4 ms·
Reading this article I'm not convinced this is the fault of the Rift. Motion parallax happens naturally when a game engine uses two cameras to render a scene (t
by Pxl_Buzzard 13y ago
Reading this article I'm not convinced this is the fault of the Rift. Motion parallax happens naturally when a game engine uses two cameras to render a scene (the 3D effect). Shape-from-shading is dependent on the lighting system the engine uses, and because developers haven't needed to program for VR they've never implemented it.
There are many changes in both game design and engine features that we'll start to see as virtual reality becomes mainstream. It's very likely we'll see a stall in graphics improvements as game engine programmers begin to enhance other systems like physics and lighting to work more realistically.
- jholman 13y agoI initially agreed. It really depends on how this purported shape-from-shading-due-to-eye-flickering thing works. I don't at all understand danah's claims about how eye flickering gives you extra data, and my first-principles impulse is to say it sounds implausible to me... but supposing she's right, depending on what she means, maybe eye trackers would be required to produce proper depth cues. In which case it's a hardware problem too (though of course the game engines would also be required to support it, blah blah).
- aaronem 13y agoNormal mapping provides exactly the sort of shape-from-shading cue boyd describes. Stereopsis isn't required for normal mapping to be beneficial; indeed, in the fifteen years since boyd did the work she summarizes in her article, normal mapping has become ubiquitous among 3D engines which purport to render a realistic result.