To keep things simple I’m making the assumption that you already know what the Oculus Rift is (and if you don’t, clicking on the hyperlink will point you toward one of the most interesting technological gizmos of the recent years) and I’m simply left with the pleasant task of telling you what it really is. We are talking about disruptive technology, so it’s not going to be simple or straightforward to assess its impact on simulation gaming (or anything else for that matter), but I can assure you it’s going to be controversial and hopefully interesting so please bear with me even though this is a rather long article.
First of all it’s important to understand that this is not a review or even a preview. I played with the very first development kit version for a few days, mostly in the DCS World simulator and that’s not nearly enough for either, especially since we have no real idea about the capabilities of the version Oculus / Facebook intend to release sometime in 2015. Furthermore, this article isn’t as much about the Oculus device itself as it is about what the technology it represents means to gaming – and more specifically, simulator gaming. After all, simulator gamers are probably the first in line every time there’s a new technology for increasing immersion, which is precisely what the Oculus Rift is all about. In this article I will shortly recap the tech specs of the device (not that important), my personal experiences using it (rather important for me) and an estimate on how the technology will impact our beloved hobby in the future (very important for all of us).
If you’re wondering; the reason I only experimented with DCS is because getting the prototype Rift to work with games is not trivial and I had fairly little time for testing. DCS has relatively good Rift support and has all kinds of aircraft in it from props to helos to jets, so once I got it working perfectly, I didn’t take the risk of messing with the settings for other simulators / games. I would absolutely love to try out for example Rise of Flight, which with its beautiful Digital Nature engine has always been my personal champion in immersiveness.
The tech- understanding where we are at currently (or were last year, but erm yeah you know etc…)
Let’s start with a simple picture. If you have any interest in VR technology, you have seen this kind of screenshots before and you will also know how completely useless these actually are – trying to make people who don’t actually have the Oculus device understand what the experience is like is like trying to define the color blue without even mentioning the terms electromagnetic radiation or wavelength. However, there are two important things to understand about the image, so perhaps it’s not utterly useless after all. Here you go, the venerable P-51D Mustang in DCS World as seen through Oculus Rift (DK1) rendering:
First of all, as you can see we’re looking at a stereoscopic rendering. This is important, because the absolutely beautiful stereo effect is the biggest thing the device has to offer for simulator gaming. However, it also means that not only does it take more rendering power to generate the double image, but due to how the Oculus technology works the resolution of the screen is also first halved and then cropped to a roughly elliptical shape. The effective resolution is actually pretty hard to determine, since we’re talking about moving stereoscopic images, which means that the human brain is heavily involved, which again means that explaining advanced quantum physics would probably be far easier. And I absolutely fail at everything related to quantum physics, simple or advanced.
The important thing to understand is that currently the resolution is lousy, even compared to regular monitors. Not only that, but the screen in the device I tested was rather bad and I was literally looking at it through a looking glass. We’ll look at resolution issues a bit later from a different perspective, but for now keep in mind that the resolution will improve (and already has done so) and you can’t fault Oculus for the bad image quality, since it was absolutely not the point of the first development kit. However, keep in mind that whenever I describe the experience of using the Oculus Rift, the bad screen always had an effect.
I think that first and foremost when people think about virtual reality goggles, they see themselves immersed in a virtual world with all of their senses (mostly vision) completely enveloped by the artificial reality. This is not the case with Oculus Rift DK1. What you get is a relatively large-ish round-ish image that spans about 110 degrees (according to the specs) surrounded by lots of darkness. It’s actually a bit like wearing rather restrictive diving goggles. It’s not bad – especially compared to many of the earlier virtual reality devices – but a full human view of about 180 x 120 degrees it is not. The vertical view is very good, especially in the center (remember the round shape of the image), but horizontally it’s lacking – at least compared to a 3 screen setup. You would be right in pointing out that most things are, but due to another feature of the device, the horizontal view is kind of important. It’s time to give the legendary TrackIR a run for its money.
One of the great features of the Oculus Rift is its 1:1 head tracking. Not because the DK1 implementation is almost as bad as the resolution, but because of how the view follows your head, it provides an insanely immersive sense of being in the cockpit of a fighter jet or a helicopter or whatever, especially combined with the brilliant stereoscopic 3D effect. finally, we will have the privilege of actually checking six by actually checking six. Except that it’s kind of hard if you take into account the rather restricted horizontal view. There’s a good reason real pilots’ helmets tend to have a very good viewing area and once dogfighting becomes possible (currently it is not, you couldn’t see a battleship at usual dogfighting ranges, let alone assess its speed and heading) due to improving resolutions there will be trouble because of the lacking horizontal view. To really see somewhere you have to point your nose there, it’s not enough to just get eyes on target.
There’s another pretty significant upside to the 1:1 head tracking, too, aside from the obvious immersion benefits. In simulators like DCS that provide clickable cockpits you now get exceptional sense of space – you really begin to perceive your surroundings and remember where the switches are in a much more visceral way compared to watching a 2D screen with TrackIR, which combined with the rock solid tracking (extrapolating a bit, the DK1 implementation is not exactly rock solid) will make it easier to access the controls. Also remember the good vertical field of view, that’s really handy when checking instruments or fiddling with buttons in-flight. This is very important, since one of the most common questions about Rift-like technologies has always been how to access our control devices – and that’s still a good question, since with the current Rift (and probably the release version as well) you really can’t see any real world objects while wearing the goggles. Luckily it seems that with a high quality HOTAS system you can have the speed-critical combat-related functions on your HOTAS and the less critical stuff is easily accessible in the cockpit (again extrapolating a bit, currently it’s a bit hard to access most functions since you can’t actually see most buttons and switches). And this is coming from someone who has never been that interested in clickable cockpits. Bigger controls such as mixture, prop pitch, possibly flaps, gear etc. as well can still be used by feel if you’re using for example CH or Saitek levers. That’s actually kind of an important point – a good HOTAS is probably going to be necessary for simulator gaming with Rift-like devices, but luckily it’s also probably going to be good enough.
A good HOTAS is probably going to be necessary for simulator gaming with Rift-like devices, but luckily it’s also probably going to be good enough.
From a very brief test with the newer DK2 version of the Oculus Rift it seems to me that the screen resolution (increased from 1280 x 800 to 1920 x 1080) is a step in the right direction, but there’s a marathon to go. More importantly for now, the screen felt more responsive, which together with the improved tracking gave me an improved sensation of movement in a demo I tested – which is potentially a big deal for simulators, but I’d have to test with DCS or other proper simulators to say anything even remotely definitive. So let’s go by gut feel in the predictions for now.
So what does it feel like to fly with Oculus Rift?
First of all, in my case the Oculus Rift is up against some serious hardware, I have a basic simpit with three screens, extended Warthog/Cougar and a few other controllers as well as a number of games setup to make the most of them. Honestly, though, that’s kind of what it’s expected to challenge head to head when it’s finally released. My educated guess based on the experience with the old development kit is that it will fall short in many areas, but also open up entirely new ground not only for immersive experiences, but also actual practical everyday flying and fighting. But let’s get back to the beginning.
I was kind of expecting to be blown away by the initial impression and then let down by the impracticality of it all. My cynicism is well founded in my previous experiences with virtual reality hardware – I’ve been following the technology for decades and even worked in the field for a few years. The Rift won me over by doing exactly the opposite – a sign of maturity I’ve never before seen in this kind of hardware.
Admittedly, it is a pretty spectacular feeling when you open your eyes and the Huey (which is what I started with) materializes around you in full 3D and convincing scale, almost real enough to touch. Definitely real enough to try to touch it, like most people eventually end up doing. Especially to anyone with no experience with virtual reality hardware the initial impression is bound to be impressive even though the horrid screen of the first development kit version of the Rift is a major detractor, but as I mentioned earlier I’m used to a much higher horizontal field of view and smoother tracking with the TrackIR so the very first contact with the device was a bit meh for me.
As the chopper takes off the Rift and the triple screen start trading punches in the battle for ultimate immersion. One thing that the triple screen with nVidia Surround does really well is the sense of speed and one thing I noticed immediately was that with the Rift I tended to occasionally slip into translational lift, probably because the massive peripheral vision of the triple screen seems to give me a gut feel of going faster compared to when I fly with the Rift. However, the superb sense of space, even with the bad resolution of the rift felt that I get a more precise (note that I specifically did not write accurate) sense of speed, especially when hovering the chopper. With the Rift I would perceive the slightest movement, giving me a great feeling of confidence and good performance while hovering. I kind of like the feel of speed I get from the widescreen setup, but it’s impossible to say which device actually gives the more correct perception. Again the Rift would probably blow the doors off any regular monitor, probably even 3D capable ones (which I haven’t tested) due to the excellent 3D capabilities of a wearable device with head tracking.
The triple screen setup kicks the Rift all over as far as horizontal viewing area goes and gives it a run for its money in sense of speed, but when it comes to sense of space – be it in or out of the cockpit – it’s game over for the triple screen before the contest even begins. Flying around looking at the scenery was pretty impressive in itself, but when I got more serious with the Huey, the Rift played its trump card. I have never experienced the kind of feeling of space in a simulator, not even close. The confidence I felt maneuvering near buildings and other objects was almost unreal, except that of course it was in fact quite real. I would hover effortlessly right next to the parked planes, almost bumping noses, all the time knowing pretty much exactly how far the stuff behind and out of sight was. Of course doing all sorts of stupid stuff ended up with lots of wrecked choppers so the end result wasn’t that different, but to me personally it was all the difference in the world to crash due to overconfidence instead of inability.
Having flown the DCS Huey with the Rift I actually no longer consider helicopter pilots to be superhuman like I did earlier. Give me a bit of actual gut feel in addition to the Rift’s sense of space (and a lot of practice) to help me hold the hover and heck I can imagine eventually becoming competent at piloting choppers. How’s that for an accomplishment; not only did the Rift make the virtual chopper feel more real, it made me feel like a pilot. Could you possibly imagine higher praise for a device that was built to immerse you in the virtual world? And I’m only half kidding here, the feeling of confidence and ability was very real and exciting. Interestingly, the sense of space extended to the P-51, F-86 and even the F-15. I really felt like I was flying really high (or really low) and when doing stupid stuff down low I had such a feel for the plane’s trajectory that I was able to predict the inevitable crashes with unprecedented accuracy. I can’t say anything about combat, since it wasn’t possible, but I had slightly less horrible performance in formation flying than what I’m used to. A wider horizontal field of view would help formation flying a lot, though.
The best thing about Oculus Rift is that not only does it give you a nice feel of being in an aircraft flying in air close to or far from different objects, it also improves your actual performance in some aspects of simulated flight. Even with terrible visuals and bad head tracking.
So… A taste of things to come? What does it actually taste like?
I really want to stress one more time that in the Oculus Rift DK1 we’re getting a raw glimpse of the future. It’s not supposed to tell you what the final version – and especially the ones following in the coming years – will be like. And it doesn’t. But it does give some idea of the strengths and challenges of the technology.
First of all, the resolution. Let’s forget about the awful resolution of the DK1 for a moment and we can exclude the newer DK2 as well, since its resolution, although better, is not worlds ahead of the first version like it needs to be. The thing is, I don’t really see the release version resolution being that great either, unless Oculus pull a mighty long-eared rabbit out of their hat. Since so far they’ve been using a single display we’re unlikely to get for example two full HD screens in the initial consumer release Rift. Since Oculus are (wisely) trying to keep the price reasonable, my guesstimate is that we’re likely to see at most a 2560 x 1440 display. Impressive in itself, but you have to remember that due to the double-image 3D you’re going to get about half that in reality and due to the roundish viewport you’re going to lose a not-insignificant number of pixels in the corners.
So, what does that actually mean? First of all, it might just be at about the point where you start being able to read the gauges and HUD. This is the very first step towards viability as a simulator platform and it has to be reached or the Rift is an absolute no go for simmers despite having impressive compensating features. If you can’t read the gauges you can’t really use the 1:1 3D rendering or the 1:1 head tracking and the Rift is either pointless or completely useless, depending on how zooming actually works with virtual goggles. But even if you can read the gauges and HUD, how far does that actually get us? That’s one of the biggest questions and I don’t have the answer. Let’s at least try to get a bit of perspective to the question.
One of the reasons Rift and the likes of it will require very high resolutions is the rather large viewing area (and don’t forget the halving of the screen due to 3D). I believe it was Matt Wagner of DCS fame who rightly wrote that the Rift has so few pixels that you’d be legally blind flying with the Rift DK2, due to the low amount of pixels spread over the whole viewable area of about 100 degrees. Regular monitors, depending on size, are something like about 40 to about 60 degrees as my own off the top of my head -estimate. Here comes the perspective: we’re not actually talking about a world of difference here. Would you be legally blind flying with a large 1080p monitor at about 60 degrees viewing area? I don’t actually know. Would you be admitted into military pilot training? Never. And I’m not even taking into account the serious, monocular tunnel vision you’d be suffering from – unless you have the Rift.
I’m in a serious danger of flying on a tangent here, so I digress. Just remember that when you read about how bad the amount of pixels per area the Rift and any devices that follow have, they’re not incredibly different from regular monitors. Let’s just say that in the best case scenario the emergence of the head mounted displays raises a question about what is actually “realistic” in terms of visual modeling in combat flight simulators as a whole. But as far as this article, the true difference we’re talking about is the ability to zoom in with regular monitors.
Currently people playing simulators are expected to zoom in and out to see targets far away or the cockpit nearby respectively. With the Rift active, DCS apparently doesn’t allow zooming at all. I will go to the physiological side of things later, but the reason zooming was removed is probably because it would feel uncomfortable. Based on my limited experience with the Rift that might be a good point. How about a different kind of a zoom, something like binoculars? Perhaps. After all, binoculars work well in reality and so far the Rift has scored pretty high on the “just like real life” -scale. But here’s the kicker: there has to be a way to make it work or the Rift can be completely ruled out as a simulator peripheral for the foreseeable future. The only alternative, if there’s absolutely no way to break the 1:1 scale for any reason even temporarily to give you the ability to see stuff far away (I’m not sure if any kind of dot or icon settings could solve the issue in a satisfactory way), the only option is to go head to head against the human eye in a direct fight. And that’s kind of hard.
Defining the resolution of the human eye is not a straight forward thing (remember the comparison with quantum physics earlier?), but let’s just say that nothing you can hook up to a computer comes close. And that’s without taking into account the fact that most of that resolution is directed at a very small area at the center of your focus. And the screen would have to have that kind of a pixel density all over. I don’t have enough imagination to come up with that kind of technology but even if you do, I challenge you to come up with a video card that will run it. In fact, due to slow technological development in the recent years, modern video cards can’t even run 4k comfortably and that is right in the ball park I’d like to see from Rift-like devices in the very near future. But no matter what, a lot is riding on the ability to make head mounted displays zoom in comfortably.
After all the questions surrounding the resolution it’s time for some good news. I flew mission after mission, maximum continuous time being a few hours – a normal flying session – without a hint of nausea. It’s highly dependent on individual physique and my inner ear may be completely screwed after years of gaming, but the Rift is absolutely not a guaranteed explosive projectile vomit projector, especially not for people used to rather large field of view displays. However, there might be a number of health-related factors that are currently poorly known but might become major issues when massive numbers of gamers start actively using the technology. Note that the tracking on the DK1 Rift is poor and DCS didn’t manage a perfect frame rate during the test – both of which are supposed to be potentially vomit-inducing factors, but ended up bothering me surprisingly little in the end. The weight of the Rift is again not insignificant, but I had no discomfort during the sessions even though my nose starts itching immediately every time it senses that it’s hard to scratch properly.
Finally, perhaps not as important as the other technological factors, but I would like to see a much higher horizontal field of view. Unfortunately that’s again not necessarily a very easy thing to achieve. Can the Oculus setup cater for wider screens? Note that it’s not simple to design lenses that work with a wide field of view screen, especially since the pupil of the eye will be moving along the lens quite a bit as you look to the sides. Maybe curved OLED screens could be the answer? Again, I don’t really know, but as I mentioned earlier, wide horizontal field of view is kind of important for a number of reasons.
In addition to the factors discussed in this article there are of course a number of unknown questions lurking in the future. For example, how will the lenses perform when we finally get proper resolutions? Currently the resolution and image in general are so bad that you wouldn’t know if the optics didn’t work at all let alone reliably assess the point at which the optics becomes the limiting factor instead of display quality.
Final words, finally
To sum it all up I’d say that the Oculus Rift development kit indicates that there is great inherent potential in its technology that is hard to achieve with any other approach. Additionally there’s a lot of straightforward technological development ahead – something the IT industry has traditionally been pretty good at given suitable motivation (like the Oculus Rift) – but also a lot of potentially difficult problems to solve. What Oculus has demonstrated beyond doubt is vast public interest in immersive, (relatively) large field of view 3D displays and others are sure to follow. What makes predictions really problematic, however, is the disruptive nature of the technology. The thing with disruptive technologies is that by the very definition they tend to change the way we approach problems or in this case fly simulators and it’s very difficult to say for example how far we’re willing to rethink current conventions – especially ones related to perceived “realism” – to get to the benefits offered by Rift and its competitors. After all, in the 90’s we were happily flying all sorts of simulators with lesser resolutions than what the Rift is capable of right now and although monitors have improved quite a bit since then in resolution and size, there’s still so far to go that it’s a matter of opinion if you want to consider the current technology a gigantic leap forward or just a few steps towards a very theoretical still far away goal.
The Oculus Rift or the devices that will surely follow it are not going to conquer the simulation world when the first wave arrives; even though they initially sound like the perfect match for simulators, the technological demands of the genre will probably be too high for now. Realistically the future of the technology will be decided by how impressive the head mounted display -enhanced future CoDs, Elder Scrolls games and Grand theft autos turn out to be. Personally I think they’re going to take the gaming world by storm and in a few years simulation games will be riding the bandwagon but I don’t even pretend to be confident about my prediction – that’s due to the very unpredictable nature of disruptive technology, which the Rift undoubtedly is.