Search Unity

  1. Welcome to the Unity Forums! Please take the time to read our Code of Conduct to familiarize yourself with the forum rules and how to post constructively.
  2. Dismiss Notice

Virtual Reality Machine - [Idea]

Discussion in 'General Discussion' started by Arowx, Sep 25, 2014.

  1. Arowx

    Arowx

    Joined:
    Nov 12, 2009
    Posts:
    8,194
    Think of a future where Virtual Reality is the new web. Game engines have morphed into the new web-browsers or presence providers and VR building apps.

    There are problems though is your VR space a Unity VR zone or Unreal or Crytek or the retro WebGL.

    But what if there was a standard for VR Space, if you will a Virtual Reality Machine. The concept here is similar to Java, instead of writing your code to run on many different CPU's, GPU's and Clouds, you write to a virtual machine that runs on any CPU, GPU or Cloud that can run the Virtual Reality Machine.

    Just in this case the Virtual Reality Machine runs on top of any game engine technology, so once you have built your VR world for a VRM it can run on any platform or 'browser'.

    In a way Unity already has a fledgeling version of this with it's web-player plugin, you can build your game and run it on any browser that has the web-player plugin. The difference would be that if browsers had a standard VRM built in then your Unity, Unreal or Crytek game could run in it.

    But just like java's Virtual Machine tech to make it work we would need a common Virtual Reality Language.

    Just an idea, what do you think?
     
  2. XGundam05

    XGundam05

    Joined:
    Mar 29, 2012
    Posts:
    473
    A domain specific language for similar things has been tried before: VRML.

    Additionally, why? We have OpenGL (and by extension, WebGL) to abstract handling specific GPUs. We have JavaScript engines and things like .Net/Mono/Java VM (as you described) to abstract away building for specific hardware.

    I'm not really seeing how this would be any different than what we already have?
     
  3. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    8,952
  4. Arowx

    Arowx

    Joined:
    Nov 12, 2009
    Posts:
    8,194
    VRML is an obsolete standard similar to the web with a combination of text based data and scripts.

    Maybe I'm not explaining this clearly enough a VRM is more like Java's Virtual Machine, is is designed to do all the things we need to present a VR experience to the user. Currently we already have VRM's or bits of one, they are just called game engines, physics libraries, graphics libraries and programming languages.

    But with a universal VRM standard, regardless of the platform you will be able to run your VR experience be it game, simulation, learning app, chat room, forum or piece of art.

    Why do we need a VRM like technology, well without it virtual reality is going to be like the old pre-www internet a group of systems/worlds that you need the relevant software and even hardware to visit.

    Use case:
    Arowx: Hey Zombiegorilla come on over to Arowx.com/NewGame I have a great new world to show off!
    Zombiegorilla: Is it Unity VR, Unreal VR, Crytek VR, Disney VR or another?
    Arowx: It's a new VR world builder/player called Quantum VR.
    Zombiegorilla: That's a 2gb download for the player maybe later b4n.

    Compared to:
    Arowx: Hey Zombiegorilla come on over to Arowx.com/NewGame I have a great new world to show off!
    Zombiegorilla: Got it I'm there, wow not bad.

    When was the last time you had to tell someone they need a specific browser to see a website?
     
  5. Taschenschieber

    Taschenschieber

    Joined:
    Jun 8, 2014
    Posts:
    238
    2GB download for a VR player? Are you kidding? That's twice as big as the whole Unity SDK, including the editor, MonoDevelop and the Web player.

    Really, any VR thing worth its salt will have 3D assets that are bigger than the "player".
     
  6. Arowx

    Arowx

    Joined:
    Nov 12, 2009
    Posts:
    8,194
    LOL wrong end of stick, although any downloading of VR players and having to have multiple players would not enable a the VR future we all want! The possibility to travel from world to world as easily as we do page to page on the web.

    And at the moment you have to download each VR experience separately and run them.
     
  7. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    8,952
    That all already exists, or more accurately exists but standards are non-standard. VR is just a display tech. Nothing additional is required for "VR". 3d display tech in browsers is about as close as you are going to get, and that is far from standard. Even those will require assets and code downloaded to run them. Having predefined assets/content and logic.

    Sure, JVM could be used for exactly that, but while JVM in theory delivers on universality, its not an accepted/popular standard. If it were there would be no Unity/Flash/JSHTML/etc. Also considering the hardware component, and that the hardware itself doesn't handle the stereoscopic portion, that means that there would have to specific drivers/convertors for whatever hardware used. Which would also have to be maintained.

    Also game/3d/display tech is constantly improving changing, vendors of those techs aren't going to constrain themselves to a standard. That would be bad for everyone. Competition in those arenas is what pushes fields forward.

    VMRL itself isn't outdated, it just never had a practical purpose because standards are limited by their very nature. There were tons of other attempts at spatial langs both for browser and stand alone, which pretty much illustrates why universal standards don't really exist for complex things like 3d environments, because people are constantly innovating.
     
  8. XGundam05

    XGundam05

    Joined:
    Mar 29, 2012
    Posts:
    473
    What would be nice would be nice would be some standard interface spec for HMD developers concerning the gyroscope input. Granted, this may already be the case...I haven't looked into any of the various specs in a while (primarily HID), nor did I look into them all that much when I did.

    But, HMD tech needs to mature to a stabilized point first.
     
  9. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    8,952
    Indeed. Gyroscope stuff is all over the place for pad/phone devices. Hopefully middle wear can account for HMD differences and provide some some consistency. Or with a little luck one device will be dominate enough that other manufacturers will match that target. Probably not though :) It is much more important to have have standards across HMD over pads/phones given the nature of the interface.