Month: November 2021

In 2016, I tried virtual reality (VR) for the first time and came away with mixed impressions. This happened at an art museum, and one of the exhibits was a VR trip into an surreal, simulated environment. After donning the VR goggles, I was immediately struck by the immersiveness of the experience, even if the graphics were relatively coarse. It was remarkable how quickly and automatically my brain accepted my virtual environment as being real, in spite of many cues to the contrary (such as the aforementioned coarse graphics, unnatural “stovepiping” of my field of view, and the feeling of the weight and pressure of heavy goggles on my face and scalp). I instinctively treated objects in the game as if they were real, which nearly caused me to fall when I tried to lean on a coffee-table sized virtual object on the ground when there was nothing there in the real world.

I came away from it thinking VR technology had major potential, but also important limitations that might not be solvable. As my own brief experience made clear, using the technology can be dangerous when your virtual and real-world surroundings don’t correspond. You could too easily walk face-first into your living room wall, or shatter the window with your fist during a VR boxing game.

Moving around VR environments on foot is an even bigger challenge for the same reason. While an omnidirectional treadmill could theoretically solve this problem, only rich people can afford them (and the prices are only declining slowly), and they’re not instantly responsive to changes in your velocity. If you are walking on such a treadmill and you decide to, say, suddenly step to the left, it takes a moment for the machine to sense the corresponding changes in the downward forces exerted by your feet, to deduce that you are starting to step right, and to start moving the treadmill belt in the opposite direction. This time lag between human action and machine reaction breaks the illusion of the virtual experience and can easily make you lose your balance (which is why omnidirectional treadmills have harnesses or circular railings for their users).

An experience I had in mid-2019 (yeah…this blog entry has been malingering for awhile as an unfinished draft) showed me that this “movement problem” might have a surprisingly easy solution. I had some spare time, so I stopped into a VR gaming arcade and played a first-person shooter game called “Arizona Sunshine.” Unlike the VR experience I had at the art museum, this was a professionally designed virtual environment. I also played it on a top-end HTC Vive device. The graphics were much better, though not nearly as good as the graphics in a game played on “2D” television connected to a modern game console like PS4.

I held controllers in either hand, which resembled barcode scanners and have several buttons. However, what struck me the most was the manner in which the player moved in the game. When you first put on the VR goggles, you must remember to stand in the middle of an open floor space measuring about 8′ x 8′. The goggles “sync” with this original starting space, and if, during the gameplay, you walk too close to the edge of the 8′ x 8′ square, a grid of lines appears across your field of view to indicate where the edge is. You’re supposed to step back when that happens.

For moving longer distances, you “teleport” by first holding down a specific button on one of the controllers, whereupon a curved, rainbow-like line emanates from the “gun barrel” of that controller. You move the controller to shift the spot where the rainbow line touches the ground to the location that you want to teleport to, then you release the button on the controller, and you’re there, instantly. Surprisingly, teleportation isn’t disorienting. It is an ingenious solution to the VR movement problem, and resolved some of my old doubts about the technology’s potential.

The remaining obstacles to VR’s mainstream adoption, and probable remedies to those obstacles are:

High costs. A pair of Oculus Rift VR goggles and two hand controllers costs about $400, and to work properly, the goggles need to be plugged into a desktop computer with high graphics processing specs. Computers meeting these requirements cost at least $600, pushing up the minimum total cost of an Oculus VR system to $1,000. The Rift’s closest competitor, the HTC Vive, also requires a powerful desktop, and has a higher total system cost. Compare that to a Playstation 4 console, which offers better graphics than either VR set and sells new for $270. You probably already have a TV to play PS4 games on, but if not, you could buy a 50″ set from an excellent brand like LG for $500, and it gives you the added benefit of being able to watch all sorts on non-game content like TV shows.

Hardcore video gamers are already willing to pay this price premium for the 3D experience, but the vastly larger number of casual gamers and poorer gamers won’t be interested until VR system costs get much lower. I think Sony’s approach shows the likeliest solution to the problem. Their newer PS4 game consoles–which can be thought of as high-end desktop computers that are relatively cheap since they are optimized just for gaming–have ports that you can plug Playstation VR headsets into. The console does all of the data processing, and the headsets merely act as displays.

Quietly integrating VR capabilities into game consoles, selling them to mainstream gamers who are, at this point, only interested in playing 2D games on TV screens, and then doing a marketing push later on to inform them that, for just $100 more, they could buy a VR headset, plug it into the console they’ve already paid for, and try out VR games, is probably the best and likeliest strategy to popularize VR technology. The current Playstation VR headsets have mediocre graphics, and the PS4 console isn’t as powerful as an Oculus Rift desktop, so I predict we’ll have to wait until the late-2020s for the price-performance of the headsets to improve enough, and for a new generation of more capable game consoles like PS5 to arrive, before VR gaming gets cheap enough for widescale adoption.

Lower-res graphics. If my own experience playing “Arizona Sunshine” is any indication, the graphics in VR versions of games lag the graphics in 2D versions of those same games by about one console generation. Wearing the headset, my zombie town surroundings had the same level of detail as what I remembered from the typical PS3 game I played on my TV, and the state of the art now is PS5. Granted, the immersive quality of VR gaming goes a long way to compensating for worse graphics, but I think more improvement is needed before VR’s customer base can get into the tens of millions.

I think if VR headsets displayed PS4 levels of graphics, then that plus the immersion factor would be awe-inspiring to enough people to make VR gaming go mainstream. I really think that being able to play Detroit: Become Human in VR would be captivating to average people. Just check out the graphics and imagine yourself immersed in this virtual world:

Detroit: Become Human is a PS4 game, so the one-console generation lag time means VR games that look that good will be available a few years after PS5 and XBox Series X are released. As mentioned, I think we’ll reach that point in the late-2020s.

Heavy, bulky headsets. This isn’t as big of a problem as the previous two, but the weight and pressure the VR headset exerted on my head and face were a little distracting, hurting the immersiveness of the VR experience (remember, there aren’t supposed to be any physical reminder that you’re not actually in the game environment). I also suspect that long-term use of this device could cause neck fatigue and compression headaches.

The problem can and will be eased by making the display screens thinner and lighter. Just as TV screens and computer monitors have gotten thinner thanks to better technology, so will VR goggle displays. Significant progress on this will surely happen by 2030, and the VR headsets of that year will be lighter than those of today, but it alone won’t solve the problem.

The human eye’s inability to focus on very close objects has also forced goggle designers to position the display screens around four inches in front of the wearer’s eyes. The resulting forward-heaviness of the goggles creates torque, making them feel even heavier, in the same way that a bag of groceries feels heavier if you try carrying it with your arm outstretched perpendicularly from your body as opposed to hanging down parallel to your body. Even if the screen itself is just a few millimeters thick, if it has to be four inches in front of your face, it will make the whole rig feel heavy.

A major innovation in VR image display technology that circumvented the limitations of the human eye and allowed the goggles to protrude less from the face is needed. Advanced glass lenses or retinal projectors are candidates, though I can’t give a timeline for when they or any other alternative will be commercialized.

Cords. The long cord that supplied my headset with electricity and data was also a bit encumbering, and in longer game sessions, there’s the risk of getting yourself tangled up and tripping and/or damaging your equipment. At the rate technology is improving, a wireless headset with the cost and performance specs I’ve listed so far should exist by 2030, though it will have noticeable limitations compared to corded headsets.

The most plausible setup would have the game console do most or all of the data processing and wirelessly transmit the data to the headset. The user would have to stay in the same room as the console to receive the data without lag. The headset would have an internal battery that provided maybe two hours of play–enough for a casual gamer who does it after school or work. Such a headset would have a port into which a power/data cable could be plugged to recharge the device or to do corded VR gaming if the user didn’t care.

I doubt VR technology will be so advanced by 2030 that it will be practical or common for people to rely solely on their goggles’ onboard batteries and computers for gaming, so it will be rare to see people wearing the goggles in public places, and those who do will struggle with some combination of short battery life and/or unsatisfying graphics. However, as I’ve argued here (partly based on my personal experiences), the goggles will be sufficiently advanced by that year for the medium to have gained widespread popularity, even if people can only really use the devices in their homes or in arcades.

All of this is to say nothing of the very best and most expensive VR goggles that 2030 will have to offer. Considering that there already exist goggles that can display nearly lifelike imagery, such as the “Varjo VR-3,” the cutting edge of technology nine year from now will be mesmerizing.

In conclusion, the 2020s will be the decade when virtual reality goes mainstream thanks to the quality and price of the technology reaching inflection points. VR goggles will not need to fix every deficiency of the VR experience (e.g. – imperfect graphics, unnatural in-game movement, cord and bulk of goggles disturbing the illusion of being in the game) before mass consumer adoption can happen. By 2030, however, there will be very expensive and very advanced VR devices that provide 99% lifelike audiovisual experiences, pointing the way to what will become widely available as that decade unfolds.