Valve's astounding SteamVR solves big problems - and poses bigger questions
Now that the VR dream is real, what do we do with it?
At this week's EGX Rezzed show in London, I had one of those experiences that journalists live for. I got a glimpse of the future - or perhaps it would be more accurate to describe it as a privileged understanding of the present. I tried Valve's SteamVR system and for half an hour was held rapt by a suite of demos that allowed me to walk around virtual spaces. I came out with my head spinning - only figuratively - and babbled excitedly at anyone I could grab. The first friend I subjected to my ravings told me I looked stoned. I did feel like my perceptions had been altered in a way that taking the headset off didn't immediately reverse. It's that good.
I've calmed down now, but a day later I remain profoundly impressed by Valve's technology. For context, I'd describe myself as a virtual reality agnostic; sceptical would be too strong a word for my position, but I'm not a true believer either. I'm interested in the possibilities of VR and happy to see some significant hardware and software innovation being driven by the games industry. But I have doubts about its applications and its mass-market appeal.
That said, the last time I felt this excited by the potential of something I was reporting on was when John Carmack introduced me to an early prototype of Oculus Rift at E3 2012. It was a fun demo, but I was mostly energised by the prospect of a new technological frontier for games to head towards. This is, after all, a tech-driven art form that has taken many of its biggest creative strides in tandem with advances in engineering, and it's been a long while now since the 3D graphics and online gaming booms flooded the medium with new experiences. Even if VR could never be so game-changing - Carmack seemed to envisage it, primarily, as a natural extension of the first-person perspective, a better way to experience the games he was already making - a little bit of that future-tech frisson could go a long way.
Could it ever. I didn't imagine the astonishing distance it would travel in less than three years. There have been important advances: in display resolution and in the range, responsiveness and smoothness of head-tracking. The related and vital battle against motion sickness has taken big strides forward, too. I'd heard that SteamVR running on HTC's Vive headset causes no sickness at all and can confirm that. (I even subjected it to the ultimate stress-test by sampling it first thing in the morning with an awards-ceremony-grade hangover. Nothing.)
These are just small steps compared to Valve's big leap, however. As Rich Leadbetter explained in his hands-on from GDC, SteamVR allows the user to walk freely around a small virtual space, demarcated by a couple of light-emitting base stations. It's no exaggeration to say that this freedom completely transforms the VR experience.
It's the difference between having a viewport into a virtual world and being in it. Your physical, sensory and emotional connection to the software is on another level altogether. You feel grounded there. My memories of playing Elite on Oculus Rift are images - the memories of playing a game. My memories of Valve's VR demos feel more like memories of places I've visited.
In a demo called The Blue, I stood on the deck of a sunken ship and waved schools of fish away with my hand. A manta ray swam by, close, and I backed away from it almost involuntarily. Then a whale approached - its size was breathtaking. I had to move and crane my neck to take it in. I felt awe in its presence, and fear of its sheer non-existent mass.
The next demo, Skyworld by Vertigo Games, was quite different. I stepped back to make room for a table in the centre of the virtual space. On the table was a miniature fantasy world, dotted with houses and castles, with a dragon sitting atop a mountain in the centre. I could use a spellbook and wand to interact with the world, but what was so enrapturing was to walk around it and lean into it - instinctively trying to stay clear of the table, although there was nothing to stop me passing through it - and examine the toylike world from my lofty viewpoint. VR thinking has tended towards first-person experiences with a realistic scale, like the undersea demo, but Skyworld illustrated the huge potential for strategy and God games, with tangible worlds you can loom over, reach into and manipulate.
What these two experiences shared was an overpowering sense of intimacy with the virtual space. This is brought home again and again in the demos, and its impact is profound. The smallness of the space you can move around - 15 feet square - is almost a virtue in this respect, because it focuses you on your immediate surroundings in a way the vast majority of video games don't, and which has much more in common with real life. In an informal talk about VR later in the day at Rezzed - which is available on YouTube, and well worth a watch - Valve's Chet Faliszek underlined this point.
"[People have] seen Star Trek and the Holodeck, and they want to run around and play a current FPS. That's where we started. We ported Team Fortress 2 over. It turns out, that's actually a bad game for VR. VR scales more like real life - things aren't just flying at you all the time. That ends up being not a limitation, but something special. It's a new opportunity for us. But there's so little we know."
Having experienced the demos, I know exactly what Faliszek means. VR demands a paradigm shift in the thinking of game designers and artists about how they build virtual space and how players should interact with it. We're only at the very beginning of this journey now. Vive's commercial launch is in a matter of months - it's due for release in 2015 - but this process will likely take years, and at the end of it the games won't resemble those we're currently used to. In short, they won't be Half-Life 3.
Were there any clues to what they might be among the demos I tried? A few. The demo that most resembled a viable commercial product was, not coincidentally, the one that made least use of SteamVR's virtual space. Dovetail Games' fishing simulator plonks you by a peaceful pond and lets you use SteamVR's two wand-like controllers to cast a fishing rod and reel the line in. You could imagine it working well, if a little less convincingly, on a fixed-point VR set-up like Sony's Morpheus.
It's my belief that the driving force behind early adoption of VR will be the passionate niche communities of simulation gaming - whether it's sim-racing fans like myself, train and flight sim diehards, or the more fantastical space pilots who'll play Elite, Star Citizen and Eve Valkyrie. With these games, a fixed player position isn't a huge issue, the interface is often handled by bespoke controllers, and the audience will spare no expense in its pursuit of veracity. So Valve's partnership with Dovetail, developers of the insanely profitable Train Simulator games, is sound thinking. But this is just a first step.
Bossa's Surgeon Simulator and Owlchemy's Job Simulator had me dashing around a room, using the controllers - which feature triggers, grip controls and circular touch pads under your thumbs - to manipulate tools and objects. These interactions still have a little way to go before they feel consistent and accurate, although your projection into virtual space is so convincing that the lack of physical feedback isn't as big an issue as you might think. They also prove that a heavily stylised cartoon world is no barrier at all to immersion - indeed, it's great fun in a VR context.
Less interactive, but more intriguing, was a demo based on Fireproof's popular puzzle game The Room. In this, you walk up to objects like an old box camera or a model ship and can lean in to examine them: a mystifyingly magical experience, as it was in Skyworld. While your gaze is focused, the whole room around you transforms into a related diorama - placing you aboard the model ship you've just studied, say. Not just a clever manipulation of the player's viewpoint, this hints at a possible solution for transitions between play-spaces.
Meanwhile, a demo called Tilt Brush allows you to create 3D light sculptures by painting in the air with the controllers. This was the best demonstration of the controllers' interactive power; it felt so natural to glance down at the two wands, replicated in the virtual space, and use the control wheels to select new options. It was also simply thrilling to build something you could move around and examine from any angle. The potential of VR for creative apps - or games with strong creative elements, like Minecraft - is clearly enormous.
Valve saved its own show-stopping demo for last. Set in Portal's Aperture labs, and created in the new Source 2 engine, it's a hilarious and visually astonishing vignette that casts you as a human test subject of Aperture's implacably mad robot intelligences. It starts from a point of tangible realism - you're opening drawers and pulling levers in a small, sterile box room - and then employs a sequence of startling and delightful Gulliver-like tricks of scale to show off the spatial and emotional range of VR.
One drawer contains a society of tiny stick-men who scatter in panic at your presence; behind a wall you find one of Portal 2's co-op robots which staggers brokenly into the room. It's full-scale, as tall as you, and the sense of its physical presence next to you is utterly remarkable. When the walls fall away to reveal the vanishing-point enormity of the Aperture installation above, around and - sickeningly - below you, it's hair-raising. When GLaDOS' armature rears over you to dispense her trademark withering sarcasm, it's absolutely terrifying. The feeling of vulnerability engendered by this familiar, huge antagonist dominating my personal space is not something I will forget in a hurry. If you're looking for the value of VR as a storytelling medium, it's in the emotional impact and overwhelming physical presence of this demo.
What John Carmack showed me three years ago was an exciting new technology mapped, exhilaratingly if awkwardly, onto video games as we know them. What Valve showed me this week was a big leap forward in that technology alongside a series of tentative baby steps into a whole new kind of gaming.
As Rich pointed out in his piece, SteamVR sets even bigger challenges for game designers than the already considerable ones posed by fixed-point VR. But it's already clear from these demos that it might provoke more innovative and radical responses that are more willing to leave gaming as we know it behind. I have no doubt that the cutting edge of VR development will all take place on Valve's system - at least until its rivals offer the same freedom of movement to players.
Back in the real world, the practical challenges are great, too - not least in persuading players to clear enough space in their homes to use this device properly, and the potential for social stigma to attach to the goofy-looking headsets and the players' withdrawal into entirely private experiences. I still think that these present major obstacles to the widespread adoption of VR, which even more practical and commercially realistic offerings like Morpheus will struggle against.
But having tried SteamVR, I now get why Faliszek - himself a former VR sceptic - eventually rolled his desk over to join Valve's VR team. Over the next few years, this is where intrepid engineers and artists will get to grapple with big questions of design and technology, of psychology, even of philosophy. This is the new frontier.