I had an interesting e-mail exchange with an old friend recently on the topic of Virtual Reality. That forced me to forced me to clarify my thinking on the subject, and that has gotten me thinking about the space a lot.
Virtual Reality has been kicking around for a long time. Anybody remember the Michael Chrichton book “Disclosure”? It came out in 1993, and an even worse movie version starring Michael Douglas came out in 1994. Both featured ‘hot’ virtual reality tech, that looked ridiculous then, and in some ways it is hard to see how much the industry has moved on since then.
However, a few things have become clearer:
- Virtual Reality (VR) is nice, but the real force will be Augmented Reality (AR)
- Content, Content, Content
- The hardware is ready, mass adoption price points by 2017
- No one is going to wear VR headsets in public
- VR will be a solid niche category for hardware makers, at least for a few years
- Some new ‘magical’ invention may surprise and prove me totally wrong
Let me walk through each of those.
My conversation initially was about the hardware required for VR. You can buy an Oculus Rift for ~$350 today. I have tried these out a few times, and found the experience to be pretty good. I think its safe to say the consensus is that the industry already has all the technology they need to make VR gear. Once we start to get more content for VR (more about content in a moment), it may turn out that headsets actually need more algorithms or something (e.g. to reduce vertigo), and that may require even more expensive processors. That scenario aside, I think next year we will start to see a lot more VR gear go on sale. Given the normal cost curves for gear, I think we will start to see price points for VR reach volume inflection points by 2017, say $200 or so. At that point, it starts to become affordable for bundling with video consoles or computers.
However, we now reach a chicken and egg problem. For prices to get truly mass-affordable, we need high volumes. On the other had, volumes will not come until consumers want VR gear. And for consumers to demand this stuff, they will want things to do with it. Which leads to the subject of content.
By far, the biggest hurdle for VR right now is the lack of things to do with it. I think it is safe to predict that the first, popular applications for VR will be video gaming. Speaking as a gamer, it is easy to see the appeal of more immersive playing experiences. My understanding is that several gaming studios and start-ups are working on creating VR-ready games. My best guess at this point, is that we will start to see some interesting VR titles next year, and a lot more by 2017. I think it is reasonable to estimate a hit VR title by that point that feeds into the falling price points for the gear, and drives adoption. Forecasting hits is very hard to do, but if had to pick a date, I would say Holiday Season 2017.
I plan to spend a lot of time at CES in January investigating all of this, so please stay tuned for more.
However, I think the adoption of VR will have problems getting beyond gaming, at least until someone can really surprise us. First, truly immersive VR is going to require headsets of some sort. And no one is going to wear those headsets in public. Second, it is unclear why anyone would wear VR gear to consume other forms of content. Watching a movie at home on the couch could be improved by watching the same movie in a 360 degree VR version. But it is going to be even longer before movies are filmed in 360. Also, unless people start watching movies alone, there is a limit to how many people will buy multiple rigs so that the whole family can watch a VR film at home. That being said, I imagine there will be a sizable market for Adult VR content. Again, it sounds like that is already in the works. But mainstream movies are probably not going to be a big VR category anytime soon.(And in all fairness, maybe Adult VR content is enough to drive a major market.)
Another interesting possibility is the consumption of specialty, user-generated video like footage from GoPro. Now, I am not exactly sure who watches (or edits) all that GoPro footage, but I can say that personally, I would enjoy watching ski and surf runs from a first-person GoPro perspective.
If gaming and other forms of video consumption remains the major source of VR adoption, then the market for VR hardware is likely to remain fairly limited. According to Venture Beat and the gaming industry body ESA, 80% of American households own a gaming console. That means the US market is about 100 million gaming consoles, so say 400 million consoles globally (being generous). If each one buys a $200 VR headset, it equates to an $80 billion hardware opportunity over five years or so. That is a great business, worth about $40 billion to the supply chain. But keep this in context, Apple sells more than that in iPhones every year. Eventually, prices will likely compress as China Inc. takes interest, but still a good market for a few years.
The trouble is that beyond gaming, we quickly run out of applications for VR. Many big manufacturers (e.g. Boeing) have been experimenting with VR applications since the time of that Disclosure movie. As far as I can tell, those experiments have not gotten very far. The utility of such applications is greatly outweighed by the complexity they require. I suppose stock market traders could trade in their 8-screen Bloomberg set-ups, for a pair of VR goggles, but it is hard to think of many other practical enterprise applications.
But notice above I said industrial applications for VR are limited by ‘complexity’. What exactly is that complexity? And here we go back to the question of content. Let’s say you want to build a VR rig for assemblers on a manufacturing line. Ideally, the system would show the workers where to install some component. But how does the system know where that component goes? Eventually, we may have some machine intelligence that can read blueprints and AutoCAD and translate that into a real world overlay, but that does not really exist yet. Translating digital information into a real world overlay is the true bottleneck in the creation of VR content.
Barring some crazy, government initiative, I think it is very unlikely that anyone is going to go out and build some great, monolithic ‘Data Layer’ for the world. I imagine Microsoft wants to have some monumental Operating System (OS) for VR, but adoption is going to remain bound by the people who actually have to go out and create the content.
Instead, I think a far more promising (and likely) approach is that the industry takes a gradual, piecemeal path towards creating this Data Layer. I picture this as starting off as taking data that is already online and then adopting it for use with goggles. Yelp has tried to do this with their ‘Monocle’ feature on their iOS app. Users could hold up their phone camera and see a visual overlay pointing the way to various restaurants. I like this idea a lot, even though Yelp has struggled with a lot of implementation details.
Going down this path, it is easy to think of Data Layer applications. Hold up your Facebook app to see which of your friends is nearby. Google StreetView overlays on Google Maps navigation. Hold up your Amazon app at the book store and see price comparisons for a product. (That last one already exists.) This way, the Internet players can gradually bring online content into Data Layers for displays on various devices.
However, it should be clear that the things I am envisioning do not need full VR gear. Instead, this fits much better with the concept of Augmented Reality, and for this all a user needs is their smartphone. Increasingly, I find the idea of AR much more compelling and timely. The good news is that AR does not need any fancy new hardware. Of course, that is the bad news too, at least for the supply chain. AR has two big advantages – the lack of need for specialized hardware and the ability to gradually add in new content using existing distribution frameworks (i.e. App Stores).
To be fair, AR faces a similar set of challenges. Namely, crafting content for a very different kind of medium. No one is yet entirely clear what AR can or should appear. As Yelp’s Monocle demonstrates, just putting existing data on top of a camera or map, is not enough. Someone is going to have to do a lot of design and User Experience (UX) work first.
I went into this exercise very skeptical. Those images of Michael Douglas wondering around a virtual filing cabinet are apparently etched into my memory. I am also wary of new hardware markets as so much of that landscape is in bad shape right now. But I have to admit that I am now much more intrigued by the potential for VR and especially AR to produce interesting use cases in coming years.
I should also add, that all of my analyses here could be proven completely wrong. As part of my research I went onto the Magic Leap website. They are reportedly raising an oddly specific $827 million in funding. That is a lot more money than a hardware company needs. They have lots of job listing around content creation. So they are clearly up to something more interesting. And their web site also hints at something entirely unexpected. I do not know much about what they are working on, but that whale jumping out of the gym floor video is pretty wild. So I want to very explicitly leave open the potential for some unexpected innovation. The whole VR/AR space has enough momentum now that we should expect to be surprised.