talking about Virtual Reality revolution in filming
The two minute VR film created by Cutting Edge for the Samsung Gear headset, lets people experience the benefits of Advil first hand, inviting them into a high stakes, hyper-real ride that leaves pain for dead. The campaign also includes a combined CGI and live-action commercial and seamless 360 online video. Cutting Edge director and VFX supervisor Simon Maddison led both the VR experience and TVC.
VFX SUPERVISOR / ANIMATION DIRECTOR
Simon was co-owner of the internationally renowned visual effects house Fuel VFX for 13 years. Through hard work, a driving passion for amazing visuals, and an uncompromising dedication to quality, Simon has built an unparalleled reputation in the industry which he now brings to Cutting Edge. Trained in fine arts and photography, Simon’s formative years in visual effects was in 3D animation.
How was your team involved with the project?
We were originally contracted to direct a 30 second TVC for Advil via agency Matterhorn. VR is something a few of us here at Cutting Edge have been interested in for quite some time and when the concept for the commercial was presented to us, we immediately saw how well it lent itself to a VR experience. During the shoot, I mentioned this to the creatives who also got pretty excited. As the post-production commenced, we also began putting together a little demo to show what it might become.
Unlike regular film production, VR IS SOMETHING THAT IS HARD TO PUT INTO WORDS. Normally when writing a treatment for a 2d project, I would use a lot of references that people are familiar with. Other films, commercials, photography etc. With VR, few people have a point of reference for it. You can describe the immersion, but anyone who has ever put on a set of goggles and experienced something great knows it’s not yet something that is easily described. So with our little demo in hand, we took it to the client with the agency, who also fell in love with the idea.
And what was your reaction when you heard your team would be working on such an unusual project, such as Advil VR?
To be honest, we weren’t sure the client would actually jump at going into production. Luckily for us, they got on-board almost straight away. This got us all pretty excited.
The concept for the world was really borne out of the brand. The things that were important to the client included the grid and the liquid – two key parts of their message. It was from here that we developed an environment. Once again, as this entire project was borne from the commercial, we already had a well-established look and feel. It certainly made the development of the VR experience more straightforward for everyone. Surprisingly, what differed in the VR experience was less conceptual and more about distances.
THE VR REVOLUTION
Did you pay attention to how people will interact with the world and what impressions are more memorable?
We sure did. Blocking the action for a two-minute journey (where people can look wherever they want) is a very different framework to composing shots and standard editing practices. In VR there is no lensing or editing. The frame is the entire world, and in my mind, cuts are very jarring so until we all discover how to get around this, they don’t really work. Inevitably what you’re left with is finding a new language for storytelling. Blocking the two minutes became a process of how to lead the eye where we wanted it to go. There still has to be a narrative. Audio was the final part to that puzzle. How do you get someone to look right or behind? One quick way is to make a sound occur in the place you want people to look towards. We worked pretty hard with the sound company on creating binaural audio that would help steer our audience at the moment the narrative needed it.
With VR, people could see the whole world and not the part that the director would like to show. Do you use any new approaches for getting final "shot”?
That’s a very interesting question. I think in this instance, because the design had already been established in the commercial, it was much easier to get the ‘look’ signed off. But because it was one long two-minute experience we had to think about it in different terms to traditional ‘shot’ production. It’s really the production of an ‘experience’ so people aren’t concentrating on the kinds of things they would normally discuss.
For example, we never once parked on a frame and discussed it in the kind of detail we do when working on a commercial or film project. The questions were much more about ‘how did it feel’; was it immersive enough, did people look where we needed them to look.
We also needed to navigate with the client how to best review this stuff. We were working with the Samsung Gear VR, which has no output to a second monitor. That means we had no way of knowing where the person reviewing it was looking. We found the best way was to have the client watch it in the Gear, then we had the unwrapped 360 mono QuickTime play on a screen so we could all review it while looking at the same thing. It was new for everyone but we all got on board pretty quickly, which was great.
THE BIGGEST CHALLENGE
The team on this has had a lot of experience in stereoscopic film making due to our history on film, so luckily we didn’t have such a steep learning curve on that front. What was a bit of a challenge was the language of stereo in a VR environment. For example, do you see the hands or are they just a disembodied point in space on the bike? If we do see the hands, do we see the rest of the body and at what point does the convergence become uncomfortable?
However, I think the real main challenge was rendering. One shot that was two-and-a-half minutes long meant any change to the look required the entire thing to be re-rendered -- at 60 frames a second. Also, you see almost the entire environment the whole time so there was a LOT of 3d geometry that had be loaded at any given time. Couple all of that with the fact we were rendering two ‘eyes’ at 2K each (that’s a total of 4000 pixels by 2000 pixels) and that you’re not getting ‘shots’ locked off along the way, means you have a gigantic rendering bottleneck at the end of production. It actually became a massive issue in the end and we ended up relying on a cloud based render farm to get us through.
Could you say that VR brings revolution not only into the advertisement industry but it could influence visual effects production too?
I think we’re at the beginning of a pretty large revolution across the board. There are the obvious things we need to tackle as a VFX community, like stitching software, volume of rendering and viewing platforms. However knowing what we went through when stereoscopic filmmaking came along, and the challenges we needed to overcome, means now we’re well equipped to deal with VR requirements.
I think the biggest revolution is going to come in the form of the content. VR is certainly not something new, it has been around for a very long time and in academic, medical and military circles its development has never really slowed. What is different now is that suddenly a lot of people are going to have access to it as an entertainment platform. In my opinion, its success is going to come down to what we as storytellers create for those audiences. How quickly can we discover what the new language is and can we shift our idea of how to present a narrative fast enough for people to stay engaged with it? I think we can and it’s a very exciting time. It’s been a while since we’ve been able to pioneer to this extent so I can’t wait to see what’s ahead.