Sunday, June 25, 2017

An updated perspective on VR (after getting a PSVR)

If you have been reading this blog you might know how much I have ranted about my disappointment about VR, and why I fear it might turn out to be a complete commercial flop and a niche technology (with the same ultimate fate as the Kinect and the PS Vita). That's not to say I didn't want a VR headset. Of course I have always wanted one. I have been wanting one for about 4 years now, and that desire has never diminished. The major reason why I haven't purchased one has been the exorbitant price (and the abysmal library of triple-A games.)

Recently I finally got myself a PSVR, because there was a decent deal in an online store here. I commented about it in my previous post.

I have now played several games with it, including Farpoint, Robinson: The Journey, Until Dawn: Rush of Blood, and Tumble VR (besides, of course, a dozen or so free demos and apps.)

In my past posts I have made several claims about VR, without actually having experienced it myself. Now I have. So, has this confirmed these claims? Was I mistaken? Was it as bad as I predicted?

"Room-scale VR"


Especially the HTC Vive has been always marketed pretty much solely for "room-scale VR". In other words, you play games standing up. I always predicted that this is a completely untenable form of gameplay. Nobody plays games standing up; not for long periods of time. 15 minutes maybe, but not much longer.

It is perfectly possible to play games standing up with the PSVR, and it works ok with many games (and a few even expect it, such as Job Simulator). The range within which you can walk around is not as large as with the HTC Vive, and turning your back to the camera will cause tracking problems for the controller(s) (because the system needs to see the controller(s) for proper tracking, else it's only a best-guess based on its accelerometers). But with those caveats there's nothing stopping you from playing standing up, and many games are completely playable that way. You can look completely around you, and even move a bit (within the area visible to the camera, which at a normal distance means something like 1-2 meters of movement range).

When I started playing the game Farpoint (which uses the Aim Controller), I decided to try it standing up, for fuller immersion. It works pretty well like that.

But it turns out I was completely right: After something like a half hour of gameplay my feet and my lower back were hurting. I couldn't play for much longer.

I'm in a relatively good physical shape, mind you. For instance, every time I go to work, I jog the stairs up to our office, on the 6th floor, without much problems. I still couldn't play the game for more than about 30 minutes standing up. It was a sit-down game for me from that point forward (and it's completely playable like that).

Room-scale VR? Thanks, but no thanks. It just doesn't work. It might make for fancy tech demos to awe people for 10 minutes, but that's it. It's not a feasible way to actually play games on the long run, especially not for the average gamer (who, let's face it, is not exactly an athlete).

Move/Aim controllers


Some/most PSVR games can be played with the standard DualShock 4 controller (which supports tracking thanks to its light bar and accelerometers). However, the tracking tends to be relatively poor, and very often the tracking will veer off (by making the in-game stand-in for the controller slowly turn to one side over time) and require readjusting every couple of minutes or so. With games that require a tracked controller, and support the DualShock 4, it's ok but far from perfect.

The Move and Aim controllers are better in this regard. While they, too, can suffer from similar tracking issues, it's not even nearly as bad. In my experience you can often play for even 30 minutes or longer without the tracking veering off too much. (It also seems to be somewhat self-adjusting, meaning that just rotating the controller around a bit will readjust automatically the tracking system, and orient it properly in-game. However, sometimes it starts veering off in position rather than orientation, especially veering closer to the camera than it should.)

I predicted, however, that playing with them (as well as the equivalents in the HTC Vive and the Oculus Rift) would be tiresome for your arms. As pictured on the right, how long do you think you would be able to hold those controllers, with your arms extended like that?

In this case I was kind of half-right. When playing Farpoint with the Aim Controller, I indeed experienced my arms getting tired after a while, especially in sections of the game with long battles, requiring constantly holding the controller up, aiming at enemies, for extensive periods of time. The longer I played like that, the more tired my arms became, until it was almost impossible to play.

The "half-right" part comes from the fact that in most games you don't actually need to be holding your arms extended constantly. (Or, at least, in the three games so far that I have played that have required tracked controllers.) During the periods between having to use the controllers, you can rest your arms on your lap (well, at least if you are playing sitting down!)

While the Move and Aim controllers increase immersion, overall I still found Robinson: The Journey to be the most enjoyable playing experience because it does not used a tracked controller at all. It's played with the normal DualShock 4 controller, without tracking, which means that you just play it as you would play any other game. Ultimately I found this the most comfortable and likeable way to play a VR game. Tracked controllers may increase immersion, but in the long run they feel gimmicky and needlessly strenuous to your arms. (But I will fully grant that a tracked controller is better for aiming and shooting in VR.)

Nausea


Nausea, or motion sickness, has always been a stated issue with VR and how games should deal with it. It has always been stated (and I fully believed it) that if the in-game camera movements do not match your physical head movement, it would very quickly cause strong nausea. Even if what's simulated is eg. you sitting in a car, it would cause nausea because you can't feel the car turning, accelerating and decelerating as normal. This has been one of the major arguments against sit-down VR and for "room-scale" VR, and it has been one of the major arguments against the idea of traditional FPS games supporting VR.

I expected that nausea would be a problem for me, but that I would get used to it. Short sessions, taking a pause when feeling nausea, getting myself accustomed to it... no matter how long it would take, I would eventually get used to it. Other people report getting used to it, so I was sure I would too.

Three of the games I have played so far have great theoretical potential to cause nausea (in the order in which I played them):

Until Dawn: Rush of Blood is essentially a rollercoaster ride, and at points it really seems to want to cause the player vertigo by having huge downslopes that are ridden at great speeds. It almost feels like the game creators wanted the player to throw up.

Farpoint resembles somewhat a traditional first-person shooter, in that you move around with the thumbstick (in the Aim Controller). The movement is not exceptionally fast, but it's normal forward/backward/strafe movement with the thumbstick, as in any FPS game. (The other thumbstick does not rotate the camera. You just look around with your head, and aim and shoot with the Aim Controller. The game has been designed so that you don't need to look nor shoot behind you.) Since in-game movement does not match your physical movement, this has great potential to cause nausea.

Robinson: The Journey resembles even more a first-person shooter in that it's played with the DualShock 4 controller, and not only do you move as normal with the left thumbstick, you also rotate the view with the right thumbstick. The rotation has been limited to only horizontal rotation (after all, you can just look up and down with your head), and the rotation happens in discrete steps rather than being smooth (although the camera very quickly rotates between these steps, rather than just jumping outright.) Certainly these limitations exist to diminish nausea, but it still has very great potential to cause it. And if you search the internet, you will find lots of people reporting feeling nauseous.

So, how much nausea did these games cause me? How horrible was it?

Nothing. No nausea. None at all. Even those vertigo-inducing rollercoaster rides had absolutely no effect. The first couple of times I played Farpoint, there might have been a bit of discomfort, but after that there was none, no matter how long sessions I played. With Robinson I never felt any sort of nausea at any point in the slightest, even though my longest contiguous playing session was several hours long (I think it was something like 3 or 4 hours of contiguous gameplay without pauses.)

I was actually very positively surprised how little nausea I have experienced with the PSVR, even in games where theoretically I should have.

I suppose that this makes me extra right: Not only was I right in that I would get used to the nausea... but in fact it turns out that I don't have the problem at all.

That's not to say that if I got to play a traditional FPS game (like Portal 2) with no nausea-reducing limitations at all, that I wouldn't get nauseous. I might well do. But this experience has convinced me that I would most probably get used to it quite quickly, assuming I would have any nausea problem in the first place.

Resolution


This is something that I actually did not predict. Low resolution has always been stated as one of the problems with current VR, but I didn't expect it to be as bad. I was more expecting it to be more of a problem with the so-called "screendoor effect" in the HTC Vive and the Oculus Rift, which is caused by the visible gaps between pixels, and which is almost nonexistent in the PSVR (because, as far as I know, the PSVR uses some other type of display panel than those other two, and which has almost no screendoor effect at all.)

After all, the PSVR has a full-HD 1920x1080 pixel display. How bad can that be? Sure, you probably can see the individual pixels, but surely it can't be that bad?

I was quite negatively surprised about how bad it really looks. It quite literally looks like you are looking at an old 640x480 pixel CRT monitor. It's that bad. Especially if there is no antialiasing, or very little antialiasing, images are really and very obviously pixelated, to the point of being outright bothering. This image (from a previous blog post) demonstrates how it looks like on the monitor, and how it approximately looks like with the PSVR (click the image for a larger clearer version):


Even when the image has extremely high antialiasing, it still looks like looking at an old CRT TV. The pixelation is much less obvious this way, but it still doesn't look sharp, like with normal displays.

That being said, heavy antialiasing does remove the bothersome nature of the pixelation. Robinson: The Journey succeeds in this quite marvelously. It indeed looks a bit like looking at an old CRT TV, but on the other hand the pixelation is almost unnoticeble, so it looks quite good.

Most PSVR games so far, however, do not use antialiasing that strong, which often makes them look quite bad.


Gameplay limitations


One of my biggest issues with VR has been how much the format limits gameplay and game mechanics (or, to be more precise, how much game developers limit themselves when creating VR games). Will VR be eternally relegated to vehicle simulators and Myst-like games (and other such games where you essentially stand or sit still, and just "teleport" around)?

In some aspects I was right, in others wrong. In the previous section I already described some aspects of this.

I was actually positively surprised that two of the games I have played so far had free movement using a thumbstick, rather than using some idiotic teleporting mechanic. I'm also surprised how little nausea or discomfort this caused (that is: None at all.)

Those games prove beyond any shadow of a doubt that you don't need a stupid teleporting mechanic for these games. I'm actually a bit surprised that the game developers dared to defy the (quite false) wisdom, and went for full-on regular old thumbstick movement. It also proves wrong all the masses of VR fanboys who kept pounding on how the teleporting mechanic is necessary, and how if the game were to move in the traditional way it would cause projectile vomiting in seconds. I very strongly suspected this to be the case (ie. them being wrong), and I was proven completely right.

That being said, both games still have their limitations, both because of VR and to reduce potential nausea.

Farpoint has been designed to be played as a sit-down experience, which is great (because, as said, playing standing up for long periods of time is just not feasible.) However, severe compromises have been made in the game because of that. All levels have been designed so that advancing in them can be made within 180 degrees. You never need to move back. Enemies never attack you from the back (which would feel quite unfair, if you are sitting down). Even if an enemy is behind you, it will move in front of you (which sometimes looks a bit ridiculous because the enemy will just run right past you, to your line of fire, rather than shooting you in the back, which would make most sense.)

Robinson: The Journey has also been designed to be playable while sitting down, but it gets rid of that limitation by allowing the view to be rotated horizontally, as with any regular FPS game. As mentioned, however, the rotation is not smooth, but done in discrete steps. While the jumps between steps are not immediate, and instead the camera rotates very rapidly between them, it's still there. (Of course given that you can look around freely with your head, this is not really of a limitation in terms of where you can look at.) This discrete camera rotation has certainly be added to reduce nausea. In my case, at least, it works like a charm for that purpose.

Neither game (or any of the other games I have played) has jumping or crouching as a game mechanic. (Of course if you are standing, you could jump and crouch if you really wanted, but there's little to gain from it. It also would probably be inadvisable, especially jumping.)

The latter game has climbing as a game mechanic, though, and it works surprisingly well. Kudos. (It can be slow at molasses, though. Of course it's more of a puzzle game than a hectic Doom-style first-person shooter, so it doesn't really matter.)

The one thing I'm most glad of is that the game developers discarded the stupid idea of teleporting. It makes the games infinitely more enjoyable and playable.

No comments:

Post a Comment