To VR or not to VR, that is the question

So I'm working on a space game and I've had it working in VR since the Rift DK1. It's been kinda cool, but the further I get into the design, the more I'm doubting that it's a good fit for VR support.

The good:

* Great sense of scale, especially when flying near capital ships
* Ability to track enemy ships zipping around you via head movement
* Ability to acquire missile lock by looking at enemy ships
* Cool stereoscopic effect (lasers and missile trails have a very nice sense of depth for instance)

The bad:

* Technical issues with DK2 and Unity VR integration: no image effects are rendering, cockpit camera not rendering the cockpit for some reason, frame rate challenges
* Future features may include a VTOL vehicle to use on planet surfaces. My current prototype is easier to control in 3rd person, which doesn't seem like a good fit for VR.
* Future features may include first-person walking on planet surfaces, which could be a recipe for VR nausea.
* UI overlays have to be very carefully placed to stay in the player's field of view in VR and probably need to be better integrated into the cockpit rather than my original concept of a "helmet HUD."

The ugly.:

* Most of the UI is in an in-game computer screen, but it's a bit text-heavy, which may not be a great fit for VR.
* Current HUD targeting overlays aren't good for stereoscopic viewing; I'd need to solve some issues with that. I'm drawing those elements on a plane just outside of the ship, which is at least somewhat VR friendly, but having targeting overlays at a different depth than the ships is not visually comfortable. I can play with solutions, but I have limited time.

Any thoughts on if it's worth working through all of the VR issues? It'd be kind of a relief to "let it go", but a little disappointing too.

 

The game sounds really cool. I would stick to whichever platform you care about the most. VR and non-VR are completely different beasts. They are SO different that you can't expect your game to play the same way between them. You can keep trying to make a hybrid but it won't be as good of a game as if you had chosen a platform and stuck with it. Maybe you could have a full game on your preferred platform and a branch-off on the secondary platform?

If you try to do BOTH platforms by yourself you might never finish. Narrow your scope

Pros for VR:
-High risk, high reward. Maaaybe you could get a hit VR game and tap into this expanding market.
-Good experience
-Fun
-Novel tech

Cons for VR:
-Very few consumers have an HMD. The market will grow but slowly.
-Its hard.

 

@SniperEvan good points, thanks for the reply.

It's primarily a hobby game that I might eventually try to release on Steam and/or PS4 someday as a $5 game or something. But it's mostly for fun, so the market side of it isn't the most important thing.

The thing is, it's "mostly" working in VR already and I've already done a lot to support both. I suppose that's the "sunk cost fallacy" in action.

It probably would make sense just to drop VR support entirely. Starting to lean more and more that direction.

It might be more fun to do a true VR game with motion controls and all that anyway I suppose. Part of it is that I go to a VR meetup and it'd be fun to have a custom project to talk about and show. A small game (or even a non-game demo of some sort) that uses the full-room feature and motion controls of the Vive I'm sure would be more interesting for most anyway.

 

That sounds really cool. Is anything stopping you from ditching the standard monitor and committing to VR?

 

The stuff I listed under The Bad and The Ugly is pretty significant. Supporting standard 2D monitors isn't the challenge. It's the VR stuff that may simply present too many roadblocks for that particular game.

In general, I wonder if gaming isn't even that great of a fit for VR. The most compelling demos I've had weren't actually games.

 

I think, (as SniperEvan says) that VR market is expanding quickly and more good VR games are welcome. With the actual hardware and interfaces, the VR games have to be adjusted and trimmed to fit the VR limitations but in the future it will be greatly improved.
Your project sounds very promising so, stay on VR.

 

Eve: Valkyrie is solving the in-cockpit issues pretty well.

The technical issues you listed (image effects, cockpit not rendering) are surmountable.

The design issues are harder.

Do you absolutely need a text-heavy UI? I'd think in practical terms that a space explorer wouldn't want to be overwhelmed with a wall of text. Can you switch to indicator icons and audio alerts (like Jarvis or Siri)?

For on-planet locomotion, what about putting the player in a rover? Then the mechanics would work similarly to the ship's cockpit.

Maybe I have a stronger stomach for VR, but traditional on-foot locomotion really isn't so bad as long as my head mostly looks up and down. Rotating left and right is what gets me. So maybe you can help prevent nausea with appropriate environment design. Or then again just use a rover to circumvent the problem.

 

Looking forward to trying that when I get my Rift CV1. Should be a good reference for what I'm trying to do at least. The technical issues I imagine are solvable, but that's time I could spend on gameplay.

During normal flying around gameplay, it's not text heavy. There are some cockpit displays that should work fine in VR, with some target overlays and some system and "IM" type messages.

But it's got some Elite and RPG type elements with inventory, item equipping, skill trees, journal entries, mission descriptions, etc that are rendered onto an in-game computer screen. I also will want text that describes what the next mission goal is and so on.

When the computer screen is active, it moves closer to the front of your face (in 2D it takes up most of the screen). With positional tracking, you can adjust how far or close your eyes are to the screen, but with that much text, it's maybe not the ideal VR experience. Still, I think Elite Dangerous has similar screens and lots of people enjoy that in VR, so it may not be a huge deal. I'm definitely curious how the Rift CV1 and Vive screen resolution plays with the size of text I have.

Here are some examples of the UI elements I'm concerned about:

https://dl.dropboxusercontent.com/u/28342849/Full Burn/UI/FullBurn Info Feed.png
https://dl.dropboxusercontent.com/u/28342849/Full Burn/UI/FullBurn Installed Components.png
https://dl.dropboxusercontent.com/u/28342849/Full Burn/UI/FullBurn Installed Weapons.png
https://dl.dropboxusercontent.com/u/28342849/Full Burn/UI/FullBurn Cargo Bay.png

For navigating large distances, I was going to have a VTOL craft, though my current prototype plays better with a 3rd person camera. Not that that's a deal-breaker - you *can* do 3rd person in VR after all.

I wanted to have explorable "dungeons" like Fallout or Skyrim with a sense that you're exploring nooks and crannies for stuff, so a surface rover wouldn't fit in with that. There's a good chance that planet surfaces won't make it into version 1 anyway, but I do enjoy playing with all of the terrain tools out there these days.

Yeah, it turns out a lot of people get sick with traditional first-person controls in VR. I've attended a VR meetup for a couple of years now, and it's a common thing. There's a VR chat app called Alt Space, and the makers of that are very, very sensitive to that issue, to the point that you have to teleport to go to a new location. Works well with the Vive controllers. Just point on the ground to where you want to go and teleport.

I played Alien: Isolation for 30-40 minutes with DK2 once. Everything was perfectly smooth, but I got VERY sick and it took a while to get back to normal. So I get why it's a big issue.

A solution could be to let the player decide between traditional FPS controls and teleport controls I suppose. More work of course.

 

That's just what I was thinking. I'm still on a DK2. I'm contributing to a horror game (The Corridor) that has a lot of readable items, and while it works just fine it's not as crisp as a 2D screen.

True, but I feel like 3rd person breaks the "you are there" immersiveness of VR.

When I played the first prototype level of The Corridor in VR, I got nauseous because it invited a lot of lateral movement and rotation. The level designers have since built the levels with lots of straight corridors and straight up/down stairs. There aren't many turns, and they're mostly 90 degrees. I mapped the gamepad triggers to rotate immediately 90 degrees while the character "blinks" (quick fade out and in), which is similar to a teleport mechanic. I don't know how you'd avoid nausea with traditional locomotion in more organic, twisty environments.

Switching between FPS controls and teleport controls probably isn't that much work. But I do see you dilemma. Everything seems like a compromise to get it to work well in both VR and 2D.

 

Yeah, my first naive thought way back when was: "hey, it's easy to turn the game cameras into VR cameras, why not?" But traditional games maybe just aren't the best fit for VR.

I may be better off just letting the VR thing go and be more of a consumer of VR content for now. It would seem that VR content needs to be very specifically tailored for that experience.

 

The CV1 is definitely looks higher resolution, quite a step up from the DK2. From looking at your screens, I think the text will be legible, but would probably look better if it were enlarged - kind of depends on how much is in your field of view. the text under "OnPoint" and "Ninja Steal Solutions" (in the logo) would be too small, in my opinion.
If you have a small sample build, I can look at it under the CV1 and let you know.

 

I used to have text much larger, but it looked kind of silly in 2D and there just wasn't enough room for the text and UI elements. For little bits like super tiny text next to logos, I'm not concerned. But the player should be able to read the normal-sized inventory item names and stats and such. It's definitely too small to be comfortable reading with DK2.

 

Deciding if something is a good fit for VR is probably the wrong way to go about it. A VR title needs to be designed for VR from the ground up. Retrofitting an existing game is not a good idea:

1. You are going to change how the camera behaves. This is mandatory.
2. You are going to have to limit movement, speed, it's a game breaking thing.
3. VR is short gaming span. You can't expect someone to wear it for more than 5 to 20 mins. Beyond this is kind of unheard of.

With that said, some games may translate well, anything where the player can't move, such a vehicle, or cockpit or room experience, that can translate well with few breaking changes. You will find in vehicles, sickness is still a thing, depending on how violent the vehicle can move.

Until VR is 120fps all the time, and essential issues are solved with resolution, rendering performance and so on, I do not think VR is a decision to take lightly beyond party game experiences.

 

Yes and no. I designed a lot of what I have around VR the same time I got a DK1. Several elements designed for VR work out just fine in 2D mode. It wasn't until I got a little further along that I realized I was drifting away from an ideal VR experience.

It's not quite unheard of. Cymatic Bruce has attended a couple of the VR meetups I've gone to. (he's a big VR YouTuber, now works for Altspace). He said that in AltspaceVR, the *average* session length for their VR D&D games is six hours. In fact, the main reason they don't support GearVR for those game sessions is that the battery life isn't long enough and there are heat issues.

You also have things like 3D sculpting and eventually even using Unity in VR. With room-scale VR especially, I could see staying in VR for extended periods.

I do generally agree with what you're saying, especially in the context of games. I'm not sure how long of a session I'd want to do with my own game.

I'm definitely prone to VR sickness, though not as much as some, and I found my space game quite comfortable and natural even during dogfights. I'm more concerned about the other game-y elements. But there isn't any jerky movement or anything while flying. Though you can do pretty fast rolls and things, so it's one of those "your mileage will vary" things I'm sure.

Definitely agree.

 

Not sure what the averages are, but I've spent 8-10 hours in it at a time more than once.

As for the VR/No-VR discussion, I haven't read all this so am just going on the original post: What I'll be doing from now on is making products that run on both. It's not too hard if you design it that way from the start. I.e., GUI is in world space and uses a world space VR cursor regardless of whether it's running in VR or not. My current project has some sales, but at this point I'm not even sure if anyone but me has run it on the Rift. It's still early of course, but I'm not expecting big sales from the VR market any time soon on my current project. Best to target both probably if you want to do VR.

 

I guess it will be higher for developers. I'm not sure what the averages are either, but everyone who isn't a developer that I know, gets "full" from it by 20 mins or so. Must depend on game. But 10 hours isn't healthy - for anyone :/

 

People that play the AltspaceVR D&D game that use the Vive are walking around and interacting with people. That game lends itself to longer sessions it sounds like, since you're playing with a virtual tabletop and just kind of casually playing. You can have private discussions in different parts of the environment, etc. Doesn't seem that much different than doing tabletop gaming in person. I've not played the D&D game, but I did do a demo of AltspaceVR using a Vive, and it was a relaxed, natural experience.

If you're talking about having a screen that close to your eyes for that long, I don't know if there's a lot of data on that.

 

That's right there's absolutely NO data on that. So by default, assume it's just not healthy for your eyes at minimum. 10 hours inches from a mobile screen? ugh!

 

But 20 minutes is fine? Is there some intuitive max that just "sounds right?"

The max I've done is around 30 minutes, but I had to quit due to nausea.

 

I think as you know, it depends on the person. I'd guess that developers become hardened to it, so whatever personal limit someone has is going to be varied wildly. Personally I'd cap out at 20 min sessions until hardware improves, or even have periods of VR which are flat to ease strain.

 

When working in front of a screen, opthalmologists recommend the 20-20-20 Rule: every 20 minutes, focus on something 20 feet away for 20 seconds. So hippocoder's 20 minutes might be ideal. For eye comfort and to keep eye muscles from atrophy, it would probably be a good idea to do the same in VR. I know my eyes tend to get dry because I unconsciously blink less when staring at a screen or wearing a headset.

Then again, Thorsten Wiedemann spent 48 hours in VR.