VR Design Discussion

News & General Discussion
1

What is your list of DO’s and DON’Ts for VR?

I was hoping we could have a thread where we can discuss the more overall general design of VR, without diving too much into specific technical details.

This is still, for many, uncharted territory. Which makes it incredibly interesting to get different opinions on.

I’ve been owning an Oculus CV1, with touch controllers, now for about a month. And in that time, I’ve transitioned from being a skeptic to overly convinced this is going to be a huge part of both high end gaming and industry applications (arch viz, industrial, training, sim etc).
It really meshed together for me when I got to try it with the touch controllers, and it had a profound “wow” effect.

For those of you working on the second wave titles, inspired by the first wave or just generally curious with R&D, there is a lot interesting data to collect by dissecting the first wave games and applications.

In addition, this genre is moving so fast currently, that a lot of the VR tips from spring and earlier are already outdated or replaced.

I know I have my personal list of things I thought worked really well, I don’t know how that list compares to other peoples experiences.
And I am also very impressed how quickly the brain adapts to unknown situations, and how you can teach the brain to find things intuitive even though they aren’t, because it just hasn’t known anything similar.

2

Here’s a big don’t I learned the hard way when testing somebody’s attempts to port an FPS to VR:

Do not include headbob if the character can move freely. Free movement in VR is tricky enough to do without inducing motion sickness as it is (hence a huge reliance on teleportation or pure stationary gameplay) but headbob and other involuntary movements are a disaster to include.

4 Likes
3

Oculus made it hard for the first wave titles to support spatial controllers since Touch wasn’t available yet. But second wave titles will need to be designed for spatial controllers. There’s no going halfway with spatial output (headset) but no spatial input (controllers). In other words, no gamepads.

I agree with @Murgilod , too. Designers will need to break the habit of using a lot of fakery, such as making the camera look like the player’s head is bobbing, and instead leave a lot of this to the player. In the same vein, you can’t take control of the camera they way you might in a non-VR cutscene. Designers will need to develop softer ways to direct the player’s attention.

1 Like
4

One of the best ways to ask people to share is to share first. What does your list consist of?

2 Likes
5

Related article;

6

I’ll try to summarize my observations based on my own experiences and observations made by demoing it to around ~20 newcomers to this Gen’s VR. And I’ve noticed impressions tend to be wildly different in terms of preferences.
But my overall user data is somewhat small, so it might not correlate with other’s experiences.

Head Clipping / Intersecting Objects.
I’ve seen 4 different ways so far of dealing with this.
I’ll list it in matter of most comfortable to least comfortable experience.

  • Head collider, head don’t intersect wall.
  • Fade to black, view gradually fade to black as you start to intersect the object/wall.
  • Don’t restrict head movement, let the user clip objects if wanted.
  • Black display on/off. A trigger that that turns the screen black when user is intersecting objects.

Quitting / Exiting / Saving
One thing I’ve noticed to be a very common factor for people that differs from non VR, is that when a user is done playing they just pull the headset off. I’ve yet to see anyone trying to enter a menu of any kind.
Which is why I believe quick save should be used frequently in VR, where it is appropriate.
Also my impression is that a “Are you sure you want to quit” screen is a big no no, as most seem to be fumbling around when trying hit UI buttons in general. And if the user is in a state of slight discomfort, warm, nausea, sweat, there is little chance of keeping the headset on to enter a menu.

Holding Objects
I’ve seen it mentioned that Vive users prefer a trigger button for holding objects, while Oculus users prefer a constant analogue button. This is similar to my own experience, overall I’d say holding a button for holding an object and letting go will let go of the object makes for a more real experience.

Lense Flares, Additive Post Effects etc
I’ve seen a few games use common post effect filters, and everytime I thought about it and found it annoying was when the “main character” was not wearing a helmet or goggles or similar.

Pitch Black / Ambient Light
Some game operate with 0 ambient lighting. Only local lights or flashlight or similar.
I’ve noticed I find this as well uncanny as in real life even the smallest amount of light will reflect of surfaces and bounce around in the environment. This of course is near impossible to replicate in games.
Maybe the solution is to have a very very small amount of constant ambient light, I don’t know.

Eyes Micro Movement
This suddenly had a huge impact for me. The common things for games have been pointing the eyes towards the main camera. A huge part of what makes an NPC “alive” in VR is the eyes focusing on the “social triangle” switching between left eye, right eye and mouth.

Main Character Sounds
Some games have the character make sounds like breathing or “ouch” like sounds when getting hit.
I’ve heard people mentioning this has been uncanny, particularly when the voice doesn’t match the user male/female deep/light voice etc.

Everything Interactable
Might be the biggest obstacle to overcome yet. But parts of expect everything in the environment to have interactivity, even if it’s just a small button on a generator, I expect to be able to push it, and become disappointed when I can’t.
This might be a learning curve from going over to VR.

Geometric surfaces
Cards and decals are experienced very different in VR than in non VR. You suddenly expect a poster on a wall to have some kind of depth to it and various obstacles on the ground (like newspapers trash etc) look really really flat.
Geometric details definitely has a lot more impact than textures.

Sound
3D sound has a lot more value than in non VR games, even if it is ambient sound, like traffic or ocean, I somewhat expect to hear which direction it comes from. And not only the “close sounds” from interactables . This is a thing I expect to see a lot more experimentation with in the future.

Invisible Walls
Maybe it’s just me, but I found this extremely jarring. On the other hand, I found it really difficult to push myself through objects that didn’t have a collider if it ment I would have crashed with it in real life.
Yet to be tested, but suggestion here would be to put visible wall in place (even a translucent one) if you need to temporary restrict an area.

Physics Mass/Weight
A general guideline in VR is not have the player interact with a lot of heavy objects, like levers etc. Because there is zero resistance so it feels like a paperweight.
I generally found it more believable when object had less mass for physics sim even if they were “supposed” to be heavy objects. So, “I think”, a good tip would be to underplay the mass of an object rather than overplay it.

5 Likes
7

Another thing I want to touch upon is the use of virtual buttons.

Generally, when you design, you have to work around the number of buttons available on the controller and/or keyboard layout.
Now we have introduced the concept of virtual pressure points, and by that I don’t mean UI, but adding trigger zones on your virtual/physical body. Which works extremely good when centered around the upper body excluding the upper arm.

For an example, using a non time critical action, such as activating a skill can be done touch your chest or a button attached to your left wrist.

I think this is very interesting, because it invites creative thinking on top of “rebinding hotkeys”.

I’ve also been thinking a lot about inventory systems and how “mount points” on your virtual body is exactly what an old school hotbar or radial wheel is supposed to emulate. So instead of thinking about it as fixed points for gear, you could potentially consider it as open hotbar slots.

8

Can you explain what you mean here? How are you detecting these touches? Are you talking about a system like Kinect, or something involving the relative positions of the Vive controllers, or what?

9

Sure, so you have colliders attached to the last bone on the index finger of your virtual hands, and collider triggers attached elsewhere on your virtual body (as shown for an example here).
So you can make a point gesture, or fist…or palm to activate a “physical-virtual” button/activation point.

The game referenced above actually has the flashlight on/off button attached to your helmet and there is another example I’ve seen which used a wrist watch as a menu activator.

edit: I should probably add that even though the only tracking points are the the controllers and the headset, and it’s relative position to the floor. It still gives a fairly accurate indication of where the rest of your body is, mostly. When used in conjunction with something like Final IK

10

OK, I was asking not so much about how it’s implemented in software, but what hardware was tracking your physical hands and body.

Yep, that clears it up nicely. I agree this raises some neat possibilities.

Personally I’m more excited for something like LeapMotion, Kinect, or whatever Apple comes out with, actually tracking the hands/fingers of the user, so maybe you don’t even need controllers. Then you could do all sorts of neat gesture-based interfaces, which would be particularly appropriate to (say) casting magic spells.

Of course you can also do gross gestures with a controller, treating it as a wand, gun, sword hilt, or (as in that video) some sort of futuristic hand-mounted hardware. And I guess this gets back to exactly your point: touching a controller to some part of your body (whose location we can reliably guess) is a very natural gesture that’s easy to recognize.

1 Like
11

You wouldn’t guess by looking at it, but the Oculus Touch controllers map very naturally to finely-detailed finger movement, or at least that’s the way it feels when you use them. It was worth the wait.

2 Likes
12

It’s great to see so much innovation and diversity in the VR marketplace. But I have to confess, part of me is looking forward to some standardization. :slight_smile: Right now the different VR systems have such wildly different capabilities, it’s really hard to design a compelling experience for more than one of them.

13

I’m really hoping the next wave of VR titles include some more top-down/strategy type games, perhaps where you’re really big and you control a number of small units in front of you. This kind of game might not be focused on immersion as a virtual experience, but instead trying to create new gameplay mechanics with VRs unique control system. VRs hand controllers give you the ability to reach out and click in 3d space, this has never been possible before. I think there’s a slew of new game mechanics, genres even, that may be possible with this. I think it’s time to go back to basics :slight_smile:

Edit: Also I love some more games that I can play sitting down or even lying in bed. Standing up and exploring is great, but sometimes I just want to be lazy

1 Like
14

I just did some very quick (a few hours) tests on grip, stability and weight.

After that I went ahead and watched the Robo Recall GDC talk.
I personally think there a lot room for trial and error here, and it feels like there is no de facto common practice here.

Just wondering what kind of problems and solutions people have regarding this,
if anyone has some insight to share?