don’t disagree, but something like that would bother the hell out of me. I’m waiting till ultra resolution becomes easier to drive.

        • Voldenuit
        • 1 year ago

        Pimax says 8K.

        We don’t have to render games at 8K, having the higher resolution screen should in theory reduce the screen door, although that is a bit of a brute force approach.

          • Chrispy_
          • 1 year ago

          An 8K HMD with Nvidia’s foveated rendering technique (even better if the foveated rendering was moved to follow where you look via eye-tracking) would be awesome.

          There was an article about that being in development a while back on here, shame we haven’t seen it in any HMDs on the market yet though.

        • Forge
        • 1 year ago

        Could people please stop using “chord” as a synonym for “cord”? The cord connects the device to the peripheral. A chord connects a note to the melody. Also, it’s a spinal cord, never a spinal chord.

          • DPete27
          • 1 year ago

          I always confuse them. Thanks for the correction.

        • psuedonymous
        • 1 year ago

        “How many pixels can they cram into a 2″x2″ screen?”

        With current technology but money-no-object? Scale up a microdisplay to that size (fab on Silicon rather than glass or plastic like commodity displays, using aligned multipatterning like with massive astronomy CCD arrays). eMagin have 2048×2048 microdisplays with a 0.3um pixel pitch. Over a 2″x2″ panel, that’s a 5480×5380 display. Using the 90mm panels in the Rift CV1 and HTC Vive, that would give 9680×9680.

        • demani
        • 1 year ago

        But that is assuming games and only games (or 3d rendered on the fly). But for 3d videos (I’m working with a company doing shark cage and other underwater videos) the resolution itself IS the biggest impediment. Its all pre-rendered, so its just playback.

        And even if the resolution is high you could always downscale the exact same way we did when we had video cards that couldn’t handle 1024×768/1920×1080/UHD etc. Render for a lower resolution and you shouldn’t have anything worse, but the display hardware can get the improvements in next year’s video cards.

      • tipoo
      • 1 year ago

      Now I can’t un-notice my nose. Jerk.

        • Chrispy_
        • 1 year ago

        You’re welcome 😀

      • Billstevens
      • 1 year ago

      I owned Oculus DK 1 and 2 and have had the final Rift since launch. The final Rift does not have any screen door effect worth talking about… they did fix this.

      What it has in its place can in some cases be more annoying. And that is a lens flare effect from the new lenses used to hide the screen door…

      However in many games and scenes this can become easy to ignore. Then we are just left with:

      1. A lack of AAA games ( I think the current crop of games gets old fast, but I haven’t played a lot of the newer titles so I am sure some owners would disagree.)
      2. Low resolution is limiting
      3. I suspect higher FOV would enhance immersion. Though I am not sure how much.

        • Chrispy_
        • 1 year ago

        I’m with you on the lack of AAA games.

        Lone Echo is probably my highlight of VR ownership so far, EVE Valkyrie was also up there and although I have many complaints about Elite: Dangerous, it’s still something I enjoy in VR.

        Elite suffers because of the resolution, the other two games are designed specifically for VR and therefore have limited text and massive fonts wherever it matters. Higher FOV would definitely help in all cases but it’s obviously going to come at a cost that even current HMDs and GPUs are struggling to pay.

    • enixenigma
    • 1 year ago

    Nice that they did something to combat SDE, but I am a little disappointed that they didn’t move to adopt the VirtualLink standard.

    • psuedonymous
    • 1 year ago

    [quote<]While the Odyssey+ keeps its predecessor's OLED display panels with a 1440x1600 resolution per eye, the Anti-SDE display apparently places a diffusion (read: softening) filter of some sort over the pixel grid of those panels that's claimed to reduce the perceived space between pixels to a negligible level. [/quote<]Sounds to me like whatever exclusivity agreement on the diffusion/diffraction filter applied to the (Samsung fabbed) Rift CV1 panels has expired and Samsung is free to apply it elsewhere. [quote<]Though the actual pixels-per-inch figure of this headset remains the same as the non-plussed Odyssey at 616 PPI, Samsung claims the new HMD has an "effective PPI" of 1233.[/quote<]Sounds to me like a load of utter bollocks cooked up by some marketing bellend.

      • Voldenuit
      • 1 year ago

      You can achieve the same effect by smearing petroleum jelly on your eyeballs before donning HMDs.

      • DavidC1
      • 1 year ago

      The people who disable anti-aliasing because of the way it blurs everything are going to hate this. Some are already not fans of Rift’s softening effect.

      Cheats are always going to be annoying.

      • Eggrenade
      • 1 year ago

      Pixels per inch is pretty useless for HMDs, with their fixed field of view. Pixels per steradian is the number I want!

    • danny e.
    • 1 year ago

    Anyone remember back when vr was the next big thing and oculus seemed promising?

    Then Facebook happened.
    No one cares.

      • Voldenuit
      • 1 year ago

      EVE Online makers found the same thing:

      [url<]https://www.techpowerup.com/248785/ccp-says-they-expected-the-vr-market-to-boom-much-sooner[/url<]

        • psuedonymous
        • 1 year ago

        Nothing at all to do with trying to build a MOBA for a platform which even the most optimistic estimates could jot sustain it. Or that at Eve Valkyries was just [i<]not very fun[/i<]. And, for those who have not tried it, has possibly the worst menu system ever implemented.

      • Puiucs
      • 1 year ago

      Facebook had nothing to do with it not getting traction early on. It was 100% the extremely high price. The 200$ Oculus Go proved just how successful VR can be if you combine a good product with the correct price as it sold in just its launch quarter more than Oculus Rift did in half a year.

      Once something similar to the Oculus Go reaches a price of around 120-130$ adoption rates will rise sharply. I want to see something better than the Oculus Quest at 200$ by the end of 2019.

        • DavidC1
        • 1 year ago

        I’m not as optimistic as you are.

        Most people view technology advancements the way Moore’s Law allowed blindingly rapid pace of development on microprocessors.

        Literally everything else is not microprocessors. The advancements elsewhere takes hard work, vision, and proper planning.

        There are things that make a headset that has a fixed price, or has very, very little room for improvement. Between that and increasing greed among mega corporations such quality headset at that price is going to take long time.

        Even Moore’s Law is nearing its end. The whole sci-fi genre and future “extrapolations” relied on the such rapid pace of development.

      • gh32zT
      • 1 year ago

      Sad but true. They just cancelled the Rift 2 — it seems clear that even Facebook/Oculus knows that PC VR is dead. It sucks, because the Rift 1 is great other than the resolution issue.

      • Krogoth
      • 1 year ago

      VR has always been overhyped. It is doomed to remain a niche. Not even Facebook’s massive marketing campaign could save it.

      The gaming genres where VR would make sense are pretty much dead. These days, it is all about FPS, MOBAs and battle royales.

      • End User
      • 1 year ago

      VR and iRacing is the most immersive gaming experience I’ve ever experienced. Fracking awesome.

        • Krogoth
        • 1 year ago

        Too bad that they are niches that hardly anyone bothers with. They are far more interested in flavor of the month FPS, MMORPG, MOBA game(s).

      • ET3D
      • 1 year ago

      No, I don’t remember VR ever being ‘the next big thing’. What I remember is that at every step of the way a lot of people said it wouldn’t catch on and would be a small market.

      I mean, sure, at some level there was VR enthusiasm, and several companies tried to get into it, but it still obviously had quite a few drawbacks. At the technological level there’s still a way to go.

      If anything, I think that Facebook is taking it the right way by creating standalone headsets. The need to tie a headset to a PC (and a powerful one at that) relegated it to an enthusiast’s toy.

      The main problem I see with Facebook is that it’s not a console company or experienced games store, and as a result the surrounding ecosystem is rather weak. (Or, to put it more bluntly, the Oculus store is shitty, and they need to invest more in content.)

        • danny e.
        • 1 year ago

        I will never buy any product from facebook. So, their potential market just dropped by … 1. =)

        standalone is not the right way to go, imo.
        I stand by facebook killing any changes Oculus had at becoming anything and, in turn, setting VR back by years.

          • ET3D
          • 1 year ago

          Just to clarify, are you in that potential market at all? Have you bought any other VR headset or are considering buying one?

          To you and a few other vocal enthusiasts, Facebook is the devil’s spawn. To most people, Facebook is a familiar name that would help sell products.

          I’m not sure what the way to go would be, in your opinion. An expensive product tied to another expensive product is a niche market. A cheaper product tied to a cheaper mass platform, like the PS VR, is a larger niche. A standalone has the widest appeal.

          Of course that’s predicated on content, and standalone is weakest and therefore would have the least amount or worst looking content, but it’s still an enabler. Just being able to take your VR with you wherever you want it is a powerful notion. Facebook is working on getting a standalone headset released that has decent games, and while that still will be hugely weaker than desktop VR, I still think that combination (decent games + portability) is powerful.

          I agree that some of it is a matter of timing. Netbooks for example were the right form factor, but were originally too weak. The thin and lights today are what most people want. Can standalone VR provide a good enough experience? Personally I think so. Even the rather weak Oculus Go has some good games, and I can definitely see the potential for even better ones. A better standalone headset could do even more.

    • DPete27
    • 1 year ago

    My wife and I went to a VR arcade this past weekend for the first time!
    The screen door effect was definitely noticeable in menus, but my brain mostly ignored it while in-game. Nonetheless, if this filter improves the perceived pixel density, I’m all for it. Until game devs and AMD/Nvidia settle on a way to not fully render both screens individually (even though they’re both showing essentially the same image) we just can’t afford a physical increase in pixel density.

      • psuedonymous
      • 1 year ago

      “Until game devs and AMD/Nvidia settle on a way to not fully render both screens individually (even though they’re both showing essentially the same image”

      There are a whole bunch of techniques already in use that do exactly that, at various levels (from avoiding duplicating draw calls to literally pulling pixels from one view and shifting them to the other).

      The problem is our visual system is literally built around finding little differences between eye views in order to perceive the environment (not just for depth perception, but things like analysing surface texture through changes in specular pattern). By ‘cheating’ and replicating things between eyes that data is lost.

      • ET3D
      • 1 year ago

      Frankly the easiest way would be to remove any 3D effect. I see VR movies, and the effect of ‘being there’ is fine without 3D. I don’t know about games, but I suspect they’d work okay without it too.

    • Voldenuit
    • 1 year ago

    I have the OG Odyssey, and Screen Door is a real problem on it. Course, their going with a Pentile OLED for the display doesn’t help matters.

    • leor
    • 1 year ago

    Tech that does double +1 is pretty badass.

Pin It on Pinterest

Share This