BREAKING: Holodeck Arrives 350 Years Ahead of Schedule

[Synexious]

WOW.

[youtube-hd]http://www.youtube.com/watch?v=VrgWH1KUDt4[/youtube-hd]

This is better than the HMZ-T1, the ST1080, the Wide5, and even the almighty Sensics piSight. Better than any HMD. This is CAVE technology for the twenty-first century. This is the arrival of the holodeck 350 years ahead of schedule. This is your portal into the metaverse.

This is the solution. Adding the wedgeroller treadmill seen in the Ultimate Battlefield 3 Simulator, and a surround sound system behind the projection walls, you have a true holodeck. No need for anything strapped to your body. No wires, no batteries, no visible technology necessary. Full visual immersion, total freedom of movement. I am floored. I cannot overstate my excitement. I can see a path to perfect VR now. It seems much closer now than ever before. I'm determined to earn enough money to build my own holodeck as soon as possible. VR enthusiasts, rejoice! Seriously. We can literally live in the metaverse full-time now!

[Fredz]

I don't really understand your excitement, CAVE systems have been available for 20 years now and I don't see how this is any different from it, apart from the projection on the coffee table which looks pretty gimmickly to me.

And the problem with such a setup has been known for a long time too : you need 5 3D projectors ($2500 at least) and a big cubic dedicated room inside a much larger one. So not really in the same league than a $800 HMD.

[Synexious]

Ha, I guess the wow factor of the video, when the walls seemed to disappear and huge virtual spaces opened up, got to me. I have never seen anything like this before. I am truly amazed. I have scoured the internet for VR technologies, and seen many videos and descriptions of CAVEs and HMDs, but nothing has ever been this utterly awesome. I realize CAVE technology itself is twenty years old, but every other CAVE I've seen looks, well, twenty years old. To cover all six surfaces would cost a minimum of $3000, using six $500 720p 3D projectors. For 1080p, the price is tripled. 2D projectors would cost less, and you wouldn't have to wear glasses. There are around 8700 hours in a year, and bulbs last 2000-4000 hours, so to run the system year-round would require thousands in bulb replacement costs. Overheating would also be problematic. A large outer cube (I'd prefer a sphere or at least slightly curved corners) isn't necessary if the projectors are hung from the ceiling, but then of course the ceiling itself cannot be projected onto. Well, a projector could be pointed up from the center of the floor... Short-throw lenses could still reduce the distance required for rear projection. For now, since I don't have $10,000+, I'll have to stick with my $800 HMD, but nonetheless I am still utterly amazed by this CAVE. Being able to live in a room like this, with transforming furniture, would be awesome. If the projectors were quiet, always-on, and 4K resolution, you'd really get the sense of inhabiting a simulated environment . A 4K CAVE would cost $180,000 today, but in ten years, could be affordable. Eventually even 8K... and beyond! Traditional bulb projectors would overheat and be too expensive to replace, but laser projectors would be ideal. By the time laser projectors are both high quality and affordable, though, lightweight, fully immersive HMDs may very well have been developed. Still, I like the idea of not having to wear anything, being able to just wake up to find yourself floating through clouds or on the Moon, going about your day as simulated environments seamlessly envelop you at all times. Interacting with others, both physically and virtually, would be more natural without the barrier of HMDs. HMDs are still the most practical, though. They last essentially forever, can be taken anywhere, and require no maintenance.

[PalmerTech]

I have used CAVEs with some pretty nice graphics being run into them, along with the kind of perspective tracking that is used in this video, but there are a lot of problems.

1) You need a large space for rear projection
2) If you go with front projection, then you cannot move away from the center, because you will cast shadows
3) No way to look down, especially with an ODT
4) There is always going to be lag with the perspective correction. The graphics they were showing are really not all that advanced! Once you get into harder stuff, warping the scene to match your angle of view is really, really hard to do in real time. Lag is already a problem with HMDs, and all they have to do is move your viewport around. This kind of tech takes the tracking delay, and then adds a significant computational load on top of it.
5) You won't be interacting with other people inside this environment. Stuff like this is strictly single user, maybe two or three if the room is really big and you make sure to all stand very close together, facing in the same direction. Otherwise, everything looks warped and distorted! The guy in the video, for example, was seeing a really warped and unrealistic view, his amazement was all acting. The only thing with the correct view was the camera.
6) You still need glasses or electronics of some kind, in order to accurately track your position and angle of view
7) 3D is hard, REALLY hard to do with the kind of perspective correction seen here. Getting convergence and separation correct across multiple surfaces in real time is a nightmare.

Another problem is that this kind of demo looks a lot more impressive when recorded on video than it does in real life. In real life, it is still pretty easy to tell that the walls are close to you, especially on the corners. CAVEs are great visualization technology for stuff like models, maps, and CAD work, but they are not as immersive as you would imagine. It is hard to believe you are in a limitless room when you can easily tell that your eyes are focused on a surface only 6 feet away, and if it is not done in 3D, then the convergence of your eyes is telling you the exact same thing!

Still really awesome, though! Look into Casio laser projectors. They have been available for a while, are very bright, and use lasers and LEDs as the light source, no bulb problems. Not too expensive, either.

[cybereality]

This is definitely amazing, but I think the ultimate HMD will beat the ultimate CAVE. Only because the HMD is totally self-contained and can be used in any environment. The CAVE requires a large and expensive studio space to do it correctly. But that video is still really cool, and I would love to try a CAVE one day.

[Synexious]

Yeah. HMDs still win. When we have affordable, comfortable, totally immersive Snow Crash HMDs and a photorealistic metaverse, the world will change because all the millions of people who can only afford tiny apartments and limited luxuries will suddenly have unlimited cyberspace open up to them. That's what excites me most about the next few decades. Anyone will be able to get in on the ground level with just the super-HMD, then they'll be able to add ODMiP systems and datasuits when they want. Combined with solar power and home automation, it will be possible to live a very fun life with very little work.

To telecommunicate [or just communicate - it's possible friends, family, and even couples (or triads, or whatever) will still interact via mediated reality, displaying their avatars even to those physically present with them] with HMDs, broadcast a hyperreal avatar linked to your facial expressions instead of your uglier self.

I often reference Snow Crash. I bet cyber and Palmer have read it. Has anyone else? If not, you should. Every VR enthusiast should. I still need to read its inspiration, the Sprawl trilogy.

[blzd1]

Definately cool. However, with all the projectors I see it as being fairly inefficient. Not everyone has the room for these setups either. I honestly don't see holodeck as the wave of the future. Beyond HMDs which may eventually evolve so thin that they are sunglasses or even contact lenses, I see virtual reality via cyberbnetic brain implants which may be so small that they are microscopic. You would enter a virtual reality through a cloud computing system whereby electromagnetic signals are sent to your brain via satellite. Therein a mind control system takes over your brain's audio, visual, and tactile centers, and you become completely immersed in your new environment.

I see it kind of like that episode in Ghost in the Shell: Stand Alone Complex called Chat, Chat, Chat (episode 9) where the major 'dives' into a virtual reality chat room through her cybernetic brain implants.

[cybereality]

Actually, we will have full immersion Virtual Reality, without implants. At least not the kind you expect from the Matrix. Future nanobots will be able to enter our bodies (for example, from an injection or a pill) and augment our senses to provide VR or AR with convincing audio/visuals and even tactile feedback. Smell and taste could also be simulated, which would be much harder with external devices. So HMDs and body suits will only get us so far, but the next 10 or 15 years should be pretty interesting.

@Aphradonis: I have read the Sprawl trilogy, and most of Snow Crash. I mean to finish it sometime soon, I even bought the audio-book. I also got the audio-book for Neuromancer, even though I've read it before. Such a great book, the best ever.

[android78]

Actually, we will have full immersion Virtual Reality, without implants. At least not the kind you expect from the Matrix. Future nanobots will be able to enter our bodies (for example, from an injection or a pill) and augment our senses to provide VR or AR with convincing audio/visuals and even tactile feedback. Smell and taste could also be simulated, which would be much harder with external devices. So HMDs and body suits will only get us so far, but the next 10 or 15 years should be pretty interesting.

@Aphradonis: I have read the Sprawl trilogy, and most of Snow Crash. I mean to finish it sometime soon, I even bought the audio-book. I also got the audio-book for Neuromancer, even though I've read it before. Such a great book, the best ever.

Have you been watching too many sci-fi movies again? If you look at the backlash to the current technology and people complaining that it's too invasive, I have trouble imagining more then 1% of people would even consider allowing nanobots that would control your brain.

I think that HMD technology is the way things will go. I imagine that we will end up with a true light field type technology so that the focus issue that we currently have will be eliminated though. I'm not sure if this may begin with an actually moving display/lens system (that would display the different parts of the image when it got to the correct depty), or multi-layered display, but I'm sure it's possible. If we can get a good implementation of roller-shoes (as discussed in another thread), then I think that's about as far as most consumers would take it.
I don't think cave technology is the way of the future. It just seems too inefficient to me. It's going the opposite way to most consumer electronics.

[cybereality]

Have you been watching too many sci-fi movies again? If you look at the backlash to the current technology and people complaining that it's too invasive, I have trouble imagining more then 1% of people would even consider allowing nanobots that would control your brain.

No, I actually have just been reading Ray Kurzweil. Just finished The Singularity is Near, very good read. You see, these nanotechnologies will first be used for medicine, and will be able to cure AIDS and cancer and all types of stuff. They will also be able to reduce or reverse aging, and make people healthier in general. So VR/entertainment will not be the first uses of this technology. People will become comfortable with them for life-saving reasons, just as we are comfortable getting vaccines or flu shots. And, sure, there will be a large group of people against the technology, but that is no different from any other era.

[CyberVillain]

Actually, we will have full immersion Virtual Reality, without implants. At least not the kind you expect from the Matrix. Future nanobots will be able to enter our bodies (for example, from an injection or a pill) and augment our senses to provide VR or AR with convincing audio/visuals and even tactile feedback. Smell and taste could also be simulated, which would be much harder with external devices. So HMDs and body suits will only get us so far, but the next 10 or 15 years should be pretty interesting.

I think jacking onto the central nervous system will be the way to go, if you can fool the brain to accept your alternative signals then vision, touch and smell will be no problem to simulate

[blzd1]

Has anyone else seen this?

http://www.youtube.com/watch?v=IeZFYu3dzgM" onclick="window.open(this.href);return false;

Kind of off the radar for most people. According to that, Intel will have a chip in the brain by 2020 that controls a cursor on the screen by thought alone. That is only the beginning in my opinion. So maybe synthetic telepathic communication by 2030. Then what, this virtual/augmented reality by 2040?

Just the way I see things going.

Of course, there could be truth to these nanobots, which would be more along the lines of Ray Kurzweil's predictions. According to wikipedia though, he has been wrong before. (Cars that control themselves by 2010?) I really know nothing about how they would operate and if they are used, see them more useful for repairing the body. Having a single chip, that connects everyone's brains wirelessly to the internet via a cybernetic infrastructure would make more sense to me. The less built in functions and more mind control is directed by software, the smaller and more efficient it could be.

[cbwan]

I've tested several Caves in my life and when calibrated correctly it is really impressive.
I've seen people bumping into the walls because they forgot about them,
I've seen people not wanting to take a step when presentend with a virtual hole in front of them,
I've seen people falling because they tried to lean on a virtual seat ..

@PalmerTech: You're mostly right, a few thoughts though :
3) Most "real" Caves have floor screens
4) Why would there be more lag in a cave than in a hmd ? the perspective correction is extremely fast. I would more worry about tracking delay which, as you mention, also exists in hmds, network sync which is really fast also. The real issue is more the huge number of pixels to display. Last Cave I setup had 11 HD projectors ! Poor graphics cards ( 1 computer with 2 QuadroPlex for 8 projectors, another with 1 QPlex for 3 projectors ).
5) Some people are working on sterescopy for multiple users: a wall for 6 users : http://www.uni-weimar.de/cms/medien/vr/ ... ments.html" onclick="window.open(this.href);return false;
7) If you know the 3d model of the environment, everything is possible with some good tools

Check this huge (9.6m wide) cave :



[youtube-hd]http://www.youtube.com/watch?v=cTd5DLe8KD0[/youtube-hd]

Other Caves (also with a ceiling) : http://cb.nowan.net/blog/2011/12/11/new ... ive-cubes/" onclick="window.open(this.href);return false;

I know HMDs will give us the VR we want, but for now I've never been immersed in a HMD as much as I've been in a Cave ..

[PalmerTech]

3) Most "real" Caves have floor screens
4) Why would there be more lag in a cave than in a hmd ? the perspective correction is extremely fast. I would more worry about tracking delay which, as you mention, also exists in hmds, network sync which is really fast also. The real issue is more the huge number of pixels to display. Last Cave I setup had 11 HD projectors ! Poor graphics cards ( 1 computer with 2 QuadroPlex for 8 projectors, another with 1 QPlex for 3 projectors ).
5) Some people are working on sterescopy for multiple users: a wall for 6 users : http://www.uni-weimar.de/cms/medien/vr/ ... ments.html" onclick="window.open(this.href);return false;
7) If you know the 3d model of the environment, everything is possible with some good tools

3) I know most real CAVE systems have a floor screen, I was just responding to the OP's idea to use an omnidirectional treadmill in a CAVE. That is a very iffy technology as it is, I can't think of many good ways to project onto an ODT at this point.
4) In my experience, the perspective correction has been pretty slow. Maybe the implementation I saw was bad for some reason? I have a lot of hours in a Wide5 using a very nice optical tracking stage, and the lag seemed noticeably higher using a CAVE like setup in the same stage. Perhaps it was just the fault of the massive resolution, as you suggest.
5) Thanks for that link, looks like multiple users could be practical at some point in the future!
7) I will admit that I am not a programmer, very much a hardware guy here. My comments on the difficulty of rendering 3D in CAVEs is based almost entirely on what I have heard from other programmers, so I am glad to hear someone saying that there are good, tools that make it possible.

I really do like CAVEs, I wish I had space for a big one like you showed!

[cbwan]

3) sorry, I read the answers too quickly
4) the perspective correction is "just" a projection matrix to compute. But if your application usually runs @ 60 hz on a regular pc, it will probably run @ 30hz in stereo, which increases the lag. On top of that some projectors will buffer a frame or two. What I would love to have is a tool to measure latency of all the subsystems to be able to diagnose exactly where it comes from. Very few people know that unfortunately.. Add to that tracker/projectors calibration problems and you can quickly get an expensive system that is is hardly immersive ! So it takes a lot of hardware and software skills to calibrate and optimize correctly, and again very few people know how to do that correctly !
7) I've setup Caves in less than 1h so yes, it's doable (but no I don't know everything (yet))

Even with the wide Cave there it's frustrating to turn around and see the real world, that's the biggest reason (and the space needed) why HMDs will rule the world
With redirected walking you don't even need such a huge space..

[bobv5]

"I often reference Snow Crash. I bet cyber and Palmer have read it. Has anyone else? If not, you should. Every VR enthusiast should. I still need to read its inspiration, the Sprawl trilogy."

Added to the list. Tis a long list though.

"Future nanobots will be able to enter our bodies (for example, from an injection or a pill)" Or a bottle of milk. Or a door handle. Perhaps an air freshener in a pubic toilet?

Anybody want some Nature Fresh milk?

"the next 10 or 15 years should be pretty interesting"

Isn't that a Chinese curse? "May you live in interesting times"

"No, I actually have just been reading Ray Kurzweil. Just finished The Singularity is Near, very good read. You see, these nanotechnologies will first be used for medicine, and will be able to cure AIDS and cancer and all types of stuff. They will also be able to reduce or reverse aging, and make people healthier in general. So VR/entertainment will not be the first uses of this technology. People will become comfortable with them for life-saving reasons, just as we are comfortable getting vaccines or flu shots. And, sure, there will be a large group of people against the technology, but that is no different from any other era."

Not read it cover to cover, but skiped through it. I would love Kurzweil's vision of the future to be true, but I can't see it happening. Would you really let sony or microsoft have a eula on your brain? Probably not, but millions would. Both company’s have recently changed the eula's for the consoles to say that you give up your right to sue them. With brain implanted playstation 6, what is to stop them including a statement that you are not allowed to think bad things about sony? With built in enforcement? That many would agree to without ever knowing?

"According to that, Intel will have a chip in the brain by 2020 that controls a cursor on the screen by thought alone."
Can't watch youtube on this machine, but guessing it involves potentially fatal surgery? For mouse control? Really?

"According to wikipedia though, he has been wrong before. (Cars that control themselves by 2010?)" Cars have been made that can control themselves. Kurzweil's dates might not be exact, but so far he doesn't seem t be too far off.

I'm not trying to urinate on peoples fires, just pointing out that if this tech happens, in the control of the wrong people it could be hell on earth. I know such phrases get thrown around a lot, but I think it is justified in this case. If you guy's haven't read "1984", read it. It is available free online, and it seems to be a legal copy. http://www.planetebook.com/1984.asp" onclick="window.open(this.href);return false; Then you might see why I worry.


As for the original post, many people are happy to sit and watch non interactive media. Three walls filed with a non interactive 3d film would be pretty awesome.

[brantlew]

I often reference Snow Crash. I bet cyber and Palmer have read it. Has anyone else? If not, you should. Every VR enthusiast should. I still need to read its inspiration, the Sprawl trilogy.

A bit off topic but if you're going to read anything by Snow Crash author Neal Stephenson, then you have to read Cryptonomicon. Stephenson is the king of geek writing. I devour everything he writes and to be honest, Snow Crash is only his third or fourth best.

[cybereality]

Not strictly cyberpunk, but 'Daemon' by Daniel Suarez was really awesome. Don't want to spoil much, but its about a sort of rouge AI that gets out of control.