CCP Games Explains Why Virtual Reality First Person Shooters Still Don't Work

An anonymous reader writes Icelandic studio CCP is better known for EVE Online, but its first foray into virtual reality with space shooter Valkyrie has caused a stir, and is widely seen as a flagship game for the Oculus Rift headset. In a new interview, Valkyrie executive producer Owen O'Brien explains what advantages the game will have when played with a headset — and gives his view on why a dogfighter is better suited to VR than a first person shooter: "People have hacked it together, but it doesn't really work," he says. "The basic problem is Simulator Sickness. In Valkyrie or any cockpit game or driving game, what you're doing in the real world, assuming you're sitting down, more or less mimics what your brain is telling you you're doing in the game. So you don't get that disconnect, and it's that disconnect that causes sickness. So, the problem with first-person shooters is that you're running or crouching or jumping in the game but not in the real world, and because it's so realistic it can make some people (not everybody) feel nauseated if they start doing it for extended periods of time."

Comments

Not if you use the Virtuix Omni

[gurps_npc]

The Virtuix Omni is basically an omnidirectional treadmill.

You use it in a VR environment and to move forward, you walk forward on the treadmilll.

This should solve the simulator sickeness issue.

Re:Go Linux!

[Anonymous Coward]

I'm sure the real reason is that it's not open source. Who is with me?

The medium is the message

[EmperorOfCanada]

Changes in the medium can have massive changes in the message that is best sent through that medium. Before TV radio plays were huge, but TV simply was a better medium. It wasn't that radio plays sucked but that telling episodic stories was done so much better on TV. Also when TV first started much of it was simply radio plays put back into a stage format and videotaped. Moving the camera through the scenery with lots of outdoor locations were a while coming and again the flat play like structure is still used in sitcoms.

Within even moving our internet browsing and gaming to mobile devices has resulted in wildly different usage patterns, there are the obvious ones such as using map tools more but Facebook does not seem to have translated to mobile as well as instagram, or twitter. Also the first person shooter largely has failed on mobile whereas I don't think that Angry Birds would have gotten much traction in a desktop only universe.

So surprise surprise VR goggles aren't turning out to be a screen you wear on your eyes but a whole new medium. I am willing to bet that there will be a genre that takes off on VR and that genre might not even really exist right now. Something really different. A simple example of different was that Wii games had a wildly different flavour than anything proceeding them. I don't remember a game prior to the Wii where I stood on a platform eagerly flapping my arms to propel what looked like a guy in a chicken suit though the air. Yet the Kinect games never caught my fancy as the games were often too serious and made me feel like a fool flapping my arms. The Just Dance game was close but was probably too late.

I am going to throw this one out there for free: Maybe the VR goggles will take off in Colorado and Washington with the blockbuster title being "The Stoner Olympics"

The obvious FPS that needs to be made then

[SuperKendall]

Wheelchair Hunter eXTreme

You're sitting down. You could even sell wheels that attached to the side of office char armrests... and a gun accessory that tracked position relative to your body to match the virtual version.

Or, a Battlezone clone where you are in an open cockpit.

Or is it unrealistic speed?

[Immerman]

Seriously - how many VR FPSes limit motion to anything resembling realistic speeds? Especially rotation. The average FPS has you running around at probably 20+mph and spinning 180* in a fraction of a second with a flick of the wrist. It takes me 3-4 seconds to complete a full rotation in real life at normal speeds, and if I spend much time turning quicker than that I start to get nauseous without any simulator needed. I can turn my head faster, but there's a lot more biofeedback maintaining orientation in that case. Spinning at FPS speeds should be generating massive accelerations of your inner ear, not to mention instantly launching from a complete standstill to a 20mph run - I'm not at all surprised that the absence of such accelerations throws people for a loop.

I haven't heard many complaints of nausea from the various VR first-person adventure games, and I can't help thinking tat that is largely because they are typically far slower paced than a twitchy FPS. An obvious solution would be more realistically paced FPSes. Or potentially even just considerably gentler accelerations. Maybe you can still run at 20mph, but it takes you 5-10 seconds to get there from a complete stop. Can't see any solution for faster spinning as radial acceleration is constant at constant speed, but then I'm not sure it's needed - being able to look around at a realistic pace should greatly reduce the need for instant spins, especially if you can aim independently from head tracking so that you can fire directly backwards at that guy just visible in the edge of your vision while looking over your shoulder.

Re:Or is it unrealistic speed?

[Anonymous Coward]

...especially if you can aim independently from head tracking so that you can fire directly backwards at that guy just visible in the edge of your vision while looking over your shoulder.

What you're suggesting presents a whole new class of problems actually (from personal experience from while messing about with the Rift). The issue is that from the VR's headset point of view, there is no difference between rotating your swivel chair and facing a direction and moving your head.
The reason this matters is for having intuitive controls. When you push forward on your d-pad, and you're facing forward, you expect to go forward. When you press forward while facing right, you expect to go right. When facing forward with your head looking right, when you press forward, you expect to go forward, but instead you go right. The problem obviously exists with only slight head movements too.
The solution seems to be to tie directional control to body position (or at least shoulder orientation or something), and view control to head position. This would imply that first person shooter, to be played how you expect it to be played, required additional sensors besides those found on the Oculus. The necessary workaround without the additional sensors is to tie movement to view direction most of the time, but that causes the problems you correctly identify.

Re:Or is it unrealistic speed?

[sexconker]

It takes me 3-4 seconds to complete a full rotation in real life at normal speeds, and if I spend much time turning quicker than that I start to get nauseous without any simulator needed

3-4 seconds? Is that 1 second per hundred pounds?

Karma to burn so fuck you.

[bmo]

I could have let this one slide, but I have a few things to say:

1. Darl, Darl McBride, is that you? When will you be testifying against Mark Shurtleff and John Swallow? You have a chance to redeem your soul! Imagine that!
2. The myth that "you can't make money on open source" is a myth so debunked that you have entire industries built on it, from servers and supercomputers to cellphones and kids' toys.
3. The myth that people don't get paid (slaves) to develop open source is belied by the fact that small companies like IBM are major contributors and specifically pay for people to work on open source code.

And even Microsoft pays people to do it now.

You can take your 20 year old arguments, write them out on oaktag, fold it until it's all sharp corners, and shove it straight up your arse.

Have a great day.

--
BMO

"You ate the poison mushroom!" reflex.

[Ungrounded Lightning]

The human body has three systems for balance - Inner ears (3-axis accellerometers and "rate gyros"), visual modeling, and muscle/tendon position & stress sensors - and needs any two to balance, stand, and walk properly.

It also has a reflex: When two of them disagree (particularly visual vs. ear), it is interpreted as "You just ate a neurotoxin! Get it out NOW and we MIGHT survive it!"

Thus nausea, projectile vomiting, explosive diahrrea, and clothes-soaking sweating if the mismatch is strong. If it's smaller - nausea. ("Whatever you just ate may have been toxic or spoiled. So you're not going to like it anymore.")

Of course other things than being poisoned can trigger it:

Diseases that temporarily incapacitate or permanently damage the inner ear are one class. (For instance, Meniere's Disease, where the pressure-relef valve for the inner ear sticks, the pressure rises, and the membranes with the sensory nerves tear. Result: Sudden extremem vertigo attack - hours on the floor - followed by days or weeks of gradually reduced incapacity until the brain maps out the change to the ear - followed by another tear and repeat indefinitely. Very high suicide rate.)

Vechicles, where you may visually fixate on the accellerating inside rather than the surroundings, are another: Cars, boats, ariplanes (and the corresponding car/sea/air sicknesses) are notorious, as are carnival rides and trains. For relief, make a point of looking at the horizon or otherwise the exterior. Eventually the brain may learn "I'm in a vehicle. Ignore the weird signals from the ears. (That's why vertigo sufferers may NOT have attacks in MOVING vehicles...)

And, of course, VR mismatches - to the point that there is a term of art: "Barfogenisis" (I hear the lengths of some of the rides at Disneyland are calibrated so they end and the crowd is out into the hall just BEFORE the effect would become pronounced.)

Which is why FAST flicker still causes vertigo.

[Ungrounded Lightning]

I find that any kind of response time lag between my inputs and the real world, especially when it varies, is what makes me sick ...

My wife has vertigo. Her attacks can be triggered by fluorescent or high-pressure arc lights where the flicker rate is above the flicker-fusion rate of the eye. (This makes trips to warehouse stores problematic - they have to be short or she'll be down for the rest of the day. That's hard at, say, Costco.)

I used to wonder how this could be, and finally realized that the "strobe light" effect produces small, but significant, errors in observed position of the background items (shelves, etc.) that she uses for reference to balance despite the damaged inner ear.

When they first began using fluorescent lights in factories - in the days before guards over moving machinery were common - the worker injury rate went 'way up. Turns out the lights made the AC-powered motors, turning at or near an integer fraction of the line frequency, look like they were stopped or only moving slowly.

The fix was to build the light fixtures in two-tube versions, with a capacitor and an extra inductor in the balast, so the "lead lamp" and "lag lamp" would light at a quarter-cycle offset. In combination with suitably persistent phosphors this made them largely fill in each other's dim times, enough to make fast-moving parts blur and look like they were moving. For large arc lights, a similar fix was to arrange them so adjacent lamps were distributed among the three phases of the power feed, rather than having rows or patches of lights all flickering in unison.

Unfortunately, this lore has apparently been lost - at least outside the specialists wiring factories full of moving parts. Warehouse stores have rows and sections of arc lighting all wired to the same phase. I'm not sure, but I don't think the new electronic ballasts for flourescent lights do the lead-lag thing, OR have enough raw filtering capacitance to power the lamp through the phase reversals. (And then there's LED lamps...)

It's not a safety hazard these days, now that OSHA rules have all the fast-spinning machinery covered with guards. But for those with vertigo it's a big problem.