Why Mobile VR: Part 2

Abhijeet Naik
CinematicVR
Published in
3 min readMay 15, 2017

“If you assume any rate of improvement at all, then the games will become indistinguishable from reality, just indistinguishable.” — Elon Musk

In Part 1 of these blog series, I wrote about desktop grade VR. In the new installment, we will discuss choosing wireless HMD’s when starting a new VR startup. You ready? Here we go.

Mobile VR like Samsung Gear VR has improved exponentially since its first innovator edition which only worked with Note 4. Unlike desktop grade VR which has its own dedicated screen, mobile VR relies on your phones display, so resolution and pixel density (screen-door effect) is all dependent on your smart phones hardware. Every year the System On a Chip’s (SOC’s) are getting more powerful and more advanced which means it opens more door for you as VR company to leverage the maximum output possible. Since the launch of VR capable smartphones, GPU’s like Adreno 530 and Snapdragon’s 821/835 can now handle more GPU overheads. That said, developing on mobile VR can be really challenging as you are still limited to a mobile grade GPU and CPU which cannot do as much as compared to a desktop grade graphics.

Currently most of the available mobile VR doesn’t support real time rendering due to their GPU’s limitation. Optimizing it using Open GL 3.1/3.2 rendering API and with proper optimization you can acheive awesome graphics on mobile VR devices as well.

Let’s start with draw calls.

With real-time rendering in mobile VR, each geometry is drawn twice which increases the GPU overhead, but with the single pass call, the entire rendering is done once instead of drawing the entire scene on each eye sequentially. Currently most of the available mobile VR doesn’t support real time rendering due to their GPU’s limitation. Optimizing it using Open GL 3.1/3.2 rendering API and with proper optimization you can acheive awesome graphics on mobile VR devices as well. It requires a lot of effort and time and in some use cases, that might be appropriate for you. For example, if you are targeting mobile games and you want to implement mobile VR, this can be challenging because for great graphics experience you need dynamic lighting and as the number of draw calls are limited, optimizing the VR game for mobile can be tough.

If you see the numbers of VR units sold, mobile VR has sold approximately 1.6 million units in 2016 which is way higher than the desktop Grade VR as it is expensive and has to be connected to a high-end gaming rig to maintain 90 hertz.

Another thing you have to consider is gameplay time, the maximum amount of time that a user can spend in VR. As per Oculus' recommendation you should not spend more than 45 mins in VR as it may cause VR sickness in a few people. Also, mobile VR like Samsung Gear VR with Samsung Galaxy S6 the max recommended time is around 45 min and with Samsung Galaxy S7 its around 1.5 to 2 hours depending on the complexity of the game. That’s the reason why games in VR currently have around 15 min to 2 hours gameplay time in mobile VR. There is only one full game currently, Resident Evil 7, which is a complete 7 hours game ported from PC gameplay to VR gameplay. Whoa.

That said, there are so many use cases which are currently unexplored because of tech limitations in mobile VR. This scenario will definitely change in the near future as the GPU’s and CPU’s will be able to handle more draw calls, and real-time global illumination and more real-time reflection/refraction post processing effects.

Questions? Please feel free to drop them in the comments below.

Like this story? Get more in the weekly newsletter from CinematicVR.
This article was first published on
VirtualSpaces.

--

--