On Friday, I received my new Oculus Go – the stand-alone mobile VR headset. The reason I bought it was so I could begin to share some of the archaeological work I’m doing in virtual reality, and also to play with a cool new toy. I’m starting to wrap my head around its usefulness (after a weekend of wrapping it around my head). First I’ll outline general impressions, then I’ll go through viewing models from 3D photogrammetry and using it for virtual tours.
It’s a very impressive experience. After watching VR from afar for years, and coming from a Google Cardboard – the resolution, field of view and head tracking are amazing. It isn’t quite as mind-blowing as the HTC Vive demonstration I’ve experienced, but from memory the image is actually clearer on the Go. The headset is comfortable, clear and snappy – it’s very easy to use and the battery lasts for about 2 and a half hours. So far everyone I’ve shown it to has the hang of it in a minute and is very impressed.
It is good for watching Netflix, but where it shines is watching some of the 360 and 3D content by creators such as National Geographic on Youtube. Documentary experiences in VR no longer feel like such a gimmick, they are genuinely easy to get lost in. Games, while not captivating, are fun – I’ve found the wingsuit racing game Rush to be the most fun – it feels like it benefits from being VR. Web browsing is useful, the picture is impressive but the typing interface is as clunky as you would expect.
At $320 Australian (including shipping), the price point is about right. The app ecosystem is clearly underdeveloped, but you can access enough media for it to be worth it. I have no doubt that the software variety will improve soon.
The most impressive feature I used up within the first few minutes was viewing 360 photos and video. I have been able to look at many of the 360 photosphere’s I’ve taken over the years, and show them to other people. These photospheres were all taken on my phone with the Google Streetview app, and they look great on the headset. It has a strong sense of presence, and the images are sharp – they look much better than viewing them on a phone or even a big desktop monitor. I think finding ways to share interactive 360 content over the web will be the main use of lower cost VR technology.
Some good examples of virtual tours for the Go include native apps like Mission ISS (about the International Space Station), and webVR examples such as a tour of Son Doong Cave – the largest cave inte the world. The ISS example is incredible, it allows you to fly through and around the Station in 3D – both exploring or doing ‘missions’. However, it was clearly expensive to develop and requires a large download to run – it also requires skills that I certainly don’t have, I presume it was developed in Unity. The Son Doong example is much more interesting to me, it runs right in the web browser, linking together simple 360-degree panoramas, and including 1 or 2 short voiceovers per scene explaining the interesting features. I completed the tour in one sitting and was engaged the whole time. I also felt like I had a good understanding of the cave by the end.
It was developed using a tool called vizor.io. I spent a little time testing it out and building a tour of the Kim Bong Carpentry Village that we recorded as part of the Vietnam Maritime Archaeology Project. It was very easy to put together a tour with multiple interlinked panoramas, and some simple image and text pop-ups. I had no problem testing this in the headset’s web browser as I edited it, and publishing it online. I think adding some voiceovers will polish the experience nicely. It doesn’t, however, have any 3D file viewing capabilities yet.
For some time in the future, I’m interested in using Facebook’s React-360 library to build similar tours showcasing the material at the WA Shipwreck Gallery, alongside some of the models we have collected at MAAWA. The library just went through a major change, it used to be React VR, which is what powers vizor.io, but the documentation, support and examples haven’t quite caught up with its transition to ‘React 360’. It’s barely over a week old, so it’s understandable – I will be watching it closely, and soon will maybe using it to put together a more interactive tour experience including 3d models.
I presume it won’t be long before vizor.io can do many more things. This blog post from late last year shows they are clearly pursing the museum direction, and it’s worth read on if you’re curious about designing museum experiences for VR.
As a low-cost and easy to use platform, the Oculus Go is a perfect end-point for these web based VR tours. I will be using it to design them and show them – and I’m sure museums will also start to pick up it or similar headsets at this price-point.
For viewing 3d models from Photogrammetry
Apart from 360 tours, the other reason I was interested in a VR headset was for viewing 3D models that I’ve recorded. I was counting on Sketchfab’s webVR function, something that has worked well on my cardboard for more than a year. However, it doesn’t run very smoothly at all – the resolution is incredibly low, there is a lot of lag on most models (very distracting when the frame is half black in VR), and the interface doesn’t use the trigger on the remote, it uses to top clickpad (very unintuitive). It is basically unusable. The issue has been flagged in the forums and they will presumably work on a solution soon. Not even Facebook, who owns Oculus, is optimised for the Go yet – so other developers need a bit of time to catch up. That being said, simply using the browser in 2D as a giant screen is very effective – you can see a lot of detail in the models and get a very good idea of what you are looking at. Currently, that is your only option.
Overall, the Go is an incredible piece of hardware. For it’s price-point, its worth for just viewing the 360 content that is available. The software is undeveloped, but it will catch up, and I look forward to seeing it much more widely used. If you’re looking for a workhorse that will change the way you work, then you’ll have to wait a few years, but if you’re looking for a way to experiment with how it can be applied to archaeology, then this is a good way in. You’ll be seeing more of this here in the coming weeks.