In our new series How To Tech we highlight the development culture of startups. We asked Avinash Changa, CEO of WeMakeVR, how a VR company deals with creative processes, rapid change in development and a huge technology stack.
First, can you tell us about the technology stack at a VR company and everything you have to take into account?
Working with VR creates a long and diverse list. But it basically comes down to both software and hardware. First, there’s user hardware we work with, like the Oculus Rift, HTC Vive , which are high-end pc-based VR devices. Then there’s Playstation VR, which is obviously console based.
Then there’s mobile VR as well, like the Samsung Gear or cardboard-like VR. For all those types of machines you need to use software to play VR on those devices.
For both cinematic VR (realistic experiences) and gaming VR (interactive in-engine experiences) we use the Unity or Unreal engines to create 3D environments. For certain applications you need to develop your own engines, especially if you want to push performances.
To make assets in 3D we use programs like Maya and Cinema4D, and import them into Unity for instance. Next to that we use anyting from Indesign to Photoshop to polish those assets.
We also developed our own stereo-3D VR-camera which captures everything around it. A lot of current 360-degree cameras leave black spots in the image, because the camera is standing on a tripod. That’s why in some 3D-environments you don’t see an image in the top or bottom.
We use several tools for ’stitching’ the captured images together. Again, a lot of possibilities here. You can use Aftereffects, Nuke, or you can do it with self-made algorithms. At WeMakeVR, there’s no standard in doing this; unique projects often require unique solutions.
You said that you sometimes develop your own engines. Why is that and how do you decide which one to use?
It all depends. For clients, every 3D-environment is customized, so we need to think about the experience we want to offer, not about the technology or skills we have. Every 3D or VR-tool has its perks and downsides, and it’s really nuanced. So we mix it up every now and then.
How about the process of creating VR-experiences?
When a concept is ready, both the tech team and the film team have a look at it. The tech team for instance comes up with new ideas on how to use a camera, and the film team thinks about which stories we want to tell, and what kind of shots we should use. Both teams go back and forth all the time, working together.
We don’t scrum because all of our projects are unique and different after all. Scrum is great for dynamic, non-standardized workflows, but you need to at least have a few standards in place. Right now we still need to customize a lot.
And how do you manage such creative processes?
Slack and Trello are very important to us to keep track of everything that needs to be done. We make tasks in Trello and the teams communicate in Slack. It’s important to log all of these activities. Google Docs is also part of our toolset. There are some standards though. For instance briefings. But next to that, everything goes very organic because we’re a small team of eight people.
Also, on project level, you need a good mix of asset and resource management. We used Shotgun for that – it’s flexible, but not for VR. So even in managing projects we tend to customize.
How does a typical workweek look like at WeMakeVR?
Every week we have status meetings (briefings), we work on production planning, creative development, post production and hardware development. Next to that, nearly every single week there’s some event going on where we do presentation and give people the opportunity to experience demos.
Can you tell me something I don’t know about working in your team?
Everyone in our team is a generalist, a jack of all trades. So everyone who plans the project can also film and upload and process the images. That really benefits our team, because that way, everyone can come up with new ideas. Even from someone you didn’t expect it from.
Because of this multidisciplinarity, we really ’pivot’ a lot. And we need to, because developments in VR are going fast. That’s why we tend to move in that same pace.