Slashdot Mirror
More on the 3D DTI Monitor
Tyana pointed us to a review at Evil3D about the DTI 3D Monitor that
we mentioned here not to long ago. They actually sat down and used it a bit and talk about the price ($12k!) the aesthetics (They like it) and play some games (it crashes a lot). This is a really fascinating technology tho: 3D without glasses, and it apparently works really well, assuming you can hold your head still while you play Q3A!
What are the potential barriers towards adopting such a technology - better connection with kinesthetics, the intuitive match between spatial awareness and body motion - between physics models like MathsEngine to express - some killerapp vertically integrated applications to reach selected markets (like telesurgery which requires precise placement) to help bring the price-point down - software/content that supports 2-3D with the marginal effort of adding 3D smaller than marginal increase in sales
Apart from the gee-whiz factor, a realistic look at what services would benefit most from such displays needs to be addressed, especially their willingness to cough up the money. Remember that hardware is only 10% of the total costs, ad another 20% for peripherals/support, 30% for software/operational consumables, and 40% for training. It's starting to looking expensive.
3D will have a role but I suspect widespread star wars type holographic displays are still a way off unless a miracle occurs.
LL
At the moment such setups are too expensive for the casual user, but if the technology would be scaled down to a similar setup like DTI's, it could well become affordable. It'd need such glasses, a head tracking system for optimal representation, and special software, just like with the DTI system. Call me optimistic, but it could be done quite soon for no more than $500 total, in the near future, if I go by prices for other goodies coming onto the market. But so long as that doen't happen, there may be a place for the DTI stuff.
Stefan.
I should get a colleague from the VR group to comment here, but I don't think they read Slashdot yet.
The truth shall make you fret. (Ankh-Morpork tImes motto)
I am sure not going to win any points for this POV, but I think these displays are DOA.
A limited FOV, the need to not move the head too much, and a high price tag - effectively puts it out of reach of consumers, and should even make the IS department of any company who might have a need for such a display to do a double-take.
One person here mentioned a CAVE solution, or an ImmersaDesk-style system. Such systems would be better for many of the same apps, but both are still quite pricey. This same poster thinks that such systems will become affordable for the home user (I imagine he was speaking of the ImmerasDesk system) in the near future, coming in at around $500.00! This person has never priced LCD projectors (and I think high-end CAVE systems use CRT projector systems, just to get the high refresh rates needed for the stereo shutter glasses) apparently - prices for these devices, which would be an intregral part of any such system - haven't come down much in the past 10 years. Indeed, what seems to be apparent is rather than keeping the older tech around from the early 90's (which would be more than adequate for home projector use), and charging less - projector companies tend to dump it, and only sell the latest, thus keeping prices high (an ok projector will run you a minimum of around $1000 - used, you might get away with $800).
So, what does that leave for the average consumer? CAVE/ImmersaDesk systems are out and this DTI tech is out. Both mainly from a cost standpoint - prices that are not likely to come down at all in the near future for either system (barring some breakthrough in display tech - and even then the price will be high because it is *NEW* and *EXCITING*).
Only two possible affordable solutions - Shutter Glasses, and HMD's.
Shutter glasses are available now, and supported by several graphics cards, and they are cheap - however, while they allow more freedom of movement of the head, they still only have a limited FOV - due to the monitor. They are, and will continue to be, the choice for most people - only due to cost and ease of use.
HMD's are still rather pricey - but one can get an HMD today for around $1000 dollars that works quite well. Still, this is rather expensive for most gamers, who have already shelled out this much and more for their system likely - and don't have much left over for exciting 3D. What to do, what to do...
Homebrew, anyone? Why has EVERYONE (ok, that's an exageration - but not by much) forgotten about homebrew VR? Once, way back in the early 90's - if we didn't have a custom 3D display tech, we built it ourselves, using cheap (and cheaper today!) Casio hand-held TVs and Fresnel lenses, cobbled together in a plastic frame. The PowerGlove allowed us to reach into our worlds, and a lightweight boom mounted system allowed us to look around (some of us got adventurous, and used strings, LED's, ultrasonics ripped from PG systems, and other tricks for tracking - but the boom arm was the most accurate).
Today, homebrew VR has all but died - a few people still play the game, but most went on to other things. But look at what is available! We have free, open-source, GPL'd display engines! Low, low cost HMDs on the used market (one can pick up a used Forte VFX-1 for about $400 on eBay - Victormaxx Stuntmasters can be had for around $50). One could buy a couple of StuntMasters and home-brew their own HMD, with good res and a 60 degree+ FOV (the point of immersion), for under $300. We have fast, low cost computers and extreme 3D cards only dreamed about by the pioneers of the early 90's. Heck, one can run Rend386 or Avril on a normal PC of today, rendering standard VGA, and see frame rates well over 150 FPS. The code to both of these tools is available, and other code is available to do similar tasks under Linux and other OS's - using today's graphics cards...
What has happened? Why aren't we satisfying the urge to explore our own 3D worlds?
In a way, we are - witness things like Q3A and UT - both are networked virtual environments - the only thing lacking is full immersion, with full tracking. I can't understand why these players aren't clammoring for a fully immersed experience, or at least why some of them haven't built their own HMD systems, or why they would rather sit in front of a monitor playing a game, rather than being *in* the game. Q3A and UT offer the possibility of extreme VE immersion only dreamed of in the Rend386 days - yet donning a homebrew HMD, reaching out with a PowerGlove, and catching a spinning bannana is still something that neither UT or Q3A can quite match...
On a final note I would like to offer a link to my website, where I am trying to inform the public about homebrew VR, and today's possibilities:
PhoenixGarage.ORG
Reason is the Path to God - Anon
I'm on Solaris 2.8 with Netscape 4.7, and it works fine...
On the other hand, if they instead used polarized light with the article's technology, you could move around as much as you want...and polarized glasses aren't as heavy as head mounted displays.
For games, the most important cue is motion parallax--the way a 3D image changes when you move your head. The "3D monitor" actually makes this worse, not better, since the stereo cues also don't behave right when you move your head.
For 3D gaming, the best thing is going to be a head mounted display. Ideally, it will have considerable peripheral vision, since that gives a sense of immersion. Used with gyroscopic sensors, a head mounted display lets you look around naturally, and if you add head tracking of some form (an accelerometer may be sufficient), it can display motion parallax as well. Once you have a head mounted display, if you also want to throw in stereo, that's pretty easy.
I have been sick the past few days - but anyhow, if I understand you correctly, you are saying that by having this display somehow track where you (move/turn) your head, it tracks with it (kinda like if I turn my head to the right, the monitor in front of me moves to the right) - would increase the FOV?
If this is what you mean, then, in a word - no.
FOV is defined by the combined angular measument of a pair of imaginary "cones" radiating from the approximate center of each eye (one cone per eye). There are both horizontal and vertical FOV measurements (generally expressed in degrees) - so each "cone" can actually have an elliptical cross section (generally, vertical FOV is smaller than horizontal FOV in HMDs). Some HMD manufacturers are beginning to use a diagonal FOV measurment - muddling the issue further (kinda like diagonal measurments don't mean jack for a monitor, until you know about the 4:3 ratio - however, in the case of an HMD, you really have no clue what the ratio is)!
Generally, when speaking about FOV measurements for immersion, a horizontal measurement is used as the baseline. Most commercial, off the shelf HMDs, that are priced under $1000 (like the I-Glasses, or the Victormaxx Cybermaxx), have an FOV of around 30 degrees. Some of the good ones can go up to 45 degrees (I believe the Forte VFX-1 is around the area, maybe less). It takes a horizontal FOV of at least 60 degrees to be in the area of full immersion. An FOV of 65 or 70 degrees is better, in that at this point the FOV begins to extend into the peripheral vision areas, thus enhancing the immersion experience. A carefully built homebrew rig can approach (and with a lot of care in design, hit) the 60 degree mark. It takes patience, and a lot of skill, but it has been done before.
But tell me if my interpretation is wrong. What you seem to describe sounds more like a moving "window" on the virtual environment being viewed - not necessarily a bad thing - certainly something that could be utilized (imagine a gauzy "plane" in front of you in an HMD, that would show a cutaway "x-ray" view of the inside of an object in the VE, as you turned your head!) just not what I was describing...
Reason is the Path to God - Anon
The technology is nice, but it's really not that new. The advent of flat panel displays on the desktop is what really makes this practical (I laugh slightly when I call it practical) on the desktop. If you have ever been to Disney, they were thinking about putting a similar device into their "innoventions" exhibit at least 5 years ago, and there were devices that did similar things too.
It's cool and all, but not AMAZING. When I was a kid I drew plans for something similar. The theory has been around since long before any of us were born (calling it theory is a bit sketchy since it was theory in the dark ages).
Also, for $12K, I can get a head mounted display, head tracker, all wireless, and still have some left for some neato controllers to hold.
Eh...
Personally, I'm looking forward to a production version of this soon. I'm a big fan of 3d gaming in general, and having written a few, I can think of nothing I'd like better than a tunnel-race type game where you're practically immersed...
I wear glasses (contacts at the moment actually, but you get the point) so I find it easier not to PUT anything on my eyes. While I appreciate the ease and availability of head-mount systems and all, I personally (and I'm fairly certain others) prefer having a physical monitor in place on my desktop.
"I'm not even supposed to BE here today!"
"When in 3D mode it produces one line of light for every two columns of pixels on the LCD. Imagine that the columns are divided into two groups. One group being the even numbered columns and the other being the odd numbered columns. When you sit directly in front of the display or in certain positions off to the side, your left eye sees these light lines through the odd columns of pixels, while your right eye sees them through the even columns. So, each eye sees only half of the pixels."
Since they are using more or lass standard flat panel LCD's this more or less halves the horizontal resolution. The unfortunate fact here is for stereo imaging you need all the horizontal depth you can get. Since the perceived depth is relative to the horizontal offset of the two images, the dumber of depth steps becomes limited.
The offset between image components is limited to the distance between eyes (this would be perceived as infinite depth), this distance is approximately 3 inches. The resolution (once halved) is about 30 pixels per inch giving a maximum offset of difference of about 90 pixels. The translates to about 90 discrete steps of depth although this can be smoothed of somewhat with anti-aliasing it could really do with some more horizontal resolution. That said, I still want one.
Anybody else out there using Solaris (I'm on 2.6..) and Netscape 4.72? My box seems to not like the evil3d page.. NS just crashed twice in a row when loading this site. Can anyone try and confirm this to be an incompatability with the site/NS 4.72 or is it just my box being a piece?
Thanks,
//Phizzy
"Most European technology just isn't worth our stealing," -- Former CIA chief James Woolsey, referring to Echelon