Dr Jonathan Waldern is standing in a meeting room in Silicon Valley, wearing a lab coat and grinning from ear to ear. “There will be people crying in this, people falling in love, people falling over. For all sorts of reasons, this strikes at the core of being a human being. It’s so compelling … this is as big an opportunity as the internet.”
He’s talking about virtual reality (VR), the immersive systems that cover your eyes (and ideally ears) and fool your brain into thinking you’re somewhere, or even someone, else. Waldern, 54, is a veteran of the business: in the 90s, his company Virtuality was a pioneer in commercial VR, offering an arcade game experience in which a bulky headset transformed the room into a prehistoric scene in which pterodactyls swooped over and around you, and you tried to shoot them with your gun.
The last time I saw Waldern was in the UK, more than 20 years ago. Virtuality, which he had brought to the London stock market, was in talks with Atari to bring its ambitious Jaguar headset to consumers. But the deal fell apart, and then so did Virtuality. The market for VR shrank to zero almost overnight; its speed and completeness foreshadowed the dot-com boom and bust of 2000-2001.
But, just as the dot-coms evolved and returned like mammals emerging from a meteor strike, VR has come back too. And that’s what Waldern, who now runs a Silicon Valley company called Digilens, is excited about. He sees the field enjoying a resurgence, its long hibernation is over. And survivors like him are emerging to capitalise on the new boom.
“It’s a very small community [in VR] around here,” Waldern continues, referring to the west coast. (Though it transpires that when he says “small”, he means “in the thousands, but geographically close”.) “The cost of development in this realm is so high that very few corporations would take it on. Only the big three – Microsoft, Google, Facebook. Apple is the fourth, but isn’t so prominent.”
Does that mean Apple is looking at VR too? “I couldn’t possibly comment,” says Waldern, and grins. (Apple filed a patent in September 2008 for putting an iPhone into a headset, and another in 2014, but there’s been little other sign of activity on that front from the company.)
Even if it isn’t, the presence of the other three big companies in this space shows how hot VR has suddenly become. Oculus Rift, bought by Facebook for $2bn in March 2014, has been the poster child for the new face of this technology; the company now aims to launch a headset for consumers in 2016. It will have challengers though. Sony is working on Project Morpheus, which has the same aim, and HTC has built Vive, unveiled at Mobile World Congress in March, which will work with the Valve system for downloadable games. Samsung has its own Gear VR, developed with Oculus. Even Google’s Cardboard, a DIY system which you literally fold together from a cardboard template before sticking your smartphone into a slot a few inches from your eyes, has had around 1.2m downloads since Google introduced it at its I/O developer conference in May 2014.
Meanwhile the parallel development of “augmented reality” (where objects and data are overlaid on to the scene you see in front of you) is growing quickly, too, with Microsoft’s Hololens the most promising offering to date, along with Google-funded Magic Leap. (Google Glass, which is undergoing a rethink, is like a subset of AR.)
But it’s VR that has everyone excited. In March 2014, Chris Dixon, a partner at the increasingly influential venture capital company Andreessen Horowitz, observed that using an early version of Oculus Rift the year before was one of “a handful of technology demos in my life that made me feel like I was glimpsing into the future” – and his list ran back as far as the Apple II computer from the 70s.
VR, he noted, “has long been a staple of science fiction” (of which Star Trek: the Next Generation’s “holodeck” was only the most recent) but hardware was now cheap enough, and engineers keen enough to explore a new frontier, that it could finally deliver on the science fiction promise. (Dixon’s team cashed out on Facebook’s acquisition of Oculus Rift.)
In the 90s, VR certainly felt like SF. Waldern’s company relied on top-end Silicon Graphics systems with six-figure price tags. There’s a video of him from 1994 at W Industries, showing how pixellated the experience was. “Our system used some of the very first Sony LCDs, with 300 by 200 pixels,” Waldern recalls. “Today you get 1,080p [1,080 horizontal lines] minimum, and by the launch next year you’ll probably have 4K by 4K. And the computational power is transformational – we were working on about a megaflop, and each machine cost about $70,000, which is a massive barrier to adoption. Now with an Nvidia GPU you’re talking about a teraflop.”
VR is a complex real-time computing problem. Ideally, the fictional scene should show no individual pixels, fill your field of view – which in normally sighted people covers 180 degrees – and change without any lag, no matter how quickly you move your head. That means the system has to be able to calculate what the scene will look like at any point, and incorporate the movement of your head, reported from the helmet, into its computation, and update the screens quickly enough to give the sensation of realistic movement.
Measurement of head movement is done using accelerometers and gyroscopes; the screens can now offer millions of pixels just inches from your eyes; computing power aims to cut lag and latency. The latter can quickly induce nausea if the scene isn’t redrawn quickly enough as the body’s balance system disagrees with what the eyes tell it they’re seeing; that was a key problem with early VR systems.
But those are being overcome, thanks to the rapid growth of the smartphone industry. That has made accelerometers and gyroscopic chips cheap, plentiful and fast, while also pushing the quality of displays to a level where you can’t distinguish pixels even in screens a few inches from your eyes. That though requires graphics processors that can handle huge numbers of floating point operations per second (flops) – but, as Waldern says, that’s available.
Among those eagerly exploiting that space is Professor Bob Stone, chair of interactive multimedia systems at the University of Birmingham, and another British veteran of the 1990s boom and bust. After another 20-year gap, I caught up with him as he travelled from a location where he had been using a drone to create a 3D “point cloud” of a church and archaeological dig in Wiltshire; in just five minutes of flight it captured enough data to create a realistic 3D model that one could “fly” through at leisure.
When we spoke, Stone initially seemed unimpressed by Oculus’s offering, suggesting that “it’s not much different from the headsets of the late 1990s”. He cited “pixel bleed” (where the pixels don’t appear sharp) and limits on the field of view. “We’re using Oculus Rift development kit 2 [the latest version] in defence projects, because it’s the best of a bad bunch,” he said. Why bad? Stone feels that if you have to wear something, “you aren’t really experiencing total immersion”; he thinks ideally VR should simply feel like life. “More than 20 years on, shouldn’t we be further ahead than this?” he asks.
But later he emails, relenting on his apparent negativity: “VR has been my career for over 28 years now, and I genuinely believe in what VR can deliver when the technologies are applied appropriately and with the needs of the end users and end organisations well and truly in mind. As I said, it seems that every seven years we see the same old issues and hype coming through (these days via the web, YouTube, etc) that people out there seem to believe without question!”
The original bust in VR came when first-generation products were crushed by the weight of expectations. So what led to the renaissance of VR – and why has it taken so long?
Waldern reckons a key reason is that games, which were the main consumer-facing use for VR, shifted to a different set of genres. “There are periodic cycles in the games industry, for technologies and genres. People want change from time to time.” After a couple of decades of 2D and pseudo-3D offerings (think of Nintendo’s games, and Quake and Doom), gamers are ready for an “immersive” experience, he argues.
Digilens’s work involves building optical waveguides, so that projected images can fill or overlay the field of vision. Most of the uses on the company site involve business, rather than consumer, application, but Waldern sees all sorts of potential. “Why do you think Microsoft bought Minecraft?” he asks. “Hololens is perfect for it.”
Stone agrees: “One thing that saved VR was that while we were running around talking about headsets and gloves, the gaming community was coming up underneath, and came up with software and hardware in the quest for the best quality games. That drove down the cost of computing with graphics, and also made available a number of toolkits for games-quality virtual worlds that people could sympathise with.” Games drove costs down: “Things that in the late 1990s cost six-figure sums you can now do on an £800 laptop.”
Brian Blau, research director at analysis company Gartner, forecast in March – before Oculus’s announcement of a tentative release date – that there would be 2m VR headsets sold in 2016, rising to 8m in 2018; in all he thinks there could be 25m in use within a few years.
That doesn’t sound quite like the gigantic opportunity that Waldern described. But VR is seeing a growing number of uses in commercial spaces, where the ability to train people, or give them a new perspective, is valuable.
Facebook’s purchase of Oculus, though, suggests that it sees the capability to transform how we interact with friends – which is what Waldern is thinking of when he talks about it potentially being bigger than the internet.
“Even back at the start, the biggest kick we got was when we used ISDN lines between Berlin and London, and we linked players up in a game. It was amazing: you could meet, play, talk with people. Now, games have been doing that virtually for a while. But when you can feel you’re really meeting someone, right there, as a jolly alien or Robin Hood …” He stretches for the words. “Immersivity is the main thing. This has many years to run, and we have to get to where we are totally convinced – but we’re looking at something which has transformational capability for society.”
guardian.co.uk © Guardian News & Media Limited 2010