The Ultimate VR Experience™. What is it?

[cybereality]

Ok, this thread is a spin off of some comments in the mega-HMZ thread (already too long so discussion is here).

So what is your Ultimate VR Experience™? What is it that current game and simulation software is lacking?

This topic is for talking about software *only*. Lets just assume that we already have high-FOV high-res HMDs, working headtrackers, data-gloves, fast optical tracking, etc.

I will start: I'd like to see richer interactions with objects and space, and the people within them. Just basic stuff that current games don't allow you to do. For example, pouring water into a glass, and then drinking it. Maybe pulling a match out of a book of matches, striking it, then lighting up a cigarette (even though I quit years ago). How about kissing a girl? I'm sure most guys would want to do that.

How about simulations that aren't all about violence? Like a simulation where you could buy computer parts and build an actual working computer inside the machine? Or what about an adventure game where you could actually talk to game characters and hold entire conversations? I am thinking like you just go into a bar in the game, order your favorite drink (and maybe way in the future actually get drunk in real life...) and then spark a conversion up with some cute girl. Or have a heated debate with a college professor. Maybe a simulation where you have to go shopping at the super-market, and you get graded by how much you can save, or how healthy the food is. And then you have to go home and cook it. Not every experience has to be going John Rambo and saving the world.

So what other kinds of things would you guys like to see?

[pierreye]

Well, violence or/and sex sells! I would prefer that in the VR we do things that we normally wouldn't do in actual life.

Other things that can be simulated like roller coaster, bungie jumping, white water rafting etc but all these need external g force simulation to make it realistic. Going to try out Virtual rollercoaster using software such as Nolimits rollercoaster which support mouse view.

http://www.nolimitscoaster.com/

[brantlew]

I would like to see a new form of theatrical entertainment emerge from VR. It would be like a hybrid of stage theater and film. Stage theater in the sense that you would be somewhat offset from the performers - watching everything from the outside of a group of actors. But also like cinema because the setting would be untethered. You would travel along with the actors from environment to environment and scene to scene, and your placement relative to the action would change.

It's really hard to imagine because most of us are so conditioned to film editing, but so many of those techniques would not transfer. For example - closeups are used all the time to focus the viewers attention on some emotional nuance of the actors performance. But in VR it would be incredibly uncomfortable to be thrust right up into someone's face. Not to mention the viewer would be free to look at whatever they wanted to in the scene. As a director, how do you focus attention on the correct part of the action. Actors might have to return to the older style of "over-acting" to grab attention and convey emotion over a larger distance.

I just think it would be really interesting to watch it evolve as an art form.

[adventurer]

To me, an ultimate VR experience not only need to stimulate audio visual part but also other things like force feedback, smelling a flower, feeling the wind fondling your face. We can do all these kind of things by current technology used in 5d cinema, cyber carpet,force feedback tech etc. We don't need to wait nanotech, brain interface to do it.

in VR environment, I want to run freely through skyline in a fantasy, fly like superman, have abilities like xmen, martial art fighting like in jackie chan or jet li kung fu movies, gun fighting like in current FPS games today, etc. Also, I don't like violence so I will want more peaceful experience but not less interesting. There are a lot of wonderful things to do with VR. All of these actions above need a tremendous amount of detail to be able to bring realistic feeling. However, all the games and softwares now are lacking the detail necessary to create VR wonderful experiences.
Another thing is that in VR environment, I want relatively realistic time system. I mean when being in a story in game, the game often fast forward to the important parts. For example, the game just say "3 weeks later" and pass all these time. It must give the player option to really live these 3 weeks in game. Or when night comes, it just fastfoward to the next morning, I really want to spend a night sleep in these games to get the feeling of really living in there.

the next thing current software and games is lacking is AI. The AI is not strong enough to stimulate different personalities, get to the point that we forget we are talking to a computer. Having ability to make real conversation an interesting person simulated by computer during an epic quest is really wonderful experience.

[WiredEarp]

I have some ideas for a FPS:
At the start, you have to use your hands to pickup the weapons from the armoury you wish to use. Then, you go to the ammo table and grab magazines individually and attach them to your belt. Looking at your gun, you can move the selector switch and look through the sights, gaining magnification if the sights are actually a scope. To change magazines, you have to remove the existing one from your gun, grab a new one from your belt, and insert it (you'll feel a click). When you fire, the weapon kicks in your hands. When under attack, you can put your gun over and around obstacles to shoot. Some levels, you have to watch your footing, for example fighting high amongst the girders of a construction site. When shot, you can hold the wound to slow the rate of bleed out until you find a medikit - where you need to grab a bandage and place it on the wound. When you do this, you'll notice blood all over your hands. To throw a grenade, you'll grab one off your vest, pull the pin, and actually throw it.

Cyber, I think your bar game idea would be big in Japan . I wouldn't have much interest myself personally tho in shopping at a supermarket haha. Saving the world sounds way more fun! I agree totally with the 'richer interactions' however.

Brantlew, your idea sounds almost identical to one i've had about soap operas which are recorded on multiple kinect type cameras, digitized to 3d environments and retexturemapped, giving the viewers the ability to move around the sets and characters, following their favorites etc, so each episode everyone could choose their own viewpoints. This idea has even more validity for things such as sporting events and in fact I believe this to be the future. Sitting back, watching a soccer game from any position you wish, following the poeple you wish to follow - or watching a boxing match with the bloody ref removed

While we are shooting ideas around as well, how about a shared movie/game combination, where you buy a ticket for an hour long 'cybermovie'. You then have a terminal that provides you with yuor view into the game. On starting, everyone is assigned a role from the plot, and you are told what your character is like. From then on, how well you play your character gets you points, and at the end of the 'movie' there are awards for 'best actor' 'worst actor' etc. During the game, the viewers all can work together or in opposition to achieve their game goals.
I'm picturing as an example, something like 'Lost', where all the viewers/characters are on a plane that crashes. Then, everyone has to work together to not only achieve the team goals, but their individual ones if possible (or choose which are most important to them). If your character is killed, you could be given a new (previously NPC) character to play, or if that is not possible, you could be put into an 'audience' of eliminated characters that could still affect the gameplay by voting on effects and pathways the plot can go down.

[FingerFlinger]

In the near future, I'm most excited about about being immersed in the popular video games that we all play, but I think that the ultimate experience will be an alternate reality, a la Snow Crash or Ready Player One. For many, the ability to virtually visit distant friends and attend events will lessen the importance of travel and allow people to stay connected in a much more intimate way than is possible with telephone or video chat. I think this type of environment would elevate VR to an essential part of life, the way that the Internet is quickly becoming.

That said, a general-purpose VR setup is really difficult and expensive to make. I personally want to focus my efforts on very focused VR experiences that are as compelling and immersive as possible. For instance, my big project coming up in the next year or so is going to be the Tour de France Google bike, see here http://www.engadget.com/2011/07/05/goog ... ance-refu/


I work for the company that makes these and have managed to get hold of one. I need to make sure I can tear it apart first, but I want to create a little demo with HMD support. Imagine a bicycle racing game where the bike follows the terrain incline, the pedal resistance changes appropriately, and you can turn your head around to see how far behind your competitors are. The only major missing piece for total immersion is simulating actual g-forces!

I also want to throw one of these http://brewersledge.com/products/treadwall/treadwall-m4 in a CAVE.

[brantlew]

As others have alluded to, telepresence will be an important application for VR and I think we are getting close to the necessary technology to create a modest implementation (keep in mind that people are perfectly happy with 320x200 @ 10fps video conferencing so it doesn't need to be perfect - just acceptable and easy to use). With a minimal HMD (Google goggles) and a Kinect-like device it might be possible to transmit a low-resolution body-tracked avatar and achieve a modest form of telepresence. Keeping the worn equipment to the bare minimum would help enable acceptance by the masses.

[CyberVillain]

I hope we will see consumer targeted compact omni directional thread mills in the coming years, something compact and light enough to store under the bed or similar. Together with gun, head and eye tracking and games with proper support for these we will have my dream of an ultimate VR experience.

[PalmerTech]

Such a fun topic.

Some concepts:

1) Long form experiences are key to building strong emotional connections. Most games focus on an adrenaline response, that can change with VR. Remember going to summer camp as a kid? Even though you were there for a relatively short period of time (Several days to several weeks), being wholly immersed in a new environment made it feel like home very quickly. When you left camp to go back home, it was usually pretty heart and mind wrenching! Imagine, if you will, being in a simulation for an entire day/weekend/week, with minimal breaks for the real world. A complex game that fosters emotional connections to the characters could be absolutely amazing!

2) Games you can play only once. You don't get second chances in the real world, that tension would add greatly to any virtual experience, in my opinion. On top of that, you have the rose tinted goggles effect: People think things were better than they actually were! There have been studies comparing sports memories between people who took photos and people and took video, and those who were referring to photos had more positive memories of the experience than people who took videos, possibly because people who saw the video had less leeway to exaggerate the memory. Being able to play and replay games is great, but I think it would be very cool to have a game that you can play one time and one time only, no second chances.

3) In most games today, dying is the worst thing that can happen to you. In real life? Not so much. People are often more terrified of letting other people die, particularly loved ones. A VR game that plays on this fear would be sweet.

Some actual game concepts:

1) A game that showcases the "boring" side of war. War is 99% sitting around and 1% actual fighting, so living through the parts we never see would be very interesting. Losing a squadmate you have been talking to and joking around with for days or weeks would really hurt.

2) Online roleplaying communities, along the line of the "interactive movie" idea that WiredEarp brought up. It would be a lot more accessible than real world roleplaying, since you have a much larger pool of like minded people, and none of the financial travel/costume/etc expenses.

3) Sex. Sex always sells. Maybe not with an AI, but cybersex via text is already extremely popular in some circles. I am sure they would jump at the chance to have something infinitely better!

[cybereality]

So I played this indie game called "Dinner Date" and its exactly the type of experience I'd like to see more of. Basically you play a bachelor in his 20's sitting in his kitchen thinking. The guy has a date, but the girl is late. So you follow this dude for like 30 minutes listening to the stuff going on in his head. You can also have basic interaction with the environment, drinking wine, eating bread, etc. Although the control scheme is pretty bad (just uses the keyboard and is unresponsive) I can imagine how cool this would be with a VR-style controller (Hydra, Leep, etc.). Worth checking out, I think the game sells for like $3 on Steam, very cheap.

[youtube-hd]http://www.youtube.com/watch?&v=VF_Mv3OzLpQ[/youtube-hd]

[brantlew]

I don't get it. It seems like "experience" - just for experience's sake, but the topic is so mundane. It's like the scene in Being John Malkovich when they enter his mind while he's brushing his teeth. The novelty is over in a couple of minutes. Now there are experiences that could simply stand on their own without a plot like VR mountain climbing or VR pornography. But with something this mundane it seems like it would need some kind of "plot" to get you involved.

[cybereality]

Yes, it might sound "mundane", but I think that's sort of the point. Not every virtual experience has to be saving the world from nazi-alien invasion.

[FingerFlinger]

Cybereality, I am with you all the way. Especially because VR is so unexplored, I think small experiential sketches like this are important. We need to develop a toolbox of techniques specifically for VR design.

One idea I have kicking around in my head is a short experience at a jazz club. You walk in, take a seat at a table, listen to the music, and have a drink. I think that that having actual table and drink props could push the immersion through the roof and really give a sense of "being there".

[Chriky]

Personally I don't think smell and taste are really something worth striving for. I have a friend who is anosmic and I don't think he feels like he is experiencing a significantly reduced version of reality.

Assuming we have sight and sound nailed, then the big one is touch. Unless something seriously game changing is invented that literally let's you send fake brain signals, touch is never going to be able to be simulated properly. If your virtual environment contains a bar, the player has to be able to do a pull-up on it. As far as I can see, this means that any VR system that uses redirected walking cannot simulate touch.

With this in mind, I would want to see VR arenas that have 1:1 movement tracking, with objects that are also tracked, that you can pick up and move, doors you can slam shut etc.

Some kind of laser tag that was VR-ed up to make it seem like you were in the middle of a warzone, with planes flying over head would be awesome. The other advantage is it let's you play co-op against an army of CPU players (you wouldn't be able to touch these guys, obviously) who wouldn't mind if they had a crap game. The problem with making paintball etc exciting is that if someone does a daring rescue mission, ten other guys have to be grunts that get boringly shot in the process.

There's a fair amount of stuff you can do with a setup like this; MGS style optical camoflague, lightsabers, magic casting, xray vision...

[brantlew]

@Chriky: Here you go. This is why PalmerTech is so lucky.

http://www.mtbs3d.com/phpBB/viewtopic.php?f=120&t=13779

One interesting caveat of this system is that they do use redirection motion, but they use "stage hands" to move props in and out of the areas when they are needed.

[FingerFlinger]

I think there is a potential for using redirected motion in combination with 1:1 translation. I imagine a warehouse that is mostly open space, but some region in the corner has tactile props such as a door, or a table and chair. If the game can understand the player's intent to interact with these objects, it can direct the player towards that region and then turn off the redirected motion in favor of a 1:1 mapping.

[JohnnyGrant]

For me the ultimate VR experience would be a Blade Runner/Shadowrun experience in a not so distant future...like being in the Shadowrun Universe and being able to log into the Matrix inside the game would be a dream come true

[Zaptruder]

I want a game that marries together Skyrim, Minecraft and Uncharted.

Sure it'll have violent elements, but the main point is really that you get an open world that you can freely explore with a wide range of motion and can create and construct your own emergent gameplay in.

Or maybe I'm just waiting for second life 2.0 with graphics and tools that aren't archaic and dilapidated. I'm all about that emergent creative possibilities, and using VR to enhance the reality of those visions.

But the ultimate vision of VR in my mind - perfect sensory simulation with isolation from real sensory inputs (i.e. The Matrix), with a VR based internet. Websites aren't just 2D planes with hyperlinks - they're virtual environments, allowing us to jump freely from one environment to another - each environment with its own set of rules, customs, regulations - but none are subject to the traditional restrictions of society - unless they so choose to be (in which case, they'd have to provide a compelling experience over their competing virtual environment/websites).

[profvr]

Personally I don't think smell and taste are really something worth striving for. I have a friend who is anosmic and I don't think he feels like he is experiencing a significantly reduced version of reality.

Assuming we have sight and sound nailed, then the big one is touch.

The group I work with has done a lot of work on reactions to immersive simulations. If there is one thing we have learnt, its that the brain is very (very) good at noticing discrepancies or lack of information. If you see a full cup of coffee and it doesn't smell like coffee, some part of the brain starts to "worry" about the inconsistency and such worries lead to a variety of response, the most damaging of which is the loss of what has been called the suspension of disbelief, i.e. the person stops playing their role in the environments. The second thing we learnt is that individual reactions and adaptability to such reactions varies a lot. Some people care very little about audio,for others its essential. Its a bit of a stretch, but worthy of study, that the reason some people don't engage with 3D games or media is because they lack some sensory cues that the person is particularly sensitive to. Your friend's normal reality contains no smell sensations, so he doesn't miss them in a virtual reality; someone else might miss them significantly.

Touch is the difficult one, but bear in mind that in an immersive system you get lots of touch feedback from your clothes, the things you are holding, etc. There is quite a bit of work in the academic community on how you (the player or user) are represented in the world. I've had someone in an experiment worry about the fact, and state that it broke all sense of illusion of the immersion, when they saw themselves wearing a jumper when in the HMD, but they actually had just a t-shirt on. This broke their illusion of immersion.

The third thing we've learnt is that the brain is easily fooled. In another experiment, sometimes the participant will come out of an immersive experience claiming that there were touch feedback cues when in fact there were no such cues. Its this type of effect which makes immersive virtual reality such a fascinating area to study.

[brantlew]

@profvr: In my free walking experiments, one thing that takes me out of the illusion are steep inclines and declines. Walking uphill in the virtual world with your head looking up but your body just moving on a flat plane just feels odd. I will sometimes adjust the camera view so that I'm looking straight ahead in the real world just so that the tactile feel of walking matches my visuals.

[profvr]

@profvr: In my free walking experiments, one thing that takes me out of the illusion are steep inclines and declines. Walking uphill in the virtual world with your head looking up but your body just moving on a flat plane just feels odd. I will sometimes adjust the camera view so that I'm looking straight ahead in the real world just so that the tactile feel of walking matches my visuals.

This is a really good example of the problem: your proprioception (balance and joint sense), don't match with the visuals. As a programmer, you try to pick the least worst solution to remove the conflict!

The impact of sensory conflict is a very ripe area for experimentation. For example, even if you discount slopes, some people are going to be put off by the fact that, e.g. they see grass but feel concrete under their feet. I think USC had a paper on redirected walking over multiple surfaces. I hope someone is working on redirected walking over uneven terrains. I think one physical slope might act as a good proxy for a range of slope angles in the world.

[Zaptruder]

@profvr: In my free walking experiments, one thing that takes me out of the illusion are steep inclines and declines. Walking uphill in the virtual world with your head looking up but your body just moving on a flat plane just feels odd. I will sometimes adjust the camera view so that I'm looking straight ahead in the real world just so that the tactile feel of walking matches my visuals.

This is a really good example of the problem: your proprioception (balance and joint sense), don't match with the visuals. As a programmer, you try to pick the least worst solution to remove the conflict!

The impact of sensory conflict is a very ripe area for experimentation. For example, even if you discount slopes, some people are going to be put off by the fact that, e.g. they see grass but feel concrete under their feet. I think USC had a paper on redirected walking over multiple surfaces. I hope someone is working on redirected walking over uneven terrains. I think one physical slope might act as a good proxy for a range of slope angles in the world.

It seems to me that until we get to a direct nerve interface that can also retard actual stimuli, we will always be in a situation where the sensory reality of the VR environment fluctuates on a moment to moment basis... and it can differ from person to person dependent on how they emphasize/value information on their senses.

So, standing, without touching, with a screen with good resolution and large field of view - is sufficiently convincing for most people. Once you start involving more angles from which information can become disparate from our expectations, the sense of immersion will fluctuate wildly. If you swing with a sword and miss, it'll seem ok. If you swing and hit - your arm will continue travelling while your VR arm should collide with an object and come to a halt. Similarly, if you move your head to the left; translationally, and your view doesn't move with you, then it's going to reduce that immersion factor.

Visual-spatially, we're really close to getting a 'good enough' solution for really high quality VR.

For the rest... maybe we can rely on the neuroplasticity of our brain to create a new sense for the information that we can't otherwise present to the user.

This talk http://www.ted.com/talks/lang/en/neil_h ... color.html
is from a guy that is completely colour blind - but has had a colour frequency to audio translator attached to create a new synesthetic sense where he perceives colours as sound and even music.

So like I was suggesting, if we can present consistent cues of the information that should be there, our brains should be able to generate a new sensory experience to bridge the gap in immersion. I suggested in my 'Ideas for full motion VR' thread that because of the impracticalities of tactile feedback (having to wear an exoskeleton), it might be possible to present some sense of that information through audio/visual cues. That is, VR will always present our limbs as it's been tracked - but when it's synced, it's invisible. When it's unsynced (in the case of collision in the VR environment stopping the virtual limbs) - your real limb is shown ghosting and tearing away - if it gets too far, your virtual limb becomes limp, and you'll need to re-engage it.

Over time, you'll come to respect the reality of the virtual environment - and almost act like a mime; stopping your limb from clipping through the virtual plane without the need to touch it*. The combination of proprioception and audio/visual indicators should nonetheless produce something that's substantially more immersive - perhaps as immersive as we can make games and interactive media until direct sensory stimulation technology (which is at this point just sci-fi).

*I'm simulating air-punching an imaginary wall as I speak. It's tactile reality does not seem ridiculous to me, even though it is imaginary. Of course my fists don't feel the impact, but the rest of the muscles in my arms and chest contract violently and suddenly as if something (my brain, rather than physics) were stopping them.

[FingerFlinger]

As an aside, I've come up with an idea for the sword thing. Imagine you've got your sword prop and attached to the tip is some high-strength cord. You've also got some crazy apparatus that reels in and lets out the cord as you wave the sword around, but as soon as you 'hit' something in the game, the apparatus drops a clutch and brakes the cord. In my head, this seems impractical and highly prone to mechanical failure, but I think it could probably work. You couldn't really make it work in a walk-around scenario, however.

EDIT:@brantlew: Regarding looking up the virtual hill, while actually walking on flat ground... Have you tried setting the camera pitch to the character's direction vector? It would automatically accommodate the terrain. Just wasn't clear if you are resetting the camera view manually or in code.

[CyberVillain]

We need tech to jack into the central neversystem, and we need it now!

[cybereality]

We need tech to jack into the central neversystem, and we need it now!

Well are not quite there yet, but making progress every day. They just recently discovered how the signals are coded from the eye, allowing implants to restore sight to blind people:

http://www.extremetech.com/extreme/1344 ... hetic-eyes

[FingerFlinger]

Interestingly, if the brain doesn't learn how to see early in life, it's exceptionally difficult, if not impossible for it to learn later on. Some scientists blindfolded kittens for a weeks after their birth and the cats were essentially blind for the rest of their lives.

[CynicalWanderer]

For me, just walking around a well crafted 3D environment and looking at the artistry of the scenery is the main VR experience I want for now. I'd love to just wander through the worlds of various games like Elder Scrolls, Coruscant in a Star Wars game, Deux Ex, Fallout 3 etc and just gaze and explore, even mod the games to turn off all combat and quests and HUDs so that all that remains is a living world. Could even potentially port older games like Quake and Painkiller into modern engines to spruce them up and relive the moody atmosphere of our gaming history. Or load up non-game worlds such as renders of actual Mars terrain based on rover images, swim with aquatic life under the sea, or simply just explore some real city one hasn't been to. It's all exciting. My point here is that the interactivity requirements for the experience that I consider ultimate VR are reduced to mostly vision and simple movements - these are more important to me than haptic realism or complex movements like jumps, ladder climbing, and fast-paced gunplay with strafing and combat rolls, all of which will be more challenging for the community to solve. Sure, others will require those elements before they're satisfied, and I'll definitely play those sorts of games too when the solutions are found that work best with them, but I'll happily take a nice casual stroll through Dear Esther, with more limited interactivity, in the meantime.

So with that in mind, once you take away the demands of twitch gameplay and other wish list items, what I'd most like to see in this first generation of the resurgence of VR is leisurely pacing (which should also minimise motion sickness), and for that, I think a good HMD and a cheap linear manual treadmill should be sufficient for decent immersion. You'd hold onto the treadmill bar for stability and then walk forwards and backwards more or less normally, giving a low cost, off the shelf solution and avoiding the problems of motorised treadmill inertia or ball bearing slipperiness that other VR walking solutions currently suffer from. Probably stick a joystick or Xbox controller on the bar for one hand to use for turning and other interactions, or go all the way and integrate the Razer Hydra or Leap up there, for more possibilities such as showing one's hands and handling virtual objects. Perhaps even construct some sort of stiff plastic hip ring that extends from the treadmill bar to around your waist, so you'd still move the treadmill by exerting force against the bar but using your torso instead of your hands, for better hands free motion. The idea of all this being to keep the costs affordable enough to still achieve the main goals of visual and basic locomotive feedback, which I believe provide the most critical elements for the feeling of bodily immersion, and the rest can come later.

Anyway, just some ideas that I've had for awhile, which I'd shelved until recently when OR made it all seem possible again. When the Rift comes out, that will cover the HMD aspect, and at that point I'll probably try experimenting with the treadmill, and maybe modding some older games to remove all of the combat and forced head bobbing and blowback etc. Smarter people than myself will no doubt solve the other aspects of haptics and free movement soon enough, and at that point I'll be happy to upgrade my expectations once again!

[android78]

I think that, for the most part, the games trying to simulate certain things can cause issues. For instance, a common example is swinging a sword. If you have actual hand tracking, then the program uses inverse kinematics based on collision detection as you slice through an avatar, that is going to confuse the senses. On the other hand, if your sword continue through the object unobstructed, it will seem like you are just slicing through a very soft object. By adding some feedback (I would like to see more electric impulse feedback being worked on), then it will feel like you are slicing something that has a 'texture' to it. While not 100% accurate, it should help with immersion, while not breaking it.
There are obvious problem with what to do for solid objects, but I still think that allowing your hands to appear to just pass through it is a better solution for VR, then having your view appear out of sync with what your feeling/doing.

[cybereality]

You can always "fake it" by actively adhering to the limits of the simulation. Should work for most cases, though obviously not nearly as good as real haptic feedback.

[android78]

@Cyber - what do you mean by 'fake it' and 'actively adhering to the limits of the simulation'? My point is that you are better off keeping your view of what is happening in sync with your limb position in the world.

[cybereality]

What I am saying is you can try to manually keep your body in sync with the game, in the cases when they would otherwise go out of sync. So, for example, in a sword-fighting game you would have to stop your swing at the point of impact, and not just wail around freely. Or if you were picking up a glass of water, you would stop gripping at the point when your virtual hands touched the glass.

[rmcclelland]

I want to put on my VR gear and have a beer with my best friend, who lives on the opposite coast. With good body and facial expression tracking, I think I could be fooled into thinking his avatar is him. We could cease to remember we are in a simulation and just hang out. I see virtual presence as one of the most transformative VR technologies. Maybe it will be the ultimate solution to DC traffic jams...

I want to put on my VR gear and walk through a beautiful landscape. Skyrim would be a start, but Lothlórien (LOTR) would be amazing.

I would also like to see convincing virtual beings (agents). It seems all the chat rooms, facebook posts, forum traffic, even phone conversations could be digested into a believable chat-bot. VR and AI are a powerful combination.

Finally, I want to fly. I always had the most wonderful dreams of flying. Not quite sure how the acceleration would be handled. A hand-glider prop would be a start.

I think a VR Arcade could be possible in he future, which could give experiences that require expensive props such as bikes, hang-gliders, ODT, etc.

[android78]

I see what you're saying. But even in that case, without actual physical constraints, wouldn't it be best if the VR world was programmed without the artificial collision detection between your limbs and the objects, for when you go that little bit too far?

[Zaptruder]

I see what you're saying. But even in that case, without actual physical constraints, wouldn't it be best if the VR world was programmed without the artificial collision detection between your limbs and the objects, for when you go that little bit too far?

The solution is to show the disjunct between your own limb and the virtual limb that is constrained by virtual environment and physics.

Obviously you can't stop your arm from wailing through with a sword - but your virtual arm could be shown as the 'real arm', held back by another sword or shield - while your real arm is shown in some abstract form (maybe a red pixelated voxel representation that glows brighter the bigger the disjunct grows).

Add in some factor to motivate players to actively minimize the disjunct - e.g. you move too far from your virtual limb, and you lose control of the limb, drop your sword, become significantly disadvantaged in battle - and soon enough people will respect the 'physical reality' of the virtual space and its objects, even without external haptic feedback.

We end up inducing a proprioception form of haptic feedback through the use of visual, audio and environmental behaviour cues.

It won't be as convincing as a full on exo suit, or even some haptic gloves - but the big advantage is that it's much more accessible to everyone, with the hardware requirements been zero.

[cadcoke5]

We end up inducing a proprioception form of haptic feedback through the use of visual, audio and environmental behavior cues.

Excellent ideas being presented, Zaptruder.

Joe Dunfee

[profvr]

Obviously you can't stop your arm from wailing through with a sword - but your virtual arm could be shown as the 'real arm', held back by another sword or shield - while your real arm is shown in some abstract form (maybe a red pixelated voxel representation that glows brighter the bigger the disjunct grows).

Add in some factor to motivate players to actively minimize the disjunct - e.g. you move too far from your virtual limb, and you lose control of the limb, drop your sword, become significantly disadvantaged in battle - and soon enough people will respect the 'physical reality' of the virtual space and its objects, even without external haptic feedback.

That's an interesting idea. Some observations and thoughts in response:

From the academic VR literature you might want to read up about a recent flurry of activity about recreating in virtual reality something called the "rubber hand illusion" and then various "body-swapping illusions".

From my own work and work of colleagues we know that the brain is not very sensitive to position offsets of the body. Latency is important, as is motion parallax, linearity, etc. but the brain is amazingly plastic when it comes to understanding the relative position of the head and hand. You can easily demonstrate this to yourself with shift prism glasses.

The type of environmental constraint you mention are reasonably frequently used so that virtual body limbs do not go through objects (though its rarer to see a "ghost limb" that does exactly follow the real limb"). However implementing that is surprisingly hard. You can imagine constraining the virtual limb to the closest point on the surface to the real limb, but then you get odd snapping effects.

I am reminded of a talk by Philip Oliver (Blitz Games) about the "Puss in Boots" games, where they observed that users (kids mainly) would mimic the action of the main character with Kinect or Move even though it bore little relation to the gesture recognition they actually needed to use to make the sword fighting actions.

You may not need to motivate people to avoid these collisions between real limb and virtual object. My experience is that many people avoid them anyway if they can, especially if the stereo and depth perception are well tuned (if they aren't then interaction in personal space is very hard anyway). Perhaps that's because collisions are "uncanny".

However, the idea of motivating people to avoid these discrepancies is an interesting this to explore. I'm sure I've seen (but I can't find) a study where they made the virtual limb turn red and make a buzzing sound. I've also seen a vibrotactile vest used to tell people they were walking through virtual objects (with the effect that they stopped doing this). More subtle ways of doing this (such as your suggestion of causing you to drop a held object) might be very effective.

[space123321]

The more I test my DIY Rift - the more I wish for an open world to simply browse around, interact and bascially chill out in. A Second Life type environment would be simply amazing! I remember a scene at the begining of The Lawnmower Man when the doctor is in his basement and using a software entitled "Flying, Falling, Floating" (or something of this nature). I really want to experience a true VR enivonment constructed by users creations and imaginations. LOL - with the current selecion of games avaliable, I am finding myself walking around taking in the environment, and in turn having the #$%^ scared out of me as something or someone usually ends up jumping out at me when my attention is on something else lol!

[FingerFlinger]

I hope that Majong actually implements Rift support for Minecraft. Set it to Peaceful and just walk around...

[brantlew]

LOL - with the current selecion of games avaliable, I am finding myself walking around taking in the environment, and in turn having the #$%^ scared out of me as something or someone usually ends up jumping out at me when my attention is on something else lol!

A while back I found the perfect game for testing/chilling out in VR. It's called theHunter. It's basically a hunting simulator - which is an incredibly boring concept on your desktop, but it's great for VR. The graphics are good and you can just walk through miles of fields and woods without having to worry about getting attacked or running out of time. Plus, it's totally free to play Now I just wish someone would do a warp driver for it. (hint, hint Cyber)

http://www.thehunter.com/pub/

[space123321]

Thanks brantlew - going to try this out tonight!