How Portal and Unreal Tournament solved VR's movement problem

How Portal and Unreal Tournament solved VR's movement problem
Matthew Jarvis

By Matthew Jarvis

April 12th 2016 at 12:31PM

HTC Vive launch title Budget Cuts is one of virtual reality’s most versatile experiences yet, requiring players to shoot, throw, duck and crawl in order to sneak past an army of murderous robots. Develop caught up with Neat Corp’s Joachim Holmér to ask why he threw virtual knives at a wall for a day and find out why stealth is the perfect fit for VR

What was the inspiration for Budget Cuts?

It came from lots of experimentation. We didn't really know what we were getting ourselves into with VR, so we did two weeks of prototyping to see what works and what doesn't.

What made us go the route of first-person stealth was simply that an enemy pointing a gun at you in VR feels very intimidating. A stealth game would enhance that feeling of being watched, triggering all the instinctive reactions for ducking and hiding.

The game uses a unique movement mechanic, combining SteamVR’s room-scale VR with a Portal-like gun. How did you come to rest on this combination as the best way to traverse Budget Cuts’ environments? What challenges did you face when building in the mechanic – such as the player’s ability to move quickly if spotted or to specifically position themselves to remain hidden?

We decided early on that anyone should be able to play the game, without motion sickness. This means joystick movement is immediately ruled out, along with any other methods using continuous acceleration.

In our case, we still wanted to be able to move around large levels, so the first test we did was a simple point and click to teleport. It works well, and doesn't cause any motion sickness, but it is pretty jarring. Simply flashing to teleport to a new location is harsh when there's no transition.

We then decided to try a combination of the translocator from the Unreal Tournament series, and the portal from the Portal series. The translocator in UT fires a small pad that flies in an arc. You can then press a button to teleport to that location.

We're doing the same thing, but when the pad lands, we also open up a portal, allowing you to see what the other side looks like, before teleporting. When you then press the button to teleport, the portal wraps around you, and you're then standing in the new location.

It has worked very well so far. The fact that you can see where you'll end up, makes sure that players know where they will stand in relation to any enemies that may be near that location. The fact that the portal pad flies physically in the air, makes for a natural limitation as for why you can't just immediately teleport anywhere how quickly you like.

"An enemy pointing a gun at you in VR feels very intimidating."

Joachim Holmér, Neat Corp

Players can use a variety of weapons to defend themselves in Budget Cuts – from a crossbow to throwing knives. How did you balance the use of the weapons by players to maintain a fair, but realistic level of accuracy in VR?

It generally works out pretty well. The crossbow is pretty straightforward, we've added iron sights if you want to aim accurately, which works well for people as long as they've seen where they sit in the weapon. The dart does have gravity and a travel time though, which is extra noticeable for longer distances, which you need to compensate for.

The knives on the other hand, could almost be an entire story in and of itself.

There are many factors that affect how well you can throw knives in VR. To name just a few: your throwing skills in reality, your ability to predict arcs and motion, the path your hand and arm takes when throwing, your sense of when – during the analog trigger release – the object is actually released and how well the code can parse the sparse data given to it during the fast hand motion.

We started out by simply reading the tracking data through SteamVR. They give accurate data, but only at a specific framerate. In reality, lots of motion happens between frames, especially when moving your hand quickly. So reading the raw data worked pretty well, but I wanted to see if there was room for improvement given how central it is in our game. 

I dedicated a whole day for improving throwing in the game, I set up a scene with 88 knives, and five metres of distance to a target on the wall. I then tried to hit the target with all knives as best I could.

I repeated this process for about eight different iterations of the code, and found some ways to improve it.

To select from the variety of weapons available, Budget Cuts uses a unique inventory system. How did you go about creating a UI that works in tandem with the Vive’s motion controllers and still communicates enough information to the player about how to perform specific actions?

The more we work with VR, the more we realise how important it is to move all interactions from a panel/UI style to a physical in-game representation instead.

Even if people are used to menus, it's simply harder to understand them in VR, because you're no longer using a screen and a cursor, you're in another world.

The inventory system is one such physical system, which almost everyone immediately understands. You hold a button to open the inventory, you release to close. While it's open, you have a bubble where you can drop items to store them. All your stored items show up physically above this bubble while it's open. Given how physical they are, we don't need to explain that you can grab them using your other hand, since this is what you would do with any other item.

Contrast this with our tool selection system, which people have a very hard time understanding. This is a simple radial menu, similar to what you have in many weapon selection systems in console shooters. But on a touchpad, showing a UI panel in VR, is simply hard to understand. We're going to redesign that shortly, simply because the panel approach, again, fails to communicate well enough.

 

Budget Cuts is (to my knowledge) one of, if not the, first stealth titles in VR. What makes the genre suitable for VR? Does it offer any design challenges unique to the genre versus other types of gameplay?

We found stealth to be perfect for VR, and especially room-scale VR, given how your body is controlled by, well, moving your body. There's no limit to how you want to stand, crouch, sit, lie down and so on. Do you want to hide under a desk? Well, just hide under the desk. There's nothing stopping you from doing all of these natural body movements, which is very fun in this genre.

It fits VR very well given how, in VR, you're no longer controlling a character, you are the character in the game. This makes a huge difference, because now, you feel presence like you've never done in games before. Sadly, it sounds like a bunch of buzzwords that most people will just shrug off as PR talk. It's near impossible to communicate how real it feels. You really have to try high-end VR with tracked controllers to feel presence yourself. It sounds like a cliché, but in this case it has the virtue of being true.

As for the design challenges of stealth versus other types of genres – the biggest difference is how locomotion works. In many stealth games, walking and sneaking in elaborate paths is commonplace. That's something we're missing out on, because there's no long-range walking at all. We have to design it with our translocator in mind, which means that the player can often teleport to locations that would be impossible to walk to undetected. We have yet work out all the level design and game design implications of it. We're still very much in an experimental stage of the game, even though we've got the core mechanics down.

As well as walking around the Vive’s play area, Budget Cuts requires players to duck and more for a variety of in-game actions. Does the game adapt to work with a reduced play area? Do you believe the space requirement for the Vive could limit the game’s audience?

The playspace we're designing for is a cylinder with a radius of one metre, which roughly translates into a 2x2m space. You can get by quite well even if you're just standing still, but the crucial part is that you need to have controller tracking in all directions and have some space to reach for items. You can probably technically play the game in a 1x1m space, but there's a very high risk of you accidentally hitting reality with your controllers or your head, and it'll be a very uncomfortable experience in general.

The space requirement is something that does limit the audience, the same goes with 360-degree controller tracking, which Oculus Rift and PlayStation VR aren't pushing for given their front-facing camera setup.

It definitely limits our audience, but we didn't start out thinking ‘What game would be commercially viable to make?’ we simply wanted to experiment with the Vive until we found out what was exciting to play. For us, VR wasn't all that interesting until we tried controllers in room-scale, so that was our starting point for Budget Cuts.

That being said, we are going to experiment to see if we can make a version of the game adapted for a front-facing camera setup, but we'll keep the Vive version as-is.

"We didn't start out thinking ‘What game would be commercially viable to make?’ – we simply wanted to experiment until we found out what was exciting to play."

Joachim Holmér, Neat Corp

Objects needed for progression are hidden around Budget Cuts’ environment – in drawers, under tables, in vents etc. How do you tutorialise the player so that they are encouraged to explore in VR and interact with a greater number of objects than they might in a ‘traditional’ title?

Usually we don't have to tutor that part very much – players naturally poke around with everything, simply because it's fun in VR. The only part we need to establish is that some of those objects matter in terms of gameplay.

In the current demo, we introduce this quite quickly and early on, where you have to find a key to open a safe. That being said, even ‘useless’ items can be used to trigger sounds to lure enemies.

What were the key tools, methods and technology you used to build Budget Cuts?

On the hardware side, we've been using the HTC Vive from the start. As for the engine, we're using Unity, with a few plugins: Valve's SteamVR plugin, our own Shader Forge plugin and DOTween. For art, we're using Maya and Photoshop primarily.

What development challenges remain for VR? How can these be overcome?

I think one of the bigger challenges is how to reach a bigger audience, how to convince people of VR. As mentioned earlier, you have to try it in order to be convinced by it. Many people who haven't played proper high-end VR with tracked controllers, simply don't see why it's such a big deal. They often compare it to the Kinect, or the Wiimote, or the previously failed VR push, or 3D TV. The list goes on.

What’s next for Neat Corp and Budget Cuts?

We're far from finished with Budget Cuts, so we're focusing 100 per cent on finishing the game, until we're hopefully done by the end of the year. Time will tell what happens after that.

This article is part of our month-long Virtual Reality Special. You can find more VR content here.