Virtual Reality For The Common Man?

[Bishop51]

Hey MTBS3D!

First time poster, long time lurker. Like most here, I've followed all things VR since seeing Lawnmower man in 1992, expecting a glorious VR future ahead. I've dabbled with homebrew projects, fresnel lenses, makeshift collimation, curved projection, shutter glasses but as you get older (and have kids) your time to experiment becomes seriously handicapped. 19 years later and I was about to give up on my VR dreams until I saw news of Sony's HMZ-T1. For the first time it would seem that a major company found the right entry level balance of high resolution, moderate FOV and attainable pricepoint.

Finally, real, consumer level, virtual reality seems to be within reach. But like the rest of you, I fear it will be a flash in the pan in terms of market success...and I want a unit before it disappears.

Anyway, since this forum is the single most knowledgeable out there, I have a few questions I was hoping you might answer:

* Headtracking: We all know that the HMZ-T1 does not include a tracker (though a future version might) so, are there better options than Track-IR within the same pricepoint?

* Gaming: Excluding consoles (focused on PC gaming here), realistically what games could take advantage of fully immersive VR with head tracking? Let me prempt that a bit by saying OTHER THAN flight sims, racing sims and ARMA (love those too but I REALLY like FPS). Are there any FPS games currently available which utilize free head movement and detached/independent body control via keyboard/mouse? Its an important question because for most working stiffs, making a $1500 investment into something only I can use is a hard sell at the dinner table. Are there resources available to modify existing FPS games to work with VR setups?

* PalmerTech: Or should I just say screw it and dedicate a month or two to PalmerTech's low cost, high FOV HMD http://www.mtbs3d.com/phpbb/viewtopic.php?f=120&t=13745" onclick="window.open(this.href);return false; ? Need those plans Palmer, you genius you!

Long story short, I'm committed to buying the HMZ-T1 over the new Cinemizer for various reasons (FOV mainly) but I want the more experienced users out there to tell me if I'm overestimating what software can actually utilize the hardware.

Thanks guys and thank you for the wonderful resource here at MTBS3D!

[Dom]

Hi, i am too interested in virtual reality but to achieve the realness it can offer seems like there has to be alot of programing and development. Most games on the pc only use a mouse for head movement 2 axis and just the "wasd" keys for self position. To edit and bind a camera movement should be as simple as editing a config file but knowing the certains game code be it .lua .xml C++ would help you a great deal. All you would have to find is code for player or camera movement but this sometimes would be in a .dll file and you would need the source code along with programing software.

I have asked the freetrack team http://www.free-track.net/" onclick="window.open(this.href);return false; to see if they could possibly enable a microsoft kinect to enable their software to perform the same and more functions that it normally does with a webcam. If somehow there was a way to use Kinect motion controller to enable headtracking on a pc then someone could even have the full freedom for standing up.

With freetrack it can emulate your mouse control along with pretty much any keyboard function from headmovements. So if you had a game and you had a button binded to your keyboard "F" button to look around a corner you could make a script to perform that movement. I have been aware of a program called "glovepie" http://glovepie.org/" onclick="window.open(this.href);return false; . I think if the device is supported you can change and manipulate that hardware to your game.

Really though someone should consider putting together a vr consumer software suite with microsoft kinect that has most of the popular games enabled for 3-6 dof headtracking. I would help start it but i am sure there are lots of other better skilled programers that could make it work.

[nrp]

I've hacked on Kinect/OpenNI/NITE stuff. While the joint position accuracy is pretty good, the joint orientation accuracy is not. It is the latter that is useful for HMDs.

Additionally, anything camera based like TrackIR or Free-Track would probably be irritating to use with an HMD. You'd be moving relative to a camera that you couldn't actually see.

For the original question of what games would be supported, while you could map to the mouse, it doesn't have the kind of immersiveness you get when you have full yaw/pitch/roll input. It seems very few non-sim games are designed for that. The supported games lists for Free-Track and TrackIR probably cover all of them that wouldn't require modification.

As for what tracker to use, I don't know of any good ones, so I'll be building my own out of a SpacePoint Fusion.

[Aeroflux]

Welcome, brother from another mother! I got started on the VR trip with the movie Lawnmower Man as well. The impact of VR in that movie was so deep I didn't even see it as a horror movie. I still long to find a directors cut in widescreen digital format, as I own the Laserdisc DC (in 4:3), VHS, and of course the DVD...also have both PC games and the System 7 soundtrack for the game remains on my iPhone to this day. I'll admit I haven't gotten far in the realm of VR--I mostly exhausted my effort back with the original i-glasses and a cardboard box gimmic that came with one of the VR books I bought.

The Sony headset has brought back a flicker of that hope that has been thoroughly trounced upon throughout the years...it appears to strike a balance between features and cost...something I haven't seen done. With the HDMI 1.4a specification, it will support the most 3D content I've ever seen on a single device. This really does feel like the market is on the precipice of something new.

On a side note:
Holy smokes...Track IR supports Descent 2 X-XL! I just got the game running on my PC. I'm still trying to work out how to get OpenGL running in stereo, but I thought tracking would be the difficult part.

[Bishop51]

Thanks for the welcome and valued input gents Quite honestly I don't see the kind of virtual reality that we have a fetish for taking hold of the market. That being said, I believe strongly that augmented reality will hit all the right buttons for Joe Blow consumer. All it will take is Apple or some other giant to release a lightweight pair of glasses, retinal laser projection or even LED based contacts. Or maybe, just maybe Microsoft will embrace VR as their next gimmick. But that's a looooong way off from now and I really like the balance Sony has struck here with the HMZ-T1 (baby steps right). Good on them for being first out of the gate!

So on the issue of tracking, I really think the Track-IR is the way to go for the kind of paired usage I have in mind. All potential usage scenarios have me seated in one position, not running around the room, ducking behind furniture or doing 360's To me ARMA has the configuration EXACTLY right and I wish more developers would focus on that kind of control scheme. In ARMA (just like in real life), your head independently pivots on your body but its range of motion is limited by your spine. The only way to see behind you is to turn your body to the new position and in ARMA body rotation is handled with the mouse and WASD for side/back stepping. That's how ALL VR based games should function because what it means is that you can sit in your chair comfortably but still have a solid sense of immersion. I mean, I love flight sims and driving so that will certainly eat half of my usage right there but it sounds like it will take some doing to modify OBLIVION or AMNESIA to be fully immersed in a way that works (but I'm willing to try dammit!).

See, consumer based HMD's are really only half of the issue. Until game developers start developing for the hardware things will never truly take off. Chicken meet egg.

[cybereality]

Hey Bishop51! Welcome to the forum. I agree that right now is a great time to be into VR. Stereo 3D is just taking off big-time, and now companies like Sony are taking an interest in HMDs. Of course there are guys like Vuzix, and Zeiss, who are furthering their product line in this space. And finally we have game developers taking an interest in niche setups like 3D monitors or triple-head wide-screen gaming. So these are exciting times.

At the same time, we are FAR away from where we should be in 2011. Even back in the 90's, the HMDs were not all that worse than what you find today (on the consumer market). Of course, we have made huge gains in computer performance and such. However the actual software has not evolved AT ALL. If you look at what we were playing back then, the original DOOM for example, and look at FPS games today and they are basically the same. Sure the graphics are better, but the interactivity and immersiveness in the environment has barely progressed at all. Its really quite sad. And whenever some innovative peripheral comes to the market (P5 glove, Novint Falcon, Razer Hydra, etc.) it never seems to get any kind of support from developers. Drives me crazy, we have so much technology today that no one is taking advantage of. At the same time, there is innovative stuff happening, for example the Nintendo Wiimote, Microsoft Kinect, etc. So there is movement in this field, it is just very limited from what you would have expected in this day and age.

But yes, the Sony headset does look like it could revitalize the industry. If it does well, maybe other vendors will get involved. I certainly plan to get one, I hope its as good as everyone thinks it will be. I do wonder what to do about the head-tracker. I do not know of any third party tracking solution that would be good to use with this (for example, one thats accurate and has a decent level of software support). Track-IR is nice, but they only really support simulation titles. I'm a big FPS guy myself, so that kind a blows. Plus they are nazis about letting indie developers / hobbyists use their SDK so that kills it for me right there. I have not tried Free-Track, but it also seems to be sims only. The only FPS title I have seen with real headtracking is UT2004 with the Vuzix mod. It works with VR920, 1200VR, etc. The tracking is actually really nice, it has a bunch of different modes and can be used with fully independent tracking of head and gun. So if you also had a tracker to use for the gun (ie a Nintendo Wiimote, etc.) then you could use this in a full-freedom 360 degrees VR setup. However this is just one game, as much as I love it.

I do have tentative plans to make a VR driver, that would allow support for a variety of VR-style devices in standard DirectX games. So you could have advanced headtracking capabilities (independent of mouse/gun movement), haptic force-feedback, stereo 3D, etc. Although I have been programming for years, it has mostly been with scripting languages on the web, so I am brushing up on my C++/DirectX skills at the moment. Not sure when I could have something like this ready, I am so busy. But I hope to do this as some point (probably next year sometime). At the moment I am trying to work on a VR demo as a proof of concept for a next-generation VR control setup. Mainly this will be using the Razer Hydra, but I may incorporate other devices as well (Vuzix Wrap 1200VR, etc.). This should be a lot easier then hacking commercial games (I already have some basic stuff working: http://www.youtube.com/watch?v=NJHxbDa_tAA" onclick="window.open(this.href);return false; ) so hopefully it will be quick. This will be more of a test-bed to see how the control scheme should work. Then I will look into porting this over into commercial titles. Again, I am not even sure I have all the skills to do this yet. But I am moving in that direction.

[Aeroflux]

Since both of you have an interest in independent viewpoint/gun reticle movement, how would you approach a careful aim? I didnt want to say anything, but I see some spacial confusion going on unless one of the movement controls temporarily binds to the other when, for example, staring down the crosshairs of a scope on a sniper rifle, or taking careful aim with a m-4. Then there is the concern of head movement going out of the binding process. Assume the exact position of the head and risk disorientation...or reset to a center and risk misalignment with the actual position of the head.

Id like to try some mech shooters with the head tracking bound to torso movement...anybody who has played Steel Battalion knows how frickin awesome that would be! Theres also Living Legends, a mechwarrior mod for the Crysis engine; and Hawken, which a very fast paced shooter in very well designed environments (healthy amount of vertical design). They're not strictly first person shooters, more on the edge of sims.

Of all the mech games Ive played Steel Battalion was the most immersive. A giant controller with independent torso/turn/aiming reticle controls...plus 40+ buttons really pushed the experience. With the projector I can say its a much better experience than Virtual Worlds was (if anyone remembers that place). I can hardly wait to play that on the sony headset. No sacrificing resolution for immersion this time! I just need figure out a component to HDMI fix now...

[cybereality]

Unfortunately, most (all) games probably don't model the camera/gun like in real life. So where you use the scope it is usually just a 2D overlay on the screen. Meaning if you moved your head but not your hand, then the reticle would be pointing to an incorrect position. In this case, I think the best solution is to simply lock the headtracking to the center position. It would be strange, but it would be the best solution probably.

[WiredEarp]

CyberReality, thanks for the info re UT2004, I'll have to download it and see if I can make it work with a Z800.

There are several problems with FPS VR games.

1) virtually none support independent head tracking. This could be added with a mod possibly
2) They don't support TrackIR or similar, so there is no way to use absolute aiming without adding it with a mod or a hack

1) can be solved by mounting your VR headset and controller to a gun. That way, you put the gun in your shoulder (sensor in gun) use the gun to turn, aim, etc. This way, it feels sort of natural (very special forces) and the gun in your vision makes total sense. Also, by moving the aiming from your head to the gun, it feels much more natural. Also, (1) can possibly be solved by mods.

2) Is harder to resolve. Naturalpoint suck, as they have held back controller development, not improved it. The only way to add absolute aiming to these games is to hack it in, by manipulating the mouse control variables or the actual view matrix/rotation variables. This is theoretically possible with every game, but virtually NO games seem to have hacks to enable this sort of thing. Annoying, and something I need to look into. As CR says, its much easier to write your own system to support stuff than hack into others. However, I really really want to hack one particular game. If I manage it, I'll do it using (VPRL? VRPL? gah can't remember the name of the library!), but it allows you to add a level of abstraction to your tracking devices) so that, while I'll be using the Hydra with for my purposes, in future hopefully it will also allow any other device to work with mimimal recoding.

To me, a good, FUN FPS (Arma is not that fun for me) that supported VR would be a godsend, and probably one of the killer apps that is required to make VR mainstream.

[nrp]

Since both of you have an interest in independent viewpoint/gun reticle movement, how would you approach a careful aim? I didnt want to say anything, but I see some spacial confusion going on unless one of the movement controls temporarily binds to the other when, for example, staring down the crosshairs of a scope on a sniper rifle, or taking careful aim with a m-4. Then there is the concern of head movement going out of the binding process. Assume the exact position of the head and risk disorientation...or reset to a center and risk misalignment with the actual position of the head.

I think it can be approached in a way that vaguely simulates real life. That is, the exit pupil of the scope appears in a semirealistic form in the field of view of the player, and it needs to be physically (well, virtual physically) aligned properly to look through it. You then move the gun as necessary keeping it in the field of view, turning your head as necessary to keep it so. If the head and gun viewpoints don't align well enough angularly, you can't use the scope, just like real life.

Of course, this depends on the IMUs in the HMD and gun peripheral being properly calibrated against each other initially and then not drifting during play.

[WiredEarp]

I agree, for REAL VR immersion, the gun should require you to independently align your head to look down the sights. This is for REAL VR, not just a FPS with a VR headset (you'd need sensors in both headset and gun at the minimum.

'Of course, this depends on the IMUs in the HMD and gun peripheral being properly calibrated against each other initially and then not drifting during play.'

This isn't really an issue if we use the time honoured technique of using magnetic trackers.Then you only need 2 trackers, one head, one gun, to do what you are wanting.

I seem to remember the Virtuality system had 3 separate trackers. Head, gun, and hips. You used the hips to get the amount of 'turn' of the user (and the 'move forward' button based the direction on hip direction). This way, you could move forwards in one direction, while looking at another, and still shooting behind you. You had to align the guns in this system (two versions of Dactyl Nightmare I played, one with a handgun (standard Dactyl), the other with a crossbow, shield and axe. The axe weapon in particular was quite immersive (you could 'cut' their heads off if you swung it into them). It was a lot of fun multiplayer... I remember crouching down, sticking my gun through a hole in a cube obstacle, and using it as a shield to shoot from

[nrp]

I agree, for REAL VR immersion, the gun should require you to independently align your head to look down the sights. This is for REAL VR, not just a FPS with a VR headset (you'd need sensors in both headset and gun at the minimum.

'Of course, this depends on the IMUs in the HMD and gun peripheral being properly calibrated against each other initially and then not drifting during play.'

This isn't really an issue if we use the time honoured technique of using magnetic trackers.Then you only need 2 trackers, one head, one gun, to do what you are wanting.

I seem to remember the Virtuality system had 3 separate trackers. Head, gun, and hips. You used the hips to get the amount of 'turn' of the user (and the 'move forward' button based the direction on hip direction). This way, you could move forwards in one direction, while looking at another, and still shooting behind you. You had to align the guns in this system (two versions of Dactyl Nightmare I played, one with a handgun (standard Dactyl), the other with a crossbow, shield and axe. The axe weapon in particular was quite immersive (you could 'cut' their heads off if you swung it into them). It was a lot of fun multiplayer... I remember crouching down, sticking my gun through a hole in a cube obstacle, and using it as a shield to shoot from

Yep, and magnetometers and single IC three axis gyros have dropped in price about 90% in the last year, making the prospect of 2 or 3 separate IMUs reasonable.

[WiredEarp]

Indeed. I think optical has potential as well for hobbyists. Freetrack has shown it can be done fairly cheaply and if this was redesigned to allow multiple cameras, you could use cheapish webcams to do the same thing and provide more wide ranging tracking. Or a hybrid approach again, using optical when possible to resynchronize the gyros data. Exciting times really. I'm most eagerly waiting for the Sony HMD - I'm SURE someone will manage to modify it to increase the FOV - then finally i'll be able to recreate the VR I experienced in the 90's.

[Bishop51]

Again, thanks to all for providing an awesome and deep pool of knowledge on this one! I think its great we have a resource like this to go into VR with our eyes wide open on what exactly is possible given the current state of the industry.

So, it seems to boil down to the following:

• Virtual reality in its current state primarily supports commercial driving & flying simulation titles.
• HMZ-T1 coupled with Track-IR will yield good results for a fairly small subset of driving & flying simulation titles: http://www.naturalpoint.com/trackir/03- ... s-all.html" onclick="window.open(this.href);return false;
• Almost all FPS games do not natively support an ARMA style or fixed seating position control scheme, where head independently rotates from body and body is controlled with mouse.
• It might be possible to create an abstraction layer in software to modify some games to accommodate independent head tracking BUT Track-IR does not allow open source SDK.
• Virtual Reality remains in the realm of mad garage scientist! Family will shun you, friends won't understand you and your bank account will be sucked dry by reams of hardware that may or may not work for the intended purpose.

I think that about sums it up right?

So, I have to tell you, even with all of the roadblocks the HMZ-T1 still excites me enough to at least attempt some gaming in VR. I mean, realistically, the kinds of games that would work best with Virtual Reality haven't been invented yet and what I mean by that is games which focus on a sense of place and environmental interaction, more than blasting someone's head off. I would love to be projected into a MYST style puzzle game that is rich with atmosphere and exploration...hell, maybe we could hack something for Myst URU Then again, that's going to depend on what tracker we can modify to do the job and it sounds like TrackIR is handicapped there (REALLY stupid move by TrackIR by the way. I mean, you're selling hardware guys! Make the damn thing approachable!).

[brantlew]

Virtual Reality remains in the realm of mad garage scientist! Family will shun you, friends won't understand you and your bank account will be sucked dry by reams of hardware that may or may not work for the intended purpose.

Ha ha ha! Amen brotha'

I mean, realistically, the kinds of games that would work best with Virtual Reality haven't been invented yet

Same could be said of media content. I can envision immersive "movies" or at least animated movies that held an entirely different experience because it would be like watching a stage play in a series of changing locations.

[Aeroflux]

I still have the dream that Dr. Angelo had. I'd like to see VR and AR make the world a better place, by educating those through the ultimate form of kinetic learning. Instead of schools and colleges, knowledge will be available to anyone. Where a few styles of mathematics or acting or any skill are available locally, with VR and AR anyone can see their full potential and explore curiosity without any limit.

How's that for a change?

[brantlew]

Awww come on. We all know the real driver of this technology is gonna be porn.

[cybereality]

@WiredEarp: Yeah, I still remember trying out Dactyl Nightmare at the old Virtuality arcades. Maybe it was just cause it was my first experience with VR, but that still seems like the most amazing thing to me. That Visette HMD they had was 60 degree horizontal FOV, which still trumps the HMZ-T1 we are all waiting for!!! What happened?

@Bishop51: I agree, I think the really immersive experiences have not been created yet. We are only in the infancy of this technology, at least in terms of mainstream usage. Games today are very limited in how you can interact with the environment, and even "sandbox"-style titles (GTA, etc.) are very limited in scope. There is a lot more than can be done. I am hoping to touch on some of this with the demo I am working on. I don't want to give too much away, but I can say it does not involve guns or killing people.

@brantlew: I am not going to link to anything, but there are several adult 3D sites out there. Not saying I did this, but imagine one were to find a 3D video with a first person point of view, say of a man sitting in a seat, and then they were to sit in a seat and look down into an HMD so that the mans legs matched up with their own legs. Now the brain is easily fooled, and will start to accept the video feed as reality. Would be a pretty immersive experience, I imagine (not that I would know...).

[pierreye]

You still need a doll to complete the VR experience for porn. Don't forget you need touch for complete experience. .

[cybereality]

You still need a doll to complete the VR experience for porn. Don't forget you need touch for complete experience. .

Or a full-body haptic suit. It could happen.

[WiredEarp]

@CyberReality - was it actually 60 FOV? I always thought it was 85 FOV, but all the specs I can find now on the net on the SU1000 don't mention the FOV in any way. It seems the Visette 2 is 60 though so its quite likely (Visette 2 was later, from the 486 based version of the VR system). I remember being disappointed that the VR display was just like a big screen (bit like IMAX) - but after playing for a few seconds I lost sight of the edges (seemed bigger than i imagine 60 FOV is though).

http://www.amigahistory.co.uk/virtuality.html" onclick="window.open(this.href);return false; has some nostalgic info on the games I remember playing Legend Quest (best of the games IMHO) and feeling a wave of excitement when I killed the skeleton with my knife and opened the chest to see a bigger sword for sticking foes at a distance! Unfortunately, this is when Legend Quest hung and I ended up having to play Grid Busters instead (which sucked pretty much).
I always remember thinking that if the Virtuality guys had just had a good games programmer, they could have done so much better. People wanted Doom, not lobbing explosive type balls at each other (with a ridiculously slow reload time even!).

I have to agree re the current state of gameplay. Seems its been stagnant since about Half Life (1!). Personally, I think it is AI where most of the games fall down now. GTA is a good example, when I go the wrong way down a street and nearly hit a cop, it would be nice if that cop at least took some notice. Crime and witnesses simulation etc would be another front where more progress should have been made. What language/system are you using to do your demo?

I think the future for VR eventually is brain computer interfaces, but in the meantime it will be interesting to see what tricks we come up with to fool the body, like better goggles, new ways of fooling touch like pulsed electric fields, etc. Certainly, before we could ever have a feasible brain/computer interface now, we'd need a way to paralyze people painlessly to prevent them from getting up and moving around inside the VR. Then, those adult sites will take on a whole new dimension

Oh, and no teledildonics discussion is complete with out this limerick somewhere:

There was a young man named Kleene,
who invented a f'cking machine.
Concave or convex,
it fit either sex,
and was exceedingly easy to clean.

[cybereality]

I thought the FOV was around 85, but looking on the net now all I see is 60 degrees (horizontal). Still, that's not bad.

Well I realize Dactyl Nightmare was a little bare-bones, but it was pretty much state-of-the art for the time. Sure DOOM had better art direction, but it wasn't true 3D (it used ray-casting) and thus would not work for a VR simulation. And Quake didn't come out until '96, years after Virtuality went bust. So I think what they had was good for what it was. Certainly the immersion with a HMD and cyberpuck control still trumps anything you can find today in arcades (not counting DIY stuff). I remember they also had a mech warrior type game that was pretty good. Can't recall the name.

I think with games today that the game design is hindered by trying to render a 3D world, but using a 2D monitor and (mostly) 2D controllers. Recently we have seen more 3D input devices on the consoles (Wii, Move, Kinect) and there has been some progress with interactivity. But it is still very basic, and not matching the technological capabilities of the hardware. It seems developers are just stuck making a few generic type of experiences, and only some indie devs really pushing the boundaries. The problem is the games are not rich enough. A lot of this has to do with AI, like you said, and also limited physics interaction. This is stuff that could probably be done with current high-end PCs, with enough effort. Its not an impossible task.

[cybereality]

@WiredEarp: For my demo I was building something with C++ and Ogre. However I just started looking into the Unity engine, and it seems promising. Still have a bunch of art assets and stuff to do (which would work in any engine) so I am not stressing it right now. Performance seems a tad slow in Unity, specifically for physics and I think I may need hardware PhysX to pull this off. So I will probably stick with Ogre, since it gives me more flexibility to use whatever libraries I need.

[Bishop51]

@cybereality: Cyber, you own the Razer Hydra is that correct? Using proper keymapping, is there any reason you couldn't strap one controller to the side of the HMZ-T1 for head tracking and use the other controller for movement? I mean, if that's possible, due to the open structured SDK, couldn't we make the Hydra work for competent head tracking AND body control with almost any game?

Is this something you've tried with your HMD's?

[cybereality]

I guess it is possible to mod the Hydra for use as a head-tracker. I did wonder about that before. The controller is relatively light, and if you dismantled it, I bet you could get the size down. However it is a wired controller, and the range is only about 3 feet. So it is of no use for a wireless full-freedom setup. But for a seated arrangement, I guess its a possibility.

[Bishop51]

Hmmm, very interesting!

From what I can tell the Hydra has excellent 1:1 or very near to it tracking. If you could head mount one controller, that would make the TrackIR totally redundant and the Hydra by contrast WAY more useful for general tracking in all games (general simulation included).

Another question though: If the Hydra was mounted that high, would it effect the tracking operation in any unseen ways?

[cybereality]

Well you could mount it on the bottom of an HMD I guess. Should be enough to give you about 3 feet of leeway to turn your head. Again, this will only work if you are seated, and in that case you could just use TrackIR / Free-Track. But with the Hydra software at least it supports a lot more popular games and you can create profiles yourself.

[WiredEarp]

@Bishop51, your idea about head tracking and movement using the Hydras is exactly why I bought them.
Hydra seems like it could become the best head tracker I've used. I just need a Hydra -> TrackIR converter. Then i'll be able to use it with all my sims etc. Optical is great until the leds get all in one line (certain angles etc) or you want to turn your head further than it will allow. Hydra seems way better... Re HMD tracking, it definitely works seated. I'll try some tests tonight of whether it will work standing and the ranges (I suspect it will be sweet).

@CyberReality: Yeah I wonder if it was 85, and its the later versions that have a 60FOV (they may have gone with lesser lenses to save money/weight). I guess you are right, and tell the truth, Dactyl, was pretty awesome as a tech demo (sticking my hand through holes in 3d virtual shapes? awesome!). Its just that they had no idea about gameplay. If they'd made the guns fire faster and had a couple of different ones, they could have had a world hit on their hands, rather than just something that people played once for the experience. Not sure re the mechwarrior game, was that a CS1000 game? Or was that Grid Busters, where you have the armored suit and the jetpack?
I can forsee one day (long time from now) new 'games' will come out, and will be like new movies. Everyone will actually join and be given a character, but rather than just changing their appearance, they will be given an entire storyline to act within, and work together with the other players in (the games would have to be designed to run in a movie length timeframe) For example, you could have a game like Lost where all the castaways had to work together but would be told different back story behind their characters, and be given different aims.

Re game engines, often the best engine is the one that you prefer working with. Its definitely a big help nowadays to have full physics engines etc that you can just use rather than having to write, i think you are doing the right thing looking at leveraging PhysX etc rather than trying to optimize. I'm a big do it myself guy but I do think i'm more productive (not more creative however) when I make use of other peoples tools and systems rather than writing my own.

[cybereality]

@WiredEarp: I'm not sure if it was Grid Busters. I don't recall having a jet pack. You were seated (in real-life) but in the game it looked like you were in a cockpit with like 3 large windows. It had an HMD, and you could look around. I think the controller was some sort of joystick/flight-stick type of thing. Its had to remember. Graphics were flat-shaded polygons, but the art may have been a little better than Dactyl Nightmare. Hard to remember at this point.

I also think the games can evolve from the rudimentary stories and situations they have today (aliens invade, save the world, zombies, etc.). I'd like a deeper interaction with characters, to feel a wide variety of emotions. I'd like to see horror that did just consist of a monster jumping out of a closet. I'd like to feel attached to characters, and be hurt if they die. I'd like to be able to kiss the girl at the end! Why can't they make a game on the level of Inception? It would be perfect material. Games are just in the stone age compared to the range of material you find in books or movies.

In terms of game engines, I like using C++ because I only seem to be limited by my own skills. And Ogre seems nice. With all-in-one packages like Unity, its really quick to get something started but there always seem to be limitations at some point (that are hard/impossible to fix without the source code). So I will probably just stick with Ogre, but I may use Unity as a prototyping tool.

[tritosine5G]

/offtopic
thanks to those bypass-clowns at nvidia forums, nvidia is pulling the plug on CRT mode w. nvidia driver. Great news...Reality for the common man...

[cybereality]

/offtopic
thanks to those bypass-clowns at nvidia forums, nvidia is pulling the plug on CRT mode w. nvidia driver. Great news...Reality for the common man...

Sucks, but its no surprise.

[tritosine5G]

I predict adapter business gets hot!
Like this:
http://www.curtpalme.com/IFB-FULLHD.shtm" onclick="window.open(this.href);return false;

[cybereality]

Please try to stay on topic. This is a discussion about VR.

[PalmerTech]

@Bishop51: Hah, thanks for the compliments. I am no genius, though! I have learned most of what I know about HMDs from taking apart and modifying the units I have in my ever growing collection, and endless reading of what is left from the 90s VR bubble. Lots of members here have been extremely helpful, and Tone of VRtifacts has been an especially great resource.

So, VR for the common man... That is tough. Here is the setup we use at my work, with a big huge stage: http://www.mtbs3d.com/phpBB/viewtopic.p ... 79&start=0" onclick="window.open(this.href);return false;

Sit down setups are all well and good, but where HMDs really have a benefit of simply using, say, curved projection screens on a desk, is when you can walk around and do stuff. Know what I want to do? Bring back Virtuality, or at least the spirit behind it. People run indoor airsoft fields, indoor paintball fields, why not an indoor VR arcade? Arcades are dying because they cannot bring an experience to the table that people cannot get at home, but a VR arcade could do that easily. Imagine going to the local VR chamber with a group of friends, paying a few dollars, then suiting up to spend the next half hour in a cooperative first person shooter, with your adrenaline pumping just as much as you would from running around in a real combat zone. You could have hundreds of games, thousands of scenarios, enough to never get old! Heck, you could have people use the suits in an environment like Garrys Mod or Minecraft, where they can build their own worlds by simply walking around and creating as they see fit. Do that on weekdays...

And on weekends, when people have entire free days, have the REAL fun stuff. Arrive in the morning, suit up, and then play through a simulation all day, or even all weekend! Look at what games can do in 6-8 hours (Halo, Gears of War, COD), imagine being totally immersed in a game for 12-24 hours of gameplay. You could even build in parts of the game with rest time out of the suit, where you go back to a fake barracks, unsuit, and spend some time in the "bunker" eating, sleeping, and talking to people. The next morning, you get "flown" right back out to the combat zone.

As yet another application, imagine a large facility where people could go on the weekend to play MMORPGS. Real life WoW raids!

I think that would be the best way to (Practically and realistically) allow the common man to experience VR.

[cybereality]

I would love it if they made some new VR arcades, but somehow I feel like they wouldn't make enough money to cover the expenses. Unless it was the real deal Holodeck/Matrix, I just can't see the mass market paying big bucks to do that. I mean, I would pay. And I'm sure a lot of people here would. But it just feels really far away from the mainstream.

[PalmerTech]

I am not so sure. I found out that the biggest costs for painball and airsoft facilities are liability insurance, and their employees. You need someone monitoring every single game, at LEAST one person per field. Customer and employee liability costs a lot when you are running something that involved shooting other people, too!

Sure, right now, with current HMD prices, it makes no sense. But what if a full wireless suit could be made for $3000? I mean, the guns they rent out for a few bucks at paintball fields cost $300-400, they make it up easily through volume rentals over the course of just a few weeks. And this would not just be standing in a ring, ala Virtuality; I am thinking along the lines of where I work, where you have a huge open area to run around in.

People pay $90 for a one day ticket to Disneyland, and $60 for a one day ticket to Six Flags. I think if people could spend $50 to run around inside the latest COD game for a day, they would take that chance. Heck, lots of teens waste $10 on fast food with their friends every single day after school, I know that at least some of them would much rather have an hour in the grid.

It would not work for everywhere, sure. But for where I live (Los Angeles) or heck, where you live (New York), I feel like it could do pretty darn well.

[WiredEarp]

@ PalmerTech - I think that a VR or simulation arcade is about the only arcade that could work nowadays.

However, if you are going to have a full stage, may as well go all the way, and make it an augmented reality (or full VR with physical walls also shown in the virtual environment, same sort of thing) lasertag type combat game. You could run around and pick up different weapons etc and shoot them at your friends. Due to the physical walls and opponents etc being mapped in the VR, there would be less danger of running into walls or other players.

Even just a simulation arcade would be cool. You could have stations with plane and helicopter setups etc and go in and fly against the other players etc, with a much better simulation experience than can be done at home currently. Either multiple monitors, or just a single VR headset, just like the old SD1000.

As Cyber pointed out however, its just getting the turnover to do it that could be the difficult part...

[PalmerTech]

I had thought of something like that. Augmented reality would be harder, since you would need to actually spend time and money on your physical environment. With full VR, you can just have an empty room, maybe some easily movable foam walls.

Google "MaxFlight simulator", one of those would be absolutely crazy. Most places that have them charge $5 for 5 minutes in one, crazy. In a perfect world, you could have some people on foot, some people in aircraft, and some people in land vehicles. Would be intense! In Japan, they have Gundam Simulation pods in arcades. They cost about $80,000 for a 4 pod set, and they still sell them years later, and apparently they have a high return on investment.

If you could not get the turnover, another option would be to contract it out as a military training center. Currently they run their own, but they could save costs by using a private company, I think.

[WiredEarp]

I dont think it would be harder (augmented vs full), both are just overlays onto foam walls. Just you can see the actual wall underneath with augmented. Really, it would just be foam walls, overlaid with graphics, whether semi visible or totally obscured, the concept is the same. I guess full VR would be better in that you could have avatars that didn't necessarily match your physical size and shape...

MaxFlight simulator looks awesome. I've seen some homebuilt motion systems online that actually are pretty good as well, for a lot less cost.
I think an arcade with a variety of options could be viable. IE, you'd have basic aircraft stations with just aircraft controls and a few monitors. Slightly more advanced ones with aircraft controls and just a VR headset. The most advanced and expensive ones could be full motion types like this. That way, everyone can fly against each other (same software) but those who pay more get a more realistic experience.

I remember MechWarrior pods - I imagine these must be very similar to the Gundam ones? Always wanted to play one

I've been wanting to do a VR arcade for a long time (entertainment VR is my main interest) but the enabling techs are only just now getting close to being economically feasible.

One good thing about the idea, is that its not going to be eclipsed by advances in home computers and consoles for a fairly long time.

[Bishop51]

I've used and created a number of desk mounted (and or sim-pit mounted) display systems. Including curved displays, fresnel displays, multi-panel displays and even rough collimation. They all fail miserably in terms of raw immersion. I mean, they come close but I have yet to use one that makes you feel like you're in the represented environment. Due to the form factor, I still believe there's a pretty healthy market out there for desk based VR (especially as it relates to modern games). The problem is that the display tech is still lagging behind by several years. Even Sony's new HMZ-T1, while the displays offer break-though consumer resolutions, I know in my cold VR-heart that the lens system will kill the immersion. Unless one of the brilliant minds hanging out here can hack the optics somehow...lol (the screens are too tiny to make that practical).

VR arcades (while we all know they'd be fantastic, amazing, FUN!) would likely find themselves in the company of Lazertag arenas and related dying novelties. That's not a reflection of how cool the idea is but rather a statement of where our culture is in terms of general tech usage. People don't go out anymore for their entertainment and if they do it is almost always focused on passive media consumption. Anything you do in the gaming tech market has to somehow relate to how people consume their entertainment right now. And right now they are willing to sit on their couch, at their desk or dedicate a few square feet in front of their TV. If you can make something work within those landscapes, that's where you'll find success.

I think we're on the verge of something really great in VR but there's a tricky balance that has to be played with the technology. The first priority is to show people that immersion is what's so different about HMD's. I want to throw an HMD on my nephews, make them play Portal 2 and have their little passive consumer jaws drop. There's a viable way into current games available to us today, we just have to tweak a few things! Once that first layer of understanding is breached with the average consumer, then you can start the dialog on VR arenas, real-life augmented MMO's and beyond.

Just my opinion of course