CGTrader at GDC 2016: Creating The Killer VR App

Game Developer Conference 2016 is over and the CGTrader expedition crew is back. We’re extremely excited and we have a lot of knowledge to share with a lot of you. This one is dedicated to you - 3D studios, agencies and VR developers out there: virtual reality is coming, and you better be prepared!

The VR killer app has not been established yet. At this stage, many conventional video game genres are getting ported to VR, but there is no clear leader in the market yet. We still don’t know which genre will benefit from VR the most. What is obvious is that free, easy to use game engines and software, as well as stock 3D model markets make the development and iteration accessible to everyone. The playing field has never been so level, and the development software gives us unparalleled access to the market. If you have an idea and the guts to carry out, VR is your oyster.

Prototyping and iteration are crucial in VR. Your assumptions about how users will interact with the VR app are probably not true. You must test, iterate and test again, all till you reach satisfaction. Use white-box testing, develop prototype apps, and playtest them as much as possible. You will have to be agile like you have never been. When developing the movie about Henry the hedgehog, Oculus Story Studio had to put prototyping after story development - something new to the film industry.

Developers Kite & Lightning had some other suggestions for developing VR apps - in their talk they suggested iterating in your mind: imagining as many details as possible, considering sitting or standing depending on the experience, trying to feel the experience both in your mind and how it would play out in real time. Kite & Lightning go so far as stating that speed is more important than skill, because you can learn faster by trying things out.

Frame rate rules the world. No amount of design or gameplay will matter if your user keeps throwing up. Great apps will have to offer 90 FPS to overcome motion sickness, and that has to be achieved on any platform. Even with the launch of Oculus and the others, it is projected that 75% of headsets will be mobile-based. This means working around the hardware requirements and optimizing assets as much as you can. Schell Games, the authors of a popular VR game “I Expect You To Die”, told about considering to create a game level in the bathroom, but deciding against it. The polygon budget was way too high, with flowing water and towel materials putting a serious strain on the system.

And don’t forget such standard tricks like occlusion culling (Umbra3D works specifically with that) and LOD (use soft like Simplygon, also the functionality also comes with some game engines).

Presence is more important than gameplay. Schell Games had one message: you need to be immersed in VR, so it is extremely important to keep that belief. Do not scare users to the extent that they would need to rip off the headset to remind themselves that it is not real. Do not make them sick. Make sure the experience is not confusing.

Also, requirements for realistic audio are much higher when you are in VR. If you drop a ball in video game, the sound effect doesn’t matter as much, but in VR, you have make sure the sound matches the material of the surface, be it metal or wood.

Make the goals clear and intuitive. The goals in the game or an app should be obvious. Be careful with switching them! The first level of “I Expect You To Die” was so successful because when you appear in a car, you intuitively want to drive it. Their other levels did not work as well and users did not know what to do, even after watching tutorials. Meanwhile, for the first level, they were using easy-to-get long term goals (drive the car), medium term goals (start the car) and short term goals (find the key).

Think about the role of the user. Who is the user playing in the game? Is he / she a ghost? Do other characters acknowledge your presence? Oculus Story Studio said that their Henry character was a lonely creature, yet they still wanted to make Henry look at the user at one point. This would have made things confusing, so Henry became not lonely anymore. Another issue is is working with hand controllers: how do you represent the hands? Male? Female? Black? White? Schell Games solved this issue by using black gloves.

New monetization models will have to emerge. Free to play does not work well in VR as the ubiquitous monetization reminders break the immersion. It is not yet clear what monetization models will find purchase in the virtual sphere.

And there you have it. The VR is here; it’s so close, you can feel its breath at the back of your neck. You just need to adapt your production processes to the vagaries of the new medium, and you’ll be set. Now is the time to get your foot in the door and set yourself up as name on the market, lest you have to fight the crowd later on!

For advice to modelers, check out this article!