Slashdot Mirror
What Ever Happened to Virtual Reality?
bergeron76 writes "It seems like it's been ages since I heard of any advances in "Virtual Reality" technology. Was Virtual Reality just hype? Are there any new or existing projects that have made any significant inroads (aside from the first-person shooter games)?
Is total virtual immersion a worthless persuit / dead industry? If not, what are the bottlenecks that are delaying it?"
See Doom 3 or Half Life newblah.
Someone set us up the bomb, so shine we are!
VR was ahead of its time , it was trying to skip a few steps in the eveloutionry chain.It really was a step beyond its ability , VR is still used for treatment of those suffering mental traumas(physical and pyschological) so it was not an entier dead end. Its jsut the entertainment industry was at the time not ready for it , and in pushing it has set it back a while as its seen as a joke. ,the Reality of Virtualy reality may soon come around . Right now though , its still a joke .
With the advances in 3d Graphics and so forth
The only things certain in war are Propaganda and Death. You can never be sure which is which though
Like most "hot" technologies of the past, virtual reality has lost its buzzword factor and instead has found real applications which save real money. It's used as a visualization tool (CAVE), primarily in the automotive industry. The buzz has moved on to "augmented reality", which combines a virtual reality with the real environment. Both technologies are still held up by the lack of affordable and lightweight high resolution displays. Virtual reality therefore typically surrounds the user with big stationary screens. That is not feasible for augmented reality. The more interesting applications are in the augmented reality field, so there's your hold-up.
If VR porn doesn't show up then this technology will never reach the masses.
Doom3, the Sims etc. these are all virtual realities. People just got over the whole helmet thingy.
I once took a class back in college which discussed some of limitations of virtual reality (this was back in '93). Until some of these things are addressed, and not just the economic factors, VR might not really ever take off.
For example, how do you address the gravity problem? How can you virtually simulate something that has physical weight, like throwing a virtual ball and catching it?
And if we have public access to VR devices (assuming it's still economically unfeasible for mass market personal purchasing), how do you cope with the "icky" factor? Would you want to use some VR helmet or gloves after some greasy, unkempt guy just used it?
Perhaps true VR may not be possible unless it was purely a sensory experience (like in The Matrix) or using artificially created matter (like holodecks), and the best we're going to get are fancy 3D displays with some amount of immersion.
My guess is they realized "Virtual Reality" won't really have much potential (which translates to profit) until there are better ways to interact with the games then a keyboard/controller/clunky motion sensors. Not to mention decent head-mounted displays are still quite costly.... I for one can't wait till input systems improve, and you aren't limited by the controller.
In undeveloped countries, the consumer controls the market. In capitalist America, the market controls you.
Not quite. Where is the "-1 not funny"? Not to mention "-1 fucking stupid"?
If I understand the parent, the problem appears to be that the OS cannot handle interrupt latencies of less than a few dozen ms.
While this may be true of OS like windoze, certainly this does not hold for real time OSes. Such Oses (like vxWorks, OSE, Lynx etc) are used in several application everyday.
And these are latencies at the software levels above the OS. Commercial systems also use DSPs that can service interrupts at severl kHz.
To me, it would seem that lack of computing power is not the reason holding back VR. Something else - perhaps the inherent complexity of applications - and the vast volumes of data involved - coupled with the lack of a 'good' way to represent that data. After all that is what MIT's touted Media Labs have been trying to do for several years.
Now that there's no real hardware obstacle to gloves-and-goggles VR, it's clear that the basic concept is flawed. There are two fundamental problems. First, eye-hand coordination in empty space sucks as an input method. And second, full-surround visual motion without physical motion makes you feel funny.
Various "haptic interfaces" have been tried, and some of them actually work. Most of them are for small motions, as in sculpting or surgery. For those applications, it's often easier to use an ordinary screen than goggles.
Working in VR? Imagine riding a rollercoaster as a job. Eight hours a day. While trying to do real work.
It's a chicken-and-egg problem: so long as $50 games won't include a few lines of code supporting head-tracking, people won't spend >$500 on a VR headset; pity game companies won't spend the few dollars needed to jumpstart the headset market.
I was so thrilled by VR I bought a headset. Magic Carpet 2 was great (although the controls were not real VR-friendly). And that's where it ended, still siting in a box in a closet - because nothing else supported it.
The problem - the ONLY problem as far as I could see - was that practically no real game would support it. Even current FPS games, with full 3D 360-degree motion range stereoscopic-supporting perfectly suited for VR, won't bother to include even minimal head-tracking support. Aside from Magic Carpet 2, only lame demos took VR seriously.
Just have a developer spend an extra day adding basic head-tracking to a game like Quake 3 or Half-Life 2 and I'm sure VR will start to return.
Can we get a "-1 Wrong" moderation option?
I see where you're going here, but it's not really true.
I was fortunate enough to use the CAVE at UIUC in early '97, just after Quake was first released. 4-walled VR environment where the user only needed to wear "polarized" glasses to see the 3D image. I assure you it was MUCH faster than 5 fps. And I can assure you that it was much more immersive than Quake.
But there were no texturemaps. Every object pretty much had a single color. Why? Because there was no reason for it to be more than that.
Quake looks good and VR looks bad because there are millions of PCs and handfuls of things like the CAVE. Developing a souped-up-graphics environment like you see in FPSs isn't HARD, it's just TIME CONSUMING.
The only thing that made FPSs look better than VR is that there was millions of dollars to be made selling video games and nothing to be made making texturemapped VR for the handful of CAVEs on the planet.
It seems silly to evaluate the success of an academic approach on commercial terms. At the end of the day, FPS's are more commercially successful because computers with monitors are cheap (widely available) and 3D displays are not. You can't sell a VR program to millions of people who don't have VR equipment.
paintball
Ive read a great deal of everyones posts here, and those of you slating and putting down VR seem to know nothing of what you're talking about, and those of you backing it up, dont 'appear' to have enough insight into this technology to form a solid argument.
;)
Im just finishing a BSc (hons) Degree in Cybernetics and Virtual Worlds, with many modules on 3D world constuction, immersion, sound etc - including the use of some of the latest VR headsets.
I can tell you some of the major downsides right now - the fact that the majority of these headsets work with VRML... and aweful, outdated and pretty much useless language! And those that work with other languages (or even games) only support res's of around 800x600, some of the headsets dont have stereo-scopic vision (both screens show the same image, rather than slightly different like your eyex), some of them _are_ cumbersome, and heavy, but the vast majority (especially the ones ive played with) weigh so little, that you get used to them.
As for trackers, they are one of the other major aspects that let the whole technology down. Short of spending another £1000 on a decent tracker (in my lab we have 3 headsets, and 2 trackers, one of the headsets is over £1000 and one of the trackers is over £1000 in cost, so ive gotten to use the best and the worst!) they might as well not even be there! The decent tracking sytems, however, work with a good range, and accuracy, so your movements, in full 3D with 6 DOF (Degrees of Freedom) are relayed with a 1mm accuracy to the PC, so the image is returned to the headset with the correct yaw/pitch/roll and X,Y,Z co-ordinates needed.
I agree with some earlier posts about the cables, and if you are paying money to use this equipment, thats the last thing you want to ruin your gaming experience. Ive been lucky enough to play on 4 or 5 different types of VR systems (the one in the pod's, where you stamp your feet, the big "sit-down" unit where you drive a tank or race car, and several *ahem* home style units). So i dont think VR's dead, i just think that its moved on. Lets see:
It IS used in military, for battle field simulation.
It IS used in the Motor industry, for assisting engineers with locations and construction of certain parts in the vehicles.
It IS used in biomedical sciences, to view protein strands and DNA
So VR in its rawest form is still used, but not very useful... so i'll ramble on about its nearest cousing AR or Augmented Reality.
This uses the same technology as VR but contains a semi-transparent view on the headsets, so you see real life items, with digital items over-layed. Check out AR-Quake by some students at a Uni in Australia, or see if you can find the UK militaries interest in it, when they tested a large GPS assisted unit at DERA (Defence Evaluation Research Agency).
I dont think ive missed anything, and im fully prepared to be flamed for the things that ive said
I think VR will only be successful when it involves all of your senses in a way that makes you wonder whether or not you're in the real world or a virtual one.
VR developers' goals need to be very close to those of 3-D game designers, i.e.: simple, easy, fun.
Back in the late 1990's I remember downloading and running an "interactive VRML viewer", the thought of which really excited me. The particular package I tried out was the worst implementation of 3-D world I have ever seen. Sure, it was painfully slow (very low FPS), that's not the part that bothered me. What really irked me was how the system handled camera-wall collisions. When "you", the viewer, encountered a wall, your movement stopped dead - even if you were pushing into the wall at an angle!
Every 3-D game ever made does better than that. You should slide along the wall, parallel with the wall! This is simple physics. Nobody would ever buy a game where you had to "move back when you hit a wall, and don't touch any more walls or you'll get stuck!"
Observe the world around you. What happens to your body when you lean against a wall at an angle? Don't you slide along the wall? I mean, come on. Won't someone please put the Reality back into Virtual Reality.