AMD’s Roy Taylor discusses what’s going to be big for VT in 2017 – location-based VR, more complex content, and experiences good enough you’ll be willing to pay for them.
Roy Taylor is ‘the VR guy’ at AMD. His official title is corporation vice president – content and technologies, but what this really involves is working with the entertainment industries to help them produce better and more engaging VR experiences. Better VR means more people and companies buying VR kit, which require much more powerful graphics chips than your average PC – so if it takes off and AMD’s marketed itself well, the company will sell more graphics cards (and chips inside laptops and the PS4, which has its own VR helmet in the PS VR).
Hollywood-based, and working with the film and game industries (and the ‘digital entertainment’ space that sits somewhere in-between), Roy has an overview of where the industry is headed. Ao on a recent trip to London, I caught up with him to hear his views on the challenges and opportunities for VR in 2017. Of course, working for AMD, his insights are coloured by AMD’s heavy presence in high-end VR but not in mobile (ie, phone-powered) VR. But in ‘desktop VR, AMD is almost platform-agnostic – its chips powering PCs that support both the HTC Vive and Oculus Rift, and inside the PS4.
Before looking to the year ahead, I asked Roy’s perspective on where we are now – for which he has a clearly well-prepared answer.
“We're at the roughly equivalent to about where the film industry was in 1905,” says Roy. “In 1905, the Kinetoscope had been out since 1891, so the first pieces of content were just starting to be made, but it was with the introduction of the Nickelodeon [early cinemas] in 500 locations across America when it took off.
“We're kind of in the same place now. We know that [VR] offers immense promise, particularly in film. – and to a lesser degree in games. But we haven't yet got the killer piece of content that makes us all rush home early because it's so awesome.
“This is going to come along – VR going to ‘happen’.’
Roy says that the biggest issue that the industry has right now is the relatively low number of users/potential customers for content creators. He quotes Jon Peddie Research’s stats that 750,000 VR headsets will ship before the end of this year – with 2.7 million more shipping next year.
“So in the next year, it will be still an installed base of around four million,” he says. “Even if he's wrong by a 100 percent, it’s still [only] eight million.
“A key issue we have is that modern content today costs around a million dollars a minute to make. Interestingly, it’s the same for whether it's [a game like] Battlefield 1 or [a movie such as] Ghostbusters. You've got to find a business model which will support those levels of investment. If you only have an installed base of four or eight million that's not big enough for you to go and get revenue of say $400-600 million.
“Roughly speaking, both games and movies are looking at a 4-to-1 ratio, so you spend a 100 million to make it you want to make 400 million. That's how you cover other costs. We need a bigger installed base.”
The best way to get more people to use VR, says Roy, is what the industry calls ‘location-based VR’. This is a catchall term for high-end VR experiences that take place outside your home using VR kit provided by the theme park, museum, shopping centre, cinema, film festival or games arcade that they’re located in. Notable examples we’ve covered include the Game of Thrones exhibition experience, the BBC’s Spacewalk (below, which debuted at DocFest in Sheffield), and the VR rollercoaster ride at Six Flags in the US.
These not only make money for the content creators – assuming that audiences are willing to pay extra to experience them, as they used to for ‘4D’ rides and arcade games – but if they’re done well, they encourage people to buy the hardware to have their own VR experiences at home. And if enough of them do, that's when VR becomes ‘a thing’ rather than a niche pastime.
Hardware alone can’t achieve mainstream success for VR – compelling experiences is what really makes the difference between VR being a fad like, arguably, 3D and a must-have-that-becomes-the-norm like HD. Loving what you play/experience builds the word-of-mouth that spreads the medium virally, like playing WipeOut on the PlayStation round a friend’s house in the early 90s made us all want to go buy PlayStations.
While we’ve seen a lot of VR content from companies with backgrounds in games, post-production, VFX and film – it’s the latter that Roy thinks likely will produce the best VR experiences in the immediate future. It’s not that the others can’t – he cites Bethesda’s VR version of Fallout 4 (below) as one of the best VR experiences around – it’s just that filmmakers have the most experience at telling the compelling stories that will really engage people.
“The film industry knows how to use technology to tell stories. They've been doing that for a 120 years. I just spent the morning with 200 of them [at an event at BAFTA] and they're very keen on the technology – as long as it’s used to help tell the story, not get in the way.
“[They get that] the greater degree of the immersion, the greater degree that we like it. One of the best pieces of VR from that point of view is Paranormal Activity (below), because even though you're in a room and you know people are staring at you, you still get frightened. It still makes you jump. I think they do a terrific job with that.”
Roy also mentions The Martian as an example of a notable film-related VR project. However, he notes, it wasn’t a complete success with audiences.
“I took an aggregate of the all the online reviews I read,” he says. “A lot of people liked it a lot. But at $20, they thought it was overpriced. They did not like the cutscenes. When I experienced it, I didn't like the cutscenes either, but I did like the interaction [and] the narrative. I thought the end of the experience when they shoot you off into space was terrific.”
This illustrates the challenge for VR content creators. We’re still experimenting with the form, but we need to keep giving customers experiences that don’t put them off. And as/if the audience grows, we’ve got to adapt to the changes in who those users are – both their interests and experience with VR. The non-interactive parts of The Martian that didn't appeal to the early-adopter gamers who played might have appealed to a wider audience.
What we have learned so far, Roy says, is that some formats work well and others don’t – and that the human brain can take in a lot more information in VR than it can in on a 2D screen.
“There are certain situations where you will never want to replace having control of the camera – romances for example. Romance doesn’t lends itself well to VR,” he says. “On the other hand, big complex scenes with lots of camera angles do lend themselves very well. [At BAFTA] I showed two clips that are very good example of where VR would work even better than a traditional camera.
“One is the donut scene from Boogie Nights (below) – one guy shoots another guy, who shoots another guy and then everyone's dead. There's nine different cameras angles in five seconds, because it's complex to follow everything that's going on."
Roy says that our brains would have processed that scene in VR better – as we would in reality – because our brains are designed to take in a lot of data from a three-dimensional world almost instantly, in a way it doesn't from a flat screen. It does that because, back when human beings were at the mercy of predators, our lives depended on it.
“The second one I show is the [Battle of The Bastards] battle scene from Game of Thrones. Towards the end of the battle, [one side has] lost and they're completely surrounded. Then the cavalry turn up.
“There's a lot of characters in that scene and you're trying to follow the plight of the characters. You have a lot of close-ups. Then you have distance shots to try and allow you to take in the fact that they're completely surrounded. Then finally when the charge arrives, there's a lot of aerial shots so you can understand what's taking place. That would have been great in VR.”
VR for families?
In contrast to those last two scenes. The last successful big leap in how we interact and experience content was with gestural control from the Wii and Kinect. A large part of its success was driven by families – but VR is very much for adults only. I’m not talking about VR porn, but that VR headsets are shaped for adult heads. There are also concerns about what VR experiences could do to children.
“Whether it's suitable for children, I think that that still needs to be worked out.” says Roy. “The answer is to commission the research and I don't believe that anybody has done that yet. They're just being cautious by saying ‘this is not for under 13 year olds.”
Physical and psychological issues aside though, Roy sees a lot of potential for VR for family entertainment and education.
“I think one of the most wonderful pieces of VR so far is Google Earth,” he says. “It's fantastic. I had a VR setup in my house over Thanksgiving. We had the usual Thanksgiving friends over and [Google Earth] was by far the most popular [experience] hands down. Everybody loved that, going in and exploring where they came from or went on holiday.
Roy mentions a story from a contact who had trekked across the Himalayas when he was younger – and liked to go back and retrace the route he took using Google Earth – and it’s easy to see how the immersion could bring those locations to life for students too.