I think you'll have serious problems with immersion and uncanny valley-like sensory problems with nearly all sensation peripherals, barring breakthroughs in MMI. Your visual information and the derived expectations and your proprioception and tactile information aren't going to match up and I bet that is incredibly jarring. Imagine playing wiimote games, but you can only see a stranger's body not quite doing what yours is.
As another example, you've got your omni-directional treadmill, but that doesn't simulate walking over anything but a smooth surface, maybe even just a flat surface. Now, unless you're playing a minimalist game, there aren't very many surfaces like a treadmill and you certainly can't walk on a treadmill the way you can on largish stationary objects like ships, train tops, and continents. You're definitely not going to recreate experiences like scrambling through a boulder field or wading through a creek. Even if it does by providing uneven surfaces, it completely misses out on the sensations of shaky rocks, cold snow and water, and the consequences of leaping and missing. To me, that is a huge deal for the following reason.
I can see the point of VR from a sensate or fluff perspective as in a gamer focused on experiencing something. However, when I put myself in that seat, I don't see the point of VR if it doesn't accurately and precisely create the environment or experience. I think, at the moment, you'll achieve way more for communicating an experience with good art (Art?) than you will with peripherals. Our inner ear, proprioception, and sense of touch make this a pain in the ass. How do you trick the inner ear and skin to feel something forces and sensations that aren't there? How do you sync your position en virtu with your own proprioceptive information? I know the peripherals aren't within my means and it strikes me as a very, very, very hard problem if you're only coming at it from an electronics hardware angle.
From a mechanics gamer point of view, the only point of VR i see is to try to make physical skill a bigger part of video games, which is a cool thing.
The WII and DDR were good advances and probably explain a lot of VG's acceptance as it moved closer to mainstream gaming, like bar games, arcade games, and cornhole
There's not much need to deceive the senses here and most of the games are casual representations of reality. Obviously that doesn't have to be the case, but it is and most people actually aren't that interested in mechanics and fluff, hard and accurate, for going postal or being robbed or any number of horrible things. You can abstract the fluff, while retaining a lot of the mechanical aspects of a game/activity a la Receiver; you shoot ultra-realistic guns at robotic turrets and drones instead of people. Obviously, most people like combinations of fluffy and crunchy in their games. Anyway, the point is that this sort of game tends to be more casual and be less focused on VIDEO GAMES and more on gaming, more generally.
I'm not trying to belittle oculus rift or specialized peripherals, but they are only one part of a solution for VR. If you're not hacking the brain, you've got a mighty task to accomplish in recreating many (might as well be all) sensory experiences on demand. Products like oculus rift and trackir have been great boons for flying and driving games, but they've dodged some of the bullets with tactile immersion. Both activities normally occur in a sitting position with little physical exertion, relative to running, fighting, etc..
I'm not too excited about VR until I see some consumer nerve to wire MMI. That's pie in the sky, but the technology is present in prosthesis and academic research (it'll be ready in 50 years). In any case, I believe that hoodwinking the central nervous system will probably be easier than hoodwinking our whole body.