Posted by Becky Gregory-Clarke on , last updated
After investigating the opportunities of Virtual Reality, we've learnt a lot over the last year. Here are a few of the things we've found out - which might help in your own experiments in VR.
2016 has been an eventful year for anybody keeping their beady eye on virtual reality. A set of high profile tech launches, which included the Oculus Rift, PlayStation VR, and most recently the Google Daydream, have provoked a lot of discussion from evangelists and skeptics alike.
In June, the BBC and our partners launched a set of VR and 360 video prototypes on BBC Taster tackling a range of genres and platforms, all aimed at being a small piece of the puzzle in a quest to work out what role VR might have in the future of storytelling, and what kind of technology we would need to do it.
If you’re hoping for a big reveal at this point…I’m afraid to disappoint you, the jury is still out! But we certainly know a lot more about the medium than we did last year. However, if you are contemplating doing some experiments with VR, below are some key insights and suggestions based on the projects we’ve done so far, that I would tell myself as a producer if I went back to the start of these projects a year ago.
1. Don’t underestimate the planning stage!
I think it’s fair to say that, for almost every VR prototype I’ve been involved with so far, we have probably underestimated the scoping and story-writing phase in each case. In fairness, this is partly because we were being ambitious and consciously trying to see what could be done in a relatively short amount of time, but in hindsight much of this is also down to there being a lot more variables to play with than for a linear, non-immersive story. For example, experimenting with the type and quantity of interaction, or blocking out an environment that will work with the story and be compelling in 360 degrees can involve a lot of trial and error.
With this in mind, allowing more time for iteration at scripting stage is a sensible strategy. It’s also important to get creative with your 360 degree storyboarding (I’ve seen some brilliant work from our partners using plasticine, lego bricks, tracing paper, cardboard models and 3D graphics!), do it early - and also get story-writers and technologists in the same room together as early as possible. Writers should expect the scripting to be a much more collaborative affair than they might be used to, but in the long run it will save a few headaches – as it will be easier to spot things that aren’t technically feasible or that might take a lot of effort for little gain.
2. Users won’t always do what you want them to.
If you’re an early VR adopter it’s easy to forget that not everybody will use the technology as naturally as you now do. A great example of this is the way we’ve observed that many new users still don’t necessarily look around very much when they’re wearing a headset, which can feel rather deflating if you have spent a lot of time creating an amazing 360 world and dreaming up ingenious methods to direct their attention to it! However, this doesn’t necessarily mean they can’t feel immersed or won’t enjoy it. In these early days, sometimes the challenge is about creating a good experience for people who don’t interact very much, but rewarding the people who do. This might feel frustrating from a creative point of view, but whilst the majority of the world has yet to try any VR, recognizing this behaviour may help to make it more accessible to the elusive mainstream audience. If you have any opportunity to observe real users trying out your VR content, or anybody else’s for that matter, it is definitely an insight worth having.
3. Photorealism is not the be-all and end-all for creating believable experiences.
Animation or stylised computer generated visuals may not be the first port of call for many forms of traditional storytelling, but for immersive experiences they are worth a second thought. We Wait (a story about a family of Syrian refugees trying to cross the Mediterranean) and Easter Rising (about the 1916 uprising in Ireland) both tackled serious topics with a visual style that might not normally be considered appropriate for a typical current affairs or documentary piece, but ultimately made it feasible for us to depict both a current and historical event in a way that we would not otherwise have been able to do. It’s true that this style is not to everybody’s taste, but in general this did not appear to be a big hindrance to the sense of immersion and emotional engagement that we observed. Viable, live-action ‘true VR’ experiences may be on their way, but for now my advice would be not to discount the freedom that animation or CG can offer on account of wanting your story to be taken seriously. Put another way – when faced with limited resources, I would suggest prioritising other things above trying to achieve accurate photorealism, such as good sound and well thought out interactions.
4. Porting between platforms doesn’t necessarily give consistency of experience.
We tried to do a ‘simple’ port of an experience built for an Oculus Rift to a 360 video that could be viewed with a Google Cardboard. Of course, we knew we would lose the interactive elements from this transition, but we still felt the story was enough without them to make it worthwhile, so that more people could see it. In this case, although this worked from a technical point of view, the experience itself did not feel consistent in terms of its feel and emotional impact. For example, characters felt smaller and further away, which made the story feel less intimate. The lighting also wasn’t quite suitable for the smartphone display, which detracted from the atmosphere, and even the length of the experience felt a little too long when viewed on a lower-end device. Since then, I have also seen colleagues trying to make the transition between other platforms and finding that everything requires a lot more effort than originally thought. To many techies I’m sure this will not be surprising, and most of the issues I describe are fixable, but to anybody setting out on a VR adventure, the lesson here is not to assume that the transition between platforms is just a technical job – it may require new editorial thinking too to make sure the experience is consistent in terms of the impact you are trying to make. From a resourcing point of view, don’t assume that building an experience for one platform or display means you will get another for free, even if you use the same game engine (e.g. Unity) to do it.
5. Spend more time on sound. R&D’s sound-lead VR experience.
The Turning Forest, has really lead the way for us in terms of showing where immersive, dynamic, binaural audio truly comes into its own. We Wait also used spatial audio elements, and many users we tested with noticed and appreciated the way the audio felt ‘accurate’ from their point of view. Personally, I would put high quality spatial sound near the top of the list for any future immersive projects. Well-crafted audio alone can have an incredibly immersive power, but if nothing else, it only makes sense that your amazing immersive visuals are accompanied by equally high-quality immersive sound – they deserve it! For more on the techniques used in The Turning Forest have a look at Chris Pike’s BBC R&D blogpost.
6. Spare a thought for the Web.
Ahh the web…let us not forget you. We have made one web-based VR pilot so far, Rome’s Invisible City VR, which used Web GL and Web VR to create an experience that could be viewed within a browser across phones, tablets, and even desktop, with no downloads required! This experiment has been very promising, although of course, there are limitations to what can be done here at the moment, with one of the main issues being the fact that not all browsers support this yet, and those that do are not all doing it consistently, so the experience isn’t always very stable. The web standard is in its early days, but a recent trip to the first W3C Workshop on Web and Virtual Reality has given me much reason for hope. The volume of thought and enthusiasm going into this area even at this very early stage is really encouraging, and the stage is set for what could be the only open VR standard going. So the takeaway here is not to overlook this possibility for your project, you never know – it might offer you the interoperability and the chance at higher audience reach that you seek.
7. Smartphone experiences fare better with a headset.
We created and tested with BBC Audiences a couple of VR prototypes that could be viewed on a smartphone either as a ‘window on the world’ experience (i.e. you can wave your phone around in 360 degrees to discover the virtual world around you, without putting on a headset) or as an experience that you could pop into a ‘smartphone adaptor’ headset such as a Google Cardboard. We noticed that users tended to rate the pilot more positively, and were more likely to say they would try more things like it in the future, when viewed through a headset. In fairness, there are probably several angles to this, such as an element of headset novelty that is difficult to discount, plus the fact that in some cases the pilots were probably slightly better optimised for use in a headset rather than simply on a smartphone screen. It should also be noted that this finding was more pronounced for more VR-like content (i.e. those that had immersive features such as spatial sound, and interactive elements that the user could control) than for straight 360 video. My personal interpretation here is that we should find it very encouraging that people are responding well to the headsets and enjoying the format, although for now, offering these two options is still a very good way to try and keep audience reach as broad as possible, while people don’t have headsets at home.
8. It’s tempting to try a lot of new innovations all at once, but it pays to limit your variables.
The same can be said for any R&D-type project of course, but it really holds true here so it’s worth laboring the point. For example, you might be interested in trying out clever user interactions, narrative structures, graphics, avatars, methods for moving people through space or between scenes, but you don’t necessarily need to play with all of these things at once, and in fact, it’s probably a bad idea. Decide what the most valuable thing is that you’re trying to explore, and then keep everything else as simple as you can (especially as you can often find that even the simple stuff can prove tricky in VR Land!). This way you’re more likely to be able to draw some meaningful conclusions from the work, but also make it more manageable and keep scope-creep down to a minimum. Honestly, unless you’ve got a team of seasoned VR pros with a Hollywood budget, then if you think you’ve kept it simple…stop…and make it even simpler.
There are lots of reasons to be positive about VR, although with so many factors at play I suspect it will be some time before we can be sure exactly where we are on the hype curve. Until then though, some cautious optimism with a good dose of pragmatism feels a sensible way forward. Happy experimenting!
More on Virtual Reality and 360 Video:
This post is part of the Immersive and Interactive Content section