Skip to main content
Ziggurat
Participant
July 1, 2013

Faster rendering of stereoscopic picture

Programming or even understanding well enough shaders is beyond me. But I was contemplating the possibility of having a shader that allowed for only rendering a picture once, and extrapolate real not stretched 3D images from this.

The reason why I thought about this, is because good graphics and high framreate is maybe even more important in VR than on a monitor based game. I'm not saying graphics trumps gameplay, but its at least nice to have good graphics and good framerate. And rendering a picture twice takes its toll on the framerate.

I google'ed and found this: http://fulldome.ning.com/forum/topics/stereoscopic-domemaster-images
Or am I mistaken?

    12 replies

    Show previous replies
    tomf
    Participant
    July 23, 2014
    Sadly the GS looks better on paper than it works in real life. In general the GS becomes the bottleneck unless it's a very low-poly scene.
    Tom Forsyth - OculusVR coder
    datenwolf
    Up and Coming
    August 24, 2014
    "tomf" wrote:
    Sadly the GS looks better on paper than it works in real life. In general the GS becomes the bottleneck unless it's a very low-poly scene.


    I wish I could upvote on this forum. Because if you hadn't written it, I had. Geometry Shaders turned out to be a huge disappointment and hardly anybody uses them. When they got introduced a lot of people (including me then) were hoping to implement stuff like subdivision surfaces in realtime rendering, i.e. tesselation. But the poor performance of GS put a stop to this. Luckily later Tesselation Shaders were introduced, which fulfilled the hopes we had for GS when applied for tesselation.

    The problem with GS is that by their very nature they disrupt data locality. And that is poison for efficient GPU operation.
    Terms & ConditionsAccessibility statement