One month in the Vive

I’ve only just started working with room-scale VR but it’s been a long time coming.

This all started for me in the 90s, mucking around with VRML on the desktop, playing Dactyl Nightmare in a games arcade (a minute’s play for my coin!), and selling my own VR “interactive film” screenplay, Ten Weeks in the Headbin.

A lot of excitement. A lot of dead ends. And ultimately a slow realisation that things take time. You never know when truly new mediums will take hold. This time around, VR looks more promising.

After spending the tail end of 2015 in a submarine I emerged to discover VR hype hitting overdrive. Mod had already bought two generations of the Oculus dev kit, for pre-vis and post-vis on our shows, but something was still missing. I wasn’t convinced the experience was compelling enough and early testing showed too many people getting nauseous. In December I had my first go on the HTC Vive (devkit) and another piece of the jigsaw dropped into place. You could walk around inside VR! Position tracking — techy as it sounds — is the key to making VR a comfortable and immersive experience. While the Oculus Rift also has position tracking, only the Vive to-date gives the user freedom to take steps and move around — within a maximum 5x5m area. We decided to take the plunge as early adopters once again.

Make no mistake VR adopters are still very much taking a punt. We didn’t get much change from AU$5K by the time we’d forked out for the HTC Vive and a new VR-ready PC. Ordering within minutes of the pre-order being opened, our kit was one of the first consumer models to arrive in the country, not counting the devkits distributed last year.

SETUP

Unboxing the Vive we got that special whiff of plasticy aroma as we unwrapped all the parts. It’s a treasure chest of components. Two Rubic’s Cube sized “lighthouses” are the secret ingredient — a very clever solution for simple, cost-effective tracking. They bathe the play space with invisible laser light, and don’t know or care what is being tracked — they only communicate with each other and do their laser thing — which leaves the field open for all kinds of interesting peripheral development. It’s down to the other components, a chunky head-mounted display (HMD) that is rather heavy but also well designed for comfort, and two hand-held, button-full controllers with lots of options. And cables, lots of cables.

The Vive software setup is rather slick. Love it or hate it, Valve’s Steam distribution platform is a marvel of software design and engineering. Steam VR is no exception.

The room-based calibration wizard requires you to trace an outline of your space with a controller and perform a few other tasks but it is easily done in a few minutes. This is handy because if anyone knocks your lighthouses out of position (the advice is to never move them while powered on) you will have to redo the process.

IMMERSION

Once we’d configured, it was time for the Room Tutorial, a rather cute vignette set in Aperture Labs, hosted by a robot resembling Wheatley from Portal 2. It is hard to overemphasise just how immersive the Vive is once you start moving around in your real space and manipulating virtual objects with the controllers. The tracking is spot-on. You do need to ensure that nothing blocks the line-of-sight between your headset and the lighthouses (causing the view to white out momentarily) but in general the tech feels rock solid — quite remarkable for a first release.

Most experiences we’ve come across utilise both Vive controllers in tandem but for our first development outing, I wanted to focus on using just one. I had a very specific concept in mind.

DEVELOPMENT

The Steam OpenVR stack gives developers a lot of room to move. Geddit? The hardware support for the two engines we use most at Mod (Unity and Touch Designer) was superb. Thanks, all you beta testers!

After a short Unity test run with guidance from VR Dev School I was ready to jump into my weapon of choice Touch Designer. For those of you who haven’t heard of it, Touch is a fork of the blockbuster VFX tool Houdini and engineered specifically as an authoring tool for real-time experiences. We built ACO Virtual in Touch several years ago and I was hankering to see how easy or otherwise it would be to port this “live” immersive experience to VR.

This was a surprisingly pleasant experience. Touch as a procedural engine is overkill for many simple applications but media artists love it because it places few restrictions on how you work with real-time video. If your hardware can handle it, Touch will probably enable it. Within a few days I was able to get a playable VR demo up and on 7 June we gave our first public exhibition of ACO Virtual VR incorporating the complete 35-min show.

There was still a way to go. We didn’t have the beautiful audio mixing in place — we literally couldn’t fit our sound card into our new VR PC so the experience was not quite up to the same level as the ACO Virtual touring show but what we did get was something that can run in our studio, that feels like you are in the experience, but without a van’s worth of kit. Very satisfying.

For those of you who haven’t visited ACO Virtual, the experience revolves around the ability to swipe across images of the musicians on a tablet and have the immersive sound and visuals respond as if you are conducting. I wanted to see if the Vive controller would make a decent substitute for a touchscreen.

THE AUDIENCE

Audiences react to room-based VR in different ways. We’ve noticed four broad types so far.

1) Avoiders

Some people just don’t want to put on a HMD. We’ve heard many excuses, including the classic, “I really want to try it. I’ll be right back…” Bad hair days abound with VR and it’s never a good look. There are also nomenclature issues. The conflation of “VR “ with “360 video” means that many people presume the Vive experience to be equivalent to low or middle tier tech like Google Cardboard or Galaxy Gear VR. Once bitten twice shy. We’ve seen people decline their first Vive demo on the basis that “VR gives me a headache”. So far we haven’t seen anyone get nauseous from a Vive. In staging terms, exhibitors tend to place the (high end) Vive experience last so it remains to be seen what percentage of users get only as far as Gear VRs before giving up.

2) Tasters

Not everyone is comfortable wearing an oversized ski-mask with dangly cables hanging off the back. For these people, a quick go and a few tentative steps is enough. The body language reveals the inner monologue: “A lot of work has gone into this… can I take it off yet without seeming rude?” There is also no getting away from the fact that once you put it on and can’t see the world around you, you’re vulnerable. There are trust issues here.

3) Collectors

This group is populated by those who have already tried multiple experiences and thus in a position to compare and critique. Obviously VR experiences will vary but it’s been interesting to see how quickly people form their conclusions and let you know what is or isn’t working for them. Aside from the obvious bug reports, I was surprised to see how quickly people will tell you “how this should work” given how new the form is. Again the reactions vary but across your typical collector, opinions are strong and fully-formed within a few minutes. No time for subtleties. No doubt we’re witnessing the emergence of new metrics, equivalent to the split-seconds in which a person viewing a web site for the first time subconsciously ‘decides’ how long they’ll stay there.

4) Explorers

The most satisfying group to share VR with so far are those who take time to explore, not just the interface, but the virtual worlds themselves. Again you can tell by the body language how engaged they are. I think what impresses me the most about the Vive is how quickly a newcomer can turn into an explorer given the right context.

These are not by any means hard-and-fast categories. No doubt we all move between them. In a trade show or demo setting it is also hard allow people sufficient time for reflection and re-visitation, especially when there is a queue and background noise. That said, with so little to go on, research-wise, getting as many observable people as possible into your shows makes a lot of sense. We’ve only had one public outing so far (thanks Screen NSW) but we’re still in conversation with many of the people who met us and our work for the first time that day.

SUSPENSION OF DISBELIEF

Encased in an HMD, different things can snap users out of their reverie. Such as tripping over the cable.

Thankfully we haven’t had this happen yet but we have seen it with peer studios. Not nice for anyone. Walking into objects or hitting them with outstretched hands is another issue — not everyone is content to follow instructions and “stay within the wireframe box”. Watch out for the macho user who has clearly decided to stress-test the gear at your expense. For more sensitive souls, other elements of the experience jar. The controllers for example. You can easily show your experience’s controllers as they physically appear but the creative possibilities are endless in terms of how the controllers can appear in VR and what each button can be programmed to do. For ACO Virtual, the controller appears in VR as a conductor’s baton (no buttons). While most people seemed to like this (you can swipe it around to spotlight musicians for a bit of fun) one user reported that it pulled him out of the experience entirely whenever he pressed buttons because what he was seeing didn’t correspond. One to watch over time.

Graphic detail. In a time when we live in rendered worlds, people are quick to judge quality by visual aesthetic. Visuals do not necessarily translate to a better experience. Once you have experienced a number of beautifully rendered (but static) virtual worlds — as some of us did back in the 1990s VRML days — what’s lacking is noticeable. For ACO Virtual, I was working with a rather rudimentary post-vis model of our preferred physical layout for the show (7 screens for projected images in a hexagon layout) but the feedback has been that the overall effect works. I believe largely because you’re in a darkened space.

Regarding audio, binaural sound (to replicate the sense of audio in specific 3D positions around you) will be a huge boon to VR creators. It’s still in its infancy and not yet available as an off-the-shelf resource across all platforms.

COMFORT

We’ve been surprised to see people staying in the Vive for over an hour if the experience grabs them, TiltBrush being the clear winner to-date.

What’s more, when needing a break, I’ve seen people rest their head, encased in the HMD, on the floor rather than take it off. The fact that a standing user of the Vive can always choose to adjust their position, or even sit or lie down while still using it, makes a huge difference for comfort.

Many users have suggested that we should have a ceiling mount for HMD cables that would reduce the risk of tripping. We haven’t taken this advice as yet and one disincentive is the limited cable length. Is it more desirable to reduce trip hazards vs give the user more cable-length to move around further? When trying out Björk’s VR experiences at Carriageworks in June I found I couldn’t sit on the floor as the cable went taut — arguably as risky as tripping.

FUN

It’s been very interesting watching the reactions of our community as they try the Vive for the first time. Not everyone has been blown away but enough people have been amazed for us to feel somewhat privileged to be working at the cutting edge of this space.

It’s not all about what’s going on in the headset. The best fun appears to be when groups are sharing the experience, one person in the holodeck and the rest watching a screen showing the HMD’s perspective. We already live vicariously through the adventures of our friends. Watching others play VR games and other types of experiences is now part of our culture.

I can honestly say that I’ve had more fun developing for the Vive than I have in years. As someone who doesn’t code full-time for a living there always has to be a big payoff if I’m going to get my hands dirty. There is something very satisfying about building and then sharing a VR experience.

We’ve had a steady stream of friends and colleagues enjoying their time in our holodeck so much that we’ve opened it up for public hire as a VR Arcade.

What’s clear is that the immersion everyone is talking about is no illusion. Or actually it is but you get what I mean! Room-scale VR, where you can walk around, is a new experience and it offers, for me at least, a tantalising suggestion that we may in fact be able to claw back some proper reflection time in creating virtual spaces. There will be no shortage of high adrenaline action adventures to be had with this gear, and I’ll be digging them, but there is mainstream potential here. You won’t have to muck around with controllers for all experiences. You won’t have to even walk around. But in the same way as museums and galleries host all manner of exhibitions and ideas, VR hardware is going to open doors to new experiences.

NEXT STEPS

For Mod, the next step is clear. We’ll continue to explore and we’ll be testing the market to see what’s out there for us. This isn’t the first time around for me (dotcom boom/bust, Web 2.0, and now whatever this is) so I wouldn’t be surprised if the hype doesn’t continue to build for a few years. The progress we’ve made in our first foray suggests a bright future for those who can effectively stage engaging mixed reality. That’s what we’re looking to do.

We’ve started collecting a bunch of useful requests for our software platform Rack&Pin’s VR support. Soon we’ll be making a sandbox available for developer testing and feedback.

There’s a lot to do. Having seen countless bemused faces over the last year grappling with VR hardware, there’s a lot we creators can do to craft more rewarding experiences. For starters, we need to facilitate more sophisticated and subtle interactions between the HMD wearer and others. So far Vive helpfully provides a “Knock Knock” feature for getting the user’s attention (Settings | General) but we can do better. Much better. It will pay off in every sense. Comfortable physical spaces, personable hosts and effective digital platforms. These will be the some of the key ingredients in getting mixed reality right. I think we can get avoiders and tasters over the hump to engage with short-form and long-form VR. But with a new medium like this we really need to use every trick in the book to communicate. Which is why I won’t stop writing just yet.