A Third of Valve Is Now Working on VR

Sounds to me more like the Apple Vs. Google "ecosystems".

Sure, Google has a Play store and a competing OS, but it's not like you can't find Chrome in the iOS store.

Until Facebook decides that it wants a cut of all the sales on Oculus and locks down the platform, hence Valve's concern. There's only one App store on iOS vs Android.

Valve's main bargaining chip would be not allowing their games to run on Oculus. The recent push back on Oculus exclusives and Valve willing to float cash to developers means they are well aware of their situation.

It doesn't matter much in practice - most people just use Unity or Unreal, which make it easy to target any major VR system. To the credit of these game engines, it's just not that hard to be compatible these days. That's why Oculus has started to use exclusivity contracts to gain an edge.

It wouldn't be the final word on VR API for all eternity. It's a starting point for the spec, which will change as the industry grows.

Well I would call DK1 and 2 more of the 0th generation. The Vive is good enough for regular use, and from what I hear the released Rift is too.

Motion sickness I think will become less of a problem as developers get better, and people just get used to VR. You don't have to be superhuman to have no problem with the current headsets.

And yes, for entertainment, people want fully immersive media. Not for all applications, but it has high appeal for games and other pure entertainment.

> I don't have time for the game stuff.

Well, ok. I guess an xbox would end up in your closet too, then.

I don't know how successful VR will be in business, but I think that's sort of tangential to this discussion. I am convinced it will be very successful as an entertainment product. Some companies have been experimenting with VR apartment tours, room planning, architecture demos, etc. Not sure how successful those will be.

I agree that Hololens is closer to "true" AR. That's why I said VR is more ready for public consumption, because you can go out and by a Vive right now, and it works great. Hololens isn't out and might not be out for a while, and it's not clear how well it actually works right now.

I totally agree about "room scale games". Has add-on benefits of getting out of the house and getting some exercise. Here's my idea (shh - don't tell anyone). Buy an abandoned roller skating rink and transform it into a VR game space. Of course this assume that some other entity will create one or more compelling games to be played in your space.

I've tried DCS with a Rift. It was mind-blowing. I hate using my screens now, I can't wait for a decent priced VR unit.

My biggest gripe was controls. There are quite a few keyboard shortcuts in DCS. Some of them just require looking at a keyboard, since I use them so infrequently. I can manipulate a HOTAS without looking though, so that's ok.

In DCS, some of the aircraft have clickable cockpits (meaning you can use the mouse to do everything, if you wish). If the Rift controllers could manipulate these, I'd be in heaven.

I may be too young to recall, but I seem to remember 30+ year olds complaining about how touch screens would never catch on around 2005-06. Something about the masses never accepting a product that's "merely tolerable."

I recall the same about the size of a phone in hand a few years after that and we haven't seen a sub-5" phone really sell well in how long?

This doesn't seem quite right either, though. People had been talking about 'virtual reality' for a long time, but only recently has there been so much activity in trying to make available to consumers. It's true that a non-trivial amount of people experience motion sickness, but I'd hardly call the people who use it without such issues the 'lucky few.'

> it is a niche technology because it works against human physiology

I'm not sure I agree. Many complained of motion sickness playing early FPS games, such as DooM or Descent. I wouldn't be surprised if it just takes an equilibration period to get used to the devices. Either way, unless there is actual research that shows a large portion of the population actually has difficulty adapting to the device, I don't believe the tech is fundamentallying flawed.

I think you are talking about sickness when you are talking about human physiology.

>only a lucky few are totally unaffected by it

That is true, but what you aren't covering here is that when software (and hardware) is properly designed, almost no one is affected by it! Vive games with physical locomotion do not have complaints of sickness.

Personal Computers had a 15 year adoption cycle. VR will be even longer. We don't really know who is affected or unaffected by it. Only a few dozen people have used real VR on a regular basis. By "real" VR I mean a facility with at least a $M price tag. Let's assume that this raises the percent who find it utterly tolerable to 50%. That's still way higher than the rate for amusement park rides - which is a multi-billion dollar business.

Vive has partially solved this problem with room scale - moving your goggles also moves your perspective in game.

Actually, as the OP states, the issue is with focal distance.

The military has been using AR/VR in heads up displays for decades now and have had to work around these issues. When a user is forced to focus on a 2d projection of a 3d field, it creates all the issues people have with VR including: vertigo, nausea, eye strain, anxiety, and malaise. Motion exasperates the issue.

I've had similar experiences. Showed Edge of Nowhere to someone a week ago, they spent 3 solid hours in.

Tried talking to them to see if they wanted to go for a walk or something. No response, too engrossed in game.

I've experimented with our own game engine and VR sickness. When the view doesn't accelerate relative to your real-world position, there is no issue. The problem starts when you're using external direct movement control, such as a mouse and keyboard. Moving forward and suddenly starting to move backwards is slightly uncomfortable. Rotating your head while rotating the mouse in the same direction is the worst for me.

It's the inverse of car sickness, and affects some people but not others. My coworker has no issue using full FPS mouse/kb controls within VR.

They are plentiful - for one example, see the article and comments here: https://news.ycombinator.com/item?id=11795913

The one with the blue people.

That's a big apples to oranges comparison, but breaking it down anyways... for an apple watch all you need is a smartphone which the majority of people have. For one of these VR HMDs you need a 300 dollar graphics card and a capable computer, which is a much smaller market segment for a device which has only been out a handful of months.

So, how will this change? Well GPU's are getting increasingly powerful, Nvidia will likely debut a GPU that is powerful enough for a Rift/Vive on a laptop this year, and AMD will release a graphics card powerful enough for VR on desktop at a price point less than 200 bucks.

My 8 year old son knows more about the Rift from watching his youtube buddies than I did. That's when I realized its gonna stick this time. He had a chance to use it at his summer camp and now he's saving his allowance so he can ask Santa for one and buy a Kinect to make VR experiences at home (Luckily his superhero dad is a PHP and C# programmer by day)

Little kids love even the simplest buggy broken demos. They go wild for it.

I think some of those will always be problems or be problems for a long time, but the core experience of presence makes them matter less.

- Resolution: It'll always be worse than a phone because for the foreseeable future VR screens will come from phone screens and the field of view on a VR HMD is many times greater. People liked doom 1 and other low res games for what they were in the day, and I think people will accept the trade off.

- Oculus Rift is pretty sharp across the FOV, I hear Vive is a bit worse in this respect.

- Moving in VR is hard... I think 3rd person view is pretty great for action/adventure games. I think the nature and style of games will inherently need to shift, which we're already seeing a bit in the Oculus Home catalog.

- True, it's the chicken and egg paradox. How can you get big games without a big user base? How do you get a big user base? I think Oculus and Valve are getting this right by directly investing in games. We have more and more fun games from large studios coming out which our current user base size really shouldn't allow for.

I think the first and second points are the strongest as it requires a change of expectations, and highlight the real fight for adoption in VR. FPS games wont work, and the most obvious metric that gamers had for high quality experience, resolution, is going to appear lower.

Only response is current gamers enjoying buying and playing games on Vive. As game developers figure new fun experiences for the platform that overcome those obstacles we'll likely start to see more success. I think a key will definitely be social aspects, as there isn't anything quite like standing next to someone or an avatar that is interacting with you in VR.

> They haven't found a decent solution to the problem of moving around in a VR world. Right now the best answer is teleportation

I would really enjoy if someone made a game in the spirit of QWOP that uses the two triggers in a gamepad to simulate steps made by each leg.

However, I don't mean that I would enjoy playing that game.

>Many people are getting sick.

Source? It runs contrary to my experience, and my sample is large enough to be data, not anecdata.

> Eyetracking

Not needed for gaming. From your text it looks like you're thinking in terms of virtual desktop, which is not the target.

> Games

Too early to tell. There are already a few very good titles and a promising pipeline anyhow. The outlook is positive, not negative.

> Price/version

It is the same logic that should have doomed the yet undoomed gaming graphic adapter market.

> FOV

Acceptable for first generation devices. Does not prevent immersion.

> Lock-in/exclusivity

Basically the only valid point in your list. True. I don't think it kills VR, but fragmenting a small market is a dumb dumb move.

Many people get sick with conventional 3d games, but it hasn't done much to the market's enthusiasm.

Your post convinced me that VR is getting big in the near future. Why? Because your complaints are so closely similar to the complaints about consumer 3d acceleration boards when they were new.

> Right now it looks like certain games won't work on every headset. This immediately tells consumers to wait for standardization. It's a fucking monitor. You don't get to have exclusives for your monitor.

Alright...I'm triggered, let's do this!

Here's what a VR consists of, which makes it substantially more complex than a monitor:

* Image warping. This is currently built into drivers, specific devices may require specific warping so that the images look correct after lens distortion.

* Latency reduction techniques - both the Rift and the Vive have different techniques for handling what happens if your FPS drops. I believe both of them take a rendered image that is out of date, and re-project it so its now up to date with your current head position. That is some complex math.

* Head position tracking. This is a combination of accelerometer/gyro and a lower latency but higher accuracy tracking solution like the lighthouse.

* Input devices. For example the Vive controllers communicate directly with the headset. The input devices also track physical position.

* Output - as well as sending position back to the computer, they may also send other information such as controller button presses, your IPD (because different IPDs require different graphical output).

In the future we will have some crazy optimizations going on to improve efficiency. Such as splitting each eye into four renders. Foveated rendering, which would mean the headset would contain a very low latency pupil tracker so that only the area you are looking at would be rendered with high detail, and so your eye direction can be known to the software.

All that said, yeah it's possible to create a standard wrapper around this stuff, but saying it's a monitor is not really true.

>As an aside, I like moving my focus with my mouse. It's incomparably more accurate than hoping my deviated septim is pointing exactly where I want to look. If you can't disable headtracking that's an immediate red flag for me.

Also talking about mouse input makes me feel like you haven't tried the Vive. It's not about giving you an advantage, the draw of these technologies is that they can make you feel like you are there. Think about the difference between playing angry birds on your phone, vs actually picking up a giant bird, putting it in a slingshot and destroying some houses.

More likely that trolls would play without VR and find a way to target VR players with nausea-inducing effects.

Valve also doesn't make Fruit Ninja, but they were just trying to make a point.

There are multiple testimonies on the internet about head mounted displays along with joystick movement inducing motion sickness. It gets worse as the pace of the movements increase. Haven't run any experiments myself, so perhaps you are right. What is the scale of the DK2 demo? If it is a room scale demo you can't infer much from it. The issue would be the addition of the joystick movement with the head movement, resulting in a final movement vector that is in conflict with your inner ear. Any of those two movements taken on their own (e.g. moving your head while standing still, or moving with joystick while keeping your head straight) would not result in any conflicting signal.

Other applications of VR still apply, but this I believe will remain a major hinderance for FPS style games.

I agree completely with the telepresence part. Depth cams + more ergonomic headsets + high speed networks can already offer a glimmer of what future VR telepresence could offer.

But I also agree with the previous comment regarding first-person locomotion in VR (at least as it's currently done). I've also spent quite a bit of time in my launch day DK2 and when you're actually walking it's OK, when you're sitting or standing in a way that mirrors your in-world avatar it's OK. But when you are standing or sitting and you use a gamepad or keyboard to handle "WASD" while using your head as "mouselook", it still either makes me queasy after a while or gives me a headache.

There are just too many little mismatches in most implementations that make it hard to take a typical FPS-style game and slap VR onto it. The simplest example I can think of is walking around with a gamepad and VR headset:

Imagine you are in a game with your DK2 on your head and an Xbox controller in your hands. You push forward on the stick and your character walks forward. OK...not too bad. Once you get over the initial dizziness it's pretty cool.

Then you want to turn right. You push the stick to the right and your perspective starts to pivot as well. Halfway through this turn, you want to look to the right. You turn your head to the right and as you do, your "turning motion" speeds up since you are turning both your body and your head.

Then when you stop pushing right on the stick, the speed of the pivot drops. When you stop moving your head, it stops completely. The whole thing is a mess of speeding up and slowing down changes in your angle of view. And this is just one simple motion. It's OK on a flat screen in front of you but when it fills your whole field of vision, it can get really disorienting and dizzying fast.

Consider that in a game, there are all sorts of combinations of movement on the X and Z axis, changes of view from pivoting your body, and changes of view on all three axes from moving your head. These are often fast and unpredictable. And all the while, if you're standing up, walking around a small room, or sitting in a chair, they don't necessarily match your actual position.

Either way, sorry for the wall of text. It's hard to describe things that are easier shown than told. I agree that finding new types of games and experiences are they key here. My main complaint in this field right now is the number of people who are still so focused on the FPS style of games because they were so immersive and popular on 2D screens. The goal should not immediately be to cram those into VR just because it would seem on the surface to be a perfect match.

But that's my point. It is still a good indication of what a modern FPS is, and of what HL3 would be. I don't see them changing the game radically to be a "native" VR title, which is what VR needs (and which may not even be what we call a FPS anymore).

So SteamVR = PC and Oculus = Console?

If so, I need to go and join the glorious SteamVR master race. ;)

This. Facebook is more interested in VR for sports, messaging, telepresence, news, ads, and entertainment media than it is for games.

Downside of that strategy is that the Valve/Vive bloc could produce radically superior technology because gaming creates seductive technical barriers to overcome. FB's intended uses do not.

We already have virtually zero latency in VR at the moment thanks to predictive tracking and rewarping of the frames to the predicted head pose and direct driver support.

The motion sickness is mostly coming from vection now, when you move in VR but not in real reality. It's an active area of research, people are experimenting with things like reducing the FOV while moving, using interactive overlaid grid to pull the scene around, etc.

No, it's not just due to rendering lag. Motion sickness is caused by a disparity between the what the eyes perceive as motion and the vestibular system of the inner ear [0].

That's not all, though. Another major problem with VR is the lack of compensation for vergence [1] and accommodation [2]. These two systems are critical for depth perception and when they are not accounted for the result is confusion of the visual system, eye strain, fatigue, and even nausea.

[0] https://en.m.wikipedia.org/wiki/Vestibular_system

[1] https://en.m.wikipedia.org/wiki/Vergence

[2] https://en.m.wikipedia.org/wiki/Accommodation_(eye)

I don't know if it is overblown, but it is certainly real. I've only experience motion sickness under two conditions: helicopter flight on moonless nights, and developing software for the original rift. It was a real bummer, after a few uncomfortable weeks I put it back in the case and haven't touched it since.

I own a Vive. Zero motion sickness on room scale games, zero when flying around in Elite:Dangerous. However, get me in something with wheels and it's hell. Rolling in the buggy in Elite is bad, driving in Project Cars is worse.

I've learned that, in Elite, if I look at where the buggy is going (it power slides a lot), then I almost avoid motion sickness. Project Cars, though, I can't play. Cornering, at high speed, completely immersed, with no lateral acceleration, triggers some alarm in me.

FPS games are inherently a million times worse for VR sickness than room-scale walking around, since the former has the mismatch of the avatar walking around with the player sitting in a chair, while the latter is 1:1

HL2 is the only game that I felt nauseous from. Wonder why? I stopped playing though instead of trying finding a solution so I don't know if it's a known problem.

I actually got really naughtius playing half life 2 on the pc many years ago. Maybe there's something about that game.

I think it's the games. Room scale games, like Fantastic Contraption, induced no motion sickness on the twenty or so people I had trying it out. Games where your avatar is moving but you physically aren't are dangerous territory. I never tried HL2, but I easily imagine it begin nauseating.

Additionally, HL3 will have to be big to wow fans, or it will be a disappointment. People have been waiting a long time, and riding the hype-train the whole way. The HL games were revolutionary, and, this would be another revolutionary step for them.

That seems like a bit of a stretch. While many of the Half-Life games have involved teleportation as a plot element, the vast majority of their gameplay has used more traditional methods of movement -- either walking or driving small vehicles.

I wouldn't play HL in VR. Headcrabs jumping at your face would be nightmarish. I'm being serious, too. If I'm playing VR, don't go for my face!