Although the full fidelity, full presence VR experience will probably stay quite expensive, what will save VR from hitting an entry wall is simpler forms of entertainment and VR video.
Being in a NASCAR or F1 pit, standing next to the coach on a soccer or baseball match, seeing a journalistic report on riots or a documentary about the amazon while feeling there is pretty awesome and doesn't require expensive hardware. Something better than a Google Cardboard but cheaper than a Gear VR+Phone can pull it off. Those are live/recorded experience that require virtually no graphics horsepower compared to drawing polygons, shading, texturing and physics. They require no movement tracking appart from headtracking... VR porn falls into this category too (straightforward VR porn, not silly Second Life-like constructed virtual worlds).
These things are cheaper, more accessible and dead easy to produce. They also appeal to a very wide public. Once VR/360 cameras get more popular you also get experiences like remotely watching your son be born if you're overseas or something, or just the Facebook video feed of your friend, giving you the chance to witness some of the party that you missed. Then there's conferencing, teaching and training... regular everyday uses that appeal to non geeks/nerds, businessmen, educators, real-estate, governments and what have you.
Beyond that you can get into actual virtual (as opposed to remote/recorded) reality. From basic applications like architectural end engineering. Even simple 3d environments that let you model an apartment already has some appeal, they also don't require heavy hardware or tricks of the trade to provide full immersion and avoid motion sickness.
Then, the icing on top of the cake is full blown virtual worlds, games. These also can be low requirement, cartoony, artsy worlds can provide immersion with attention to detail, interactivity and some secret sauce techniques to avoid motion sickness. Even on Occulus DK1 you can avoid it if you limit movement speed, etc. The entry level is mobile-like games, there's plenty of market for that. With all this basis holding up the market, more specialized and expensive set-ups like Occulus consumer eddition and Steam VR hardware can provide premium experiences, taking full advantage of developments in graphics hardware, DX12, modern source engines, etc.
As this massive market comes into it's own more factories and processes will start to dedicate themselves to VR hardware or components, more software and tricks of the trade will be developped and new ideas might overcome some of the limitations of interface on the higher end. What's more, core tech for VR is already good enough, but sinergy from highly important components is great right now, moblie is pushing compact displays to ridiculous levels of PPI, CPUs and GPUs are getting ever more compact and energy efficient.
Think about Gear VR, it's effectively a cobbled together thing... it's primarily a phone with a barebones VR headset strapped in. This makes sense because VR is still a risky prospect for Samsung. But just think of what they could do if they made a dedicated VR headset with the hardware specifically desinged. 2 high PPI displays, one for each eye, instead of 1, slap about 4 NVIDIA Tegra X1's in the thing, full sensors, no LTE or other useless phone components, propper cooling... with current battery and fast charge technology you could probably pull 3h battery life with a 30min recharge time.
In a couple of years, techonogy that's already here or about to surface will be viable to make wirless or headset contained VR hardware.
Compared to high-end PCs and graphics cards, that cater to a narrow public, VR is much more likely to succeed and develop. Sure, high-end VR will be costly and nich at first, experiences won't be stellar compared to well established and experienced platforms, and diversity of quality games and software will be lacking.
VR is sort of doomed to succeed, it might take a few years for the tech to be universally compelling and developers and creative people to fully embrace production for it, but unlike 3d television that the industry tried make us think we wanted it, this is something we always wanted, but the industry is now about to arrive at a point where it can sell it.
.
