For 2/3 are you talking about developing new VR hardware, or the VR software part of your game?
There's no need for a game-dev to be making VR hardware - there's going to be a handful of very good VR HMD brands servicing that market with high quality products; Oculus and HTC to start with. Plus there's the PS VR, which is the only hardware choice for Playstation developers. If you do want to though, there's DIY HMD kits available for ~US$30, favored by drone racers and those with shoebox apartments :lol:
The hardware cost to developers is fairly minimal for any game-dev as a business (as opposed to game-dev as an expensive hobby). An amazing HMD such as the Vive is equivalent in cost to hiring a contract programmer for a day or two.
I've been supporting VR since the Oculus DK1 in 2012, and the dev-kits have been around $400 each (or much cheaper 2nd hand!), which is half the price of the final consumer version. We have an Oculus DK1, two Oculus DK2's, an Oculus CV1 (current consumer version) two HTC Vive prototypes and a Sony PS VR prototype -- all up costing us about US$1100. That figure is so low because some of those companies have been nice enough to give us free hardware, as they really want gamedevs making products that support them. Just look at the dev list that Sony put out for PS VR - that's a lot of PS VR units in the hands of devs already.
As for the software side, if you're using a decent engine, such as Unity or Unreal, then most of the work has already been done for you.
The hard parts are:
- Making VR-friendly gameplay -- usually that means cockpit based games, or ones with minimal movement... Although a bundled Oculus title is a 3rd person adventure, so that doesn't always apply.
- Making a VR-friendly camera. It should be mostly HMD controlled and never surprise the player.
- Never drawing any UI stuff directly to the screen. The easiest solution is doing traditional 2D GUI's, but rendering them them to a texture that is then placed on a quad hovering half a meter in front of the player where they can focus on it easily.
- Making sure that your game always runs at a solid 90 frames per second on your minimum-spec PC hardware, with two full-HD viewports... That's a challenge if porting to VR, but performance budgets are manageable if you're aware of them during the entirety of the project's lifecycle.
If you keep those pitfalls in mind from the beginning, then the impact on your software is pretty minimal.
If you're using your own engine, then a simple integration with the Valve or Oculus SDKs (the Valve SDK works with both Oculus and HTC hardware, so you can just use the one) only takes about a day. Call it a week if you want to get it perfect. Of course, if your game already has a bunch of GUIs / cutscenes / camera-controllers / etc, then you've probably also got a bunch of re-work to do.
For 4 - The price is pretty high at the moment. IIRC, Vive is $800, Oculus is $600 (but the later to be released "Oculus Touch" controllers will probably bring it up closer to the Vive's price) and PS VR is $400. For the Vive/Oculus, you should probably also spend $1000 on a new GPU if you haven't already! :lol: This is going to restrict the userbase in the near-term... although the number of orders for Vive/Oculus are still pretty impressive so far!
IMHO, this means that PS VR is the one that's most likely to see anything close to mass market penetration any time soon. For $800 total you can get a console, HMD and motion controllers -- compared to more like $2800 for a high end gaming PC, HMD and motion controllers. Playstation is also selling a product that anyone can easily set up in their living room - it's a lot more mass-market friendly than anything attached to PC-gaming.
These high prices are to be expected though -- adjusted for inflation, the original NES, PS1 and PS2 were all about $400, and the PS3 closer to $600. Early adopters pay a premium, and in the later years, the prices start to fall.
As for mass market penetration, the big picture is not actually about gaming. Oculus and HTC are not gaming companies. HTC is a consumer electronics company. Oculus is a VR company that's owned by an advertising / social media megacorp. Oculus' business is not to make a mass market gaming device; it's to bring VR products to a mass market in general. Gamers are simply their early adopters who are willing to pay a premium for the first iterations of their products. You can be sure that they want games to eventually only be a small part of their business, as they expand into every other kind of business activity that can be supported by VR -- largely social experiences. They're also working towards full body VR, with eyes, ears, head and three fingers being just their very first stepping stone towards that goal.
[edit]Or check out this picture from google today -- Sports, TV and everyday applications sharing equal placement with VR gaming[/edit]
As well as the above HMD's, there's also the GearVR (and "cardboard"/knock-offs), which are $100 for the real deal, or $2 otherwise. These are going to also be part of the mainstreamification of VR, as the barrier of entry is much lower (assuming you've got a nice new Samsung phone in your pocket already) and there's already hundreds of "experiences" that interest non-gamers available on there.
As for AR, there's of course the hololens, but smartphones are also going to seamlessly grow into this marketplace. The GPU, internet, sensors, camera and screen in a modern smartphone makes it perfect for AR applications. Future hardware and software advances will cement it.
That's going to create two kinds of AR -- the hand-held window type you can get from a phone/tablet, and the head-mounted type. The latter is going to be more expensive / less accessible, probably piggybacking off VR for the near future... At least until people decide that google glass is cool :D
/cdn0.vox-cdn.com/uploads/chorus_asset/file/6501873/DaydreamHome.0.png){kind=link}